From 65cc02a8e1d4449daedf3a84e9dfdda2e44a5d97 Mon Sep 17 00:00:00 2001 From: Colin Ian King Date: Fri, 23 Mar 2018 22:53:50 +0000 Subject: [PATCH 001/190] batman-adv: don't pass a NULL hard_iface to batadv_hardif_put In the case where hard_iface is NULL, the error path may pass a null pointer to batadv_hardif_put causing a null pointer dereference error. Avoid this by only calling the function if hard_iface not null. Detected by CoverityScan, CID#1466456 ("Explicit null dereferenced") Fixes: 53dd9a68ba68 ("batman-adv: add multicast flags netlink support") Signed-off-by: Colin Ian King Signed-off-by: Sven Eckelmann Signed-off-by: Simon Wunderlich --- net/batman-adv/multicast.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/batman-adv/multicast.c b/net/batman-adv/multicast.c index a11d3d89f012..a35f597e8c8b 100644 --- a/net/batman-adv/multicast.c +++ b/net/batman-adv/multicast.c @@ -1536,7 +1536,7 @@ batadv_mcast_netlink_get_primary(struct netlink_callback *cb, if (!ret && primary_if) *primary_if = hard_iface; - else + else if (hard_iface) batadv_hardif_put(hard_iface); return ret; From e76b6312391bdd62e31dc86cb65e478b07b7909e Mon Sep 17 00:00:00 2001 From: Ilya Dryomov Date: Wed, 2 May 2018 20:16:56 +0200 Subject: [PATCH 002/190] iov_iter: fix return type of __pipe_get_pages() It returns -EFAULT and happens to be a helper for pipe_get_pages() whose return type is ssize_t. Signed-off-by: Ilya Dryomov Signed-off-by: Al Viro --- lib/iov_iter.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/lib/iov_iter.c b/lib/iov_iter.c index 970212670b6a..4d5bf40d399d 100644 --- a/lib/iov_iter.c +++ b/lib/iov_iter.c @@ -1012,7 +1012,7 @@ unsigned long iov_iter_gap_alignment(const struct iov_iter *i) } EXPORT_SYMBOL(iov_iter_gap_alignment); -static inline size_t __pipe_get_pages(struct iov_iter *i, +static inline ssize_t __pipe_get_pages(struct iov_iter *i, size_t maxsize, struct page **pages, int idx, From d7760d638b140d53c6390a2fbee9b06460b43e9e Mon Sep 17 00:00:00 2001 From: Ilya Dryomov Date: Wed, 2 May 2018 20:16:57 +0200 Subject: [PATCH 003/190] iov_iter: fix memory leak in pipe_get_pages_alloc() Make n signed to avoid leaking the pages array if __pipe_get_pages() fails to allocate any pages. Signed-off-by: Ilya Dryomov Signed-off-by: Al Viro --- lib/iov_iter.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/lib/iov_iter.c b/lib/iov_iter.c index 4d5bf40d399d..fdae394172fa 100644 --- a/lib/iov_iter.c +++ b/lib/iov_iter.c @@ -1102,7 +1102,7 @@ static ssize_t pipe_get_pages_alloc(struct iov_iter *i, size_t *start) { struct page **p; - size_t n; + ssize_t n; int idx; int npages; From 1aa7a5735a41418d8e01fa7c9565eb2657e2ea3f Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Tue, 1 May 2018 15:55:51 +0200 Subject: [PATCH 004/190] x86/nospec: Simplify alternative_msr_write() The macro is not type safe and I did look for why that "g" constraint for the asm doesn't work: it's because the asm is more fundamentally wrong. It does movl %[val], %%eax but "val" isn't a 32-bit value, so then gcc will pass it in a register, and generate code like movl %rsi, %eax and gas will complain about a nonsensical 'mov' instruction (it's moving a 64-bit register to a 32-bit one). Passing it through memory will just hide the real bug - gcc still thinks the memory location is 64-bit, but the "movl" will only load the first 32 bits and it all happens to work because x86 is little-endian. Convert it to a type safe inline function with a little trick which hands the feature into the ALTERNATIVE macro. Signed-off-by: Linus Torvalds Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar --- arch/x86/include/asm/nospec-branch.h | 19 ++++++++++--------- 1 file changed, 10 insertions(+), 9 deletions(-) diff --git a/arch/x86/include/asm/nospec-branch.h b/arch/x86/include/asm/nospec-branch.h index f928ad9b143f..870acfc2fd34 100644 --- a/arch/x86/include/asm/nospec-branch.h +++ b/arch/x86/include/asm/nospec-branch.h @@ -241,15 +241,16 @@ static inline void vmexit_fill_RSB(void) #endif } -#define alternative_msr_write(_msr, _val, _feature) \ - asm volatile(ALTERNATIVE("", \ - "movl %[msr], %%ecx\n\t" \ - "movl %[val], %%eax\n\t" \ - "movl $0, %%edx\n\t" \ - "wrmsr", \ - _feature) \ - : : [msr] "i" (_msr), [val] "i" (_val) \ - : "eax", "ecx", "edx", "memory") +static __always_inline +void alternative_msr_write(unsigned int msr, u64 val, unsigned int feature) +{ + asm volatile(ALTERNATIVE("", "wrmsr", %c[feature]) + : : "c" (msr), + "a" (val), + "d" (val >> 32), + [feature] "i" (feature) + : "memory"); +} static inline void indirect_branch_prediction_barrier(void) { From 4a28bfe3267b68e22c663ac26185aa16c9b879ef Mon Sep 17 00:00:00 2001 From: Konrad Rzeszutek Wilk Date: Wed, 25 Apr 2018 22:04:16 -0400 Subject: [PATCH 005/190] x86/bugs: Concentrate bug detection into a separate function Combine the various logic which goes through all those x86_cpu_id matching structures in one function. Suggested-by: Borislav Petkov Signed-off-by: Konrad Rzeszutek Wilk Signed-off-by: Thomas Gleixner Reviewed-by: Borislav Petkov Reviewed-by: Ingo Molnar --- arch/x86/kernel/cpu/common.c | 21 +++++++++++---------- 1 file changed, 11 insertions(+), 10 deletions(-) diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index ce243f7d2d4e..9ed0a18ff6f6 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -927,21 +927,27 @@ static const __initconst struct x86_cpu_id cpu_no_meltdown[] = { {} }; -static bool __init cpu_vulnerable_to_meltdown(struct cpuinfo_x86 *c) +static void __init cpu_set_bug_bits(struct cpuinfo_x86 *c) { u64 ia32_cap = 0; + if (x86_match_cpu(cpu_no_speculation)) + return; + + setup_force_cpu_bug(X86_BUG_SPECTRE_V1); + setup_force_cpu_bug(X86_BUG_SPECTRE_V2); + if (x86_match_cpu(cpu_no_meltdown)) - return false; + return; if (cpu_has(c, X86_FEATURE_ARCH_CAPABILITIES)) rdmsrl(MSR_IA32_ARCH_CAPABILITIES, ia32_cap); /* Rogue Data Cache Load? No! */ if (ia32_cap & ARCH_CAP_RDCL_NO) - return false; + return; - return true; + setup_force_cpu_bug(X86_BUG_CPU_MELTDOWN); } /* @@ -992,12 +998,7 @@ static void __init early_identify_cpu(struct cpuinfo_x86 *c) setup_force_cpu_cap(X86_FEATURE_ALWAYS); - if (!x86_match_cpu(cpu_no_speculation)) { - if (cpu_vulnerable_to_meltdown(c)) - setup_force_cpu_bug(X86_BUG_CPU_MELTDOWN); - setup_force_cpu_bug(X86_BUG_SPECTRE_V1); - setup_force_cpu_bug(X86_BUG_SPECTRE_V2); - } + cpu_set_bug_bits(c); fpu__init_system(c); From d1059518b4789cabe34bb4b714d07e6089c82ca1 Mon Sep 17 00:00:00 2001 From: Konrad Rzeszutek Wilk Date: Wed, 25 Apr 2018 22:04:17 -0400 Subject: [PATCH 006/190] x86/bugs: Concentrate bug reporting into a separate function Those SysFS functions have a similar preamble, as such make common code to handle them. Suggested-by: Borislav Petkov Signed-off-by: Konrad Rzeszutek Wilk Signed-off-by: Thomas Gleixner Reviewed-by: Borislav Petkov Reviewed-by: Ingo Molnar --- arch/x86/kernel/cpu/bugs.c | 48 ++++++++++++++++++++++++++------------ 1 file changed, 33 insertions(+), 15 deletions(-) diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index bfca937bdcc3..ad613f73d071 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -314,30 +314,48 @@ static void __init spectre_v2_select_mitigation(void) #undef pr_fmt #ifdef CONFIG_SYSFS + +ssize_t cpu_show_common(struct device *dev, struct device_attribute *attr, + char *buf, unsigned int bug) +{ + if (!boot_cpu_has_bug(bug)) + return sprintf(buf, "Not affected\n"); + + switch (bug) { + case X86_BUG_CPU_MELTDOWN: + if (boot_cpu_has(X86_FEATURE_PTI)) + return sprintf(buf, "Mitigation: PTI\n"); + + break; + + case X86_BUG_SPECTRE_V1: + return sprintf(buf, "Mitigation: __user pointer sanitization\n"); + + case X86_BUG_SPECTRE_V2: + return sprintf(buf, "%s%s%s%s\n", spectre_v2_strings[spectre_v2_enabled], + boot_cpu_has(X86_FEATURE_USE_IBPB) ? ", IBPB" : "", + boot_cpu_has(X86_FEATURE_USE_IBRS_FW) ? ", IBRS_FW" : "", + spectre_v2_module_string()); + + default: + break; + } + + return sprintf(buf, "Vulnerable\n"); +} + ssize_t cpu_show_meltdown(struct device *dev, struct device_attribute *attr, char *buf) { - if (!boot_cpu_has_bug(X86_BUG_CPU_MELTDOWN)) - return sprintf(buf, "Not affected\n"); - if (boot_cpu_has(X86_FEATURE_PTI)) - return sprintf(buf, "Mitigation: PTI\n"); - return sprintf(buf, "Vulnerable\n"); + return cpu_show_common(dev, attr, buf, X86_BUG_CPU_MELTDOWN); } ssize_t cpu_show_spectre_v1(struct device *dev, struct device_attribute *attr, char *buf) { - if (!boot_cpu_has_bug(X86_BUG_SPECTRE_V1)) - return sprintf(buf, "Not affected\n"); - return sprintf(buf, "Mitigation: __user pointer sanitization\n"); + return cpu_show_common(dev, attr, buf, X86_BUG_SPECTRE_V1); } ssize_t cpu_show_spectre_v2(struct device *dev, struct device_attribute *attr, char *buf) { - if (!boot_cpu_has_bug(X86_BUG_SPECTRE_V2)) - return sprintf(buf, "Not affected\n"); - - return sprintf(buf, "%s%s%s%s\n", spectre_v2_strings[spectre_v2_enabled], - boot_cpu_has(X86_FEATURE_USE_IBPB) ? ", IBPB" : "", - boot_cpu_has(X86_FEATURE_USE_IBRS_FW) ? ", IBRS_FW" : "", - spectre_v2_module_string()); + return cpu_show_common(dev, attr, buf, X86_BUG_SPECTRE_V2); } #endif From 1b86883ccb8d5d9506529d42dbe1a5257cb30b18 Mon Sep 17 00:00:00 2001 From: Konrad Rzeszutek Wilk Date: Wed, 25 Apr 2018 22:04:18 -0400 Subject: [PATCH 007/190] x86/bugs: Read SPEC_CTRL MSR during boot and re-use reserved bits The 336996-Speculative-Execution-Side-Channel-Mitigations.pdf refers to all the other bits as reserved. The Intel SDM glossary defines reserved as implementation specific - aka unknown. As such at bootup this must be taken it into account and proper masking for the bits in use applied. A copy of this document is available at https://bugzilla.kernel.org/show_bug.cgi?id=199511 [ tglx: Made x86_spec_ctrl_base __ro_after_init ] Suggested-by: Jon Masters Signed-off-by: Konrad Rzeszutek Wilk Signed-off-by: Thomas Gleixner Reviewed-by: Borislav Petkov Reviewed-by: Ingo Molnar --- arch/x86/include/asm/nospec-branch.h | 24 ++++++++++++++++++++---- arch/x86/kernel/cpu/bugs.c | 28 ++++++++++++++++++++++++++++ 2 files changed, 48 insertions(+), 4 deletions(-) diff --git a/arch/x86/include/asm/nospec-branch.h b/arch/x86/include/asm/nospec-branch.h index 870acfc2fd34..9ec3d4d448cd 100644 --- a/arch/x86/include/asm/nospec-branch.h +++ b/arch/x86/include/asm/nospec-branch.h @@ -217,6 +217,17 @@ enum spectre_v2_mitigation { SPECTRE_V2_IBRS, }; +/* + * The Intel specification for the SPEC_CTRL MSR requires that we + * preserve any already set reserved bits at boot time (e.g. for + * future additions that this kernel is not currently aware of). + * We then set any additional mitigation bits that we want + * ourselves and always use this as the base for SPEC_CTRL. + * We also use this when handling guest entry/exit as below. + */ +extern void x86_spec_ctrl_set(u64); +extern u64 x86_spec_ctrl_get_default(void); + extern char __indirect_thunk_start[]; extern char __indirect_thunk_end[]; @@ -254,8 +265,9 @@ void alternative_msr_write(unsigned int msr, u64 val, unsigned int feature) static inline void indirect_branch_prediction_barrier(void) { - alternative_msr_write(MSR_IA32_PRED_CMD, PRED_CMD_IBPB, - X86_FEATURE_USE_IBPB); + u64 val = PRED_CMD_IBPB; + + alternative_msr_write(MSR_IA32_PRED_CMD, val, X86_FEATURE_USE_IBPB); } /* @@ -266,14 +278,18 @@ static inline void indirect_branch_prediction_barrier(void) */ #define firmware_restrict_branch_speculation_start() \ do { \ + u64 val = x86_spec_ctrl_get_default() | SPEC_CTRL_IBRS; \ + \ preempt_disable(); \ - alternative_msr_write(MSR_IA32_SPEC_CTRL, SPEC_CTRL_IBRS, \ + alternative_msr_write(MSR_IA32_SPEC_CTRL, val, \ X86_FEATURE_USE_IBRS_FW); \ } while (0) #define firmware_restrict_branch_speculation_end() \ do { \ - alternative_msr_write(MSR_IA32_SPEC_CTRL, 0, \ + u64 val = x86_spec_ctrl_get_default(); \ + \ + alternative_msr_write(MSR_IA32_SPEC_CTRL, val, \ X86_FEATURE_USE_IBRS_FW); \ preempt_enable(); \ } while (0) diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index ad613f73d071..6ed84f50d74a 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -28,6 +28,12 @@ static void __init spectre_v2_select_mitigation(void); +/* + * Our boot-time value of the SPEC_CTRL MSR. We read it once so that any + * writes to SPEC_CTRL contain whatever reserved bits have been set. + */ +static u64 __ro_after_init x86_spec_ctrl_base; + void __init check_bugs(void) { identify_boot_cpu(); @@ -37,6 +43,13 @@ void __init check_bugs(void) print_cpu_info(&boot_cpu_data); } + /* + * Read the SPEC_CTRL MSR to account for reserved bits which may + * have unknown values. + */ + if (boot_cpu_has(X86_FEATURE_IBRS)) + rdmsrl(MSR_IA32_SPEC_CTRL, x86_spec_ctrl_base); + /* Select the proper spectre mitigation before patching alternatives */ spectre_v2_select_mitigation(); @@ -95,6 +108,21 @@ static const char *spectre_v2_strings[] = { static enum spectre_v2_mitigation spectre_v2_enabled = SPECTRE_V2_NONE; +void x86_spec_ctrl_set(u64 val) +{ + if (val & ~SPEC_CTRL_IBRS) + WARN_ONCE(1, "SPEC_CTRL MSR value 0x%16llx is unknown.\n", val); + else + wrmsrl(MSR_IA32_SPEC_CTRL, x86_spec_ctrl_base | val); +} +EXPORT_SYMBOL_GPL(x86_spec_ctrl_set); + +u64 x86_spec_ctrl_get_default(void) +{ + return x86_spec_ctrl_base; +} +EXPORT_SYMBOL_GPL(x86_spec_ctrl_get_default); + #ifdef RETPOLINE static bool spectre_v2_bad_module; From 5cf687548705412da47c9cec342fd952d71ed3d5 Mon Sep 17 00:00:00 2001 From: Konrad Rzeszutek Wilk Date: Wed, 25 Apr 2018 22:04:19 -0400 Subject: [PATCH 008/190] x86/bugs, KVM: Support the combination of guest and host IBRS A guest may modify the SPEC_CTRL MSR from the value used by the kernel. Since the kernel doesn't use IBRS, this means a value of zero is what is needed in the host. But the 336996-Speculative-Execution-Side-Channel-Mitigations.pdf refers to the other bits as reserved so the kernel should respect the boot time SPEC_CTRL value and use that. This allows to deal with future extensions to the SPEC_CTRL interface if any at all. Note: This uses wrmsrl() instead of native_wrmsl(). I does not make any difference as paravirt will over-write the callq *0xfff.. with the wrmsrl assembler code. Signed-off-by: Konrad Rzeszutek Wilk Signed-off-by: Thomas Gleixner Reviewed-by: Borislav Petkov Reviewed-by: Ingo Molnar --- arch/x86/include/asm/nospec-branch.h | 10 ++++++++++ arch/x86/kernel/cpu/bugs.c | 18 ++++++++++++++++++ arch/x86/kvm/svm.c | 6 ++---- arch/x86/kvm/vmx.c | 6 ++---- 4 files changed, 32 insertions(+), 8 deletions(-) diff --git a/arch/x86/include/asm/nospec-branch.h b/arch/x86/include/asm/nospec-branch.h index 9ec3d4d448cd..d1c2630922da 100644 --- a/arch/x86/include/asm/nospec-branch.h +++ b/arch/x86/include/asm/nospec-branch.h @@ -228,6 +228,16 @@ enum spectre_v2_mitigation { extern void x86_spec_ctrl_set(u64); extern u64 x86_spec_ctrl_get_default(void); +/* + * On VMENTER we must preserve whatever view of the SPEC_CTRL MSR + * the guest has, while on VMEXIT we restore the host view. This + * would be easier if SPEC_CTRL were architecturally maskable or + * shadowable for guests but this is not (currently) the case. + * Takes the guest view of SPEC_CTRL MSR as a parameter. + */ +extern void x86_spec_ctrl_set_guest(u64); +extern void x86_spec_ctrl_restore_host(u64); + extern char __indirect_thunk_start[]; extern char __indirect_thunk_end[]; diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index 6ed84f50d74a..38a8626c894c 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -123,6 +123,24 @@ u64 x86_spec_ctrl_get_default(void) } EXPORT_SYMBOL_GPL(x86_spec_ctrl_get_default); +void x86_spec_ctrl_set_guest(u64 guest_spec_ctrl) +{ + if (!boot_cpu_has(X86_FEATURE_IBRS)) + return; + if (x86_spec_ctrl_base != guest_spec_ctrl) + wrmsrl(MSR_IA32_SPEC_CTRL, guest_spec_ctrl); +} +EXPORT_SYMBOL_GPL(x86_spec_ctrl_set_guest); + +void x86_spec_ctrl_restore_host(u64 guest_spec_ctrl) +{ + if (!boot_cpu_has(X86_FEATURE_IBRS)) + return; + if (x86_spec_ctrl_base != guest_spec_ctrl) + wrmsrl(MSR_IA32_SPEC_CTRL, x86_spec_ctrl_base); +} +EXPORT_SYMBOL_GPL(x86_spec_ctrl_restore_host); + #ifdef RETPOLINE static bool spectre_v2_bad_module; diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 1fc05e428aba..8bfc8fe26851 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -5557,8 +5557,7 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu) * is no need to worry about the conditional branch over the wrmsr * being speculatively taken. */ - if (svm->spec_ctrl) - native_wrmsrl(MSR_IA32_SPEC_CTRL, svm->spec_ctrl); + x86_spec_ctrl_set_guest(svm->spec_ctrl); asm volatile ( "push %%" _ASM_BP "; \n\t" @@ -5670,8 +5669,7 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu) if (unlikely(!msr_write_intercepted(vcpu, MSR_IA32_SPEC_CTRL))) svm->spec_ctrl = native_read_msr(MSR_IA32_SPEC_CTRL); - if (svm->spec_ctrl) - native_wrmsrl(MSR_IA32_SPEC_CTRL, 0); + x86_spec_ctrl_restore_host(svm->spec_ctrl); /* Eliminate branch target predictions from guest mode */ vmexit_fill_RSB(); diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index c7668806163f..3091dc98a451 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -9720,8 +9720,7 @@ static void __noclone vmx_vcpu_run(struct kvm_vcpu *vcpu) * is no need to worry about the conditional branch over the wrmsr * being speculatively taken. */ - if (vmx->spec_ctrl) - native_wrmsrl(MSR_IA32_SPEC_CTRL, vmx->spec_ctrl); + x86_spec_ctrl_set_guest(vmx->spec_ctrl); vmx->__launched = vmx->loaded_vmcs->launched; @@ -9869,8 +9868,7 @@ static void __noclone vmx_vcpu_run(struct kvm_vcpu *vcpu) if (unlikely(!msr_write_intercepted(vcpu, MSR_IA32_SPEC_CTRL))) vmx->spec_ctrl = native_read_msr(MSR_IA32_SPEC_CTRL); - if (vmx->spec_ctrl) - native_wrmsrl(MSR_IA32_SPEC_CTRL, 0); + x86_spec_ctrl_restore_host(vmx->spec_ctrl); /* Eliminate branch target predictions from guest mode */ vmexit_fill_RSB(); From c456442cd3a59eeb1d60293c26cbe2ff2c4e42cf Mon Sep 17 00:00:00 2001 From: Konrad Rzeszutek Wilk Date: Wed, 25 Apr 2018 22:04:20 -0400 Subject: [PATCH 009/190] x86/bugs: Expose /sys/../spec_store_bypass Add the sysfs file for the new vulerability. It does not do much except show the words 'Vulnerable' for recent x86 cores. Intel cores prior to family 6 are known not to be vulnerable, and so are some Atoms and some Xeon Phi. It assumes that older Cyrix, Centaur, etc. cores are immune. Signed-off-by: Konrad Rzeszutek Wilk Signed-off-by: Thomas Gleixner Reviewed-by: Borislav Petkov Reviewed-by: Ingo Molnar --- .../ABI/testing/sysfs-devices-system-cpu | 1 + arch/x86/include/asm/cpufeatures.h | 1 + arch/x86/kernel/cpu/bugs.c | 5 ++++ arch/x86/kernel/cpu/common.c | 23 +++++++++++++++++++ drivers/base/cpu.c | 8 +++++++ include/linux/cpu.h | 2 ++ 6 files changed, 40 insertions(+) diff --git a/Documentation/ABI/testing/sysfs-devices-system-cpu b/Documentation/ABI/testing/sysfs-devices-system-cpu index 025b7cf3768d..bd4975e132d3 100644 --- a/Documentation/ABI/testing/sysfs-devices-system-cpu +++ b/Documentation/ABI/testing/sysfs-devices-system-cpu @@ -478,6 +478,7 @@ What: /sys/devices/system/cpu/vulnerabilities /sys/devices/system/cpu/vulnerabilities/meltdown /sys/devices/system/cpu/vulnerabilities/spectre_v1 /sys/devices/system/cpu/vulnerabilities/spectre_v2 + /sys/devices/system/cpu/vulnerabilities/spec_store_bypass Date: January 2018 Contact: Linux kernel mailing list Description: Information about CPU vulnerabilities diff --git a/arch/x86/include/asm/cpufeatures.h b/arch/x86/include/asm/cpufeatures.h index 578793e97431..3493b58ef2cc 100644 --- a/arch/x86/include/asm/cpufeatures.h +++ b/arch/x86/include/asm/cpufeatures.h @@ -363,5 +363,6 @@ #define X86_BUG_CPU_MELTDOWN X86_BUG(14) /* CPU is affected by meltdown attack and needs kernel page table isolation */ #define X86_BUG_SPECTRE_V1 X86_BUG(15) /* CPU is affected by Spectre variant 1 attack with conditional branches */ #define X86_BUG_SPECTRE_V2 X86_BUG(16) /* CPU is affected by Spectre variant 2 attack with indirect branches */ +#define X86_BUG_SPEC_STORE_BYPASS X86_BUG(17) /* CPU is affected by speculative store bypass attack */ #endif /* _ASM_X86_CPUFEATURES_H */ diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index 38a8626c894c..59a9f3aa9f7a 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -404,4 +404,9 @@ ssize_t cpu_show_spectre_v2(struct device *dev, struct device_attribute *attr, c { return cpu_show_common(dev, attr, buf, X86_BUG_SPECTRE_V2); } + +ssize_t cpu_show_spec_store_bypass(struct device *dev, struct device_attribute *attr, char *buf) +{ + return cpu_show_common(dev, attr, buf, X86_BUG_SPEC_STORE_BYPASS); +} #endif diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index 9ed0a18ff6f6..6ff0cc441a59 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -927,10 +927,33 @@ static const __initconst struct x86_cpu_id cpu_no_meltdown[] = { {} }; +static const __initconst struct x86_cpu_id cpu_no_spec_store_bypass[] = { + { X86_VENDOR_INTEL, 6, INTEL_FAM6_ATOM_PINEVIEW }, + { X86_VENDOR_INTEL, 6, INTEL_FAM6_ATOM_LINCROFT }, + { X86_VENDOR_INTEL, 6, INTEL_FAM6_ATOM_PENWELL }, + { X86_VENDOR_INTEL, 6, INTEL_FAM6_ATOM_CLOVERVIEW }, + { X86_VENDOR_INTEL, 6, INTEL_FAM6_ATOM_CEDARVIEW }, + { X86_VENDOR_INTEL, 6, INTEL_FAM6_ATOM_SILVERMONT1 }, + { X86_VENDOR_INTEL, 6, INTEL_FAM6_ATOM_AIRMONT }, + { X86_VENDOR_INTEL, 6, INTEL_FAM6_ATOM_SILVERMONT2 }, + { X86_VENDOR_INTEL, 6, INTEL_FAM6_ATOM_MERRIFIELD }, + { X86_VENDOR_INTEL, 6, INTEL_FAM6_CORE_YONAH }, + { X86_VENDOR_INTEL, 6, INTEL_FAM6_XEON_PHI_KNL }, + { X86_VENDOR_INTEL, 6, INTEL_FAM6_XEON_PHI_KNM }, + { X86_VENDOR_CENTAUR, 5, }, + { X86_VENDOR_INTEL, 5, }, + { X86_VENDOR_NSC, 5, }, + { X86_VENDOR_ANY, 4, }, + {} +}; + static void __init cpu_set_bug_bits(struct cpuinfo_x86 *c) { u64 ia32_cap = 0; + if (!x86_match_cpu(cpu_no_spec_store_bypass)) + setup_force_cpu_bug(X86_BUG_SPEC_STORE_BYPASS); + if (x86_match_cpu(cpu_no_speculation)) return; diff --git a/drivers/base/cpu.c b/drivers/base/cpu.c index 2da998baa75c..30cc9c877ebb 100644 --- a/drivers/base/cpu.c +++ b/drivers/base/cpu.c @@ -534,14 +534,22 @@ ssize_t __weak cpu_show_spectre_v2(struct device *dev, return sprintf(buf, "Not affected\n"); } +ssize_t __weak cpu_show_spec_store_bypass(struct device *dev, + struct device_attribute *attr, char *buf) +{ + return sprintf(buf, "Not affected\n"); +} + static DEVICE_ATTR(meltdown, 0444, cpu_show_meltdown, NULL); static DEVICE_ATTR(spectre_v1, 0444, cpu_show_spectre_v1, NULL); static DEVICE_ATTR(spectre_v2, 0444, cpu_show_spectre_v2, NULL); +static DEVICE_ATTR(spec_store_bypass, 0444, cpu_show_spec_store_bypass, NULL); static struct attribute *cpu_root_vulnerabilities_attrs[] = { &dev_attr_meltdown.attr, &dev_attr_spectre_v1.attr, &dev_attr_spectre_v2.attr, + &dev_attr_spec_store_bypass.attr, NULL }; diff --git a/include/linux/cpu.h b/include/linux/cpu.h index 7b01bc11c692..a97a63eef59f 100644 --- a/include/linux/cpu.h +++ b/include/linux/cpu.h @@ -53,6 +53,8 @@ extern ssize_t cpu_show_spectre_v1(struct device *dev, struct device_attribute *attr, char *buf); extern ssize_t cpu_show_spectre_v2(struct device *dev, struct device_attribute *attr, char *buf); +extern ssize_t cpu_show_spec_store_bypass(struct device *dev, + struct device_attribute *attr, char *buf); extern __printf(4, 5) struct device *cpu_device_create(struct device *parent, void *drvdata, From 0cc5fa00b0a88dad140b4e5c2cead9951ad36822 Mon Sep 17 00:00:00 2001 From: Konrad Rzeszutek Wilk Date: Sat, 28 Apr 2018 22:34:17 +0200 Subject: [PATCH 010/190] x86/cpufeatures: Add X86_FEATURE_RDS Add the CPU feature bit CPUID.7.0.EDX[31] which indicates whether the CPU supports Reduced Data Speculation. [ tglx: Split it out from a later patch ] Signed-off-by: Konrad Rzeszutek Wilk Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar --- arch/x86/include/asm/cpufeatures.h | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/x86/include/asm/cpufeatures.h b/arch/x86/include/asm/cpufeatures.h index 3493b58ef2cc..48305a1ddfd6 100644 --- a/arch/x86/include/asm/cpufeatures.h +++ b/arch/x86/include/asm/cpufeatures.h @@ -334,6 +334,7 @@ #define X86_FEATURE_SPEC_CTRL (18*32+26) /* "" Speculation Control (IBRS + IBPB) */ #define X86_FEATURE_INTEL_STIBP (18*32+27) /* "" Single Thread Indirect Branch Predictors */ #define X86_FEATURE_ARCH_CAPABILITIES (18*32+29) /* IA32_ARCH_CAPABILITIES MSR (Intel) */ +#define X86_FEATURE_RDS (18*32+31) /* Reduced Data Speculation */ /* * BUG word(s) From 24f7fc83b9204d20f878c57cb77d261ae825e033 Mon Sep 17 00:00:00 2001 From: Konrad Rzeszutek Wilk Date: Wed, 25 Apr 2018 22:04:21 -0400 Subject: [PATCH 011/190] x86/bugs: Provide boot parameters for the spec_store_bypass_disable mitigation Contemporary high performance processors use a common industry-wide optimization known as "Speculative Store Bypass" in which loads from addresses to which a recent store has occurred may (speculatively) see an older value. Intel refers to this feature as "Memory Disambiguation" which is part of their "Smart Memory Access" capability. Memory Disambiguation can expose a cache side-channel attack against such speculatively read values. An attacker can create exploit code that allows them to read memory outside of a sandbox environment (for example, malicious JavaScript in a web page), or to perform more complex attacks against code running within the same privilege level, e.g. via the stack. As a first step to mitigate against such attacks, provide two boot command line control knobs: nospec_store_bypass_disable spec_store_bypass_disable=[off,auto,on] By default affected x86 processors will power on with Speculative Store Bypass enabled. Hence the provided kernel parameters are written from the point of view of whether to enable a mitigation or not. The parameters are as follows: - auto - Kernel detects whether your CPU model contains an implementation of Speculative Store Bypass and picks the most appropriate mitigation. - on - disable Speculative Store Bypass - off - enable Speculative Store Bypass [ tglx: Reordered the checks so that the whole evaluation is not done when the CPU does not support RDS ] Signed-off-by: Konrad Rzeszutek Wilk Signed-off-by: Thomas Gleixner Reviewed-by: Borislav Petkov Reviewed-by: Ingo Molnar --- .../admin-guide/kernel-parameters.txt | 33 ++++++ arch/x86/include/asm/cpufeatures.h | 1 + arch/x86/include/asm/nospec-branch.h | 6 + arch/x86/kernel/cpu/bugs.c | 103 ++++++++++++++++++ 4 files changed, 143 insertions(+) diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt index 11fc28ecdb6d..f32112c55f7e 100644 --- a/Documentation/admin-guide/kernel-parameters.txt +++ b/Documentation/admin-guide/kernel-parameters.txt @@ -2680,6 +2680,9 @@ allow data leaks with this option, which is equivalent to spectre_v2=off. + nospec_store_bypass_disable + [HW] Disable all mitigations for the Speculative Store Bypass vulnerability + noxsave [BUGS=X86] Disables x86 extended register state save and restore using xsave. The kernel will fallback to enabling legacy floating-point and sse state. @@ -4025,6 +4028,36 @@ Not specifying this option is equivalent to spectre_v2=auto. + spec_store_bypass_disable= + [HW] Control Speculative Store Bypass (SSB) Disable mitigation + (Speculative Store Bypass vulnerability) + + Certain CPUs are vulnerable to an exploit against a + a common industry wide performance optimization known + as "Speculative Store Bypass" in which recent stores + to the same memory location may not be observed by + later loads during speculative execution. The idea + is that such stores are unlikely and that they can + be detected prior to instruction retirement at the + end of a particular speculation execution window. + + In vulnerable processors, the speculatively forwarded + store can be used in a cache side channel attack, for + example to read memory to which the attacker does not + directly have access (e.g. inside sandboxed code). + + This parameter controls whether the Speculative Store + Bypass optimization is used. + + on - Unconditionally disable Speculative Store Bypass + off - Unconditionally enable Speculative Store Bypass + auto - Kernel detects whether the CPU model contains an + implementation of Speculative Store Bypass and + picks the most appropriate mitigation + + Not specifying this option is equivalent to + spec_store_bypass_disable=auto. + spia_io_base= [HW,MTD] spia_fio_base= spia_pedr= diff --git a/arch/x86/include/asm/cpufeatures.h b/arch/x86/include/asm/cpufeatures.h index 48305a1ddfd6..58689ed969bb 100644 --- a/arch/x86/include/asm/cpufeatures.h +++ b/arch/x86/include/asm/cpufeatures.h @@ -214,6 +214,7 @@ #define X86_FEATURE_USE_IBPB ( 7*32+21) /* "" Indirect Branch Prediction Barrier enabled */ #define X86_FEATURE_USE_IBRS_FW ( 7*32+22) /* "" Use IBRS during runtime firmware calls */ +#define X86_FEATURE_SPEC_STORE_BYPASS_DISABLE ( 7*32+23) /* "" Disable Speculative Store Bypass. */ /* Virtualization flags: Linux defined, word 8 */ #define X86_FEATURE_TPR_SHADOW ( 8*32+ 0) /* Intel TPR Shadow */ diff --git a/arch/x86/include/asm/nospec-branch.h b/arch/x86/include/asm/nospec-branch.h index d1c2630922da..7b9eacfc03f3 100644 --- a/arch/x86/include/asm/nospec-branch.h +++ b/arch/x86/include/asm/nospec-branch.h @@ -238,6 +238,12 @@ extern u64 x86_spec_ctrl_get_default(void); extern void x86_spec_ctrl_set_guest(u64); extern void x86_spec_ctrl_restore_host(u64); +/* The Speculative Store Bypass disable variants */ +enum ssb_mitigation { + SPEC_STORE_BYPASS_NONE, + SPEC_STORE_BYPASS_DISABLE, +}; + extern char __indirect_thunk_start[]; extern char __indirect_thunk_end[]; diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index 59a9f3aa9f7a..2d1f1cb604aa 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -27,6 +27,7 @@ #include static void __init spectre_v2_select_mitigation(void); +static void __init ssb_select_mitigation(void); /* * Our boot-time value of the SPEC_CTRL MSR. We read it once so that any @@ -53,6 +54,12 @@ void __init check_bugs(void) /* Select the proper spectre mitigation before patching alternatives */ spectre_v2_select_mitigation(); + /* + * Select proper mitigation for any exposure to the Speculative Store + * Bypass vulnerability. + */ + ssb_select_mitigation(); + #ifdef CONFIG_X86_32 /* * Check whether we are able to run this kernel safely on SMP. @@ -357,6 +364,99 @@ static void __init spectre_v2_select_mitigation(void) } } +#undef pr_fmt +#define pr_fmt(fmt) "Speculative Store Bypass: " fmt + +static enum ssb_mitigation ssb_mode = SPEC_STORE_BYPASS_NONE; + +/* The kernel command line selection */ +enum ssb_mitigation_cmd { + SPEC_STORE_BYPASS_CMD_NONE, + SPEC_STORE_BYPASS_CMD_AUTO, + SPEC_STORE_BYPASS_CMD_ON, +}; + +static const char *ssb_strings[] = { + [SPEC_STORE_BYPASS_NONE] = "Vulnerable", + [SPEC_STORE_BYPASS_DISABLE] = "Mitigation: Speculative Store Bypass disabled" +}; + +static const struct { + const char *option; + enum ssb_mitigation_cmd cmd; +} ssb_mitigation_options[] = { + { "auto", SPEC_STORE_BYPASS_CMD_AUTO }, /* Platform decides */ + { "on", SPEC_STORE_BYPASS_CMD_ON }, /* Disable Speculative Store Bypass */ + { "off", SPEC_STORE_BYPASS_CMD_NONE }, /* Don't touch Speculative Store Bypass */ +}; + +static enum ssb_mitigation_cmd __init ssb_parse_cmdline(void) +{ + enum ssb_mitigation_cmd cmd = SPEC_STORE_BYPASS_CMD_AUTO; + char arg[20]; + int ret, i; + + if (cmdline_find_option_bool(boot_command_line, "nospec_store_bypass_disable")) { + return SPEC_STORE_BYPASS_CMD_NONE; + } else { + ret = cmdline_find_option(boot_command_line, "spec_store_bypass_disable", + arg, sizeof(arg)); + if (ret < 0) + return SPEC_STORE_BYPASS_CMD_AUTO; + + for (i = 0; i < ARRAY_SIZE(ssb_mitigation_options); i++) { + if (!match_option(arg, ret, ssb_mitigation_options[i].option)) + continue; + + cmd = ssb_mitigation_options[i].cmd; + break; + } + + if (i >= ARRAY_SIZE(ssb_mitigation_options)) { + pr_err("unknown option (%s). Switching to AUTO select\n", arg); + return SPEC_STORE_BYPASS_CMD_AUTO; + } + } + + return cmd; +} + +static enum ssb_mitigation_cmd __init __ssb_select_mitigation(void) +{ + enum ssb_mitigation mode = SPEC_STORE_BYPASS_NONE; + enum ssb_mitigation_cmd cmd; + + if (!boot_cpu_has(X86_FEATURE_RDS)) + return mode; + + cmd = ssb_parse_cmdline(); + if (!boot_cpu_has_bug(X86_BUG_SPEC_STORE_BYPASS) && + (cmd == SPEC_STORE_BYPASS_CMD_NONE || + cmd == SPEC_STORE_BYPASS_CMD_AUTO)) + return mode; + + switch (cmd) { + case SPEC_STORE_BYPASS_CMD_AUTO: + case SPEC_STORE_BYPASS_CMD_ON: + mode = SPEC_STORE_BYPASS_DISABLE; + break; + case SPEC_STORE_BYPASS_CMD_NONE: + break; + } + + if (mode != SPEC_STORE_BYPASS_NONE) + setup_force_cpu_cap(X86_FEATURE_SPEC_STORE_BYPASS_DISABLE); + return mode; +} + +static void ssb_select_mitigation() +{ + ssb_mode = __ssb_select_mitigation(); + + if (boot_cpu_has_bug(X86_BUG_SPEC_STORE_BYPASS)) + pr_info("%s\n", ssb_strings[ssb_mode]); +} + #undef pr_fmt #ifdef CONFIG_SYSFS @@ -383,6 +483,9 @@ ssize_t cpu_show_common(struct device *dev, struct device_attribute *attr, boot_cpu_has(X86_FEATURE_USE_IBRS_FW) ? ", IBRS_FW" : "", spectre_v2_module_string()); + case X86_BUG_SPEC_STORE_BYPASS: + return sprintf(buf, "%s\n", ssb_strings[ssb_mode]); + default: break; } From 772439717dbf703b39990be58d8d4e3e4ad0598a Mon Sep 17 00:00:00 2001 From: Konrad Rzeszutek Wilk Date: Wed, 25 Apr 2018 22:04:22 -0400 Subject: [PATCH 012/190] x86/bugs/intel: Set proper CPU features and setup RDS Intel CPUs expose methods to: - Detect whether RDS capability is available via CPUID.7.0.EDX[31], - The SPEC_CTRL MSR(0x48), bit 2 set to enable RDS. - MSR_IA32_ARCH_CAPABILITIES, Bit(4) no need to enable RRS. With that in mind if spec_store_bypass_disable=[auto,on] is selected set at boot-time the SPEC_CTRL MSR to enable RDS if the platform requires it. Note that this does not fix the KVM case where the SPEC_CTRL is exposed to guests which can muck with it, see patch titled : KVM/SVM/VMX/x86/spectre_v2: Support the combination of guest and host IBRS. And for the firmware (IBRS to be set), see patch titled: x86/spectre_v2: Read SPEC_CTRL MSR during boot and re-use reserved bits [ tglx: Distangled it from the intel implementation and kept the call order ] Signed-off-by: Konrad Rzeszutek Wilk Signed-off-by: Thomas Gleixner Reviewed-by: Borislav Petkov Reviewed-by: Ingo Molnar --- arch/x86/include/asm/msr-index.h | 6 ++++++ arch/x86/kernel/cpu/bugs.c | 30 ++++++++++++++++++++++++++++-- arch/x86/kernel/cpu/common.c | 10 ++++++---- arch/x86/kernel/cpu/cpu.h | 2 ++ arch/x86/kernel/cpu/intel.c | 1 + 5 files changed, 43 insertions(+), 6 deletions(-) diff --git a/arch/x86/include/asm/msr-index.h b/arch/x86/include/asm/msr-index.h index 53d5b1b9255e..21e1a6df9907 100644 --- a/arch/x86/include/asm/msr-index.h +++ b/arch/x86/include/asm/msr-index.h @@ -42,6 +42,7 @@ #define MSR_IA32_SPEC_CTRL 0x00000048 /* Speculation Control */ #define SPEC_CTRL_IBRS (1 << 0) /* Indirect Branch Restricted Speculation */ #define SPEC_CTRL_STIBP (1 << 1) /* Single Thread Indirect Branch Predictors */ +#define SPEC_CTRL_RDS (1 << 2) /* Reduced Data Speculation */ #define MSR_IA32_PRED_CMD 0x00000049 /* Prediction Command */ #define PRED_CMD_IBPB (1 << 0) /* Indirect Branch Prediction Barrier */ @@ -68,6 +69,11 @@ #define MSR_IA32_ARCH_CAPABILITIES 0x0000010a #define ARCH_CAP_RDCL_NO (1 << 0) /* Not susceptible to Meltdown */ #define ARCH_CAP_IBRS_ALL (1 << 1) /* Enhanced IBRS support */ +#define ARCH_CAP_RDS_NO (1 << 4) /* + * Not susceptible to Speculative Store Bypass + * attack, so no Reduced Data Speculation control + * required. + */ #define MSR_IA32_BBL_CR_CTL 0x00000119 #define MSR_IA32_BBL_CR_CTL3 0x0000011e diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index 2d1f1cb604aa..17edf5bdc361 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -117,7 +117,7 @@ static enum spectre_v2_mitigation spectre_v2_enabled = SPECTRE_V2_NONE; void x86_spec_ctrl_set(u64 val) { - if (val & ~SPEC_CTRL_IBRS) + if (val & ~(SPEC_CTRL_IBRS | SPEC_CTRL_RDS)) WARN_ONCE(1, "SPEC_CTRL MSR value 0x%16llx is unknown.\n", val); else wrmsrl(MSR_IA32_SPEC_CTRL, x86_spec_ctrl_base | val); @@ -444,8 +444,28 @@ static enum ssb_mitigation_cmd __init __ssb_select_mitigation(void) break; } - if (mode != SPEC_STORE_BYPASS_NONE) + /* + * We have three CPU feature flags that are in play here: + * - X86_BUG_SPEC_STORE_BYPASS - CPU is susceptible. + * - X86_FEATURE_RDS - CPU is able to turn off speculative store bypass + * - X86_FEATURE_SPEC_STORE_BYPASS_DISABLE - engage the mitigation + */ + if (mode != SPEC_STORE_BYPASS_NONE) { setup_force_cpu_cap(X86_FEATURE_SPEC_STORE_BYPASS_DISABLE); + /* + * Intel uses the SPEC CTRL MSR Bit(2) for this, while AMD uses + * a completely different MSR and bit dependent on family. + */ + switch (boot_cpu_data.x86_vendor) { + case X86_VENDOR_INTEL: + x86_spec_ctrl_base |= SPEC_CTRL_RDS; + x86_spec_ctrl_set(SPEC_CTRL_RDS); + break; + case X86_VENDOR_AMD: + break; + } + } + return mode; } @@ -459,6 +479,12 @@ static void ssb_select_mitigation() #undef pr_fmt +void x86_spec_ctrl_setup_ap(void) +{ + if (boot_cpu_has(X86_FEATURE_IBRS)) + x86_spec_ctrl_set(x86_spec_ctrl_base & (SPEC_CTRL_IBRS | SPEC_CTRL_RDS)); +} + #ifdef CONFIG_SYSFS ssize_t cpu_show_common(struct device *dev, struct device_attribute *attr, diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index 6ff0cc441a59..d6dc71d616ea 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -951,7 +951,11 @@ static void __init cpu_set_bug_bits(struct cpuinfo_x86 *c) { u64 ia32_cap = 0; - if (!x86_match_cpu(cpu_no_spec_store_bypass)) + if (cpu_has(c, X86_FEATURE_ARCH_CAPABILITIES)) + rdmsrl(MSR_IA32_ARCH_CAPABILITIES, ia32_cap); + + if (!x86_match_cpu(cpu_no_spec_store_bypass) && + !(ia32_cap & ARCH_CAP_RDS_NO)) setup_force_cpu_bug(X86_BUG_SPEC_STORE_BYPASS); if (x86_match_cpu(cpu_no_speculation)) @@ -963,9 +967,6 @@ static void __init cpu_set_bug_bits(struct cpuinfo_x86 *c) if (x86_match_cpu(cpu_no_meltdown)) return; - if (cpu_has(c, X86_FEATURE_ARCH_CAPABILITIES)) - rdmsrl(MSR_IA32_ARCH_CAPABILITIES, ia32_cap); - /* Rogue Data Cache Load? No! */ if (ia32_cap & ARCH_CAP_RDCL_NO) return; @@ -1383,6 +1384,7 @@ void identify_secondary_cpu(struct cpuinfo_x86 *c) #endif mtrr_ap_init(); validate_apic_and_package_id(c); + x86_spec_ctrl_setup_ap(); } static __init int setup_noclflush(char *arg) diff --git a/arch/x86/kernel/cpu/cpu.h b/arch/x86/kernel/cpu/cpu.h index e806b11a99af..37672d299e35 100644 --- a/arch/x86/kernel/cpu/cpu.h +++ b/arch/x86/kernel/cpu/cpu.h @@ -50,4 +50,6 @@ extern void cpu_detect_cache_sizes(struct cpuinfo_x86 *c); unsigned int aperfmperf_get_khz(int cpu); +extern void x86_spec_ctrl_setup_ap(void); + #endif /* ARCH_X86_CPU_H */ diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c index 60d1897041da..ef3f9c01c274 100644 --- a/arch/x86/kernel/cpu/intel.c +++ b/arch/x86/kernel/cpu/intel.c @@ -189,6 +189,7 @@ static void early_init_intel(struct cpuinfo_x86 *c) setup_clear_cpu_cap(X86_FEATURE_STIBP); setup_clear_cpu_cap(X86_FEATURE_SPEC_CTRL); setup_clear_cpu_cap(X86_FEATURE_INTEL_STIBP); + setup_clear_cpu_cap(X86_FEATURE_RDS); } /* From 1115a859f33276fe8afb31c60cf9d8e657872558 Mon Sep 17 00:00:00 2001 From: Konrad Rzeszutek Wilk Date: Wed, 25 Apr 2018 22:04:23 -0400 Subject: [PATCH 013/190] x86/bugs: Whitelist allowed SPEC_CTRL MSR values Intel and AMD SPEC_CTRL (0x48) MSR semantics may differ in the future (or in fact use different MSRs for the same functionality). As such a run-time mechanism is required to whitelist the appropriate MSR values. [ tglx: Made the variable __ro_after_init ] Signed-off-by: Konrad Rzeszutek Wilk Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar --- arch/x86/kernel/cpu/bugs.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index 17edf5bdc361..d809d398675a 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -35,6 +35,12 @@ static void __init ssb_select_mitigation(void); */ static u64 __ro_after_init x86_spec_ctrl_base; +/* + * The vendor and possibly platform specific bits which can be modified in + * x86_spec_ctrl_base. + */ +static u64 __ro_after_init x86_spec_ctrl_mask = ~SPEC_CTRL_IBRS; + void __init check_bugs(void) { identify_boot_cpu(); @@ -117,7 +123,7 @@ static enum spectre_v2_mitigation spectre_v2_enabled = SPECTRE_V2_NONE; void x86_spec_ctrl_set(u64 val) { - if (val & ~(SPEC_CTRL_IBRS | SPEC_CTRL_RDS)) + if (val & x86_spec_ctrl_mask) WARN_ONCE(1, "SPEC_CTRL MSR value 0x%16llx is unknown.\n", val); else wrmsrl(MSR_IA32_SPEC_CTRL, x86_spec_ctrl_base | val); @@ -459,6 +465,7 @@ static enum ssb_mitigation_cmd __init __ssb_select_mitigation(void) switch (boot_cpu_data.x86_vendor) { case X86_VENDOR_INTEL: x86_spec_ctrl_base |= SPEC_CTRL_RDS; + x86_spec_ctrl_mask &= ~SPEC_CTRL_RDS; x86_spec_ctrl_set(SPEC_CTRL_RDS); break; case X86_VENDOR_AMD: @@ -482,7 +489,7 @@ static void ssb_select_mitigation() void x86_spec_ctrl_setup_ap(void) { if (boot_cpu_has(X86_FEATURE_IBRS)) - x86_spec_ctrl_set(x86_spec_ctrl_base & (SPEC_CTRL_IBRS | SPEC_CTRL_RDS)); + x86_spec_ctrl_set(x86_spec_ctrl_base & ~x86_spec_ctrl_mask); } #ifdef CONFIG_SYSFS From 764f3c21588a059cd783c6ba0734d4db2d72822d Mon Sep 17 00:00:00 2001 From: Konrad Rzeszutek Wilk Date: Wed, 25 Apr 2018 22:04:24 -0400 Subject: [PATCH 014/190] x86/bugs/AMD: Add support to disable RDS on Fam[15,16,17]h if requested AMD does not need the Speculative Store Bypass mitigation to be enabled. The parameters for this are already available and can be done via MSR C001_1020. Each family uses a different bit in that MSR for this. [ tglx: Expose the bit mask via a variable and move the actual MSR fiddling into the bugs code as that's the right thing to do and also required to prepare for dynamic enable/disable ] Suggested-by: Borislav Petkov Signed-off-by: Konrad Rzeszutek Wilk Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar --- arch/x86/include/asm/cpufeatures.h | 1 + arch/x86/include/asm/nospec-branch.h | 4 ++++ arch/x86/kernel/cpu/amd.c | 26 ++++++++++++++++++++++++++ arch/x86/kernel/cpu/bugs.c | 27 ++++++++++++++++++++++++++- arch/x86/kernel/cpu/common.c | 4 ++++ 5 files changed, 61 insertions(+), 1 deletion(-) diff --git a/arch/x86/include/asm/cpufeatures.h b/arch/x86/include/asm/cpufeatures.h index 58689ed969bb..b2464c1787df 100644 --- a/arch/x86/include/asm/cpufeatures.h +++ b/arch/x86/include/asm/cpufeatures.h @@ -215,6 +215,7 @@ #define X86_FEATURE_USE_IBPB ( 7*32+21) /* "" Indirect Branch Prediction Barrier enabled */ #define X86_FEATURE_USE_IBRS_FW ( 7*32+22) /* "" Use IBRS during runtime firmware calls */ #define X86_FEATURE_SPEC_STORE_BYPASS_DISABLE ( 7*32+23) /* "" Disable Speculative Store Bypass. */ +#define X86_FEATURE_AMD_RDS (7*32+24) /* "" AMD RDS implementation */ /* Virtualization flags: Linux defined, word 8 */ #define X86_FEATURE_TPR_SHADOW ( 8*32+ 0) /* Intel TPR Shadow */ diff --git a/arch/x86/include/asm/nospec-branch.h b/arch/x86/include/asm/nospec-branch.h index 7b9eacfc03f3..3a1541cf32de 100644 --- a/arch/x86/include/asm/nospec-branch.h +++ b/arch/x86/include/asm/nospec-branch.h @@ -244,6 +244,10 @@ enum ssb_mitigation { SPEC_STORE_BYPASS_DISABLE, }; +/* AMD specific Speculative Store Bypass MSR data */ +extern u64 x86_amd_ls_cfg_base; +extern u64 x86_amd_ls_cfg_rds_mask; + extern char __indirect_thunk_start[]; extern char __indirect_thunk_end[]; diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index 12bc0a1139da..3d553fa5075d 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -10,6 +10,7 @@ #include #include #include +#include #include #include #include @@ -554,6 +555,26 @@ static void bsp_init_amd(struct cpuinfo_x86 *c) rdmsrl(MSR_FAM10H_NODE_ID, value); nodes_per_socket = ((value >> 3) & 7) + 1; } + + if (c->x86 >= 0x15 && c->x86 <= 0x17) { + unsigned int bit; + + switch (c->x86) { + case 0x15: bit = 54; break; + case 0x16: bit = 33; break; + case 0x17: bit = 10; break; + default: return; + } + /* + * Try to cache the base value so further operations can + * avoid RMW. If that faults, do not enable RDS. + */ + if (!rdmsrl_safe(MSR_AMD64_LS_CFG, &x86_amd_ls_cfg_base)) { + setup_force_cpu_cap(X86_FEATURE_RDS); + setup_force_cpu_cap(X86_FEATURE_AMD_RDS); + x86_amd_ls_cfg_rds_mask = 1ULL << bit; + } + } } static void early_detect_mem_encrypt(struct cpuinfo_x86 *c) @@ -898,6 +919,11 @@ static void init_amd(struct cpuinfo_x86 *c) /* AMD CPUs don't reset SS attributes on SYSRET, Xen does. */ if (!cpu_has(c, X86_FEATURE_XENPV)) set_cpu_bug(c, X86_BUG_SYSRET_SS_ATTRS); + + if (boot_cpu_has(X86_FEATURE_AMD_RDS)) { + set_cpu_cap(c, X86_FEATURE_RDS); + set_cpu_cap(c, X86_FEATURE_AMD_RDS); + } } #ifdef CONFIG_X86_32 diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index d809d398675a..86cdf635636d 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -41,6 +41,13 @@ static u64 __ro_after_init x86_spec_ctrl_base; */ static u64 __ro_after_init x86_spec_ctrl_mask = ~SPEC_CTRL_IBRS; +/* + * AMD specific MSR info for Speculative Store Bypass control. + * x86_amd_ls_cfg_rds_mask is initialized in identify_boot_cpu(). + */ +u64 __ro_after_init x86_amd_ls_cfg_base; +u64 __ro_after_init x86_amd_ls_cfg_rds_mask; + void __init check_bugs(void) { identify_boot_cpu(); @@ -52,7 +59,8 @@ void __init check_bugs(void) /* * Read the SPEC_CTRL MSR to account for reserved bits which may - * have unknown values. + * have unknown values. AMD64_LS_CFG MSR is cached in the early AMD + * init code as it is not enumerated and depends on the family. */ if (boot_cpu_has(X86_FEATURE_IBRS)) rdmsrl(MSR_IA32_SPEC_CTRL, x86_spec_ctrl_base); @@ -154,6 +162,14 @@ void x86_spec_ctrl_restore_host(u64 guest_spec_ctrl) } EXPORT_SYMBOL_GPL(x86_spec_ctrl_restore_host); +static void x86_amd_rds_enable(void) +{ + u64 msrval = x86_amd_ls_cfg_base | x86_amd_ls_cfg_rds_mask; + + if (boot_cpu_has(X86_FEATURE_AMD_RDS)) + wrmsrl(MSR_AMD64_LS_CFG, msrval); +} + #ifdef RETPOLINE static bool spectre_v2_bad_module; @@ -443,6 +459,11 @@ static enum ssb_mitigation_cmd __init __ssb_select_mitigation(void) switch (cmd) { case SPEC_STORE_BYPASS_CMD_AUTO: + /* + * AMD platforms by default don't need SSB mitigation. + */ + if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD) + break; case SPEC_STORE_BYPASS_CMD_ON: mode = SPEC_STORE_BYPASS_DISABLE; break; @@ -469,6 +490,7 @@ static enum ssb_mitigation_cmd __init __ssb_select_mitigation(void) x86_spec_ctrl_set(SPEC_CTRL_RDS); break; case X86_VENDOR_AMD: + x86_amd_rds_enable(); break; } } @@ -490,6 +512,9 @@ void x86_spec_ctrl_setup_ap(void) { if (boot_cpu_has(X86_FEATURE_IBRS)) x86_spec_ctrl_set(x86_spec_ctrl_base & ~x86_spec_ctrl_mask); + + if (ssb_mode == SPEC_STORE_BYPASS_DISABLE) + x86_amd_rds_enable(); } #ifdef CONFIG_SYSFS diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index d6dc71d616ea..e0517bcee446 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -943,6 +943,10 @@ static const __initconst struct x86_cpu_id cpu_no_spec_store_bypass[] = { { X86_VENDOR_CENTAUR, 5, }, { X86_VENDOR_INTEL, 5, }, { X86_VENDOR_NSC, 5, }, + { X86_VENDOR_AMD, 0x12, }, + { X86_VENDOR_AMD, 0x11, }, + { X86_VENDOR_AMD, 0x10, }, + { X86_VENDOR_AMD, 0xf, }, { X86_VENDOR_ANY, 4, }, {} }; From da39556f66f5cfe8f9c989206974f1cb16ca5d7c Mon Sep 17 00:00:00 2001 From: Konrad Rzeszutek Wilk Date: Wed, 25 Apr 2018 22:04:25 -0400 Subject: [PATCH 015/190] x86/KVM/VMX: Expose SPEC_CTRL Bit(2) to the guest Expose the CPUID.7.EDX[31] bit to the guest, and also guard against various combinations of SPEC_CTRL MSR values. The handling of the MSR (to take into account the host value of SPEC_CTRL Bit(2)) is taken care of in patch: KVM/SVM/VMX/x86/spectre_v2: Support the combination of guest and host IBRS Signed-off-by: Konrad Rzeszutek Wilk Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar --- arch/x86/kvm/cpuid.c | 2 +- arch/x86/kvm/vmx.c | 8 +++++--- 2 files changed, 6 insertions(+), 4 deletions(-) diff --git a/arch/x86/kvm/cpuid.c b/arch/x86/kvm/cpuid.c index 82055b90a8b3..376ac9a2a2b9 100644 --- a/arch/x86/kvm/cpuid.c +++ b/arch/x86/kvm/cpuid.c @@ -407,7 +407,7 @@ static inline int __do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 function, /* cpuid 7.0.edx*/ const u32 kvm_cpuid_7_0_edx_x86_features = - F(AVX512_4VNNIW) | F(AVX512_4FMAPS) | F(SPEC_CTRL) | + F(AVX512_4VNNIW) | F(AVX512_4FMAPS) | F(SPEC_CTRL) | F(RDS) | F(ARCH_CAPABILITIES); /* all calls to cpuid_count() should be made on the same cpu */ diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 3091dc98a451..4197942b6d9b 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -3524,7 +3524,8 @@ static int vmx_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) case MSR_IA32_SPEC_CTRL: if (!msr_info->host_initiated && !guest_cpuid_has(vcpu, X86_FEATURE_IBRS) && - !guest_cpuid_has(vcpu, X86_FEATURE_SPEC_CTRL)) + !guest_cpuid_has(vcpu, X86_FEATURE_SPEC_CTRL) && + !guest_cpuid_has(vcpu, X86_FEATURE_RDS)) return 1; msr_info->data = to_vmx(vcpu)->spec_ctrl; @@ -3643,11 +3644,12 @@ static int vmx_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) case MSR_IA32_SPEC_CTRL: if (!msr_info->host_initiated && !guest_cpuid_has(vcpu, X86_FEATURE_IBRS) && - !guest_cpuid_has(vcpu, X86_FEATURE_SPEC_CTRL)) + !guest_cpuid_has(vcpu, X86_FEATURE_SPEC_CTRL) && + !guest_cpuid_has(vcpu, X86_FEATURE_RDS)) return 1; /* The STIBP bit doesn't fault even if it's not advertised */ - if (data & ~(SPEC_CTRL_IBRS | SPEC_CTRL_STIBP)) + if (data & ~(SPEC_CTRL_IBRS | SPEC_CTRL_STIBP | SPEC_CTRL_RDS)) return 1; vmx->spec_ctrl = data; From 28a2775217b17208811fa43a9e96bd1fdf417b86 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Sun, 29 Apr 2018 15:01:37 +0200 Subject: [PATCH 016/190] x86/speculation: Create spec-ctrl.h to avoid include hell Having everything in nospec-branch.h creates a hell of dependencies when adding the prctl based switching mechanism. Move everything which is not required in nospec-branch.h to spec-ctrl.h and fix up the includes in the relevant files. Signed-off-by: Thomas Gleixner Reviewed-by: Konrad Rzeszutek Wilk Reviewed-by: Ingo Molnar --- arch/x86/include/asm/nospec-branch.h | 14 -------------- arch/x86/include/asm/spec-ctrl.h | 21 +++++++++++++++++++++ arch/x86/kernel/cpu/amd.c | 2 +- arch/x86/kernel/cpu/bugs.c | 2 +- arch/x86/kvm/svm.c | 2 +- arch/x86/kvm/vmx.c | 2 +- 6 files changed, 25 insertions(+), 18 deletions(-) create mode 100644 arch/x86/include/asm/spec-ctrl.h diff --git a/arch/x86/include/asm/nospec-branch.h b/arch/x86/include/asm/nospec-branch.h index 3a1541cf32de..1119f14bc883 100644 --- a/arch/x86/include/asm/nospec-branch.h +++ b/arch/x86/include/asm/nospec-branch.h @@ -228,26 +228,12 @@ enum spectre_v2_mitigation { extern void x86_spec_ctrl_set(u64); extern u64 x86_spec_ctrl_get_default(void); -/* - * On VMENTER we must preserve whatever view of the SPEC_CTRL MSR - * the guest has, while on VMEXIT we restore the host view. This - * would be easier if SPEC_CTRL were architecturally maskable or - * shadowable for guests but this is not (currently) the case. - * Takes the guest view of SPEC_CTRL MSR as a parameter. - */ -extern void x86_spec_ctrl_set_guest(u64); -extern void x86_spec_ctrl_restore_host(u64); - /* The Speculative Store Bypass disable variants */ enum ssb_mitigation { SPEC_STORE_BYPASS_NONE, SPEC_STORE_BYPASS_DISABLE, }; -/* AMD specific Speculative Store Bypass MSR data */ -extern u64 x86_amd_ls_cfg_base; -extern u64 x86_amd_ls_cfg_rds_mask; - extern char __indirect_thunk_start[]; extern char __indirect_thunk_end[]; diff --git a/arch/x86/include/asm/spec-ctrl.h b/arch/x86/include/asm/spec-ctrl.h new file mode 100644 index 000000000000..3ad64420a06e --- /dev/null +++ b/arch/x86/include/asm/spec-ctrl.h @@ -0,0 +1,21 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _ASM_X86_SPECCTRL_H_ +#define _ASM_X86_SPECCTRL_H_ + +#include + +/* + * On VMENTER we must preserve whatever view of the SPEC_CTRL MSR + * the guest has, while on VMEXIT we restore the host view. This + * would be easier if SPEC_CTRL were architecturally maskable or + * shadowable for guests but this is not (currently) the case. + * Takes the guest view of SPEC_CTRL MSR as a parameter. + */ +extern void x86_spec_ctrl_set_guest(u64); +extern void x86_spec_ctrl_restore_host(u64); + +/* AMD specific Speculative Store Bypass MSR data */ +extern u64 x86_amd_ls_cfg_base; +extern u64 x86_amd_ls_cfg_rds_mask; + +#endif diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index 3d553fa5075d..18efc33a8d2e 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -10,7 +10,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index 86cdf635636d..ec171873167a 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -13,7 +13,7 @@ #include #include -#include +#include #include #include #include diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 8bfc8fe26851..437c1b371129 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -49,7 +49,7 @@ #include #include #include -#include +#include #include #include "trace.h" diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 4197942b6d9b..16a111e44691 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -51,7 +51,7 @@ #include #include #include -#include +#include #include #include "trace.h" From b617cfc858161140d69cc0b5cc211996b557a1c7 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Sun, 29 Apr 2018 15:20:11 +0200 Subject: [PATCH 017/190] prctl: Add speculation control prctls Add two new prctls to control aspects of speculation related vulnerabilites and their mitigations to provide finer grained control over performance impacting mitigations. PR_GET_SPECULATION_CTRL returns the state of the speculation misfeature which is selected with arg2 of prctl(2). The return value uses bit 0-2 with the following meaning: Bit Define Description 0 PR_SPEC_PRCTL Mitigation can be controlled per task by PR_SET_SPECULATION_CTRL 1 PR_SPEC_ENABLE The speculation feature is enabled, mitigation is disabled 2 PR_SPEC_DISABLE The speculation feature is disabled, mitigation is enabled If all bits are 0 the CPU is not affected by the speculation misfeature. If PR_SPEC_PRCTL is set, then the per task control of the mitigation is available. If not set, prctl(PR_SET_SPECULATION_CTRL) for the speculation misfeature will fail. PR_SET_SPECULATION_CTRL allows to control the speculation misfeature, which is selected by arg2 of prctl(2) per task. arg3 is used to hand in the control value, i.e. either PR_SPEC_ENABLE or PR_SPEC_DISABLE. The common return values are: EINVAL prctl is not implemented by the architecture or the unused prctl() arguments are not 0 ENODEV arg2 is selecting a not supported speculation misfeature PR_SET_SPECULATION_CTRL has these additional return values: ERANGE arg3 is incorrect, i.e. it's not either PR_SPEC_ENABLE or PR_SPEC_DISABLE ENXIO prctl control of the selected speculation misfeature is disabled The first supported controlable speculation misfeature is PR_SPEC_STORE_BYPASS. Add the define so this can be shared between architectures. Based on an initial patch from Tim Chen and mostly rewritten. Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar Reviewed-by: Konrad Rzeszutek Wilk --- Documentation/userspace-api/index.rst | 1 + Documentation/userspace-api/spec_ctrl.rst | 86 +++++++++++++++++++++++ include/linux/nospec.h | 5 ++ include/uapi/linux/prctl.h | 11 +++ kernel/sys.c | 22 ++++++ 5 files changed, 125 insertions(+) create mode 100644 Documentation/userspace-api/spec_ctrl.rst diff --git a/Documentation/userspace-api/index.rst b/Documentation/userspace-api/index.rst index 7b2eb1b7d4ca..a3233da7fa88 100644 --- a/Documentation/userspace-api/index.rst +++ b/Documentation/userspace-api/index.rst @@ -19,6 +19,7 @@ place where this information is gathered. no_new_privs seccomp_filter unshare + spec_ctrl .. only:: subproject and html diff --git a/Documentation/userspace-api/spec_ctrl.rst b/Documentation/userspace-api/spec_ctrl.rst new file mode 100644 index 000000000000..ddbebcd01208 --- /dev/null +++ b/Documentation/userspace-api/spec_ctrl.rst @@ -0,0 +1,86 @@ +=================== +Speculation Control +=================== + +Quite some CPUs have speculation related misfeatures which are in fact +vulnerabilites causing data leaks in various forms even accross privilege +domains. + +The kernel provides mitigation for such vulnerabilities in various +forms. Some of these mitigations are compile time configurable and some on +the kernel command line. + +There is also a class of mitigations which are very expensive, but they can +be restricted to a certain set of processes or tasks in controlled +environments. The mechanism to control these mitigations is via +:manpage:`prctl(2)`. + +There are two prctl options which are related to this: + + * PR_GET_SPECULATION_CTRL + + * PR_SET_SPECULATION_CTRL + +PR_GET_SPECULATION_CTRL +----------------------- + +PR_GET_SPECULATION_CTRL returns the state of the speculation misfeature +which is selected with arg2 of prctl(2). The return value uses bits 0-2 with +the following meaning: + +==== ================ =================================================== +Bit Define Description +==== ================ =================================================== +0 PR_SPEC_PRCTL Mitigation can be controlled per task by + PR_SET_SPECULATION_CTRL +1 PR_SPEC_ENABLE The speculation feature is enabled, mitigation is + disabled +2 PR_SPEC_DISABLE The speculation feature is disabled, mitigation is + enabled +==== ================ =================================================== + +If all bits are 0 the CPU is not affected by the speculation misfeature. + +If PR_SPEC_PRCTL is set, then the per task control of the mitigation is +available. If not set, prctl(PR_SET_SPECULATION_CTRL) for the speculation +misfeature will fail. + +PR_SET_SPECULATION_CTRL +----------------------- +PR_SET_SPECULATION_CTRL allows to control the speculation misfeature, which +is selected by arg2 of :manpage:`prctl(2)` per task. arg3 is used to hand +in the control value, i.e. either PR_SPEC_ENABLE or PR_SPEC_DISABLE. + +Common error codes +------------------ +======= ================================================================= +Value Meaning +======= ================================================================= +EINVAL The prctl is not implemented by the architecture or unused + prctl(2) arguments are not 0 + +ENODEV arg2 is selecting a not supported speculation misfeature +======= ================================================================= + +PR_SET_SPECULATION_CTRL error codes +----------------------------------- +======= ================================================================= +Value Meaning +======= ================================================================= +0 Success + +ERANGE arg3 is incorrect, i.e. it's neither PR_SPEC_ENABLE nor + PR_SPEC_DISABLE + +ENXIO Control of the selected speculation misfeature is not possible. + See PR_GET_SPECULATION_CTRL. +======= ================================================================= + +Speculation misfeature controls +------------------------------- +- PR_SPEC_STORE_BYPASS: Speculative Store Bypass + + Invocations: + * prctl(PR_GET_SPECULATION_CTRL, PR_SPEC_STORE_BYPASS, 0, 0, 0); + * prctl(PR_SET_SPECULATION_CTRL, PR_SPEC_STORE_BYPASS, PR_SPEC_ENABLE, 0, 0); + * prctl(PR_SET_SPECULATION_CTRL, PR_SPEC_STORE_BYPASS, PR_SPEC_DISABLE, 0, 0); diff --git a/include/linux/nospec.h b/include/linux/nospec.h index e791ebc65c9c..700bb8a4e4ea 100644 --- a/include/linux/nospec.h +++ b/include/linux/nospec.h @@ -55,4 +55,9 @@ static inline unsigned long array_index_mask_nospec(unsigned long index, \ (typeof(_i)) (_i & _mask); \ }) + +/* Speculation control prctl */ +int arch_prctl_spec_ctrl_get(unsigned long which); +int arch_prctl_spec_ctrl_set(unsigned long which, unsigned long ctrl); + #endif /* _LINUX_NOSPEC_H */ diff --git a/include/uapi/linux/prctl.h b/include/uapi/linux/prctl.h index af5f8c2df87a..ebf057ac1346 100644 --- a/include/uapi/linux/prctl.h +++ b/include/uapi/linux/prctl.h @@ -207,4 +207,15 @@ struct prctl_mm_map { # define PR_SVE_VL_LEN_MASK 0xffff # define PR_SVE_VL_INHERIT (1 << 17) /* inherit across exec */ +/* Per task speculation control */ +#define PR_GET_SPECULATION_CTRL 52 +#define PR_SET_SPECULATION_CTRL 53 +/* Speculation control variants */ +# define PR_SPEC_STORE_BYPASS 0 +/* Return and control values for PR_SET/GET_SPECULATION_CTRL */ +# define PR_SPEC_NOT_AFFECTED 0 +# define PR_SPEC_PRCTL (1UL << 0) +# define PR_SPEC_ENABLE (1UL << 1) +# define PR_SPEC_DISABLE (1UL << 2) + #endif /* _LINUX_PRCTL_H */ diff --git a/kernel/sys.c b/kernel/sys.c index ad692183dfe9..b76dee23bdc9 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -61,6 +61,8 @@ #include #include +#include + #include /* Move somewhere else to avoid recompiling? */ #include @@ -2242,6 +2244,16 @@ static int propagate_has_child_subreaper(struct task_struct *p, void *data) return 1; } +int __weak arch_prctl_spec_ctrl_get(unsigned long which) +{ + return -EINVAL; +} + +int __weak arch_prctl_spec_ctrl_set(unsigned long which, unsigned long ctrl) +{ + return -EINVAL; +} + SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3, unsigned long, arg4, unsigned long, arg5) { @@ -2450,6 +2462,16 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3, case PR_SVE_GET_VL: error = SVE_GET_VL(); break; + case PR_GET_SPECULATION_CTRL: + if (arg3 || arg4 || arg5) + return -EINVAL; + error = arch_prctl_spec_ctrl_get(arg2); + break; + case PR_SET_SPECULATION_CTRL: + if (arg4 || arg5) + return -EINVAL; + error = arch_prctl_spec_ctrl_set(arg2, arg3); + break; default: error = -EINVAL; break; From 885f82bfbc6fefb6664ea27965c3ab9ac4194b8c Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Sun, 29 Apr 2018 15:21:42 +0200 Subject: [PATCH 018/190] x86/process: Allow runtime control of Speculative Store Bypass The Speculative Store Bypass vulnerability can be mitigated with the Reduced Data Speculation (RDS) feature. To allow finer grained control of this eventually expensive mitigation a per task mitigation control is required. Add a new TIF_RDS flag and put it into the group of TIF flags which are evaluated for mismatch in switch_to(). If these bits differ in the previous and the next task, then the slow path function __switch_to_xtra() is invoked. Implement the TIF_RDS dependent mitigation control in the slow path. If the prctl for controlling Speculative Store Bypass is disabled or no task uses the prctl then there is no overhead in the switch_to() fast path. Update the KVM related speculation control functions to take TID_RDS into account as well. Based on a patch from Tim Chen. Completely rewritten. Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar Reviewed-by: Konrad Rzeszutek Wilk --- arch/x86/include/asm/msr-index.h | 3 ++- arch/x86/include/asm/spec-ctrl.h | 17 +++++++++++++++++ arch/x86/include/asm/thread_info.h | 4 +++- arch/x86/kernel/cpu/bugs.c | 26 +++++++++++++++++++++----- arch/x86/kernel/process.c | 22 ++++++++++++++++++++++ 5 files changed, 65 insertions(+), 7 deletions(-) diff --git a/arch/x86/include/asm/msr-index.h b/arch/x86/include/asm/msr-index.h index 21e1a6df9907..810f50bb338d 100644 --- a/arch/x86/include/asm/msr-index.h +++ b/arch/x86/include/asm/msr-index.h @@ -42,7 +42,8 @@ #define MSR_IA32_SPEC_CTRL 0x00000048 /* Speculation Control */ #define SPEC_CTRL_IBRS (1 << 0) /* Indirect Branch Restricted Speculation */ #define SPEC_CTRL_STIBP (1 << 1) /* Single Thread Indirect Branch Predictors */ -#define SPEC_CTRL_RDS (1 << 2) /* Reduced Data Speculation */ +#define SPEC_CTRL_RDS_SHIFT 2 /* Reduced Data Speculation bit */ +#define SPEC_CTRL_RDS (1 << SPEC_CTRL_RDS_SHIFT) /* Reduced Data Speculation */ #define MSR_IA32_PRED_CMD 0x00000049 /* Prediction Command */ #define PRED_CMD_IBPB (1 << 0) /* Indirect Branch Prediction Barrier */ diff --git a/arch/x86/include/asm/spec-ctrl.h b/arch/x86/include/asm/spec-ctrl.h index 3ad64420a06e..45ef00ad5105 100644 --- a/arch/x86/include/asm/spec-ctrl.h +++ b/arch/x86/include/asm/spec-ctrl.h @@ -2,6 +2,7 @@ #ifndef _ASM_X86_SPECCTRL_H_ #define _ASM_X86_SPECCTRL_H_ +#include #include /* @@ -18,4 +19,20 @@ extern void x86_spec_ctrl_restore_host(u64); extern u64 x86_amd_ls_cfg_base; extern u64 x86_amd_ls_cfg_rds_mask; +/* The Intel SPEC CTRL MSR base value cache */ +extern u64 x86_spec_ctrl_base; + +static inline u64 rds_tif_to_spec_ctrl(u64 tifn) +{ + BUILD_BUG_ON(TIF_RDS < SPEC_CTRL_RDS_SHIFT); + return (tifn & _TIF_RDS) >> (TIF_RDS - SPEC_CTRL_RDS_SHIFT); +} + +static inline u64 rds_tif_to_amd_ls_cfg(u64 tifn) +{ + return (tifn & _TIF_RDS) ? x86_amd_ls_cfg_rds_mask : 0ULL; +} + +extern void speculative_store_bypass_update(void); + #endif diff --git a/arch/x86/include/asm/thread_info.h b/arch/x86/include/asm/thread_info.h index a5d9521bb2cb..e5c26cc59619 100644 --- a/arch/x86/include/asm/thread_info.h +++ b/arch/x86/include/asm/thread_info.h @@ -79,6 +79,7 @@ struct thread_info { #define TIF_SIGPENDING 2 /* signal pending */ #define TIF_NEED_RESCHED 3 /* rescheduling necessary */ #define TIF_SINGLESTEP 4 /* reenable singlestep on user return*/ +#define TIF_RDS 5 /* Reduced data speculation */ #define TIF_SYSCALL_EMU 6 /* syscall emulation active */ #define TIF_SYSCALL_AUDIT 7 /* syscall auditing active */ #define TIF_SECCOMP 8 /* secure computing */ @@ -105,6 +106,7 @@ struct thread_info { #define _TIF_SIGPENDING (1 << TIF_SIGPENDING) #define _TIF_NEED_RESCHED (1 << TIF_NEED_RESCHED) #define _TIF_SINGLESTEP (1 << TIF_SINGLESTEP) +#define _TIF_RDS (1 << TIF_RDS) #define _TIF_SYSCALL_EMU (1 << TIF_SYSCALL_EMU) #define _TIF_SYSCALL_AUDIT (1 << TIF_SYSCALL_AUDIT) #define _TIF_SECCOMP (1 << TIF_SECCOMP) @@ -144,7 +146,7 @@ struct thread_info { /* flags to check in __switch_to() */ #define _TIF_WORK_CTXSW \ - (_TIF_IO_BITMAP|_TIF_NOCPUID|_TIF_NOTSC|_TIF_BLOCKSTEP) + (_TIF_IO_BITMAP|_TIF_NOCPUID|_TIF_NOTSC|_TIF_BLOCKSTEP|_TIF_RDS) #define _TIF_WORK_CTXSW_PREV (_TIF_WORK_CTXSW|_TIF_USER_RETURN_NOTIFY) #define _TIF_WORK_CTXSW_NEXT (_TIF_WORK_CTXSW) diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index ec171873167a..2bc109d0f8ae 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -33,7 +33,7 @@ static void __init ssb_select_mitigation(void); * Our boot-time value of the SPEC_CTRL MSR. We read it once so that any * writes to SPEC_CTRL contain whatever reserved bits have been set. */ -static u64 __ro_after_init x86_spec_ctrl_base; +u64 __ro_after_init x86_spec_ctrl_base; /* * The vendor and possibly platform specific bits which can be modified in @@ -140,25 +140,41 @@ EXPORT_SYMBOL_GPL(x86_spec_ctrl_set); u64 x86_spec_ctrl_get_default(void) { - return x86_spec_ctrl_base; + u64 msrval = x86_spec_ctrl_base; + + if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) + msrval |= rds_tif_to_spec_ctrl(current_thread_info()->flags); + return msrval; } EXPORT_SYMBOL_GPL(x86_spec_ctrl_get_default); void x86_spec_ctrl_set_guest(u64 guest_spec_ctrl) { + u64 host = x86_spec_ctrl_base; + if (!boot_cpu_has(X86_FEATURE_IBRS)) return; - if (x86_spec_ctrl_base != guest_spec_ctrl) + + if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) + host |= rds_tif_to_spec_ctrl(current_thread_info()->flags); + + if (host != guest_spec_ctrl) wrmsrl(MSR_IA32_SPEC_CTRL, guest_spec_ctrl); } EXPORT_SYMBOL_GPL(x86_spec_ctrl_set_guest); void x86_spec_ctrl_restore_host(u64 guest_spec_ctrl) { + u64 host = x86_spec_ctrl_base; + if (!boot_cpu_has(X86_FEATURE_IBRS)) return; - if (x86_spec_ctrl_base != guest_spec_ctrl) - wrmsrl(MSR_IA32_SPEC_CTRL, x86_spec_ctrl_base); + + if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) + host |= rds_tif_to_spec_ctrl(current_thread_info()->flags); + + if (host != guest_spec_ctrl) + wrmsrl(MSR_IA32_SPEC_CTRL, host); } EXPORT_SYMBOL_GPL(x86_spec_ctrl_restore_host); diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index 03408b942adb..397342725046 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -38,6 +38,7 @@ #include #include #include +#include /* * per-CPU TSS segments. Threads are completely 'soft' on Linux, @@ -278,6 +279,24 @@ static inline void switch_to_bitmap(struct tss_struct *tss, } } +static __always_inline void __speculative_store_bypass_update(unsigned long tifn) +{ + u64 msr; + + if (static_cpu_has(X86_FEATURE_AMD_RDS)) { + msr = x86_amd_ls_cfg_base | rds_tif_to_amd_ls_cfg(tifn); + wrmsrl(MSR_AMD64_LS_CFG, msr); + } else { + msr = x86_spec_ctrl_base | rds_tif_to_spec_ctrl(tifn); + wrmsrl(MSR_IA32_SPEC_CTRL, msr); + } +} + +void speculative_store_bypass_update(void) +{ + __speculative_store_bypass_update(current_thread_info()->flags); +} + void __switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p, struct tss_struct *tss) { @@ -309,6 +328,9 @@ void __switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p, if ((tifp ^ tifn) & _TIF_NOCPUID) set_cpuid_faulting(!!(tifn & _TIF_NOCPUID)); + + if ((tifp ^ tifn) & _TIF_RDS) + __speculative_store_bypass_update(tifn); } /* From a73ec77ee17ec556fe7f165d00314cb7c047b1ac Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Sun, 29 Apr 2018 15:26:40 +0200 Subject: [PATCH 019/190] x86/speculation: Add prctl for Speculative Store Bypass mitigation Add prctl based control for Speculative Store Bypass mitigation and make it the default mitigation for Intel and AMD. Andi Kleen provided the following rationale (slightly redacted): There are multiple levels of impact of Speculative Store Bypass: 1) JITed sandbox. It cannot invoke system calls, but can do PRIME+PROBE and may have call interfaces to other code 2) Native code process. No protection inside the process at this level. 3) Kernel. 4) Between processes. The prctl tries to protect against case (1) doing attacks. If the untrusted code can do random system calls then control is already lost in a much worse way. So there needs to be system call protection in some way (using a JIT not allowing them or seccomp). Or rather if the process can subvert its environment somehow to do the prctl it can already execute arbitrary code, which is much worse than SSB. To put it differently, the point of the prctl is to not allow JITed code to read data it shouldn't read from its JITed sandbox. If it already has escaped its sandbox then it can already read everything it wants in its address space, and do much worse. The ability to control Speculative Store Bypass allows to enable the protection selectively without affecting overall system performance. Based on an initial patch from Tim Chen. Completely rewritten. Signed-off-by: Thomas Gleixner Reviewed-by: Konrad Rzeszutek Wilk --- .../admin-guide/kernel-parameters.txt | 6 +- arch/x86/include/asm/nospec-branch.h | 1 + arch/x86/kernel/cpu/bugs.c | 83 ++++++++++++++++--- 3 files changed, 79 insertions(+), 11 deletions(-) diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt index f32112c55f7e..a8d2ae1e335b 100644 --- a/Documentation/admin-guide/kernel-parameters.txt +++ b/Documentation/admin-guide/kernel-parameters.txt @@ -4053,7 +4053,11 @@ off - Unconditionally enable Speculative Store Bypass auto - Kernel detects whether the CPU model contains an implementation of Speculative Store Bypass and - picks the most appropriate mitigation + picks the most appropriate mitigation. + prctl - Control Speculative Store Bypass per thread + via prctl. Speculative Store Bypass is enabled + for a process by default. The state of the control + is inherited on fork. Not specifying this option is equivalent to spec_store_bypass_disable=auto. diff --git a/arch/x86/include/asm/nospec-branch.h b/arch/x86/include/asm/nospec-branch.h index 1119f14bc883..71ad01422655 100644 --- a/arch/x86/include/asm/nospec-branch.h +++ b/arch/x86/include/asm/nospec-branch.h @@ -232,6 +232,7 @@ extern u64 x86_spec_ctrl_get_default(void); enum ssb_mitigation { SPEC_STORE_BYPASS_NONE, SPEC_STORE_BYPASS_DISABLE, + SPEC_STORE_BYPASS_PRCTL, }; extern char __indirect_thunk_start[]; diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index 2bc109d0f8ae..fc9187b6fae7 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -12,6 +12,8 @@ #include #include #include +#include +#include #include #include @@ -412,20 +414,23 @@ enum ssb_mitigation_cmd { SPEC_STORE_BYPASS_CMD_NONE, SPEC_STORE_BYPASS_CMD_AUTO, SPEC_STORE_BYPASS_CMD_ON, + SPEC_STORE_BYPASS_CMD_PRCTL, }; static const char *ssb_strings[] = { [SPEC_STORE_BYPASS_NONE] = "Vulnerable", - [SPEC_STORE_BYPASS_DISABLE] = "Mitigation: Speculative Store Bypass disabled" + [SPEC_STORE_BYPASS_DISABLE] = "Mitigation: Speculative Store Bypass disabled", + [SPEC_STORE_BYPASS_PRCTL] = "Mitigation: Speculative Store Bypass disabled via prctl" }; static const struct { const char *option; enum ssb_mitigation_cmd cmd; } ssb_mitigation_options[] = { - { "auto", SPEC_STORE_BYPASS_CMD_AUTO }, /* Platform decides */ - { "on", SPEC_STORE_BYPASS_CMD_ON }, /* Disable Speculative Store Bypass */ - { "off", SPEC_STORE_BYPASS_CMD_NONE }, /* Don't touch Speculative Store Bypass */ + { "auto", SPEC_STORE_BYPASS_CMD_AUTO }, /* Platform decides */ + { "on", SPEC_STORE_BYPASS_CMD_ON }, /* Disable Speculative Store Bypass */ + { "off", SPEC_STORE_BYPASS_CMD_NONE }, /* Don't touch Speculative Store Bypass */ + { "prctl", SPEC_STORE_BYPASS_CMD_PRCTL }, /* Disable Speculative Store Bypass via prctl */ }; static enum ssb_mitigation_cmd __init ssb_parse_cmdline(void) @@ -475,14 +480,15 @@ static enum ssb_mitigation_cmd __init __ssb_select_mitigation(void) switch (cmd) { case SPEC_STORE_BYPASS_CMD_AUTO: - /* - * AMD platforms by default don't need SSB mitigation. - */ - if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD) - break; + /* Choose prctl as the default mode */ + mode = SPEC_STORE_BYPASS_PRCTL; + break; case SPEC_STORE_BYPASS_CMD_ON: mode = SPEC_STORE_BYPASS_DISABLE; break; + case SPEC_STORE_BYPASS_CMD_PRCTL: + mode = SPEC_STORE_BYPASS_PRCTL; + break; case SPEC_STORE_BYPASS_CMD_NONE: break; } @@ -493,7 +499,7 @@ static enum ssb_mitigation_cmd __init __ssb_select_mitigation(void) * - X86_FEATURE_RDS - CPU is able to turn off speculative store bypass * - X86_FEATURE_SPEC_STORE_BYPASS_DISABLE - engage the mitigation */ - if (mode != SPEC_STORE_BYPASS_NONE) { + if (mode == SPEC_STORE_BYPASS_DISABLE) { setup_force_cpu_cap(X86_FEATURE_SPEC_STORE_BYPASS_DISABLE); /* * Intel uses the SPEC CTRL MSR Bit(2) for this, while AMD uses @@ -524,6 +530,63 @@ static void ssb_select_mitigation() #undef pr_fmt +static int ssb_prctl_set(unsigned long ctrl) +{ + bool rds = !!test_tsk_thread_flag(current, TIF_RDS); + + if (ssb_mode != SPEC_STORE_BYPASS_PRCTL) + return -ENXIO; + + if (ctrl == PR_SPEC_ENABLE) + clear_tsk_thread_flag(current, TIF_RDS); + else + set_tsk_thread_flag(current, TIF_RDS); + + if (rds != !!test_tsk_thread_flag(current, TIF_RDS)) + speculative_store_bypass_update(); + + return 0; +} + +static int ssb_prctl_get(void) +{ + switch (ssb_mode) { + case SPEC_STORE_BYPASS_DISABLE: + return PR_SPEC_DISABLE; + case SPEC_STORE_BYPASS_PRCTL: + if (test_tsk_thread_flag(current, TIF_RDS)) + return PR_SPEC_PRCTL | PR_SPEC_DISABLE; + return PR_SPEC_PRCTL | PR_SPEC_ENABLE; + default: + if (boot_cpu_has_bug(X86_BUG_SPEC_STORE_BYPASS)) + return PR_SPEC_ENABLE; + return PR_SPEC_NOT_AFFECTED; + } +} + +int arch_prctl_spec_ctrl_set(unsigned long which, unsigned long ctrl) +{ + if (ctrl != PR_SPEC_ENABLE && ctrl != PR_SPEC_DISABLE) + return -ERANGE; + + switch (which) { + case PR_SPEC_STORE_BYPASS: + return ssb_prctl_set(ctrl); + default: + return -ENODEV; + } +} + +int arch_prctl_spec_ctrl_get(unsigned long which) +{ + switch (which) { + case PR_SPEC_STORE_BYPASS: + return ssb_prctl_get(); + default: + return -ENODEV; + } +} + void x86_spec_ctrl_setup_ap(void) { if (boot_cpu_has(X86_FEATURE_IBRS)) From 7bbf1373e228840bb0295a2ca26d548ef37f448e Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Tue, 1 May 2018 15:19:04 -0700 Subject: [PATCH 020/190] nospec: Allow getting/setting on non-current task Adjust arch_prctl_get/set_spec_ctrl() to operate on tasks other than current. This is needed both for /proc/$pid/status queries and for seccomp (since thread-syncing can trigger seccomp in non-current threads). Signed-off-by: Kees Cook Signed-off-by: Thomas Gleixner --- arch/x86/kernel/cpu/bugs.c | 27 ++++++++++++++++----------- include/linux/nospec.h | 7 +++++-- kernel/sys.c | 9 +++++---- 3 files changed, 26 insertions(+), 17 deletions(-) diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index fc9187b6fae7..e3afb610f2ad 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -530,31 +530,35 @@ static void ssb_select_mitigation() #undef pr_fmt -static int ssb_prctl_set(unsigned long ctrl) +static int ssb_prctl_set(struct task_struct *task, unsigned long ctrl) { - bool rds = !!test_tsk_thread_flag(current, TIF_RDS); + bool rds = !!test_tsk_thread_flag(task, TIF_RDS); if (ssb_mode != SPEC_STORE_BYPASS_PRCTL) return -ENXIO; if (ctrl == PR_SPEC_ENABLE) - clear_tsk_thread_flag(current, TIF_RDS); + clear_tsk_thread_flag(task, TIF_RDS); else - set_tsk_thread_flag(current, TIF_RDS); + set_tsk_thread_flag(task, TIF_RDS); - if (rds != !!test_tsk_thread_flag(current, TIF_RDS)) + /* + * If being set on non-current task, delay setting the CPU + * mitigation until it is next scheduled. + */ + if (task == current && rds != !!test_tsk_thread_flag(task, TIF_RDS)) speculative_store_bypass_update(); return 0; } -static int ssb_prctl_get(void) +static int ssb_prctl_get(struct task_struct *task) { switch (ssb_mode) { case SPEC_STORE_BYPASS_DISABLE: return PR_SPEC_DISABLE; case SPEC_STORE_BYPASS_PRCTL: - if (test_tsk_thread_flag(current, TIF_RDS)) + if (test_tsk_thread_flag(task, TIF_RDS)) return PR_SPEC_PRCTL | PR_SPEC_DISABLE; return PR_SPEC_PRCTL | PR_SPEC_ENABLE; default: @@ -564,24 +568,25 @@ static int ssb_prctl_get(void) } } -int arch_prctl_spec_ctrl_set(unsigned long which, unsigned long ctrl) +int arch_prctl_spec_ctrl_set(struct task_struct *task, unsigned long which, + unsigned long ctrl) { if (ctrl != PR_SPEC_ENABLE && ctrl != PR_SPEC_DISABLE) return -ERANGE; switch (which) { case PR_SPEC_STORE_BYPASS: - return ssb_prctl_set(ctrl); + return ssb_prctl_set(task, ctrl); default: return -ENODEV; } } -int arch_prctl_spec_ctrl_get(unsigned long which) +int arch_prctl_spec_ctrl_get(struct task_struct *task, unsigned long which) { switch (which) { case PR_SPEC_STORE_BYPASS: - return ssb_prctl_get(); + return ssb_prctl_get(task); default: return -ENODEV; } diff --git a/include/linux/nospec.h b/include/linux/nospec.h index 700bb8a4e4ea..a908c954484d 100644 --- a/include/linux/nospec.h +++ b/include/linux/nospec.h @@ -7,6 +7,8 @@ #define _LINUX_NOSPEC_H #include +struct task_struct; + /** * array_index_mask_nospec() - generate a ~0 mask when index < size, 0 otherwise * @index: array element index @@ -57,7 +59,8 @@ static inline unsigned long array_index_mask_nospec(unsigned long index, }) /* Speculation control prctl */ -int arch_prctl_spec_ctrl_get(unsigned long which); -int arch_prctl_spec_ctrl_set(unsigned long which, unsigned long ctrl); +int arch_prctl_spec_ctrl_get(struct task_struct *task, unsigned long which); +int arch_prctl_spec_ctrl_set(struct task_struct *task, unsigned long which, + unsigned long ctrl); #endif /* _LINUX_NOSPEC_H */ diff --git a/kernel/sys.c b/kernel/sys.c index b76dee23bdc9..b0eee418ee0d 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -2244,12 +2244,13 @@ static int propagate_has_child_subreaper(struct task_struct *p, void *data) return 1; } -int __weak arch_prctl_spec_ctrl_get(unsigned long which) +int __weak arch_prctl_spec_ctrl_get(struct task_struct *t, unsigned long which) { return -EINVAL; } -int __weak arch_prctl_spec_ctrl_set(unsigned long which, unsigned long ctrl) +int __weak arch_prctl_spec_ctrl_set(struct task_struct *t, unsigned long which, + unsigned long ctrl) { return -EINVAL; } @@ -2465,12 +2466,12 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3, case PR_GET_SPECULATION_CTRL: if (arg3 || arg4 || arg5) return -EINVAL; - error = arch_prctl_spec_ctrl_get(arg2); + error = arch_prctl_spec_ctrl_get(me, arg2); break; case PR_SET_SPECULATION_CTRL: if (arg4 || arg5) return -EINVAL; - error = arch_prctl_spec_ctrl_set(arg2, arg3); + error = arch_prctl_spec_ctrl_set(me, arg2, arg3); break; default: error = -EINVAL; From fae1fa0fc6cca8beee3ab8ed71d54f9a78fa3f64 Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Tue, 1 May 2018 15:31:45 -0700 Subject: [PATCH 021/190] proc: Provide details on speculation flaw mitigations As done with seccomp and no_new_privs, also show speculation flaw mitigation state in /proc/$pid/status. Signed-off-by: Kees Cook Signed-off-by: Thomas Gleixner --- fs/proc/array.c | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+) diff --git a/fs/proc/array.c b/fs/proc/array.c index ae2c807fd719..303c155f9b04 100644 --- a/fs/proc/array.c +++ b/fs/proc/array.c @@ -85,6 +85,7 @@ #include #include #include +#include #include #include #include @@ -335,6 +336,27 @@ static inline void task_seccomp(struct seq_file *m, struct task_struct *p) #ifdef CONFIG_SECCOMP seq_put_decimal_ull(m, "\nSeccomp:\t", p->seccomp.mode); #endif + seq_printf(m, "\nSpeculation Store Bypass:\t"); + switch (arch_prctl_spec_ctrl_get(p, PR_SPEC_STORE_BYPASS)) { + case -EINVAL: + seq_printf(m, "unknown"); + break; + case PR_SPEC_NOT_AFFECTED: + seq_printf(m, "not vulnerable"); + break; + case PR_SPEC_PRCTL | PR_SPEC_DISABLE: + seq_printf(m, "thread mitigated"); + break; + case PR_SPEC_PRCTL | PR_SPEC_ENABLE: + seq_printf(m, "thread vulnerable"); + break; + case PR_SPEC_DISABLE: + seq_printf(m, "globally mitigated"); + break; + default: + seq_printf(m, "vulnerable"); + break; + } seq_putc(m, '\n'); } From 5c3070890d06ff82eecb808d02d2ca39169533ef Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Tue, 1 May 2018 15:07:31 -0700 Subject: [PATCH 022/190] seccomp: Enable speculation flaw mitigations When speculation flaw mitigations are opt-in (via prctl), using seccomp will automatically opt-in to these protections, since using seccomp indicates at least some level of sandboxing is desired. Signed-off-by: Kees Cook Signed-off-by: Thomas Gleixner --- kernel/seccomp.c | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) diff --git a/kernel/seccomp.c b/kernel/seccomp.c index dc77548167ef..9f34533046aa 100644 --- a/kernel/seccomp.c +++ b/kernel/seccomp.c @@ -19,6 +19,8 @@ #include #include #include +#include +#include #include #include #include @@ -227,6 +229,19 @@ static inline bool seccomp_may_assign_mode(unsigned long seccomp_mode) return true; } +/* + * If a given speculation mitigation is opt-in (prctl()-controlled), + * select it, by disabling speculation (enabling mitigation). + */ +static inline void spec_mitigate(struct task_struct *task, + unsigned long which) +{ + int state = arch_prctl_spec_ctrl_get(task, which); + + if (state > 0 && (state & PR_SPEC_PRCTL)) + arch_prctl_spec_ctrl_set(task, which, PR_SPEC_DISABLE); +} + static inline void seccomp_assign_mode(struct task_struct *task, unsigned long seccomp_mode) { @@ -238,6 +253,8 @@ static inline void seccomp_assign_mode(struct task_struct *task, * filter) is set. */ smp_mb__before_atomic(); + /* Assume seccomp processes want speculation flaw mitigation. */ + spec_mitigate(task, PR_SPEC_STORE_BYPASS); set_tsk_thread_flag(task, TIF_SECCOMP); } From f9544b2b076ca90d887c5ae5d74fab4c21bb7c13 Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Thu, 3 May 2018 15:03:30 -0700 Subject: [PATCH 023/190] x86/bugs: Make boot modes __ro_after_init There's no reason for these to be changed after boot. Signed-off-by: Kees Cook Signed-off-by: Thomas Gleixner --- arch/x86/kernel/cpu/bugs.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index e3afb610f2ad..f8d9be0e86b1 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -129,7 +129,8 @@ static const char *spectre_v2_strings[] = { #undef pr_fmt #define pr_fmt(fmt) "Spectre V2 : " fmt -static enum spectre_v2_mitigation spectre_v2_enabled = SPECTRE_V2_NONE; +static enum spectre_v2_mitigation spectre_v2_enabled __ro_after_init = + SPECTRE_V2_NONE; void x86_spec_ctrl_set(u64 val) { @@ -407,7 +408,7 @@ static void __init spectre_v2_select_mitigation(void) #undef pr_fmt #define pr_fmt(fmt) "Speculative Store Bypass: " fmt -static enum ssb_mitigation ssb_mode = SPEC_STORE_BYPASS_NONE; +static enum ssb_mitigation ssb_mode __ro_after_init = SPEC_STORE_BYPASS_NONE; /* The kernel command line selection */ enum ssb_mitigation_cmd { From 356e4bfff2c5489e016fdb925adbf12a1e3950ee Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Thu, 3 May 2018 22:09:15 +0200 Subject: [PATCH 024/190] prctl: Add force disable speculation For certain use cases it is desired to enforce mitigations so they cannot be undone afterwards. That's important for loader stubs which want to prevent a child from disabling the mitigation again. Will also be used for seccomp(). The extra state preserving of the prctl state for SSB is a preparatory step for EBPF dymanic speculation control. Signed-off-by: Thomas Gleixner --- Documentation/userspace-api/spec_ctrl.rst | 34 +++++++++++++--------- arch/x86/kernel/cpu/bugs.c | 35 ++++++++++++++++------- fs/proc/array.c | 3 ++ include/linux/sched.h | 10 ++++++- include/uapi/linux/prctl.h | 1 + 5 files changed, 59 insertions(+), 24 deletions(-) diff --git a/Documentation/userspace-api/spec_ctrl.rst b/Documentation/userspace-api/spec_ctrl.rst index ddbebcd01208..1b3690d30943 100644 --- a/Documentation/userspace-api/spec_ctrl.rst +++ b/Documentation/userspace-api/spec_ctrl.rst @@ -25,19 +25,21 @@ PR_GET_SPECULATION_CTRL ----------------------- PR_GET_SPECULATION_CTRL returns the state of the speculation misfeature -which is selected with arg2 of prctl(2). The return value uses bits 0-2 with +which is selected with arg2 of prctl(2). The return value uses bits 0-3 with the following meaning: -==== ================ =================================================== -Bit Define Description -==== ================ =================================================== -0 PR_SPEC_PRCTL Mitigation can be controlled per task by - PR_SET_SPECULATION_CTRL -1 PR_SPEC_ENABLE The speculation feature is enabled, mitigation is - disabled -2 PR_SPEC_DISABLE The speculation feature is disabled, mitigation is - enabled -==== ================ =================================================== +==== ===================== =================================================== +Bit Define Description +==== ===================== =================================================== +0 PR_SPEC_PRCTL Mitigation can be controlled per task by + PR_SET_SPECULATION_CTRL +1 PR_SPEC_ENABLE The speculation feature is enabled, mitigation is + disabled +2 PR_SPEC_DISABLE The speculation feature is disabled, mitigation is + enabled +3 PR_SPEC_FORCE_DISABLE Same as PR_SPEC_DISABLE, but cannot be undone. A + subsequent prctl(..., PR_SPEC_ENABLE) will fail. +==== ===================== =================================================== If all bits are 0 the CPU is not affected by the speculation misfeature. @@ -47,9 +49,11 @@ misfeature will fail. PR_SET_SPECULATION_CTRL ----------------------- + PR_SET_SPECULATION_CTRL allows to control the speculation misfeature, which is selected by arg2 of :manpage:`prctl(2)` per task. arg3 is used to hand -in the control value, i.e. either PR_SPEC_ENABLE or PR_SPEC_DISABLE. +in the control value, i.e. either PR_SPEC_ENABLE or PR_SPEC_DISABLE or +PR_SPEC_FORCE_DISABLE. Common error codes ------------------ @@ -70,10 +74,13 @@ Value Meaning 0 Success ERANGE arg3 is incorrect, i.e. it's neither PR_SPEC_ENABLE nor - PR_SPEC_DISABLE + PR_SPEC_DISABLE nor PR_SPEC_FORCE_DISABLE ENXIO Control of the selected speculation misfeature is not possible. See PR_GET_SPECULATION_CTRL. + +EPERM Speculation was disabled with PR_SPEC_FORCE_DISABLE and caller + tried to enable it again. ======= ================================================================= Speculation misfeature controls @@ -84,3 +91,4 @@ Speculation misfeature controls * prctl(PR_GET_SPECULATION_CTRL, PR_SPEC_STORE_BYPASS, 0, 0, 0); * prctl(PR_SET_SPECULATION_CTRL, PR_SPEC_STORE_BYPASS, PR_SPEC_ENABLE, 0, 0); * prctl(PR_SET_SPECULATION_CTRL, PR_SPEC_STORE_BYPASS, PR_SPEC_DISABLE, 0, 0); + * prctl(PR_SET_SPECULATION_CTRL, PR_SPEC_STORE_BYPASS, PR_SPEC_FORCE_DISABLE, 0, 0); diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index f8d9be0e86b1..7e0f28160e5e 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -533,21 +533,37 @@ static void ssb_select_mitigation() static int ssb_prctl_set(struct task_struct *task, unsigned long ctrl) { - bool rds = !!test_tsk_thread_flag(task, TIF_RDS); + bool update; if (ssb_mode != SPEC_STORE_BYPASS_PRCTL) return -ENXIO; - if (ctrl == PR_SPEC_ENABLE) - clear_tsk_thread_flag(task, TIF_RDS); - else - set_tsk_thread_flag(task, TIF_RDS); + switch (ctrl) { + case PR_SPEC_ENABLE: + /* If speculation is force disabled, enable is not allowed */ + if (task_spec_ssb_force_disable(task)) + return -EPERM; + task_clear_spec_ssb_disable(task); + update = test_and_clear_tsk_thread_flag(task, TIF_RDS); + break; + case PR_SPEC_DISABLE: + task_set_spec_ssb_disable(task); + update = !test_and_set_tsk_thread_flag(task, TIF_RDS); + break; + case PR_SPEC_FORCE_DISABLE: + task_set_spec_ssb_disable(task); + task_set_spec_ssb_force_disable(task); + update = !test_and_set_tsk_thread_flag(task, TIF_RDS); + break; + default: + return -ERANGE; + } /* * If being set on non-current task, delay setting the CPU * mitigation until it is next scheduled. */ - if (task == current && rds != !!test_tsk_thread_flag(task, TIF_RDS)) + if (task == current && update) speculative_store_bypass_update(); return 0; @@ -559,7 +575,9 @@ static int ssb_prctl_get(struct task_struct *task) case SPEC_STORE_BYPASS_DISABLE: return PR_SPEC_DISABLE; case SPEC_STORE_BYPASS_PRCTL: - if (test_tsk_thread_flag(task, TIF_RDS)) + if (task_spec_ssb_force_disable(task)) + return PR_SPEC_PRCTL | PR_SPEC_FORCE_DISABLE; + if (task_spec_ssb_disable(task)) return PR_SPEC_PRCTL | PR_SPEC_DISABLE; return PR_SPEC_PRCTL | PR_SPEC_ENABLE; default: @@ -572,9 +590,6 @@ static int ssb_prctl_get(struct task_struct *task) int arch_prctl_spec_ctrl_set(struct task_struct *task, unsigned long which, unsigned long ctrl) { - if (ctrl != PR_SPEC_ENABLE && ctrl != PR_SPEC_DISABLE) - return -ERANGE; - switch (which) { case PR_SPEC_STORE_BYPASS: return ssb_prctl_set(task, ctrl); diff --git a/fs/proc/array.c b/fs/proc/array.c index 303c155f9b04..d178a0236514 100644 --- a/fs/proc/array.c +++ b/fs/proc/array.c @@ -344,6 +344,9 @@ static inline void task_seccomp(struct seq_file *m, struct task_struct *p) case PR_SPEC_NOT_AFFECTED: seq_printf(m, "not vulnerable"); break; + case PR_SPEC_PRCTL | PR_SPEC_FORCE_DISABLE: + seq_printf(m, "thread force mitigated"); + break; case PR_SPEC_PRCTL | PR_SPEC_DISABLE: seq_printf(m, "thread mitigated"); break; diff --git a/include/linux/sched.h b/include/linux/sched.h index b3d697f3b573..e4218d4deba0 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1393,7 +1393,8 @@ static inline bool is_percpu_thread(void) #define PFA_NO_NEW_PRIVS 0 /* May not gain new privileges. */ #define PFA_SPREAD_PAGE 1 /* Spread page cache over cpuset */ #define PFA_SPREAD_SLAB 2 /* Spread some slab caches over cpuset */ - +#define PFA_SPEC_SSB_DISABLE 3 /* Speculative Store Bypass disabled */ +#define PFA_SPEC_SSB_FORCE_DISABLE 4 /* Speculative Store Bypass force disabled*/ #define TASK_PFA_TEST(name, func) \ static inline bool task_##func(struct task_struct *p) \ @@ -1418,6 +1419,13 @@ TASK_PFA_TEST(SPREAD_SLAB, spread_slab) TASK_PFA_SET(SPREAD_SLAB, spread_slab) TASK_PFA_CLEAR(SPREAD_SLAB, spread_slab) +TASK_PFA_TEST(SPEC_SSB_DISABLE, spec_ssb_disable) +TASK_PFA_SET(SPEC_SSB_DISABLE, spec_ssb_disable) +TASK_PFA_CLEAR(SPEC_SSB_DISABLE, spec_ssb_disable) + +TASK_PFA_TEST(SPEC_SSB_FORCE_DISABLE, spec_ssb_force_disable) +TASK_PFA_SET(SPEC_SSB_FORCE_DISABLE, spec_ssb_force_disable) + static inline void current_restore_flags(unsigned long orig_flags, unsigned long flags) { diff --git a/include/uapi/linux/prctl.h b/include/uapi/linux/prctl.h index ebf057ac1346..db9f15f5db04 100644 --- a/include/uapi/linux/prctl.h +++ b/include/uapi/linux/prctl.h @@ -217,5 +217,6 @@ struct prctl_mm_map { # define PR_SPEC_PRCTL (1UL << 0) # define PR_SPEC_ENABLE (1UL << 1) # define PR_SPEC_DISABLE (1UL << 2) +# define PR_SPEC_FORCE_DISABLE (1UL << 3) #endif /* _LINUX_PRCTL_H */ From b849a812f7eb92e96d1c8239b06581b2cfd8b275 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Fri, 4 May 2018 09:40:03 +0200 Subject: [PATCH 025/190] seccomp: Use PR_SPEC_FORCE_DISABLE Use PR_SPEC_FORCE_DISABLE in seccomp() because seccomp does not allow to widen restrictions. Signed-off-by: Thomas Gleixner --- kernel/seccomp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/seccomp.c b/kernel/seccomp.c index 9f34533046aa..2c819d65e15f 100644 --- a/kernel/seccomp.c +++ b/kernel/seccomp.c @@ -239,7 +239,7 @@ static inline void spec_mitigate(struct task_struct *task, int state = arch_prctl_spec_ctrl_get(task, which); if (state > 0 && (state & PR_SPEC_PRCTL)) - arch_prctl_spec_ctrl_set(task, which, PR_SPEC_DISABLE); + arch_prctl_spec_ctrl_set(task, which, PR_SPEC_FORCE_DISABLE); } static inline void seccomp_assign_mode(struct task_struct *task, From 00a02d0c502a06d15e07b857f8ff921e3e402675 Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Thu, 3 May 2018 14:56:12 -0700 Subject: [PATCH 026/190] seccomp: Add filter flag to opt-out of SSB mitigation If a seccomp user is not interested in Speculative Store Bypass mitigation by default, it can set the new SECCOMP_FILTER_FLAG_SPEC_ALLOW flag when adding filters. Signed-off-by: Kees Cook Signed-off-by: Thomas Gleixner --- include/linux/seccomp.h | 5 +++-- include/uapi/linux/seccomp.h | 5 +++-- kernel/seccomp.c | 19 +++++++++------- tools/testing/selftests/seccomp/seccomp_bpf.c | 22 ++++++++++++++++--- 4 files changed, 36 insertions(+), 15 deletions(-) diff --git a/include/linux/seccomp.h b/include/linux/seccomp.h index c723a5c4e3ff..e5320f6c8654 100644 --- a/include/linux/seccomp.h +++ b/include/linux/seccomp.h @@ -4,8 +4,9 @@ #include -#define SECCOMP_FILTER_FLAG_MASK (SECCOMP_FILTER_FLAG_TSYNC | \ - SECCOMP_FILTER_FLAG_LOG) +#define SECCOMP_FILTER_FLAG_MASK (SECCOMP_FILTER_FLAG_TSYNC | \ + SECCOMP_FILTER_FLAG_LOG | \ + SECCOMP_FILTER_FLAG_SPEC_ALLOW) #ifdef CONFIG_SECCOMP diff --git a/include/uapi/linux/seccomp.h b/include/uapi/linux/seccomp.h index 2a0bd9dd104d..9efc0e73d50b 100644 --- a/include/uapi/linux/seccomp.h +++ b/include/uapi/linux/seccomp.h @@ -17,8 +17,9 @@ #define SECCOMP_GET_ACTION_AVAIL 2 /* Valid flags for SECCOMP_SET_MODE_FILTER */ -#define SECCOMP_FILTER_FLAG_TSYNC 1 -#define SECCOMP_FILTER_FLAG_LOG 2 +#define SECCOMP_FILTER_FLAG_TSYNC (1UL << 0) +#define SECCOMP_FILTER_FLAG_LOG (1UL << 1) +#define SECCOMP_FILTER_FLAG_SPEC_ALLOW (1UL << 2) /* * All BPF programs must return a 32-bit value. diff --git a/kernel/seccomp.c b/kernel/seccomp.c index 2c819d65e15f..53eb946120c1 100644 --- a/kernel/seccomp.c +++ b/kernel/seccomp.c @@ -243,7 +243,8 @@ static inline void spec_mitigate(struct task_struct *task, } static inline void seccomp_assign_mode(struct task_struct *task, - unsigned long seccomp_mode) + unsigned long seccomp_mode, + unsigned long flags) { assert_spin_locked(&task->sighand->siglock); @@ -253,8 +254,9 @@ static inline void seccomp_assign_mode(struct task_struct *task, * filter) is set. */ smp_mb__before_atomic(); - /* Assume seccomp processes want speculation flaw mitigation. */ - spec_mitigate(task, PR_SPEC_STORE_BYPASS); + /* Assume default seccomp processes want spec flaw mitigation. */ + if ((flags & SECCOMP_FILTER_FLAG_SPEC_ALLOW) == 0) + spec_mitigate(task, PR_SPEC_STORE_BYPASS); set_tsk_thread_flag(task, TIF_SECCOMP); } @@ -322,7 +324,7 @@ static inline pid_t seccomp_can_sync_threads(void) * without dropping the locks. * */ -static inline void seccomp_sync_threads(void) +static inline void seccomp_sync_threads(unsigned long flags) { struct task_struct *thread, *caller; @@ -363,7 +365,8 @@ static inline void seccomp_sync_threads(void) * allow one thread to transition the other. */ if (thread->seccomp.mode == SECCOMP_MODE_DISABLED) - seccomp_assign_mode(thread, SECCOMP_MODE_FILTER); + seccomp_assign_mode(thread, SECCOMP_MODE_FILTER, + flags); } } @@ -486,7 +489,7 @@ static long seccomp_attach_filter(unsigned int flags, /* Now that the new filter is in place, synchronize to all threads. */ if (flags & SECCOMP_FILTER_FLAG_TSYNC) - seccomp_sync_threads(); + seccomp_sync_threads(flags); return 0; } @@ -835,7 +838,7 @@ static long seccomp_set_mode_strict(void) #ifdef TIF_NOTSC disable_TSC(); #endif - seccomp_assign_mode(current, seccomp_mode); + seccomp_assign_mode(current, seccomp_mode, 0); ret = 0; out: @@ -893,7 +896,7 @@ static long seccomp_set_mode_filter(unsigned int flags, /* Do not free the successfully attached filter. */ prepared = NULL; - seccomp_assign_mode(current, seccomp_mode); + seccomp_assign_mode(current, seccomp_mode, flags); out: spin_unlock_irq(¤t->sighand->siglock); if (flags & SECCOMP_FILTER_FLAG_TSYNC) diff --git a/tools/testing/selftests/seccomp/seccomp_bpf.c b/tools/testing/selftests/seccomp/seccomp_bpf.c index 168c66d74fc5..e1473234968d 100644 --- a/tools/testing/selftests/seccomp/seccomp_bpf.c +++ b/tools/testing/selftests/seccomp/seccomp_bpf.c @@ -134,11 +134,15 @@ struct seccomp_data { #endif #ifndef SECCOMP_FILTER_FLAG_TSYNC -#define SECCOMP_FILTER_FLAG_TSYNC 1 +#define SECCOMP_FILTER_FLAG_TSYNC (1UL << 0) #endif #ifndef SECCOMP_FILTER_FLAG_LOG -#define SECCOMP_FILTER_FLAG_LOG 2 +#define SECCOMP_FILTER_FLAG_LOG (1UL << 1) +#endif + +#ifndef SECCOMP_FILTER_FLAG_SPEC_ALLOW +#define SECCOMP_FILTER_FLAG_SPEC_ALLOW (1UL << 2) #endif #ifndef PTRACE_SECCOMP_GET_METADATA @@ -2072,14 +2076,26 @@ TEST(seccomp_syscall_mode_lock) TEST(detect_seccomp_filter_flags) { unsigned int flags[] = { SECCOMP_FILTER_FLAG_TSYNC, - SECCOMP_FILTER_FLAG_LOG }; + SECCOMP_FILTER_FLAG_LOG, + SECCOMP_FILTER_FLAG_SPEC_ALLOW }; unsigned int flag, all_flags; int i; long ret; /* Test detection of known-good filter flags */ for (i = 0, all_flags = 0; i < ARRAY_SIZE(flags); i++) { + int bits = 0; + flag = flags[i]; + /* Make sure the flag is a single bit! */ + while (flag) { + if (flag & 0x1) + bits ++; + flag >>= 1; + } + ASSERT_EQ(1, bits); + flag = flags[i]; + ret = seccomp(SECCOMP_SET_MODE_FILTER, flag, NULL); ASSERT_NE(ENOSYS, errno) { TH_LOG("Kernel does not support seccomp syscall!"); From 8bf37d8c067bb7eb8e7c381bdadf9bd89182b6bc Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Fri, 4 May 2018 15:12:06 +0200 Subject: [PATCH 027/190] seccomp: Move speculation migitation control to arch code The migitation control is simpler to implement in architecture code as it avoids the extra function call to check the mode. Aside of that having an explicit seccomp enabled mode in the architecture mitigations would require even more workarounds. Move it into architecture code and provide a weak function in the seccomp code. Remove the 'which' argument as this allows the architecture to decide which mitigations are relevant for seccomp. Signed-off-by: Thomas Gleixner --- arch/x86/kernel/cpu/bugs.c | 29 ++++++++++++++++++----------- include/linux/nospec.h | 2 ++ kernel/seccomp.c | 15 ++------------- 3 files changed, 22 insertions(+), 24 deletions(-) diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index 7e0f28160e5e..5dab4c3d26e7 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -569,6 +569,24 @@ static int ssb_prctl_set(struct task_struct *task, unsigned long ctrl) return 0; } +int arch_prctl_spec_ctrl_set(struct task_struct *task, unsigned long which, + unsigned long ctrl) +{ + switch (which) { + case PR_SPEC_STORE_BYPASS: + return ssb_prctl_set(task, ctrl); + default: + return -ENODEV; + } +} + +#ifdef CONFIG_SECCOMP +void arch_seccomp_spec_mitigate(struct task_struct *task) +{ + ssb_prctl_set(task, PR_SPEC_FORCE_DISABLE); +} +#endif + static int ssb_prctl_get(struct task_struct *task) { switch (ssb_mode) { @@ -587,17 +605,6 @@ static int ssb_prctl_get(struct task_struct *task) } } -int arch_prctl_spec_ctrl_set(struct task_struct *task, unsigned long which, - unsigned long ctrl) -{ - switch (which) { - case PR_SPEC_STORE_BYPASS: - return ssb_prctl_set(task, ctrl); - default: - return -ENODEV; - } -} - int arch_prctl_spec_ctrl_get(struct task_struct *task, unsigned long which) { switch (which) { diff --git a/include/linux/nospec.h b/include/linux/nospec.h index a908c954484d..0c5ef54fd416 100644 --- a/include/linux/nospec.h +++ b/include/linux/nospec.h @@ -62,5 +62,7 @@ static inline unsigned long array_index_mask_nospec(unsigned long index, int arch_prctl_spec_ctrl_get(struct task_struct *task, unsigned long which); int arch_prctl_spec_ctrl_set(struct task_struct *task, unsigned long which, unsigned long ctrl); +/* Speculation control for seccomp enforced mitigation */ +void arch_seccomp_spec_mitigate(struct task_struct *task); #endif /* _LINUX_NOSPEC_H */ diff --git a/kernel/seccomp.c b/kernel/seccomp.c index 53eb946120c1..e691d9a6c58d 100644 --- a/kernel/seccomp.c +++ b/kernel/seccomp.c @@ -229,18 +229,7 @@ static inline bool seccomp_may_assign_mode(unsigned long seccomp_mode) return true; } -/* - * If a given speculation mitigation is opt-in (prctl()-controlled), - * select it, by disabling speculation (enabling mitigation). - */ -static inline void spec_mitigate(struct task_struct *task, - unsigned long which) -{ - int state = arch_prctl_spec_ctrl_get(task, which); - - if (state > 0 && (state & PR_SPEC_PRCTL)) - arch_prctl_spec_ctrl_set(task, which, PR_SPEC_FORCE_DISABLE); -} +void __weak arch_seccomp_spec_mitigate(struct task_struct *task) { } static inline void seccomp_assign_mode(struct task_struct *task, unsigned long seccomp_mode, @@ -256,7 +245,7 @@ static inline void seccomp_assign_mode(struct task_struct *task, smp_mb__before_atomic(); /* Assume default seccomp processes want spec flaw mitigation. */ if ((flags & SECCOMP_FILTER_FLAG_SPEC_ALLOW) == 0) - spec_mitigate(task, PR_SPEC_STORE_BYPASS); + arch_seccomp_spec_mitigate(task); set_tsk_thread_flag(task, TIF_SECCOMP); } From f21b53b20c754021935ea43364dbf53778eeba32 Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Thu, 3 May 2018 14:37:54 -0700 Subject: [PATCH 028/190] x86/speculation: Make "seccomp" the default mode for Speculative Store Bypass Unless explicitly opted out of, anything running under seccomp will have SSB mitigations enabled. Choosing the "prctl" mode will disable this. [ tglx: Adjusted it to the new arch_seccomp_spec_mitigate() mechanism ] Signed-off-by: Kees Cook Signed-off-by: Thomas Gleixner --- .../admin-guide/kernel-parameters.txt | 26 +++++++++------ arch/x86/include/asm/nospec-branch.h | 1 + arch/x86/kernel/cpu/bugs.c | 32 +++++++++++++------ 3 files changed, 41 insertions(+), 18 deletions(-) diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt index a8d2ae1e335b..f2040d46f095 100644 --- a/Documentation/admin-guide/kernel-parameters.txt +++ b/Documentation/admin-guide/kernel-parameters.txt @@ -4049,19 +4049,27 @@ This parameter controls whether the Speculative Store Bypass optimization is used. - on - Unconditionally disable Speculative Store Bypass - off - Unconditionally enable Speculative Store Bypass - auto - Kernel detects whether the CPU model contains an - implementation of Speculative Store Bypass and - picks the most appropriate mitigation. - prctl - Control Speculative Store Bypass per thread - via prctl. Speculative Store Bypass is enabled - for a process by default. The state of the control - is inherited on fork. + on - Unconditionally disable Speculative Store Bypass + off - Unconditionally enable Speculative Store Bypass + auto - Kernel detects whether the CPU model contains an + implementation of Speculative Store Bypass and + picks the most appropriate mitigation. If the + CPU is not vulnerable, "off" is selected. If the + CPU is vulnerable the default mitigation is + architecture and Kconfig dependent. See below. + prctl - Control Speculative Store Bypass per thread + via prctl. Speculative Store Bypass is enabled + for a process by default. The state of the control + is inherited on fork. + seccomp - Same as "prctl" above, but all seccomp threads + will disable SSB unless they explicitly opt out. Not specifying this option is equivalent to spec_store_bypass_disable=auto. + Default mitigations: + X86: If CONFIG_SECCOMP=y "seccomp", otherwise "prctl" + spia_io_base= [HW,MTD] spia_fio_base= spia_pedr= diff --git a/arch/x86/include/asm/nospec-branch.h b/arch/x86/include/asm/nospec-branch.h index 71ad01422655..328ea3cb769f 100644 --- a/arch/x86/include/asm/nospec-branch.h +++ b/arch/x86/include/asm/nospec-branch.h @@ -233,6 +233,7 @@ enum ssb_mitigation { SPEC_STORE_BYPASS_NONE, SPEC_STORE_BYPASS_DISABLE, SPEC_STORE_BYPASS_PRCTL, + SPEC_STORE_BYPASS_SECCOMP, }; extern char __indirect_thunk_start[]; diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index 5dab4c3d26e7..563d8e54c863 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -416,22 +416,25 @@ enum ssb_mitigation_cmd { SPEC_STORE_BYPASS_CMD_AUTO, SPEC_STORE_BYPASS_CMD_ON, SPEC_STORE_BYPASS_CMD_PRCTL, + SPEC_STORE_BYPASS_CMD_SECCOMP, }; static const char *ssb_strings[] = { [SPEC_STORE_BYPASS_NONE] = "Vulnerable", [SPEC_STORE_BYPASS_DISABLE] = "Mitigation: Speculative Store Bypass disabled", - [SPEC_STORE_BYPASS_PRCTL] = "Mitigation: Speculative Store Bypass disabled via prctl" + [SPEC_STORE_BYPASS_PRCTL] = "Mitigation: Speculative Store Bypass disabled via prctl", + [SPEC_STORE_BYPASS_SECCOMP] = "Mitigation: Speculative Store Bypass disabled via prctl and seccomp", }; static const struct { const char *option; enum ssb_mitigation_cmd cmd; } ssb_mitigation_options[] = { - { "auto", SPEC_STORE_BYPASS_CMD_AUTO }, /* Platform decides */ - { "on", SPEC_STORE_BYPASS_CMD_ON }, /* Disable Speculative Store Bypass */ - { "off", SPEC_STORE_BYPASS_CMD_NONE }, /* Don't touch Speculative Store Bypass */ - { "prctl", SPEC_STORE_BYPASS_CMD_PRCTL }, /* Disable Speculative Store Bypass via prctl */ + { "auto", SPEC_STORE_BYPASS_CMD_AUTO }, /* Platform decides */ + { "on", SPEC_STORE_BYPASS_CMD_ON }, /* Disable Speculative Store Bypass */ + { "off", SPEC_STORE_BYPASS_CMD_NONE }, /* Don't touch Speculative Store Bypass */ + { "prctl", SPEC_STORE_BYPASS_CMD_PRCTL }, /* Disable Speculative Store Bypass via prctl */ + { "seccomp", SPEC_STORE_BYPASS_CMD_SECCOMP }, /* Disable Speculative Store Bypass via prctl and seccomp */ }; static enum ssb_mitigation_cmd __init ssb_parse_cmdline(void) @@ -481,8 +484,15 @@ static enum ssb_mitigation_cmd __init __ssb_select_mitigation(void) switch (cmd) { case SPEC_STORE_BYPASS_CMD_AUTO: - /* Choose prctl as the default mode */ - mode = SPEC_STORE_BYPASS_PRCTL; + case SPEC_STORE_BYPASS_CMD_SECCOMP: + /* + * Choose prctl+seccomp as the default mode if seccomp is + * enabled. + */ + if (IS_ENABLED(CONFIG_SECCOMP)) + mode = SPEC_STORE_BYPASS_SECCOMP; + else + mode = SPEC_STORE_BYPASS_PRCTL; break; case SPEC_STORE_BYPASS_CMD_ON: mode = SPEC_STORE_BYPASS_DISABLE; @@ -530,12 +540,14 @@ static void ssb_select_mitigation() } #undef pr_fmt +#define pr_fmt(fmt) "Speculation prctl: " fmt static int ssb_prctl_set(struct task_struct *task, unsigned long ctrl) { bool update; - if (ssb_mode != SPEC_STORE_BYPASS_PRCTL) + if (ssb_mode != SPEC_STORE_BYPASS_PRCTL && + ssb_mode != SPEC_STORE_BYPASS_SECCOMP) return -ENXIO; switch (ctrl) { @@ -583,7 +595,8 @@ int arch_prctl_spec_ctrl_set(struct task_struct *task, unsigned long which, #ifdef CONFIG_SECCOMP void arch_seccomp_spec_mitigate(struct task_struct *task) { - ssb_prctl_set(task, PR_SPEC_FORCE_DISABLE); + if (ssb_mode == SPEC_STORE_BYPASS_SECCOMP) + ssb_prctl_set(task, PR_SPEC_FORCE_DISABLE); } #endif @@ -592,6 +605,7 @@ static int ssb_prctl_get(struct task_struct *task) switch (ssb_mode) { case SPEC_STORE_BYPASS_DISABLE: return PR_SPEC_DISABLE; + case SPEC_STORE_BYPASS_SECCOMP: case SPEC_STORE_BYPASS_PRCTL: if (task_spec_ssb_force_disable(task)) return PR_SPEC_PRCTL | PR_SPEC_FORCE_DISABLE; From fa89adba1941e4f3b213399b81732a5c12fd9131 Mon Sep 17 00:00:00 2001 From: Jens Remus Date: Thu, 3 May 2018 13:52:47 +0200 Subject: [PATCH 029/190] scsi: zfcp: fix infinite iteration on ERP ready list zfcp_erp_adapter_reopen() schedules blocking of all of the adapter's rports via zfcp_scsi_schedule_rports_block() and enqueues a reopen adapter ERP action via zfcp_erp_action_enqueue(). Both are separately processed asynchronously and concurrently. Blocking of rports is done in a kworker by zfcp_scsi_rport_work(). It calls zfcp_scsi_rport_block(), which then traces a DBF REC "scpdely" via zfcp_dbf_rec_trig(). zfcp_dbf_rec_trig() acquires the DBF REC spin lock and then iterates with list_for_each() over the adapter's ERP ready list without holding the ERP lock. This opens a race window in which the current list entry can be moved to another list, causing list_for_each() to iterate forever on the wrong list, as the erp_ready_head is never encountered as terminal condition. Meanwhile the ERP action can be processed in the ERP thread by zfcp_erp_thread(). It calls zfcp_erp_strategy(), which acquires the ERP lock and then calls zfcp_erp_action_to_running() to move the ERP action from the ready to the running list. zfcp_erp_action_to_running() can move the ERP action using list_move() just during the aforementioned race window. It then traces a REC RUN "erator1" via zfcp_dbf_rec_run(). zfcp_dbf_rec_run() tries to acquire the DBF REC spin lock. If this is held by the infinitely looping kworker, it effectively spins forever. Example Sequence Diagram: Process ERP Thread rport_work ------------------- ------------------- ------------------- zfcp_erp_adapter_reopen() zfcp_erp_adapter_block() zfcp_scsi_schedule_rports_block() lock ERP zfcp_scsi_rport_work() zfcp_erp_action_enqueue(ZFCP_ERP_ACTION_REOPEN_ADAPTER) list_add_tail() on ready !(rport_task==RPORT_ADD) wake_up() ERP thread zfcp_scsi_rport_block() zfcp_dbf_rec_trig() zfcp_erp_strategy() zfcp_dbf_rec_trig() unlock ERP lock DBF REC zfcp_erp_wait() lock ERP | zfcp_erp_action_to_running() | list_for_each() ready | list_move() current entry | ready to running | zfcp_dbf_rec_run() endless loop over running | zfcp_dbf_rec_run_lvl() | lock DBF REC spins forever Any adapter recovery can trigger this, such as setting the device offline or reboot. V4.9 commit 4eeaa4f3f1d6 ("zfcp: close window with unblocked rport during rport gone") introduced additional tracing of (un)blocking of rports. It missed that the adapter->erp_lock must be held when calling zfcp_dbf_rec_trig(). This fix uses the approach formerly introduced by commit aa0fec62391c ("[SCSI] zfcp: Fix sparse warning by providing new entry in dbf") that got later removed by commit ae0904f60fab ("[SCSI] zfcp: Redesign of the debug tracing for recovery actions."). Introduce zfcp_dbf_rec_trig_lock(), a wrapper for zfcp_dbf_rec_trig() that acquires and releases the adapter->erp_lock for read. Reported-by: Sebastian Ott Signed-off-by: Jens Remus Fixes: 4eeaa4f3f1d6 ("zfcp: close window with unblocked rport during rport gone") Cc: # 2.6.32+ Reviewed-by: Benjamin Block Signed-off-by: Steffen Maier Signed-off-by: Martin K. Petersen --- drivers/s390/scsi/zfcp_dbf.c | 23 ++++++++++++++++++++++- drivers/s390/scsi/zfcp_ext.h | 5 ++++- drivers/s390/scsi/zfcp_scsi.c | 14 +++++++------- 3 files changed, 33 insertions(+), 9 deletions(-) diff --git a/drivers/s390/scsi/zfcp_dbf.c b/drivers/s390/scsi/zfcp_dbf.c index a8b831000b2d..18c4f933e8b9 100644 --- a/drivers/s390/scsi/zfcp_dbf.c +++ b/drivers/s390/scsi/zfcp_dbf.c @@ -4,7 +4,7 @@ * * Debug traces for zfcp. * - * Copyright IBM Corp. 2002, 2017 + * Copyright IBM Corp. 2002, 2018 */ #define KMSG_COMPONENT "zfcp" @@ -308,6 +308,27 @@ void zfcp_dbf_rec_trig(char *tag, struct zfcp_adapter *adapter, spin_unlock_irqrestore(&dbf->rec_lock, flags); } +/** + * zfcp_dbf_rec_trig_lock - trace event related to triggered recovery with lock + * @tag: identifier for event + * @adapter: adapter on which the erp_action should run + * @port: remote port involved in the erp_action + * @sdev: scsi device involved in the erp_action + * @want: wanted erp_action + * @need: required erp_action + * + * The adapter->erp_lock must not be held. + */ +void zfcp_dbf_rec_trig_lock(char *tag, struct zfcp_adapter *adapter, + struct zfcp_port *port, struct scsi_device *sdev, + u8 want, u8 need) +{ + unsigned long flags; + + read_lock_irqsave(&adapter->erp_lock, flags); + zfcp_dbf_rec_trig(tag, adapter, port, sdev, want, need); + read_unlock_irqrestore(&adapter->erp_lock, flags); +} /** * zfcp_dbf_rec_run_lvl - trace event related to running recovery diff --git a/drivers/s390/scsi/zfcp_ext.h b/drivers/s390/scsi/zfcp_ext.h index bf8ea4df2bb8..e5eed8aac0ce 100644 --- a/drivers/s390/scsi/zfcp_ext.h +++ b/drivers/s390/scsi/zfcp_ext.h @@ -4,7 +4,7 @@ * * External function declarations. * - * Copyright IBM Corp. 2002, 2016 + * Copyright IBM Corp. 2002, 2018 */ #ifndef ZFCP_EXT_H @@ -35,6 +35,9 @@ extern int zfcp_dbf_adapter_register(struct zfcp_adapter *); extern void zfcp_dbf_adapter_unregister(struct zfcp_adapter *); extern void zfcp_dbf_rec_trig(char *, struct zfcp_adapter *, struct zfcp_port *, struct scsi_device *, u8, u8); +extern void zfcp_dbf_rec_trig_lock(char *tag, struct zfcp_adapter *adapter, + struct zfcp_port *port, + struct scsi_device *sdev, u8 want, u8 need); extern void zfcp_dbf_rec_run(char *, struct zfcp_erp_action *); extern void zfcp_dbf_rec_run_lvl(int level, char *tag, struct zfcp_erp_action *erp); diff --git a/drivers/s390/scsi/zfcp_scsi.c b/drivers/s390/scsi/zfcp_scsi.c index 4d2ba5682493..22f9562f415c 100644 --- a/drivers/s390/scsi/zfcp_scsi.c +++ b/drivers/s390/scsi/zfcp_scsi.c @@ -4,7 +4,7 @@ * * Interface to Linux SCSI midlayer. * - * Copyright IBM Corp. 2002, 2017 + * Copyright IBM Corp. 2002, 2018 */ #define KMSG_COMPONENT "zfcp" @@ -618,9 +618,9 @@ static void zfcp_scsi_rport_register(struct zfcp_port *port) ids.port_id = port->d_id; ids.roles = FC_RPORT_ROLE_FCP_TARGET; - zfcp_dbf_rec_trig("scpaddy", port->adapter, port, NULL, - ZFCP_PSEUDO_ERP_ACTION_RPORT_ADD, - ZFCP_PSEUDO_ERP_ACTION_RPORT_ADD); + zfcp_dbf_rec_trig_lock("scpaddy", port->adapter, port, NULL, + ZFCP_PSEUDO_ERP_ACTION_RPORT_ADD, + ZFCP_PSEUDO_ERP_ACTION_RPORT_ADD); rport = fc_remote_port_add(port->adapter->scsi_host, 0, &ids); if (!rport) { dev_err(&port->adapter->ccw_device->dev, @@ -642,9 +642,9 @@ static void zfcp_scsi_rport_block(struct zfcp_port *port) struct fc_rport *rport = port->rport; if (rport) { - zfcp_dbf_rec_trig("scpdely", port->adapter, port, NULL, - ZFCP_PSEUDO_ERP_ACTION_RPORT_DEL, - ZFCP_PSEUDO_ERP_ACTION_RPORT_DEL); + zfcp_dbf_rec_trig_lock("scpdely", port->adapter, port, NULL, + ZFCP_PSEUDO_ERP_ACTION_RPORT_DEL, + ZFCP_PSEUDO_ERP_ACTION_RPORT_DEL); fc_remote_port_delete(rport); port->rport = NULL; } From 322579dcc865b94b47345ad1b6002ad167f85405 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 8 May 2018 14:21:56 -0700 Subject: [PATCH 030/190] libata: Blacklist some Sandisk SSDs for NCQ Sandisk SSDs SD7SN6S256G and SD8SN8U256G are regularly locking up regularly under sustained moderate load with NCQ enabled. Blacklist for now. Signed-off-by: Tejun Heo Reported-by: Dave Jones Cc: stable@vger.kernel.org --- drivers/ata/libata-core.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index 68596bd4cf06..a2498f5cfb28 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -4493,6 +4493,10 @@ static const struct ata_blacklist_entry ata_device_blacklist [] = { /* https://bugzilla.kernel.org/show_bug.cgi?id=15573 */ { "C300-CTFDDAC128MAG", "0001", ATA_HORKAGE_NONCQ, }, + /* Some Sandisk SSDs lock up hard with NCQ enabled. Reported on + SD7SN6S256G and SD8SN8U256G */ + { "SanDisk SD[78]SN*G", NULL, ATA_HORKAGE_NONCQ, }, + /* devices which puke on READ_NATIVE_MAX */ { "HDS724040KLSA80", "KFAOA20N", ATA_HORKAGE_BROKEN_HPA, }, { "WDC WD3200JD-00KLB0", "WD-WCAMR1130137", ATA_HORKAGE_BROKEN_HPA }, From 9533b292a7acc62c294ebcbd9e1f9f9d52adb10b Mon Sep 17 00:00:00 2001 From: Greg Thelen Date: Thu, 3 May 2018 20:29:19 -0700 Subject: [PATCH 031/190] IB: remove redundant INFINIBAND kconfig dependencies INFINIBAND_ADDR_TRANS depends on INFINIBAND. So there's no need for options which depend INFINIBAND_ADDR_TRANS to also depend on INFINIBAND. Remove the unnecessary INFINIBAND depends. Signed-off-by: Greg Thelen Signed-off-by: Doug Ledford --- drivers/infiniband/ulp/srpt/Kconfig | 2 +- drivers/nvme/host/Kconfig | 2 +- drivers/nvme/target/Kconfig | 2 +- drivers/staging/lustre/lnet/Kconfig | 2 +- fs/cifs/Kconfig | 2 +- net/9p/Kconfig | 2 +- net/rds/Kconfig | 2 +- net/sunrpc/Kconfig | 2 +- 8 files changed, 8 insertions(+), 8 deletions(-) diff --git a/drivers/infiniband/ulp/srpt/Kconfig b/drivers/infiniband/ulp/srpt/Kconfig index fb8b7182f05e..25bf6955b6d0 100644 --- a/drivers/infiniband/ulp/srpt/Kconfig +++ b/drivers/infiniband/ulp/srpt/Kconfig @@ -1,6 +1,6 @@ config INFINIBAND_SRPT tristate "InfiniBand SCSI RDMA Protocol target support" - depends on INFINIBAND && INFINIBAND_ADDR_TRANS && TARGET_CORE + depends on INFINIBAND_ADDR_TRANS && TARGET_CORE ---help--- Support for the SCSI RDMA Protocol (SRP) Target driver. The diff --git a/drivers/nvme/host/Kconfig b/drivers/nvme/host/Kconfig index 88a8b5916624..dbb7464c018c 100644 --- a/drivers/nvme/host/Kconfig +++ b/drivers/nvme/host/Kconfig @@ -27,7 +27,7 @@ config NVME_FABRICS config NVME_RDMA tristate "NVM Express over Fabrics RDMA host driver" - depends on INFINIBAND && INFINIBAND_ADDR_TRANS && BLOCK + depends on INFINIBAND_ADDR_TRANS && BLOCK select NVME_CORE select NVME_FABRICS select SG_POOL diff --git a/drivers/nvme/target/Kconfig b/drivers/nvme/target/Kconfig index 3c7b61ddb0d1..7595664ee753 100644 --- a/drivers/nvme/target/Kconfig +++ b/drivers/nvme/target/Kconfig @@ -27,7 +27,7 @@ config NVME_TARGET_LOOP config NVME_TARGET_RDMA tristate "NVMe over Fabrics RDMA target support" - depends on INFINIBAND && INFINIBAND_ADDR_TRANS + depends on INFINIBAND_ADDR_TRANS depends on NVME_TARGET select SGL_ALLOC help diff --git a/drivers/staging/lustre/lnet/Kconfig b/drivers/staging/lustre/lnet/Kconfig index ad049e6f24e4..f3b1ad4bd3dc 100644 --- a/drivers/staging/lustre/lnet/Kconfig +++ b/drivers/staging/lustre/lnet/Kconfig @@ -34,7 +34,7 @@ config LNET_SELFTEST config LNET_XPRT_IB tristate "LNET infiniband support" - depends on LNET && PCI && INFINIBAND && INFINIBAND_ADDR_TRANS + depends on LNET && PCI && INFINIBAND_ADDR_TRANS default LNET && INFINIBAND help This option allows the LNET users to use infiniband as an diff --git a/fs/cifs/Kconfig b/fs/cifs/Kconfig index 5f132d59dfc2..d61e2de8d0eb 100644 --- a/fs/cifs/Kconfig +++ b/fs/cifs/Kconfig @@ -197,7 +197,7 @@ config CIFS_SMB311 config CIFS_SMB_DIRECT bool "SMB Direct support (Experimental)" - depends on CIFS=m && INFINIBAND && INFINIBAND_ADDR_TRANS || CIFS=y && INFINIBAND=y && INFINIBAND_ADDR_TRANS=y + depends on CIFS=m && INFINIBAND_ADDR_TRANS || CIFS=y && INFINIBAND_ADDR_TRANS=y help Enables SMB Direct experimental support for SMB 3.0, 3.02 and 3.1.1. SMB Direct allows transferring SMB packets over RDMA. If unsure, diff --git a/net/9p/Kconfig b/net/9p/Kconfig index e6014e0e51f7..46c39f7da444 100644 --- a/net/9p/Kconfig +++ b/net/9p/Kconfig @@ -32,7 +32,7 @@ config NET_9P_XEN config NET_9P_RDMA - depends on INET && INFINIBAND && INFINIBAND_ADDR_TRANS + depends on INET && INFINIBAND_ADDR_TRANS tristate "9P RDMA Transport (Experimental)" help This builds support for an RDMA transport. diff --git a/net/rds/Kconfig b/net/rds/Kconfig index bffde4b46c5d..1a31502ee7db 100644 --- a/net/rds/Kconfig +++ b/net/rds/Kconfig @@ -8,7 +8,7 @@ config RDS config RDS_RDMA tristate "RDS over Infiniband" - depends on RDS && INFINIBAND && INFINIBAND_ADDR_TRANS + depends on RDS && INFINIBAND_ADDR_TRANS ---help--- Allow RDS to use Infiniband as a transport. This transport supports RDMA operations. diff --git a/net/sunrpc/Kconfig b/net/sunrpc/Kconfig index ac09ca803296..6358e5271070 100644 --- a/net/sunrpc/Kconfig +++ b/net/sunrpc/Kconfig @@ -50,7 +50,7 @@ config SUNRPC_DEBUG config SUNRPC_XPRT_RDMA tristate "RPC-over-RDMA transport" - depends on SUNRPC && INFINIBAND && INFINIBAND_ADDR_TRANS + depends on SUNRPC && INFINIBAND_ADDR_TRANS default SUNRPC && INFINIBAND select SG_POOL help From 27f70620faf02a897123eb36b9e614e8d540f751 Mon Sep 17 00:00:00 2001 From: Leon Romanovsky Date: Thu, 3 May 2018 21:37:46 +0300 Subject: [PATCH 032/190] MAINTAINERS: Remove bouncing @mellanox.com addresses Delete non-existent @mellanox.com addresses from MAINTAINERS file. Signed-off-by: Leon Romanovsky Signed-off-by: Doug Ledford --- MAINTAINERS | 6 ------ 1 file changed, 6 deletions(-) diff --git a/MAINTAINERS b/MAINTAINERS index 0a1410d5a621..1b52c7e7fc7d 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -5391,7 +5391,6 @@ S: Maintained F: drivers/iommu/exynos-iommu.c EZchip NPS platform support -M: Elad Kanfi M: Vineet Gupta S: Supported F: arch/arc/plat-eznps @@ -9012,7 +9011,6 @@ Q: http://patchwork.ozlabs.org/project/netdev/list/ F: drivers/net/ethernet/mellanox/mlx5/core/en_* MELLANOX ETHERNET INNOVA DRIVER -M: Ilan Tayari R: Boris Pismenny L: netdev@vger.kernel.org S: Supported @@ -9022,7 +9020,6 @@ F: drivers/net/ethernet/mellanox/mlx5/core/fpga/* F: include/linux/mlx5/mlx5_ifc_fpga.h MELLANOX ETHERNET INNOVA IPSEC DRIVER -M: Ilan Tayari R: Boris Pismenny L: netdev@vger.kernel.org S: Supported @@ -9078,7 +9075,6 @@ F: include/uapi/rdma/mlx4-abi.h MELLANOX MLX5 core VPI driver M: Saeed Mahameed -M: Matan Barak M: Leon Romanovsky L: netdev@vger.kernel.org L: linux-rdma@vger.kernel.org @@ -9089,7 +9085,6 @@ F: drivers/net/ethernet/mellanox/mlx5/core/ F: include/linux/mlx5/ MELLANOX MLX5 IB driver -M: Matan Barak M: Leon Romanovsky L: linux-rdma@vger.kernel.org W: http://www.mellanox.com @@ -9821,7 +9816,6 @@ F: net/netfilter/xt_CONNSECMARK.c F: net/netfilter/xt_SECMARK.c NETWORKING [TLS] -M: Ilya Lesokhin M: Aviad Yehezkel M: Dave Watson L: netdev@vger.kernel.org From f9e76ca3771bf23d2142a81a88ddd8f31f5c4c03 Mon Sep 17 00:00:00 2001 From: "Michael J. Ruhl" Date: Wed, 2 May 2018 06:42:51 -0700 Subject: [PATCH 033/190] IB/hfi1: Use after free race condition in send context error path A pio send egress error can occur when the PSM library attempts to to send a bad packet. That issue is still being investigated. The pio error interrupt handler then attempts to progress the recovery of the errored pio send context. Code inspection reveals that the handling lacks the necessary locking if that recovery interleaves with a PSM close of the "context" object contains the pio send context. The lack of the locking can cause the recovery to access the already freed pio send context object and incorrectly deduce that the pio send context is actually a kernel pio send context as shown by the NULL deref stack below: [] _dev_info+0x6c/0x90 [] sc_restart+0x70/0x1f0 [hfi1] [] ? __schedule+0x424/0x9b0 [] sc_halted+0x15/0x20 [hfi1] [] process_one_work+0x17a/0x440 [] worker_thread+0x126/0x3c0 [] ? manage_workers.isra.24+0x2a0/0x2a0 [] kthread+0xcf/0xe0 [] ? insert_kthread_work+0x40/0x40 [] ret_from_fork+0x58/0x90 [] ? insert_kthread_work+0x40/0x40 This is the best case scenario and other scenarios can corrupt the already freed memory. Fix by adding the necessary locking in the pio send context error handler. Cc: # 4.9.x Reviewed-by: Mike Marciniszyn Reviewed-by: Dennis Dalessandro Signed-off-by: Michael J. Ruhl Signed-off-by: Dennis Dalessandro Signed-off-by: Doug Ledford --- drivers/infiniband/hw/hfi1/chip.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/infiniband/hw/hfi1/chip.c b/drivers/infiniband/hw/hfi1/chip.c index e6a60fa59f2b..e6bdd0c1e80a 100644 --- a/drivers/infiniband/hw/hfi1/chip.c +++ b/drivers/infiniband/hw/hfi1/chip.c @@ -5944,6 +5944,7 @@ static void is_sendctxt_err_int(struct hfi1_devdata *dd, u64 status; u32 sw_index; int i = 0; + unsigned long irq_flags; sw_index = dd->hw_to_sw[hw_context]; if (sw_index >= dd->num_send_contexts) { @@ -5953,10 +5954,12 @@ static void is_sendctxt_err_int(struct hfi1_devdata *dd, return; } sci = &dd->send_contexts[sw_index]; + spin_lock_irqsave(&dd->sc_lock, irq_flags); sc = sci->sc; if (!sc) { dd_dev_err(dd, "%s: context %u(%u): no sc?\n", __func__, sw_index, hw_context); + spin_unlock_irqrestore(&dd->sc_lock, irq_flags); return; } @@ -5978,6 +5981,7 @@ static void is_sendctxt_err_int(struct hfi1_devdata *dd, */ if (sc->type != SC_USER) queue_work(dd->pport->hfi1_wq, &sc->halt_work); + spin_unlock_irqrestore(&dd->sc_lock, irq_flags); /* * Update the counters for the corresponding status bits. From 18b0362e87dfa09e355093b897b9db854e360d28 Mon Sep 17 00:00:00 2001 From: Yishai Hadas Date: Mon, 7 May 2018 10:20:01 +0300 Subject: [PATCH 034/190] RDMA/mlx5: Don't assume that medium blueFlame register exists User can leave system without medium BlueFlames registers, however the code assumed that at least one such register exists. This patch fixes that assumption. Fixes: c1be5232d21d ("IB/mlx5: Fix micro UAR allocator") Reported-by: Rohit Zambre Signed-off-by: Yishai Hadas Signed-off-by: Leon Romanovsky Signed-off-by: Doug Ledford --- drivers/infiniband/hw/mlx5/qp.c | 18 +++++++++++------- 1 file changed, 11 insertions(+), 7 deletions(-) diff --git a/drivers/infiniband/hw/mlx5/qp.c b/drivers/infiniband/hw/mlx5/qp.c index 87b7c1be2a11..2193dc1765fb 100644 --- a/drivers/infiniband/hw/mlx5/qp.c +++ b/drivers/infiniband/hw/mlx5/qp.c @@ -484,11 +484,6 @@ static int qp_has_rq(struct ib_qp_init_attr *attr) return 1; } -static int first_med_bfreg(void) -{ - return 1; -} - enum { /* this is the first blue flame register in the array of bfregs assigned * to a processes. Since we do not use it for blue flame but rather @@ -514,6 +509,12 @@ static int num_med_bfreg(struct mlx5_ib_dev *dev, return n >= 0 ? n : 0; } +static int first_med_bfreg(struct mlx5_ib_dev *dev, + struct mlx5_bfreg_info *bfregi) +{ + return num_med_bfreg(dev, bfregi) ? 1 : -ENOMEM; +} + static int first_hi_bfreg(struct mlx5_ib_dev *dev, struct mlx5_bfreg_info *bfregi) { @@ -541,10 +542,13 @@ static int alloc_high_class_bfreg(struct mlx5_ib_dev *dev, static int alloc_med_class_bfreg(struct mlx5_ib_dev *dev, struct mlx5_bfreg_info *bfregi) { - int minidx = first_med_bfreg(); + int minidx = first_med_bfreg(dev, bfregi); int i; - for (i = first_med_bfreg(); i < first_hi_bfreg(dev, bfregi); i++) { + if (minidx < 0) + return minidx; + + for (i = minidx; i < first_hi_bfreg(dev, bfregi); i++) { if (bfregi->count[i] < bfregi->count[minidx]) minidx = i; if (!bfregi->count[minidx]) From 37da2a03c036538a5a79766d74bfcf5b873e5cad Mon Sep 17 00:00:00 2001 From: Daria Velikovsky Date: Mon, 7 May 2018 10:20:02 +0300 Subject: [PATCH 035/190] RDMA/mlx5: Use proper spec flow label type Flow label is defined as u32 in the in ipv6 flow spec, but used internally in the flow specs parsing as u8. That was causing loss of part of flow_label value. Fixes: 2d1e697e9b716 ('IB/mlx5: Add support to match inner packet fields') Reviewed-by: Maor Gottlieb Signed-off-by: Daria Velikovsky Signed-off-by: Leon Romanovsky Signed-off-by: Doug Ledford --- drivers/infiniband/hw/mlx5/main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c index a42c6b1cdb5a..cd1c342be6e3 100644 --- a/drivers/infiniband/hw/mlx5/main.c +++ b/drivers/infiniband/hw/mlx5/main.c @@ -2416,7 +2416,7 @@ static void set_proto(void *outer_c, void *outer_v, u8 mask, u8 val) MLX5_SET(fte_match_set_lyr_2_4, outer_v, ip_protocol, val); } -static void set_flow_label(void *misc_c, void *misc_v, u8 mask, u8 val, +static void set_flow_label(void *misc_c, void *misc_v, u32 mask, u32 val, bool inner) { if (inner) { From 5a7189d529cd146cd5838af97b32fcac4122b471 Mon Sep 17 00:00:00 2001 From: Mustafa Ismail Date: Mon, 7 May 2018 12:52:17 -0500 Subject: [PATCH 036/190] i40iw: Fix memory leak in error path of create QP If i40iw_allocate_dma_mem fails when creating a QP, the memory allocated for the QP structure using kzalloc is not freed because iwqp->allocated_buffer is used to free the memory and it is not setup until later. Fix this by setting iwqp->allocated_buffer before allocating the dma memory. Fixes: d37498417947 ("i40iw: add files for iwarp interface") Signed-off-by: Mustafa Ismail Signed-off-by: Shiraz Saleem Signed-off-by: Doug Ledford --- drivers/infiniband/hw/i40iw/i40iw_verbs.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/infiniband/hw/i40iw/i40iw_verbs.c b/drivers/infiniband/hw/i40iw/i40iw_verbs.c index 40e4f5ab2b46..3ebe91438523 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_verbs.c +++ b/drivers/infiniband/hw/i40iw/i40iw_verbs.c @@ -614,6 +614,7 @@ static struct ib_qp *i40iw_create_qp(struct ib_pd *ibpd, return ERR_PTR(-ENOMEM); iwqp = (struct i40iw_qp *)mem; + iwqp->allocated_buffer = mem; qp = &iwqp->sc_qp; qp->back_qp = (void *)iwqp; qp->push_idx = I40IW_INVALID_PUSH_PAGE_INDEX; @@ -642,7 +643,6 @@ static struct ib_qp *i40iw_create_qp(struct ib_pd *ibpd, goto error; } - iwqp->allocated_buffer = mem; iwqp->iwdev = iwdev; iwqp->iwpd = iwpd; iwqp->ibqp.qp_num = qp_num; From eeb1af4f53fa74fd41d288b113eebcdfca4d311c Mon Sep 17 00:00:00 2001 From: Mustafa Ismail Date: Mon, 7 May 2018 12:52:18 -0500 Subject: [PATCH 037/190] i40iw: Use correct address in dst_neigh_lookup for IPv6 Use of incorrect structure address for IPv6 neighbor lookup causes connections to IPv6 addresses to fail. Fix this by using correct address in call to dst_neigh_lookup. Fixes: f27b4746f378 ("i40iw: add connection management code") Signed-off-by: Mustafa Ismail Signed-off-by: Shiraz Saleem Signed-off-by: Doug Ledford --- drivers/infiniband/hw/i40iw/i40iw_cm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/infiniband/hw/i40iw/i40iw_cm.c b/drivers/infiniband/hw/i40iw/i40iw_cm.c index 4cfa8f4647e2..f7c6fd9ff6e2 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_cm.c +++ b/drivers/infiniband/hw/i40iw/i40iw_cm.c @@ -2093,7 +2093,7 @@ static int i40iw_addr_resolve_neigh_ipv6(struct i40iw_device *iwdev, if (netif_is_bond_slave(netdev)) netdev = netdev_master_upper_dev_get(netdev); - neigh = dst_neigh_lookup(dst, &dst_addr); + neigh = dst_neigh_lookup(dst, dst_addr.sin6_addr.in6_u.u6_addr32); rcu_read_lock(); if (neigh) { From 1661d3b0e2183ce90f6611641c350a5aa02aaa80 Mon Sep 17 00:00:00 2001 From: Alexandru Moise <00moses.alexander00@gmail.com> Date: Tue, 8 May 2018 11:02:02 +0200 Subject: [PATCH 038/190] nvmet,rxe: defer ip datagram sending to tasklet This addresses 3 separate problems: 1. When using NVME over Fabrics we may end up sending IP packets in interrupt context, we should defer this work to a tasklet. [ 50.939957] WARNING: CPU: 3 PID: 0 at kernel/softirq.c:161 __local_bh_enable_ip+0x1f/0xa0 [ 50.942602] CPU: 3 PID: 0 Comm: swapper/3 Kdump: loaded Tainted: G W 4.17.0-rc3-ARCH+ #104 [ 50.945466] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.11.0-20171110_100015-anatol 04/01/2014 [ 50.948163] RIP: 0010:__local_bh_enable_ip+0x1f/0xa0 [ 50.949631] RSP: 0018:ffff88009c183900 EFLAGS: 00010006 [ 50.951029] RAX: 0000000080010403 RBX: 0000000000000200 RCX: 0000000000000001 [ 50.952636] RDX: 0000000000000000 RSI: 0000000000000200 RDI: ffffffff817e04ec [ 50.954278] RBP: ffff88009c183910 R08: 0000000000000001 R09: 0000000000000614 [ 50.956000] R10: ffffea00021d5500 R11: 0000000000000001 R12: ffffffff817e04ec [ 50.957779] R13: 0000000000000000 R14: ffff88009566f400 R15: ffff8800956c7000 [ 50.959402] FS: 0000000000000000(0000) GS:ffff88009c180000(0000) knlGS:0000000000000000 [ 50.961552] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 50.963798] CR2: 000055c4ec0ccac0 CR3: 0000000002209001 CR4: 00000000000606e0 [ 50.966121] Call Trace: [ 50.966845] [ 50.967497] __dev_queue_xmit+0x62d/0x690 [ 50.968722] dev_queue_xmit+0x10/0x20 [ 50.969894] neigh_resolve_output+0x173/0x190 [ 50.971244] ip_finish_output2+0x2b8/0x370 [ 50.972527] ip_finish_output+0x1d2/0x220 [ 50.973785] ? ip_finish_output+0x1d2/0x220 [ 50.975010] ip_output+0xd4/0x100 [ 50.975903] ip_local_out+0x3b/0x50 [ 50.976823] rxe_send+0x74/0x120 [ 50.977702] rxe_requester+0xe3b/0x10b0 [ 50.978881] ? ip_local_deliver_finish+0xd1/0xe0 [ 50.980260] rxe_do_task+0x85/0x100 [ 50.981386] rxe_run_task+0x2f/0x40 [ 50.982470] rxe_post_send+0x51a/0x550 [ 50.983591] nvmet_rdma_queue_response+0x10a/0x170 [ 50.985024] __nvmet_req_complete+0x95/0xa0 [ 50.986287] nvmet_req_complete+0x15/0x60 [ 50.987469] nvmet_bio_done+0x2d/0x40 [ 50.988564] bio_endio+0x12c/0x140 [ 50.989654] blk_update_request+0x185/0x2a0 [ 50.990947] blk_mq_end_request+0x1e/0x80 [ 50.991997] nvme_complete_rq+0x1cc/0x1e0 [ 50.993171] nvme_pci_complete_rq+0x117/0x120 [ 50.994355] __blk_mq_complete_request+0x15e/0x180 [ 50.995988] blk_mq_complete_request+0x6f/0xa0 [ 50.997304] nvme_process_cq+0xe0/0x1b0 [ 50.998494] nvme_irq+0x28/0x50 [ 50.999572] __handle_irq_event_percpu+0xa2/0x1c0 [ 51.000986] handle_irq_event_percpu+0x32/0x80 [ 51.002356] handle_irq_event+0x3c/0x60 [ 51.003463] handle_edge_irq+0x1c9/0x200 [ 51.004473] handle_irq+0x23/0x30 [ 51.005363] do_IRQ+0x46/0xd0 [ 51.006182] common_interrupt+0xf/0xf [ 51.007129] 2. Work must always be offloaded to tasklet for rxe_post_send_kernel() when using NVMEoF in order to solve lock ordering between neigh->ha_lock seqlock and the nvme queue lock: [ 77.833783] Possible interrupt unsafe locking scenario: [ 77.833783] [ 77.835831] CPU0 CPU1 [ 77.837129] ---- ---- [ 77.838313] lock(&(&n->ha_lock)->seqcount); [ 77.839550] local_irq_disable(); [ 77.841377] lock(&(&nvmeq->q_lock)->rlock); [ 77.843222] lock(&(&n->ha_lock)->seqcount); [ 77.845178] [ 77.846298] lock(&(&nvmeq->q_lock)->rlock); [ 77.847986] [ 77.847986] *** DEADLOCK *** 3. Same goes for the lock ordering between sch->q.lock and nvme queue lock: [ 47.634271] Possible interrupt unsafe locking scenario: [ 47.634271] [ 47.636452] CPU0 CPU1 [ 47.637861] ---- ---- [ 47.639285] lock(&(&sch->q.lock)->rlock); [ 47.640654] local_irq_disable(); [ 47.642451] lock(&(&nvmeq->q_lock)->rlock); [ 47.644521] lock(&(&sch->q.lock)->rlock); [ 47.646480] [ 47.647263] lock(&(&nvmeq->q_lock)->rlock); [ 47.648492] [ 47.648492] *** DEADLOCK *** Using NVMEoF after this patch seems to finally be stable, without it, rxe eventually deadlocks the whole system and causes RCU stalls. Signed-off-by: Alexandru Moise <00moses.alexander00@gmail.com> Reviewed-by: Zhu Yanjun Signed-off-by: Doug Ledford --- drivers/infiniband/sw/rxe/rxe_verbs.c | 10 +--------- 1 file changed, 1 insertion(+), 9 deletions(-) diff --git a/drivers/infiniband/sw/rxe/rxe_verbs.c b/drivers/infiniband/sw/rxe/rxe_verbs.c index 2cb52fd48cf1..73a00a1c06f6 100644 --- a/drivers/infiniband/sw/rxe/rxe_verbs.c +++ b/drivers/infiniband/sw/rxe/rxe_verbs.c @@ -761,7 +761,6 @@ static int rxe_post_send_kernel(struct rxe_qp *qp, struct ib_send_wr *wr, unsigned int mask; unsigned int length = 0; int i; - int must_sched; while (wr) { mask = wr_opcode_mask(wr->opcode, qp); @@ -791,14 +790,7 @@ static int rxe_post_send_kernel(struct rxe_qp *qp, struct ib_send_wr *wr, wr = wr->next; } - /* - * Must sched in case of GSI QP because ib_send_mad() hold irq lock, - * and the requester call ip_local_out_sk() that takes spin_lock_bh. - */ - must_sched = (qp_type(qp) == IB_QPT_GSI) || - (queue_count(qp->sq.queue) > 1); - - rxe_run_task(&qp->req.task, must_sched); + rxe_run_task(&qp->req.task, 1); if (unlikely(qp->req.state == QP_STATE_ERROR)) rxe_run_task(&qp->comp.task, 1); From ecaaf1e26a37ddf7cba4e425cf68ae7ce1869bc0 Mon Sep 17 00:00:00 2001 From: oulijun Date: Fri, 4 May 2018 10:57:03 +0800 Subject: [PATCH 039/190] RDMA/hns: Add rq inline flags judgement It needs to set the rqie field of qp context by configured rq inline flags. Besides, it need to decide whether posting inline rqwqe by judged rq inline flags. Signed-off-by: Lijun Ou Signed-off-by: Doug Ledford --- drivers/infiniband/hw/hns/hns_roce_hw_v2.c | 18 ++++++++++++------ 1 file changed, 12 insertions(+), 6 deletions(-) diff --git a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c index 25916e8522ed..162d42c6530a 100644 --- a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c +++ b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c @@ -552,11 +552,15 @@ static int hns_roce_v2_post_recv(struct ib_qp *ibqp, struct ib_recv_wr *wr, } /* rq support inline data */ - sge_list = hr_qp->rq_inl_buf.wqe_list[ind].sg_list; - hr_qp->rq_inl_buf.wqe_list[ind].sge_cnt = (u32)wr->num_sge; - for (i = 0; i < wr->num_sge; i++) { - sge_list[i].addr = (void *)(u64)wr->sg_list[i].addr; - sge_list[i].len = wr->sg_list[i].length; + if (hr_dev->caps.flags & HNS_ROCE_CAP_FLAG_RQ_INLINE) { + sge_list = hr_qp->rq_inl_buf.wqe_list[ind].sg_list; + hr_qp->rq_inl_buf.wqe_list[ind].sge_cnt = + (u32)wr->num_sge; + for (i = 0; i < wr->num_sge; i++) { + sge_list[i].addr = + (void *)(u64)wr->sg_list[i].addr; + sge_list[i].len = wr->sg_list[i].length; + } } hr_qp->rq.wrid[ind] = wr->wr_id; @@ -2169,6 +2173,7 @@ static void modify_qp_reset_to_init(struct ib_qp *ibqp, struct hns_roce_v2_qp_context *context, struct hns_roce_v2_qp_context *qpc_mask) { + struct hns_roce_dev *hr_dev = to_hr_dev(ibqp->device); struct hns_roce_qp *hr_qp = to_hr_qp(ibqp); /* @@ -2281,7 +2286,8 @@ static void modify_qp_reset_to_init(struct ib_qp *ibqp, context->rq_db_record_addr = hr_qp->rdb.dma >> 32; qpc_mask->rq_db_record_addr = 0; - roce_set_bit(context->byte_76_srqn_op_en, V2_QPC_BYTE_76_RQIE_S, 1); + roce_set_bit(context->byte_76_srqn_op_en, V2_QPC_BYTE_76_RQIE_S, + (hr_dev->caps.flags & HNS_ROCE_CAP_FLAG_RQ_INLINE) ? 1 : 0); roce_set_bit(qpc_mask->byte_76_srqn_op_en, V2_QPC_BYTE_76_RQIE_S, 0); roce_set_field(context->byte_80_rnr_rx_cqn, V2_QPC_BYTE_80_RX_CQN_M, From 3a39bbecc88fa9a99a80de588c8f1fe16aba3446 Mon Sep 17 00:00:00 2001 From: oulijun Date: Fri, 4 May 2018 10:57:04 +0800 Subject: [PATCH 040/190] RDMA/hns: Bugfix for rq record db for kernel When used rq record db for kernel, it needs to set the rdb_en of hr_qp to 1 and configures the dma address of record rq db of qp context. Signed-off-by: Lijun Ou Signed-off-by: Doug Ledford --- drivers/infiniband/hw/hns/hns_roce_qp.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/infiniband/hw/hns/hns_roce_qp.c b/drivers/infiniband/hw/hns/hns_roce_qp.c index d4aad34c21e2..cdfb774e28ff 100644 --- a/drivers/infiniband/hw/hns/hns_roce_qp.c +++ b/drivers/infiniband/hw/hns/hns_roce_qp.c @@ -660,6 +660,7 @@ static int hns_roce_create_qp_common(struct hns_roce_dev *hr_dev, goto err_rq_sge_list; } *hr_qp->rdb.db_record = 0; + hr_qp->rdb_en = 1; } /* Allocate QP buf */ From f97a62c39423575c62649721657e5cc04f67c0ac Mon Sep 17 00:00:00 2001 From: oulijun Date: Fri, 4 May 2018 10:57:05 +0800 Subject: [PATCH 041/190] RDMA/hns: Load the RoCE dirver automatically To enable the linux-kernel system to load the hns-roce-hw-v2 driver automatically when hns-roce-hw-v2 is plugged in pci bus, it need to create a MODULE_DEVICE_TABLE for expose the pci_table of hns-roce-hw-v2 to user. Signed-off-by: Lijun Ou Reported-by: Zhou Wang Tested-by: Xiaojun Tan Signed-off-by: Doug Ledford --- drivers/infiniband/hw/hns/hns_roce_hw_v2.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c index 162d42c6530a..50f55bd2c75f 100644 --- a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c +++ b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c @@ -4709,6 +4709,8 @@ static const struct pci_device_id hns_roce_hw_v2_pci_tbl[] = { {0, } }; +MODULE_DEVICE_TABLE(pci, hns_roce_hw_v2_pci_tbl); + static int hns_roce_hw_v2_get_cfg(struct hns_roce_dev *hr_dev, struct hnae3_handle *handle) { From ad18e20ba2887e221e903d311f4c9a1586eacffb Mon Sep 17 00:00:00 2001 From: oulijun Date: Fri, 4 May 2018 10:57:06 +0800 Subject: [PATCH 042/190] RDMA/hns: Update convert function of endian format Because the sys_image_guid of ib_device_attr structure is __be64, it need to use cpu_to_be64 for converting. Signed-off-by: Lijun Ou Signed-off-by: Doug Ledford --- drivers/infiniband/hw/hns/hns_roce_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/infiniband/hw/hns/hns_roce_main.c b/drivers/infiniband/hw/hns/hns_roce_main.c index 9d48bc07a9e6..96fb6a9ed93c 100644 --- a/drivers/infiniband/hw/hns/hns_roce_main.c +++ b/drivers/infiniband/hw/hns/hns_roce_main.c @@ -199,7 +199,7 @@ static int hns_roce_query_device(struct ib_device *ib_dev, memset(props, 0, sizeof(*props)); - props->sys_image_guid = cpu_to_be32(hr_dev->sys_image_guid); + props->sys_image_guid = cpu_to_be64(hr_dev->sys_image_guid); props->max_mr_size = (u64)(~(0ULL)); props->page_size_cap = hr_dev->caps.page_size_cap; props->vendor_id = hr_dev->vendor_id; From 2349fdd483ea933b223f3eca53ed42835383d316 Mon Sep 17 00:00:00 2001 From: oulijun Date: Fri, 4 May 2018 10:57:07 +0800 Subject: [PATCH 043/190] RDMA/hns: Add return operation when configured global param fail When configure global param function run fail, it should directly return and the initial flow will stop. Signed-off-by: Lijun Ou Signed-off-by: Doug Ledford --- drivers/infiniband/hw/hns/hns_roce_hw_v2.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c index 50f55bd2c75f..539b0caacb9f 100644 --- a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c +++ b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c @@ -1085,6 +1085,7 @@ static int hns_roce_v2_profile(struct hns_roce_dev *hr_dev) if (ret) { dev_err(hr_dev->dev, "Configure global param fail, ret = %d.\n", ret); + return ret; } /* Get pf resource owned by every pf */ From 391bd5fc7de54a5cb866e8897d60ee1d76b8840a Mon Sep 17 00:00:00 2001 From: oulijun Date: Fri, 4 May 2018 10:57:08 +0800 Subject: [PATCH 044/190] RDMA/hns: Not support qp transition from reset to reset for hip06 Because hip06 hardware is not support for qp transition from reset to reset state, it need to return errno when qp transited from reset to reset. This patch fixes it. Signed-off-by: Lijun Ou Signed-off-by: Doug Ledford --- drivers/infiniband/hw/hns/hns_roce_qp.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/drivers/infiniband/hw/hns/hns_roce_qp.c b/drivers/infiniband/hw/hns/hns_roce_qp.c index cdfb774e28ff..baaf906f7c2e 100644 --- a/drivers/infiniband/hw/hns/hns_roce_qp.c +++ b/drivers/infiniband/hw/hns/hns_roce_qp.c @@ -956,7 +956,14 @@ int hns_roce_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, } if (cur_state == new_state && cur_state == IB_QPS_RESET) { - ret = 0; + if (hr_dev->caps.min_wqes) { + ret = -EPERM; + dev_err(dev, "cur_state=%d new_state=%d\n", cur_state, + new_state); + } else { + ret = 0; + } + goto out; } From 778cc5a8b75eee62d330059a2655b515cda43278 Mon Sep 17 00:00:00 2001 From: oulijun Date: Fri, 4 May 2018 10:57:09 +0800 Subject: [PATCH 045/190] RDMA/hns: Fix the bug with rq sge When received multiply rq sge, it should tag the invalid lkey for the last non-zero length sge when have some sges' length are zero. This patch fixes it. Signed-off-by: Lijun Ou Signed-off-by: Doug Ledford --- drivers/infiniband/hw/hns/hns_roce_hw_v2.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c index 539b0caacb9f..a40978bbc1d0 100644 --- a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c +++ b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c @@ -547,8 +547,8 @@ static int hns_roce_v2_post_recv(struct ib_qp *ibqp, struct ib_recv_wr *wr, } if (i < hr_qp->rq.max_gs) { - dseg[i].lkey = cpu_to_le32(HNS_ROCE_INVALID_LKEY); - dseg[i].addr = 0; + dseg->lkey = cpu_to_le32(HNS_ROCE_INVALID_LKEY); + dseg->addr = 0; } /* rq support inline data */ From 90e7a4d5066240b75cdfd1bf8944ca36622153b1 Mon Sep 17 00:00:00 2001 From: oulijun Date: Fri, 4 May 2018 10:57:10 +0800 Subject: [PATCH 046/190] RDMA/hns: Set desc_dma_addr for zero when free cmq desc In order to avoid illegal use for desc_dma_addr of ring, it needs to set it zero when free cmq desc. Signed-off-by: Lijun Ou Signed-off-by: Doug Ledford --- drivers/infiniband/hw/hns/hns_roce_hw_v2.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c index a40978bbc1d0..46c3ab97e1ba 100644 --- a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c +++ b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c @@ -617,6 +617,8 @@ static void hns_roce_free_cmq_desc(struct hns_roce_dev *hr_dev, dma_unmap_single(hr_dev->dev, ring->desc_dma_addr, ring->desc_num * sizeof(struct hns_roce_cmq_desc), DMA_BIDIRECTIONAL); + + ring->desc_dma_addr = 0; kfree(ring->desc); } From 85e0274dc66430b0d0fad7ad01cbc0e0cbebf6dc Mon Sep 17 00:00:00 2001 From: oulijun Date: Fri, 4 May 2018 10:57:11 +0800 Subject: [PATCH 047/190] RDMA/hns: Enable inner_pa_vld filed of mpt When enabled inner_pa_vld field of mpt, The pa0 and pa1 will be valid and the hardware will use it directly and not use base address of pbl. As a result, it can reduce the delay. Signed-off-by: Lijun Ou Signed-off-by: Doug Ledford --- drivers/infiniband/hw/hns/hns_roce_hw_v2.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c index 46c3ab97e1ba..7ecf0c911808 100644 --- a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c +++ b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c @@ -1379,6 +1379,8 @@ static int hns_roce_v2_write_mtpt(void *mb_buf, struct hns_roce_mr *mr, roce_set_bit(mpt_entry->byte_12_mw_pa, V2_MPT_BYTE_12_PA_S, mr->type == MR_TYPE_MR ? 0 : 1); + roce_set_bit(mpt_entry->byte_12_mw_pa, V2_MPT_BYTE_12_INNER_PA_VLD_S, + 1); mpt_entry->byte_12_mw_pa = cpu_to_le32(mpt_entry->byte_12_mw_pa); mpt_entry->len_l = cpu_to_le32(lower_32_bits(mr->size)); From 79d442071a733057e4d9f28c90fbdb4f39d9fc23 Mon Sep 17 00:00:00 2001 From: oulijun Date: Fri, 4 May 2018 10:57:12 +0800 Subject: [PATCH 048/190] RDMA/hns: Set NULL for __internal_mr This patch mainly configure value for __internal_mr of mr_free_pd. Signed-off-by: Lijun Ou Signed-off-by: Doug Ledford --- drivers/infiniband/hw/hns/hns_roce_hw_v1.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/infiniband/hw/hns/hns_roce_hw_v1.c b/drivers/infiniband/hw/hns/hns_roce_hw_v1.c index 47e1b6ac1e1a..b3417a9a49de 100644 --- a/drivers/infiniband/hw/hns/hns_roce_hw_v1.c +++ b/drivers/infiniband/hw/hns/hns_roce_hw_v1.c @@ -722,6 +722,7 @@ static int hns_roce_v1_rsv_lp_qp(struct hns_roce_dev *hr_dev) free_mr->mr_free_pd = to_hr_pd(pd); free_mr->mr_free_pd->ibpd.device = &hr_dev->ib_dev; free_mr->mr_free_pd->ibpd.uobject = NULL; + free_mr->mr_free_pd->ibpd.__internal_mr = NULL; atomic_set(&free_mr->mr_free_pd->ibpd.usecnt, 0); attr.qp_access_flags = IB_ACCESS_REMOTE_WRITE; From a0403be8af338c319b5176c1d2975d94a930a0bf Mon Sep 17 00:00:00 2001 From: oulijun Date: Fri, 4 May 2018 10:57:13 +0800 Subject: [PATCH 049/190] RDMA/hns: Fix the bug with NULL pointer When the last QP of eight QPs is not exist in hns_roce_v1_mr_free_work_fn function, the print for qpn of hr_qp may introduce a calltrace for NULL pointer. Signed-off-by: Lijun Ou Signed-off-by: Doug Ledford --- drivers/infiniband/hw/hns/hns_roce_hw_v1.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/infiniband/hw/hns/hns_roce_hw_v1.c b/drivers/infiniband/hw/hns/hns_roce_hw_v1.c index b3417a9a49de..8013d69c5ac4 100644 --- a/drivers/infiniband/hw/hns/hns_roce_hw_v1.c +++ b/drivers/infiniband/hw/hns/hns_roce_hw_v1.c @@ -1037,7 +1037,7 @@ static void hns_roce_v1_mr_free_work_fn(struct work_struct *work) do { ret = hns_roce_v1_poll_cq(&mr_free_cq->ib_cq, ne, wc); - if (ret < 0) { + if (ret < 0 && hr_qp) { dev_err(dev, "(qp:0x%lx) starts, Poll cqe failed(%d) for mr 0x%x free! Remain %d cqe\n", hr_qp->qpn, ret, hr_mr->key, ne); From a75895b1ebd944a7873cbf76d30de8720955f8b3 Mon Sep 17 00:00:00 2001 From: Andrew Boyer Date: Mon, 7 May 2018 13:23:36 -0400 Subject: [PATCH 050/190] RDMA/i40iw: Avoid panic when objects are being created and destroyed A panic occurs when there is a newly-registered element on the QP/CQ MR list waiting to be attached, but a different MR is deregistered. The current code only checks for whether the list is empty, not whether the element being deregistered is actually on the list. Fix the panic by adding a boolean to track if the object is on the list. Fixes: d37498417947 ("i40iw: add files for iwarp interface") Signed-off-by: Andrew Boyer Reviewed-by: Shiraz Saleem Signed-off-by: Doug Ledford --- drivers/infiniband/hw/i40iw/i40iw_verbs.c | 11 +++++++++-- drivers/infiniband/hw/i40iw/i40iw_verbs.h | 1 + 2 files changed, 10 insertions(+), 2 deletions(-) diff --git a/drivers/infiniband/hw/i40iw/i40iw_verbs.c b/drivers/infiniband/hw/i40iw/i40iw_verbs.c index 3ebe91438523..68679ad4c6da 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_verbs.c +++ b/drivers/infiniband/hw/i40iw/i40iw_verbs.c @@ -394,6 +394,7 @@ static struct i40iw_pbl *i40iw_get_pbl(unsigned long va, list_for_each_entry(iwpbl, pbl_list, list) { if (iwpbl->user_base == va) { + iwpbl->on_list = false; list_del(&iwpbl->list); return iwpbl; } @@ -1898,6 +1899,7 @@ static struct ib_mr *i40iw_reg_user_mr(struct ib_pd *pd, goto error; spin_lock_irqsave(&ucontext->qp_reg_mem_list_lock, flags); list_add_tail(&iwpbl->list, &ucontext->qp_reg_mem_list); + iwpbl->on_list = true; spin_unlock_irqrestore(&ucontext->qp_reg_mem_list_lock, flags); break; case IW_MEMREG_TYPE_CQ: @@ -1908,6 +1910,7 @@ static struct ib_mr *i40iw_reg_user_mr(struct ib_pd *pd, spin_lock_irqsave(&ucontext->cq_reg_mem_list_lock, flags); list_add_tail(&iwpbl->list, &ucontext->cq_reg_mem_list); + iwpbl->on_list = true; spin_unlock_irqrestore(&ucontext->cq_reg_mem_list_lock, flags); break; case IW_MEMREG_TYPE_MEM: @@ -2045,14 +2048,18 @@ static void i40iw_del_memlist(struct i40iw_mr *iwmr, switch (iwmr->type) { case IW_MEMREG_TYPE_CQ: spin_lock_irqsave(&ucontext->cq_reg_mem_list_lock, flags); - if (!list_empty(&ucontext->cq_reg_mem_list)) + if (iwpbl->on_list) { + iwpbl->on_list = false; list_del(&iwpbl->list); + } spin_unlock_irqrestore(&ucontext->cq_reg_mem_list_lock, flags); break; case IW_MEMREG_TYPE_QP: spin_lock_irqsave(&ucontext->qp_reg_mem_list_lock, flags); - if (!list_empty(&ucontext->qp_reg_mem_list)) + if (iwpbl->on_list) { + iwpbl->on_list = false; list_del(&iwpbl->list); + } spin_unlock_irqrestore(&ucontext->qp_reg_mem_list_lock, flags); break; default: diff --git a/drivers/infiniband/hw/i40iw/i40iw_verbs.h b/drivers/infiniband/hw/i40iw/i40iw_verbs.h index 9067443cd311..76cf173377ab 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_verbs.h +++ b/drivers/infiniband/hw/i40iw/i40iw_verbs.h @@ -78,6 +78,7 @@ struct i40iw_pbl { }; bool pbl_allocated; + bool on_list; u64 user_base; struct i40iw_pble_alloc pble_alloc; struct i40iw_mr *iwmr; From 9f7b16afab9b47de471f4ef6a0c6c337f0a53566 Mon Sep 17 00:00:00 2001 From: Andrew Boyer Date: Mon, 7 May 2018 13:23:37 -0400 Subject: [PATCH 051/190] RDMA/i40iw: Avoid reference leaks when processing the AEQ In this switch there is a reference held on the QP. 'continue' will grab the next event without releasing the reference, causing a leak. Change it to 'break' to drop the reference before grabbing the next event. Fixes: 4e9042e647ff ("i40iw: add hw and utils files") Signed-off-by: Andrew Boyer Reviewed-by: Shiraz Saleem Signed-off-by: Doug Ledford --- drivers/infiniband/hw/i40iw/i40iw_hw.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/infiniband/hw/i40iw/i40iw_hw.c b/drivers/infiniband/hw/i40iw/i40iw_hw.c index 6139836fb533..c9f62ca7643c 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_hw.c +++ b/drivers/infiniband/hw/i40iw/i40iw_hw.c @@ -331,7 +331,7 @@ void i40iw_process_aeq(struct i40iw_device *iwdev) switch (info->ae_id) { case I40IW_AE_LLP_FIN_RECEIVED: if (qp->term_flags) - continue; + break; if (atomic_inc_return(&iwqp->close_timer_started) == 1) { iwqp->hw_tcp_state = I40IW_TCP_STATE_CLOSE_WAIT; if ((iwqp->hw_tcp_state == I40IW_TCP_STATE_CLOSE_WAIT) && @@ -360,7 +360,7 @@ void i40iw_process_aeq(struct i40iw_device *iwdev) break; case I40IW_AE_LLP_CONNECTION_RESET: if (atomic_read(&iwqp->close_timer_started)) - continue; + break; i40iw_cm_disconn(iwqp); break; case I40IW_AE_QP_SUSPEND_COMPLETE: From 43731753c4b7d832775cf6b2301dd0447a5a1851 Mon Sep 17 00:00:00 2001 From: Andrew Boyer Date: Mon, 7 May 2018 13:23:38 -0400 Subject: [PATCH 052/190] RDMA/i40iw: Avoid panic when reading back the IRQ affinity hint The current code sets an affinity hint with a cpumask_t stored on the stack. This value can then be accessed through /proc/irq/*/affinity_hint/, causing a segfault or returning corrupt data. Move the cpumask_t into struct i40iw_msix_vector so it is available later. Backtrace: BUG: unable to handle kernel paging request at ffffb16e600e7c90 IP: irq_affinity_hint_proc_show+0x60/0xf0 PGD 17c0c6d067 PUD 17c0c6e067 PMD 15d4a0e067 PTE 0 Oops: 0000 [#1] SMP Modules linked in: ... CPU: 3 PID: 172543 Comm: grep Tainted: G OE ... #1 Hardware name: ... task: ffff9a5caee08000 task.stack: ffffb16e659d8000 RIP: 0010:irq_affinity_hint_proc_show+0x60/0xf0 RSP: 0018:ffffb16e659dbd20 EFLAGS: 00010086 RAX: 0000000000000246 RBX: ffffb16e659dbd20 RCX: 0000000000000000 RDX: ffffb16e600e7c90 RSI: 0000000000000003 RDI: 0000000000000046 RBP: ffffb16e659dbd88 R08: 0000000000000038 R09: 0000000000000001 R10: 0000000070803079 R11: 0000000000000000 R12: ffff9a59d1d97a00 R13: ffff9a5da47a6cd8 R14: ffff9a5da47a6c00 R15: ffff9a59d1d97a00 FS: 00007f946c31d740(0000) GS:ffff9a5dc1800000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: ffffb16e600e7c90 CR3: 00000016a4339000 CR4: 00000000007406e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 PKRU: 55555554 Call Trace: seq_read+0x12d/0x430 ? sched_clock_cpu+0x11/0xb0 proc_reg_read+0x48/0x70 __vfs_read+0x37/0x140 ? security_file_permission+0xa0/0xc0 vfs_read+0x96/0x140 SyS_read+0x58/0xc0 do_syscall_64+0x5a/0x190 entry_SYSCALL64_slow_path+0x25/0x25 RIP: 0033:0x7f946bbc97e0 RSP: 002b:00007ffdd0c4ae08 EFLAGS: 00000246 ORIG_RAX: 0000000000000000 RAX: ffffffffffffffda RBX: 000000000096b000 RCX: 00007f946bbc97e0 RDX: 000000000096b000 RSI: 00007f946a2f0000 RDI: 0000000000000004 RBP: 0000000000001000 R08: 00007f946a2ef011 R09: 000000000000000a R10: 0000000000001000 R11: 0000000000000246 R12: 00007f946a2f0000 R13: 0000000000000004 R14: 0000000000000000 R15: 00007f946a2f0000 Code: b9 08 00 00 00 49 89 c6 48 89 df 31 c0 4d 8d ae d8 00 00 00 f3 48 ab 4c 89 ef e8 6c 9a 56 00 49 8b 96 30 01 00 00 48 85 d2 74 3f <48> 8b 0a 48 89 4d 98 48 8b 4a 08 48 89 4d a0 48 8b 4a 10 48 89 RIP: irq_affinity_hint_proc_show+0x60/0xf0 RSP: ffffb16e659dbd20 CR2: ffffb16e600e7c90 Fixes: 8e06af711bf2 ("i40iw: add main, hdr, status") Signed-off-by: Andrew Boyer Reviewed-by: Shiraz Saleem Signed-off-by: Doug Ledford --- drivers/infiniband/hw/i40iw/i40iw.h | 1 + drivers/infiniband/hw/i40iw/i40iw_main.c | 7 +++---- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/infiniband/hw/i40iw/i40iw.h b/drivers/infiniband/hw/i40iw/i40iw.h index d5d8c1be345a..2f2b4426ded7 100644 --- a/drivers/infiniband/hw/i40iw/i40iw.h +++ b/drivers/infiniband/hw/i40iw/i40iw.h @@ -207,6 +207,7 @@ struct i40iw_msix_vector { u32 irq; u32 cpu_affinity; u32 ceq_id; + cpumask_t mask; }; struct l2params_work { diff --git a/drivers/infiniband/hw/i40iw/i40iw_main.c b/drivers/infiniband/hw/i40iw/i40iw_main.c index 9cd0d3ef9057..05001e6da1f8 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_main.c +++ b/drivers/infiniband/hw/i40iw/i40iw_main.c @@ -687,7 +687,6 @@ static enum i40iw_status_code i40iw_configure_ceq_vector(struct i40iw_device *iw struct i40iw_msix_vector *msix_vec) { enum i40iw_status_code status; - cpumask_t mask; if (iwdev->msix_shared && !ceq_id) { tasklet_init(&iwdev->dpc_tasklet, i40iw_dpc, (unsigned long)iwdev); @@ -697,9 +696,9 @@ static enum i40iw_status_code i40iw_configure_ceq_vector(struct i40iw_device *iw status = request_irq(msix_vec->irq, i40iw_ceq_handler, 0, "CEQ", iwceq); } - cpumask_clear(&mask); - cpumask_set_cpu(msix_vec->cpu_affinity, &mask); - irq_set_affinity_hint(msix_vec->irq, &mask); + cpumask_clear(&msix_vec->mask); + cpumask_set_cpu(msix_vec->cpu_affinity, &msix_vec->mask); + irq_set_affinity_hint(msix_vec->irq, &msix_vec->mask); if (status) { i40iw_pr_err("ceq irq config fail\n"); From 3d69191086fc87f202c79eb8873b9c82c2bb065a Mon Sep 17 00:00:00 2001 From: Christophe Jaillet Date: Tue, 8 May 2018 07:44:27 +0200 Subject: [PATCH 053/190] iw_cxgb4: Fix an error handling path in 'c4iw_get_dma_mr()' The error handling path of 'c4iw_get_dma_mr()' does not free resources in the correct order. If an error occures, it can leak 'mhp->wr_waitp'. Fixes: a3f12da0e99a ("iw_cxgb4: allocate wait object for each memory object") Signed-off-by: Christophe JAILLET Signed-off-by: Doug Ledford --- drivers/infiniband/hw/cxgb4/mem.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/infiniband/hw/cxgb4/mem.c b/drivers/infiniband/hw/cxgb4/mem.c index e90f2fd8dc16..1445918e3239 100644 --- a/drivers/infiniband/hw/cxgb4/mem.c +++ b/drivers/infiniband/hw/cxgb4/mem.c @@ -489,10 +489,10 @@ struct ib_mr *c4iw_get_dma_mr(struct ib_pd *pd, int acc) err_dereg_mem: dereg_mem(&rhp->rdev, mhp->attr.stag, mhp->attr.pbl_size, mhp->attr.pbl_addr, mhp->dereg_skb, mhp->wr_waitp); -err_free_wr_wait: - c4iw_put_wr_wait(mhp->wr_waitp); err_free_skb: kfree_skb(mhp->dereg_skb); +err_free_wr_wait: + c4iw_put_wr_wait(mhp->wr_waitp); err_free_mhp: kfree(mhp); return ERR_PTR(ret); From 9f65fb29374ee37856dbad847b4e121aab72b510 Mon Sep 17 00:00:00 2001 From: Konrad Rzeszutek Wilk Date: Wed, 9 May 2018 21:41:38 +0200 Subject: [PATCH 054/190] x86/bugs: Rename _RDS to _SSBD Intel collateral will reference the SSB mitigation bit in IA32_SPEC_CTL[2] as SSBD (Speculative Store Bypass Disable). Hence changing it. It is unclear yet what the MSR_IA32_ARCH_CAPABILITIES (0x10a) Bit(4) name is going to be. Following the rename it would be SSBD_NO but that rolls out to Speculative Store Bypass Disable No. Also fixed the missing space in X86_FEATURE_AMD_SSBD. [ tglx: Fixup x86_amd_rds_enable() and rds_tif_to_amd_ls_cfg() as well ] Signed-off-by: Konrad Rzeszutek Wilk Signed-off-by: Thomas Gleixner --- arch/x86/include/asm/cpufeatures.h | 4 ++-- arch/x86/include/asm/msr-index.h | 10 ++++----- arch/x86/include/asm/spec-ctrl.h | 12 +++++----- arch/x86/include/asm/thread_info.h | 6 ++--- arch/x86/kernel/cpu/amd.c | 14 ++++++------ arch/x86/kernel/cpu/bugs.c | 36 +++++++++++++++--------------- arch/x86/kernel/cpu/common.c | 2 +- arch/x86/kernel/cpu/intel.c | 2 +- arch/x86/kernel/process.c | 8 +++---- arch/x86/kvm/cpuid.c | 2 +- arch/x86/kvm/vmx.c | 6 ++--- 11 files changed, 51 insertions(+), 51 deletions(-) diff --git a/arch/x86/include/asm/cpufeatures.h b/arch/x86/include/asm/cpufeatures.h index b2464c1787df..4e1c747acbf8 100644 --- a/arch/x86/include/asm/cpufeatures.h +++ b/arch/x86/include/asm/cpufeatures.h @@ -215,7 +215,7 @@ #define X86_FEATURE_USE_IBPB ( 7*32+21) /* "" Indirect Branch Prediction Barrier enabled */ #define X86_FEATURE_USE_IBRS_FW ( 7*32+22) /* "" Use IBRS during runtime firmware calls */ #define X86_FEATURE_SPEC_STORE_BYPASS_DISABLE ( 7*32+23) /* "" Disable Speculative Store Bypass. */ -#define X86_FEATURE_AMD_RDS (7*32+24) /* "" AMD RDS implementation */ +#define X86_FEATURE_AMD_SSBD ( 7*32+24) /* "" AMD SSBD implementation */ /* Virtualization flags: Linux defined, word 8 */ #define X86_FEATURE_TPR_SHADOW ( 8*32+ 0) /* Intel TPR Shadow */ @@ -336,7 +336,7 @@ #define X86_FEATURE_SPEC_CTRL (18*32+26) /* "" Speculation Control (IBRS + IBPB) */ #define X86_FEATURE_INTEL_STIBP (18*32+27) /* "" Single Thread Indirect Branch Predictors */ #define X86_FEATURE_ARCH_CAPABILITIES (18*32+29) /* IA32_ARCH_CAPABILITIES MSR (Intel) */ -#define X86_FEATURE_RDS (18*32+31) /* Reduced Data Speculation */ +#define X86_FEATURE_SSBD (18*32+31) /* Speculative Store Bypass Disable */ /* * BUG word(s) diff --git a/arch/x86/include/asm/msr-index.h b/arch/x86/include/asm/msr-index.h index 810f50bb338d..0da3ca260b06 100644 --- a/arch/x86/include/asm/msr-index.h +++ b/arch/x86/include/asm/msr-index.h @@ -42,8 +42,8 @@ #define MSR_IA32_SPEC_CTRL 0x00000048 /* Speculation Control */ #define SPEC_CTRL_IBRS (1 << 0) /* Indirect Branch Restricted Speculation */ #define SPEC_CTRL_STIBP (1 << 1) /* Single Thread Indirect Branch Predictors */ -#define SPEC_CTRL_RDS_SHIFT 2 /* Reduced Data Speculation bit */ -#define SPEC_CTRL_RDS (1 << SPEC_CTRL_RDS_SHIFT) /* Reduced Data Speculation */ +#define SPEC_CTRL_SSBD_SHIFT 2 /* Speculative Store Bypass Disable bit */ +#define SPEC_CTRL_SSBD (1 << SPEC_CTRL_SSBD_SHIFT) /* Speculative Store Bypass Disable */ #define MSR_IA32_PRED_CMD 0x00000049 /* Prediction Command */ #define PRED_CMD_IBPB (1 << 0) /* Indirect Branch Prediction Barrier */ @@ -70,10 +70,10 @@ #define MSR_IA32_ARCH_CAPABILITIES 0x0000010a #define ARCH_CAP_RDCL_NO (1 << 0) /* Not susceptible to Meltdown */ #define ARCH_CAP_IBRS_ALL (1 << 1) /* Enhanced IBRS support */ -#define ARCH_CAP_RDS_NO (1 << 4) /* +#define ARCH_CAP_SSBD_NO (1 << 4) /* * Not susceptible to Speculative Store Bypass - * attack, so no Reduced Data Speculation control - * required. + * attack, so no Speculative Store Bypass + * control required. */ #define MSR_IA32_BBL_CR_CTL 0x00000119 diff --git a/arch/x86/include/asm/spec-ctrl.h b/arch/x86/include/asm/spec-ctrl.h index 45ef00ad5105..dc21209790bf 100644 --- a/arch/x86/include/asm/spec-ctrl.h +++ b/arch/x86/include/asm/spec-ctrl.h @@ -17,20 +17,20 @@ extern void x86_spec_ctrl_restore_host(u64); /* AMD specific Speculative Store Bypass MSR data */ extern u64 x86_amd_ls_cfg_base; -extern u64 x86_amd_ls_cfg_rds_mask; +extern u64 x86_amd_ls_cfg_ssbd_mask; /* The Intel SPEC CTRL MSR base value cache */ extern u64 x86_spec_ctrl_base; -static inline u64 rds_tif_to_spec_ctrl(u64 tifn) +static inline u64 ssbd_tif_to_spec_ctrl(u64 tifn) { - BUILD_BUG_ON(TIF_RDS < SPEC_CTRL_RDS_SHIFT); - return (tifn & _TIF_RDS) >> (TIF_RDS - SPEC_CTRL_RDS_SHIFT); + BUILD_BUG_ON(TIF_SSBD < SPEC_CTRL_SSBD_SHIFT); + return (tifn & _TIF_SSBD) >> (TIF_SSBD - SPEC_CTRL_SSBD_SHIFT); } -static inline u64 rds_tif_to_amd_ls_cfg(u64 tifn) +static inline u64 ssbd_tif_to_amd_ls_cfg(u64 tifn) { - return (tifn & _TIF_RDS) ? x86_amd_ls_cfg_rds_mask : 0ULL; + return (tifn & _TIF_SSBD) ? x86_amd_ls_cfg_ssbd_mask : 0ULL; } extern void speculative_store_bypass_update(void); diff --git a/arch/x86/include/asm/thread_info.h b/arch/x86/include/asm/thread_info.h index e5c26cc59619..2ff2a30a264f 100644 --- a/arch/x86/include/asm/thread_info.h +++ b/arch/x86/include/asm/thread_info.h @@ -79,7 +79,7 @@ struct thread_info { #define TIF_SIGPENDING 2 /* signal pending */ #define TIF_NEED_RESCHED 3 /* rescheduling necessary */ #define TIF_SINGLESTEP 4 /* reenable singlestep on user return*/ -#define TIF_RDS 5 /* Reduced data speculation */ +#define TIF_SSBD 5 /* Reduced data speculation */ #define TIF_SYSCALL_EMU 6 /* syscall emulation active */ #define TIF_SYSCALL_AUDIT 7 /* syscall auditing active */ #define TIF_SECCOMP 8 /* secure computing */ @@ -106,7 +106,7 @@ struct thread_info { #define _TIF_SIGPENDING (1 << TIF_SIGPENDING) #define _TIF_NEED_RESCHED (1 << TIF_NEED_RESCHED) #define _TIF_SINGLESTEP (1 << TIF_SINGLESTEP) -#define _TIF_RDS (1 << TIF_RDS) +#define _TIF_SSBD (1 << TIF_SSBD) #define _TIF_SYSCALL_EMU (1 << TIF_SYSCALL_EMU) #define _TIF_SYSCALL_AUDIT (1 << TIF_SYSCALL_AUDIT) #define _TIF_SECCOMP (1 << TIF_SECCOMP) @@ -146,7 +146,7 @@ struct thread_info { /* flags to check in __switch_to() */ #define _TIF_WORK_CTXSW \ - (_TIF_IO_BITMAP|_TIF_NOCPUID|_TIF_NOTSC|_TIF_BLOCKSTEP|_TIF_RDS) + (_TIF_IO_BITMAP|_TIF_NOCPUID|_TIF_NOTSC|_TIF_BLOCKSTEP|_TIF_SSBD) #define _TIF_WORK_CTXSW_PREV (_TIF_WORK_CTXSW|_TIF_USER_RETURN_NOTIFY) #define _TIF_WORK_CTXSW_NEXT (_TIF_WORK_CTXSW) diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index 18efc33a8d2e..7bde990b0385 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -567,12 +567,12 @@ static void bsp_init_amd(struct cpuinfo_x86 *c) } /* * Try to cache the base value so further operations can - * avoid RMW. If that faults, do not enable RDS. + * avoid RMW. If that faults, do not enable SSBD. */ if (!rdmsrl_safe(MSR_AMD64_LS_CFG, &x86_amd_ls_cfg_base)) { - setup_force_cpu_cap(X86_FEATURE_RDS); - setup_force_cpu_cap(X86_FEATURE_AMD_RDS); - x86_amd_ls_cfg_rds_mask = 1ULL << bit; + setup_force_cpu_cap(X86_FEATURE_SSBD); + setup_force_cpu_cap(X86_FEATURE_AMD_SSBD); + x86_amd_ls_cfg_ssbd_mask = 1ULL << bit; } } } @@ -920,9 +920,9 @@ static void init_amd(struct cpuinfo_x86 *c) if (!cpu_has(c, X86_FEATURE_XENPV)) set_cpu_bug(c, X86_BUG_SYSRET_SS_ATTRS); - if (boot_cpu_has(X86_FEATURE_AMD_RDS)) { - set_cpu_cap(c, X86_FEATURE_RDS); - set_cpu_cap(c, X86_FEATURE_AMD_RDS); + if (boot_cpu_has(X86_FEATURE_AMD_SSBD)) { + set_cpu_cap(c, X86_FEATURE_SSBD); + set_cpu_cap(c, X86_FEATURE_AMD_SSBD); } } diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index 563d8e54c863..09b116b7f3bf 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -45,10 +45,10 @@ static u64 __ro_after_init x86_spec_ctrl_mask = ~SPEC_CTRL_IBRS; /* * AMD specific MSR info for Speculative Store Bypass control. - * x86_amd_ls_cfg_rds_mask is initialized in identify_boot_cpu(). + * x86_amd_ls_cfg_ssbd_mask is initialized in identify_boot_cpu(). */ u64 __ro_after_init x86_amd_ls_cfg_base; -u64 __ro_after_init x86_amd_ls_cfg_rds_mask; +u64 __ro_after_init x86_amd_ls_cfg_ssbd_mask; void __init check_bugs(void) { @@ -146,7 +146,7 @@ u64 x86_spec_ctrl_get_default(void) u64 msrval = x86_spec_ctrl_base; if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) - msrval |= rds_tif_to_spec_ctrl(current_thread_info()->flags); + msrval |= ssbd_tif_to_spec_ctrl(current_thread_info()->flags); return msrval; } EXPORT_SYMBOL_GPL(x86_spec_ctrl_get_default); @@ -159,7 +159,7 @@ void x86_spec_ctrl_set_guest(u64 guest_spec_ctrl) return; if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) - host |= rds_tif_to_spec_ctrl(current_thread_info()->flags); + host |= ssbd_tif_to_spec_ctrl(current_thread_info()->flags); if (host != guest_spec_ctrl) wrmsrl(MSR_IA32_SPEC_CTRL, guest_spec_ctrl); @@ -174,18 +174,18 @@ void x86_spec_ctrl_restore_host(u64 guest_spec_ctrl) return; if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) - host |= rds_tif_to_spec_ctrl(current_thread_info()->flags); + host |= ssbd_tif_to_spec_ctrl(current_thread_info()->flags); if (host != guest_spec_ctrl) wrmsrl(MSR_IA32_SPEC_CTRL, host); } EXPORT_SYMBOL_GPL(x86_spec_ctrl_restore_host); -static void x86_amd_rds_enable(void) +static void x86_amd_ssb_disable(void) { - u64 msrval = x86_amd_ls_cfg_base | x86_amd_ls_cfg_rds_mask; + u64 msrval = x86_amd_ls_cfg_base | x86_amd_ls_cfg_ssbd_mask; - if (boot_cpu_has(X86_FEATURE_AMD_RDS)) + if (boot_cpu_has(X86_FEATURE_AMD_SSBD)) wrmsrl(MSR_AMD64_LS_CFG, msrval); } @@ -473,7 +473,7 @@ static enum ssb_mitigation_cmd __init __ssb_select_mitigation(void) enum ssb_mitigation mode = SPEC_STORE_BYPASS_NONE; enum ssb_mitigation_cmd cmd; - if (!boot_cpu_has(X86_FEATURE_RDS)) + if (!boot_cpu_has(X86_FEATURE_SSBD)) return mode; cmd = ssb_parse_cmdline(); @@ -507,7 +507,7 @@ static enum ssb_mitigation_cmd __init __ssb_select_mitigation(void) /* * We have three CPU feature flags that are in play here: * - X86_BUG_SPEC_STORE_BYPASS - CPU is susceptible. - * - X86_FEATURE_RDS - CPU is able to turn off speculative store bypass + * - X86_FEATURE_SSBD - CPU is able to turn off speculative store bypass * - X86_FEATURE_SPEC_STORE_BYPASS_DISABLE - engage the mitigation */ if (mode == SPEC_STORE_BYPASS_DISABLE) { @@ -518,12 +518,12 @@ static enum ssb_mitigation_cmd __init __ssb_select_mitigation(void) */ switch (boot_cpu_data.x86_vendor) { case X86_VENDOR_INTEL: - x86_spec_ctrl_base |= SPEC_CTRL_RDS; - x86_spec_ctrl_mask &= ~SPEC_CTRL_RDS; - x86_spec_ctrl_set(SPEC_CTRL_RDS); + x86_spec_ctrl_base |= SPEC_CTRL_SSBD; + x86_spec_ctrl_mask &= ~SPEC_CTRL_SSBD; + x86_spec_ctrl_set(SPEC_CTRL_SSBD); break; case X86_VENDOR_AMD: - x86_amd_rds_enable(); + x86_amd_ssb_disable(); break; } } @@ -556,16 +556,16 @@ static int ssb_prctl_set(struct task_struct *task, unsigned long ctrl) if (task_spec_ssb_force_disable(task)) return -EPERM; task_clear_spec_ssb_disable(task); - update = test_and_clear_tsk_thread_flag(task, TIF_RDS); + update = test_and_clear_tsk_thread_flag(task, TIF_SSBD); break; case PR_SPEC_DISABLE: task_set_spec_ssb_disable(task); - update = !test_and_set_tsk_thread_flag(task, TIF_RDS); + update = !test_and_set_tsk_thread_flag(task, TIF_SSBD); break; case PR_SPEC_FORCE_DISABLE: task_set_spec_ssb_disable(task); task_set_spec_ssb_force_disable(task); - update = !test_and_set_tsk_thread_flag(task, TIF_RDS); + update = !test_and_set_tsk_thread_flag(task, TIF_SSBD); break; default: return -ERANGE; @@ -635,7 +635,7 @@ void x86_spec_ctrl_setup_ap(void) x86_spec_ctrl_set(x86_spec_ctrl_base & ~x86_spec_ctrl_mask); if (ssb_mode == SPEC_STORE_BYPASS_DISABLE) - x86_amd_rds_enable(); + x86_amd_ssb_disable(); } #ifdef CONFIG_SYSFS diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index e0517bcee446..9fbb388fadac 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -959,7 +959,7 @@ static void __init cpu_set_bug_bits(struct cpuinfo_x86 *c) rdmsrl(MSR_IA32_ARCH_CAPABILITIES, ia32_cap); if (!x86_match_cpu(cpu_no_spec_store_bypass) && - !(ia32_cap & ARCH_CAP_RDS_NO)) + !(ia32_cap & ARCH_CAP_SSBD_NO)) setup_force_cpu_bug(X86_BUG_SPEC_STORE_BYPASS); if (x86_match_cpu(cpu_no_speculation)) diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c index ef3f9c01c274..0eab6c89c8d9 100644 --- a/arch/x86/kernel/cpu/intel.c +++ b/arch/x86/kernel/cpu/intel.c @@ -189,7 +189,7 @@ static void early_init_intel(struct cpuinfo_x86 *c) setup_clear_cpu_cap(X86_FEATURE_STIBP); setup_clear_cpu_cap(X86_FEATURE_SPEC_CTRL); setup_clear_cpu_cap(X86_FEATURE_INTEL_STIBP); - setup_clear_cpu_cap(X86_FEATURE_RDS); + setup_clear_cpu_cap(X86_FEATURE_SSBD); } /* diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index 397342725046..b77a091bf3b8 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -283,11 +283,11 @@ static __always_inline void __speculative_store_bypass_update(unsigned long tifn { u64 msr; - if (static_cpu_has(X86_FEATURE_AMD_RDS)) { - msr = x86_amd_ls_cfg_base | rds_tif_to_amd_ls_cfg(tifn); + if (static_cpu_has(X86_FEATURE_AMD_SSBD)) { + msr = x86_amd_ls_cfg_base | ssbd_tif_to_amd_ls_cfg(tifn); wrmsrl(MSR_AMD64_LS_CFG, msr); } else { - msr = x86_spec_ctrl_base | rds_tif_to_spec_ctrl(tifn); + msr = x86_spec_ctrl_base | ssbd_tif_to_spec_ctrl(tifn); wrmsrl(MSR_IA32_SPEC_CTRL, msr); } } @@ -329,7 +329,7 @@ void __switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p, if ((tifp ^ tifn) & _TIF_NOCPUID) set_cpuid_faulting(!!(tifn & _TIF_NOCPUID)); - if ((tifp ^ tifn) & _TIF_RDS) + if ((tifp ^ tifn) & _TIF_SSBD) __speculative_store_bypass_update(tifn); } diff --git a/arch/x86/kvm/cpuid.c b/arch/x86/kvm/cpuid.c index 376ac9a2a2b9..865c9a769864 100644 --- a/arch/x86/kvm/cpuid.c +++ b/arch/x86/kvm/cpuid.c @@ -407,7 +407,7 @@ static inline int __do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 function, /* cpuid 7.0.edx*/ const u32 kvm_cpuid_7_0_edx_x86_features = - F(AVX512_4VNNIW) | F(AVX512_4FMAPS) | F(SPEC_CTRL) | F(RDS) | + F(AVX512_4VNNIW) | F(AVX512_4FMAPS) | F(SPEC_CTRL) | F(SSBD) | F(ARCH_CAPABILITIES); /* all calls to cpuid_count() should be made on the same cpu */ diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 16a111e44691..9b8d80bf3889 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -3525,7 +3525,7 @@ static int vmx_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) if (!msr_info->host_initiated && !guest_cpuid_has(vcpu, X86_FEATURE_IBRS) && !guest_cpuid_has(vcpu, X86_FEATURE_SPEC_CTRL) && - !guest_cpuid_has(vcpu, X86_FEATURE_RDS)) + !guest_cpuid_has(vcpu, X86_FEATURE_SSBD)) return 1; msr_info->data = to_vmx(vcpu)->spec_ctrl; @@ -3645,11 +3645,11 @@ static int vmx_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) if (!msr_info->host_initiated && !guest_cpuid_has(vcpu, X86_FEATURE_IBRS) && !guest_cpuid_has(vcpu, X86_FEATURE_SPEC_CTRL) && - !guest_cpuid_has(vcpu, X86_FEATURE_RDS)) + !guest_cpuid_has(vcpu, X86_FEATURE_SSBD)) return 1; /* The STIBP bit doesn't fault even if it's not advertised */ - if (data & ~(SPEC_CTRL_IBRS | SPEC_CTRL_STIBP | SPEC_CTRL_RDS)) + if (data & ~(SPEC_CTRL_IBRS | SPEC_CTRL_STIBP | SPEC_CTRL_SSBD)) return 1; vmx->spec_ctrl = data; From e96f46ee8587607a828f783daa6eb5b44d25004d Mon Sep 17 00:00:00 2001 From: Konrad Rzeszutek Wilk Date: Wed, 9 May 2018 21:41:38 +0200 Subject: [PATCH 055/190] proc: Use underscores for SSBD in 'status' The style for the 'status' file is CamelCase or this. _. Fixes: fae1fa0fc ("proc: Provide details on speculation flaw mitigations") Signed-off-by: Konrad Rzeszutek Wilk Signed-off-by: Thomas Gleixner --- fs/proc/array.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/proc/array.c b/fs/proc/array.c index d178a0236514..72391b3f6927 100644 --- a/fs/proc/array.c +++ b/fs/proc/array.c @@ -336,7 +336,7 @@ static inline void task_seccomp(struct seq_file *m, struct task_struct *p) #ifdef CONFIG_SECCOMP seq_put_decimal_ull(m, "\nSeccomp:\t", p->seccomp.mode); #endif - seq_printf(m, "\nSpeculation Store Bypass:\t"); + seq_printf(m, "\nSpeculation_Store_Bypass:\t"); switch (arch_prctl_spec_ctrl_get(p, PR_SPEC_STORE_BYPASS)) { case -EINVAL: seq_printf(m, "unknown"); From dd0792699c4058e63c0715d9a7c2d40226fcdddc Mon Sep 17 00:00:00 2001 From: Borislav Petkov Date: Tue, 8 May 2018 15:43:45 +0200 Subject: [PATCH 056/190] Documentation/spec_ctrl: Do some minor cleanups Fix some typos, improve formulations, end sentences with a fullstop. Signed-off-by: Borislav Petkov Signed-off-by: Thomas Gleixner --- Documentation/userspace-api/spec_ctrl.rst | 24 +++++++++++------------ 1 file changed, 12 insertions(+), 12 deletions(-) diff --git a/Documentation/userspace-api/spec_ctrl.rst b/Documentation/userspace-api/spec_ctrl.rst index 1b3690d30943..32f3d55c54b7 100644 --- a/Documentation/userspace-api/spec_ctrl.rst +++ b/Documentation/userspace-api/spec_ctrl.rst @@ -2,13 +2,13 @@ Speculation Control =================== -Quite some CPUs have speculation related misfeatures which are in fact -vulnerabilites causing data leaks in various forms even accross privilege -domains. +Quite some CPUs have speculation-related misfeatures which are in +fact vulnerabilities causing data leaks in various forms even across +privilege domains. The kernel provides mitigation for such vulnerabilities in various -forms. Some of these mitigations are compile time configurable and some on -the kernel command line. +forms. Some of these mitigations are compile-time configurable and some +can be supplied on the kernel command line. There is also a class of mitigations which are very expensive, but they can be restricted to a certain set of processes or tasks in controlled @@ -32,18 +32,18 @@ the following meaning: Bit Define Description ==== ===================== =================================================== 0 PR_SPEC_PRCTL Mitigation can be controlled per task by - PR_SET_SPECULATION_CTRL + PR_SET_SPECULATION_CTRL. 1 PR_SPEC_ENABLE The speculation feature is enabled, mitigation is - disabled + disabled. 2 PR_SPEC_DISABLE The speculation feature is disabled, mitigation is - enabled + enabled. 3 PR_SPEC_FORCE_DISABLE Same as PR_SPEC_DISABLE, but cannot be undone. A subsequent prctl(..., PR_SPEC_ENABLE) will fail. ==== ===================== =================================================== If all bits are 0 the CPU is not affected by the speculation misfeature. -If PR_SPEC_PRCTL is set, then the per task control of the mitigation is +If PR_SPEC_PRCTL is set, then the per-task control of the mitigation is available. If not set, prctl(PR_SET_SPECULATION_CTRL) for the speculation misfeature will fail. @@ -61,9 +61,9 @@ Common error codes Value Meaning ======= ================================================================= EINVAL The prctl is not implemented by the architecture or unused - prctl(2) arguments are not 0 + prctl(2) arguments are not 0. -ENODEV arg2 is selecting a not supported speculation misfeature +ENODEV arg2 is selecting a not supported speculation misfeature. ======= ================================================================= PR_SET_SPECULATION_CTRL error codes @@ -74,7 +74,7 @@ Value Meaning 0 Success ERANGE arg3 is incorrect, i.e. it's neither PR_SPEC_ENABLE nor - PR_SPEC_DISABLE nor PR_SPEC_FORCE_DISABLE + PR_SPEC_DISABLE nor PR_SPEC_FORCE_DISABLE. ENXIO Control of the selected speculation misfeature is not possible. See PR_GET_SPECULATION_CTRL. From 8ba0f9bd3bdea1058c2b2676bec7905724418e40 Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Wed, 9 May 2018 21:07:40 +0200 Subject: [PATCH 057/190] batman-adv: Avoid race in TT TVLV allocator helper The functions batadv_tt_prepare_tvlv_local_data and batadv_tt_prepare_tvlv_global_data are responsible for preparing a buffer which can be used to store the TVLV container for TT and add the VLAN information to it. This will be done in three phases: 1. count the number of VLANs and their entries 2. allocate the buffer using the counters from the previous step and limits from the caller (parameter tt_len) 3. insert the VLAN information to the buffer The step 1 and 3 operate on a list which contains the VLANs. The access to these lists must be protected with an appropriate lock or otherwise they might operate on on different entries. This could for example happen when another context is adding VLAN entries to this list. This could lead to a buffer overflow in these functions when enough entries were added between step 1 and 3 to the VLAN lists that the buffer room for the entries (*tt_change) is smaller then the now required extra buffer for new VLAN entries. Fixes: 7ea7b4a14275 ("batman-adv: make the TT CRC logic VLAN specific") Signed-off-by: Sven Eckelmann Acked-by: Antonio Quartulli Signed-off-by: Simon Wunderlich --- net/batman-adv/translation-table.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c index 0225616d5771..7fa3a0a0524a 100644 --- a/net/batman-adv/translation-table.c +++ b/net/batman-adv/translation-table.c @@ -862,7 +862,7 @@ batadv_tt_prepare_tvlv_global_data(struct batadv_orig_node *orig_node, struct batadv_orig_node_vlan *vlan; u8 *tt_change_ptr; - rcu_read_lock(); + spin_lock_bh(&orig_node->vlan_list_lock); hlist_for_each_entry_rcu(vlan, &orig_node->vlan_list, list) { num_vlan++; num_entries += atomic_read(&vlan->tt.num_entries); @@ -900,7 +900,7 @@ batadv_tt_prepare_tvlv_global_data(struct batadv_orig_node *orig_node, *tt_change = (struct batadv_tvlv_tt_change *)tt_change_ptr; out: - rcu_read_unlock(); + spin_unlock_bh(&orig_node->vlan_list_lock); return tvlv_len; } @@ -936,7 +936,7 @@ batadv_tt_prepare_tvlv_local_data(struct batadv_priv *bat_priv, u8 *tt_change_ptr; int change_offset; - rcu_read_lock(); + spin_lock_bh(&bat_priv->softif_vlan_list_lock); hlist_for_each_entry_rcu(vlan, &bat_priv->softif_vlan_list, list) { num_vlan++; num_entries += atomic_read(&vlan->tt.num_entries); @@ -974,7 +974,7 @@ batadv_tt_prepare_tvlv_local_data(struct batadv_priv *bat_priv, *tt_change = (struct batadv_tvlv_tt_change *)tt_change_ptr; out: - rcu_read_unlock(); + spin_unlock_bh(&bat_priv->softif_vlan_list_lock); return tvlv_len; } From d66d8ff3d21667b41eddbe86b35ab411e40d8c5f Mon Sep 17 00:00:00 2001 From: Jiri Kosina Date: Thu, 10 May 2018 22:47:18 +0200 Subject: [PATCH 058/190] x86/bugs: Fix __ssb_select_mitigation() return type __ssb_select_mitigation() returns one of the members of enum ssb_mitigation, not ssb_mitigation_cmd; fix the prototype to reflect that. Fixes: 24f7fc83b9204 ("x86/bugs: Provide boot parameters for the spec_store_bypass_disable mitigation") Signed-off-by: Jiri Kosina Signed-off-by: Thomas Gleixner --- arch/x86/kernel/cpu/bugs.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index 09b116b7f3bf..3287a42d4df4 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -468,7 +468,7 @@ static enum ssb_mitigation_cmd __init ssb_parse_cmdline(void) return cmd; } -static enum ssb_mitigation_cmd __init __ssb_select_mitigation(void) +static enum ssb_mitigation __init __ssb_select_mitigation(void) { enum ssb_mitigation mode = SPEC_STORE_BYPASS_NONE; enum ssb_mitigation_cmd cmd; From 7bb4d366cba992904bffa4820d24e70a3de93e76 Mon Sep 17 00:00:00 2001 From: Jiri Kosina Date: Thu, 10 May 2018 22:47:32 +0200 Subject: [PATCH 059/190] x86/bugs: Make cpu_show_common() static cpu_show_common() is not used outside of arch/x86/kernel/cpu/bugs.c, so make it static. Signed-off-by: Jiri Kosina Signed-off-by: Thomas Gleixner --- arch/x86/kernel/cpu/bugs.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index 3287a42d4df4..784753f0b41e 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -640,7 +640,7 @@ void x86_spec_ctrl_setup_ap(void) #ifdef CONFIG_SYSFS -ssize_t cpu_show_common(struct device *dev, struct device_attribute *attr, +static ssize_t cpu_show_common(struct device *dev, struct device_attribute *attr, char *buf, unsigned int bug) { if (!boot_cpu_has_bug(bug)) From 1e2e547a93a00ebc21582c06ca3c6cfea2a309ee Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 4 May 2018 08:23:01 -0400 Subject: [PATCH 060/190] do d_instantiate/unlock_new_inode combinations safely For anything NFS-exported we do _not_ want to unlock new inode before it has grown an alias; original set of fixes got the ordering right, but missed the nasty complication in case of lockdep being enabled - unlock_new_inode() does lockdep_annotate_inode_mutex_key(inode) which can only be done before anyone gets a chance to touch ->i_mutex. Unfortunately, flipping the order and doing unlock_new_inode() before d_instantiate() opens a window when mkdir can race with open-by-fhandle on a guessed fhandle, leading to multiple aliases for a directory inode and all the breakage that follows from that. Correct solution: a new primitive (d_instantiate_new()) combining these two in the right order - lockdep annotate, then d_instantiate(), then the rest of unlock_new_inode(). All combinations of d_instantiate() with unlock_new_inode() should be converted to that. Cc: stable@kernel.org # 2.6.29 and later Tested-by: Mike Marshall Reviewed-by: Andreas Dilger Signed-off-by: Al Viro --- fs/btrfs/inode.c | 16 ++++------------ fs/dcache.c | 22 ++++++++++++++++++++++ fs/ecryptfs/inode.c | 3 +-- fs/ext2/namei.c | 6 ++---- fs/ext4/namei.c | 6 ++---- fs/f2fs/namei.c | 12 ++++-------- fs/jffs2/dir.c | 12 ++++-------- fs/jfs/namei.c | 12 ++++-------- fs/nilfs2/namei.c | 6 ++---- fs/orangefs/namei.c | 9 +++------ fs/reiserfs/namei.c | 12 ++++-------- fs/udf/namei.c | 6 ++---- fs/ufs/namei.c | 6 ++---- include/linux/dcache.h | 1 + 14 files changed, 57 insertions(+), 72 deletions(-) diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index e064c49c9a9a..9e97cbb4f006 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -6575,8 +6575,7 @@ static int btrfs_mknod(struct inode *dir, struct dentry *dentry, goto out_unlock_inode; } else { btrfs_update_inode(trans, root, inode); - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); } out_unlock: @@ -6652,8 +6651,7 @@ static int btrfs_create(struct inode *dir, struct dentry *dentry, goto out_unlock_inode; BTRFS_I(inode)->io_tree.ops = &btrfs_extent_io_ops; - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); out_unlock: btrfs_end_transaction(trans); @@ -6798,12 +6796,7 @@ static int btrfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) if (err) goto out_fail_inode; - d_instantiate(dentry, inode); - /* - * mkdir is special. We're unlocking after we call d_instantiate - * to avoid a race with nfsd calling d_instantiate. - */ - unlock_new_inode(inode); + d_instantiate_new(dentry, inode); drop_on_err = 0; out_fail: @@ -10246,8 +10239,7 @@ static int btrfs_symlink(struct inode *dir, struct dentry *dentry, goto out_unlock_inode; } - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); out_unlock: btrfs_end_transaction(trans); diff --git a/fs/dcache.c b/fs/dcache.c index 86d2de63461e..2acfc69878f5 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -1899,6 +1899,28 @@ void d_instantiate(struct dentry *entry, struct inode * inode) } EXPORT_SYMBOL(d_instantiate); +/* + * This should be equivalent to d_instantiate() + unlock_new_inode(), + * with lockdep-related part of unlock_new_inode() done before + * anything else. Use that instead of open-coding d_instantiate()/ + * unlock_new_inode() combinations. + */ +void d_instantiate_new(struct dentry *entry, struct inode *inode) +{ + BUG_ON(!hlist_unhashed(&entry->d_u.d_alias)); + BUG_ON(!inode); + lockdep_annotate_inode_mutex_key(inode); + security_d_instantiate(entry, inode); + spin_lock(&inode->i_lock); + __d_instantiate(entry, inode); + WARN_ON(!(inode->i_state & I_NEW)); + inode->i_state &= ~I_NEW; + smp_mb(); + wake_up_bit(&inode->i_state, __I_NEW); + spin_unlock(&inode->i_lock); +} +EXPORT_SYMBOL(d_instantiate_new); + /** * d_instantiate_no_diralias - instantiate a non-aliased dentry * @entry: dentry to complete diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c index 847904aa63a9..7bba8f2693b2 100644 --- a/fs/ecryptfs/inode.c +++ b/fs/ecryptfs/inode.c @@ -283,8 +283,7 @@ ecryptfs_create(struct inode *directory_inode, struct dentry *ecryptfs_dentry, iget_failed(ecryptfs_inode); goto out; } - unlock_new_inode(ecryptfs_inode); - d_instantiate(ecryptfs_dentry, ecryptfs_inode); + d_instantiate_new(ecryptfs_dentry, ecryptfs_inode); out: return rc; } diff --git a/fs/ext2/namei.c b/fs/ext2/namei.c index 55f7caadb093..152453a91877 100644 --- a/fs/ext2/namei.c +++ b/fs/ext2/namei.c @@ -41,8 +41,7 @@ static inline int ext2_add_nondir(struct dentry *dentry, struct inode *inode) { int err = ext2_add_link(dentry, inode); if (!err) { - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); return 0; } inode_dec_link_count(inode); @@ -255,8 +254,7 @@ static int ext2_mkdir(struct inode * dir, struct dentry * dentry, umode_t mode) if (err) goto out_fail; - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); out: return err; diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index b1f21e3a0763..4a09063ce1d2 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -2411,8 +2411,7 @@ static int ext4_add_nondir(handle_t *handle, int err = ext4_add_entry(handle, dentry, inode); if (!err) { ext4_mark_inode_dirty(handle, inode); - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); return 0; } drop_nlink(inode); @@ -2651,8 +2650,7 @@ static int ext4_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) err = ext4_mark_inode_dirty(handle, dir); if (err) goto out_clear_inode; - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); if (IS_DIRSYNC(dir)) ext4_handle_sync(handle); diff --git a/fs/f2fs/namei.c b/fs/f2fs/namei.c index d5098efe577c..75e37fd720b2 100644 --- a/fs/f2fs/namei.c +++ b/fs/f2fs/namei.c @@ -294,8 +294,7 @@ static int f2fs_create(struct inode *dir, struct dentry *dentry, umode_t mode, alloc_nid_done(sbi, ino); - d_instantiate(dentry, inode); - unlock_new_inode(inode); + d_instantiate_new(dentry, inode); if (IS_DIRSYNC(dir)) f2fs_sync_fs(sbi->sb, 1); @@ -597,8 +596,7 @@ static int f2fs_symlink(struct inode *dir, struct dentry *dentry, err = page_symlink(inode, disk_link.name, disk_link.len); err_out: - d_instantiate(dentry, inode); - unlock_new_inode(inode); + d_instantiate_new(dentry, inode); /* * Let's flush symlink data in order to avoid broken symlink as much as @@ -661,8 +659,7 @@ static int f2fs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) alloc_nid_done(sbi, inode->i_ino); - d_instantiate(dentry, inode); - unlock_new_inode(inode); + d_instantiate_new(dentry, inode); if (IS_DIRSYNC(dir)) f2fs_sync_fs(sbi->sb, 1); @@ -713,8 +710,7 @@ static int f2fs_mknod(struct inode *dir, struct dentry *dentry, alloc_nid_done(sbi, inode->i_ino); - d_instantiate(dentry, inode); - unlock_new_inode(inode); + d_instantiate_new(dentry, inode); if (IS_DIRSYNC(dir)) f2fs_sync_fs(sbi->sb, 1); diff --git a/fs/jffs2/dir.c b/fs/jffs2/dir.c index 0a754f38462e..e5a6deb38e1e 100644 --- a/fs/jffs2/dir.c +++ b/fs/jffs2/dir.c @@ -209,8 +209,7 @@ static int jffs2_create(struct inode *dir_i, struct dentry *dentry, __func__, inode->i_ino, inode->i_mode, inode->i_nlink, f->inocache->pino_nlink, inode->i_mapping->nrpages); - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); return 0; fail: @@ -430,8 +429,7 @@ static int jffs2_symlink (struct inode *dir_i, struct dentry *dentry, const char mutex_unlock(&dir_f->sem); jffs2_complete_reservation(c); - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); return 0; fail: @@ -575,8 +573,7 @@ static int jffs2_mkdir (struct inode *dir_i, struct dentry *dentry, umode_t mode mutex_unlock(&dir_f->sem); jffs2_complete_reservation(c); - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); return 0; fail: @@ -747,8 +744,7 @@ static int jffs2_mknod (struct inode *dir_i, struct dentry *dentry, umode_t mode mutex_unlock(&dir_f->sem); jffs2_complete_reservation(c); - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); return 0; fail: diff --git a/fs/jfs/namei.c b/fs/jfs/namei.c index b41596d71858..56c3fcbfe80e 100644 --- a/fs/jfs/namei.c +++ b/fs/jfs/namei.c @@ -178,8 +178,7 @@ static int jfs_create(struct inode *dip, struct dentry *dentry, umode_t mode, unlock_new_inode(ip); iput(ip); } else { - unlock_new_inode(ip); - d_instantiate(dentry, ip); + d_instantiate_new(dentry, ip); } out2: @@ -313,8 +312,7 @@ static int jfs_mkdir(struct inode *dip, struct dentry *dentry, umode_t mode) unlock_new_inode(ip); iput(ip); } else { - unlock_new_inode(ip); - d_instantiate(dentry, ip); + d_instantiate_new(dentry, ip); } out2: @@ -1059,8 +1057,7 @@ static int jfs_symlink(struct inode *dip, struct dentry *dentry, unlock_new_inode(ip); iput(ip); } else { - unlock_new_inode(ip); - d_instantiate(dentry, ip); + d_instantiate_new(dentry, ip); } out2: @@ -1447,8 +1444,7 @@ static int jfs_mknod(struct inode *dir, struct dentry *dentry, unlock_new_inode(ip); iput(ip); } else { - unlock_new_inode(ip); - d_instantiate(dentry, ip); + d_instantiate_new(dentry, ip); } out1: diff --git a/fs/nilfs2/namei.c b/fs/nilfs2/namei.c index 1a2894aa0194..dd52d3f82e8d 100644 --- a/fs/nilfs2/namei.c +++ b/fs/nilfs2/namei.c @@ -46,8 +46,7 @@ static inline int nilfs_add_nondir(struct dentry *dentry, struct inode *inode) int err = nilfs_add_link(dentry, inode); if (!err) { - d_instantiate(dentry, inode); - unlock_new_inode(inode); + d_instantiate_new(dentry, inode); return 0; } inode_dec_link_count(inode); @@ -243,8 +242,7 @@ static int nilfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) goto out_fail; nilfs_mark_inode_dirty(inode); - d_instantiate(dentry, inode); - unlock_new_inode(inode); + d_instantiate_new(dentry, inode); out: if (!err) err = nilfs_transaction_commit(dir->i_sb); diff --git a/fs/orangefs/namei.c b/fs/orangefs/namei.c index 6e3134e6d98a..1b5707c44c3f 100644 --- a/fs/orangefs/namei.c +++ b/fs/orangefs/namei.c @@ -75,8 +75,7 @@ static int orangefs_create(struct inode *dir, get_khandle_from_ino(inode), dentry); - d_instantiate(dentry, inode); - unlock_new_inode(inode); + d_instantiate_new(dentry, inode); orangefs_set_timeout(dentry); ORANGEFS_I(inode)->getattr_time = jiffies - 1; ORANGEFS_I(inode)->getattr_mask = STATX_BASIC_STATS; @@ -332,8 +331,7 @@ static int orangefs_symlink(struct inode *dir, "Assigned symlink inode new number of %pU\n", get_khandle_from_ino(inode)); - d_instantiate(dentry, inode); - unlock_new_inode(inode); + d_instantiate_new(dentry, inode); orangefs_set_timeout(dentry); ORANGEFS_I(inode)->getattr_time = jiffies - 1; ORANGEFS_I(inode)->getattr_mask = STATX_BASIC_STATS; @@ -402,8 +400,7 @@ static int orangefs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode "Assigned dir inode new number of %pU\n", get_khandle_from_ino(inode)); - d_instantiate(dentry, inode); - unlock_new_inode(inode); + d_instantiate_new(dentry, inode); orangefs_set_timeout(dentry); ORANGEFS_I(inode)->getattr_time = jiffies - 1; ORANGEFS_I(inode)->getattr_mask = STATX_BASIC_STATS; diff --git a/fs/reiserfs/namei.c b/fs/reiserfs/namei.c index bd39a998843d..5089dac02660 100644 --- a/fs/reiserfs/namei.c +++ b/fs/reiserfs/namei.c @@ -687,8 +687,7 @@ static int reiserfs_create(struct inode *dir, struct dentry *dentry, umode_t mod reiserfs_update_inode_transaction(inode); reiserfs_update_inode_transaction(dir); - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); retval = journal_end(&th); out_failed: @@ -771,8 +770,7 @@ static int reiserfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode goto out_failed; } - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); retval = journal_end(&th); out_failed: @@ -871,8 +869,7 @@ static int reiserfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode /* the above add_entry did not update dir's stat data */ reiserfs_update_sd(&th, dir); - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); retval = journal_end(&th); out_failed: reiserfs_write_unlock(dir->i_sb); @@ -1187,8 +1184,7 @@ static int reiserfs_symlink(struct inode *parent_dir, goto out_failed; } - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); retval = journal_end(&th); out_failed: reiserfs_write_unlock(parent_dir->i_sb); diff --git a/fs/udf/namei.c b/fs/udf/namei.c index 0458dd47e105..c586026508db 100644 --- a/fs/udf/namei.c +++ b/fs/udf/namei.c @@ -622,8 +622,7 @@ static int udf_add_nondir(struct dentry *dentry, struct inode *inode) if (fibh.sbh != fibh.ebh) brelse(fibh.ebh); brelse(fibh.sbh); - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); return 0; } @@ -733,8 +732,7 @@ static int udf_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) inc_nlink(dir); dir->i_ctime = dir->i_mtime = current_time(dir); mark_inode_dirty(dir); - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); if (fibh.sbh != fibh.ebh) brelse(fibh.ebh); brelse(fibh.sbh); diff --git a/fs/ufs/namei.c b/fs/ufs/namei.c index 32545cd00ceb..d5f43ba76c59 100644 --- a/fs/ufs/namei.c +++ b/fs/ufs/namei.c @@ -39,8 +39,7 @@ static inline int ufs_add_nondir(struct dentry *dentry, struct inode *inode) { int err = ufs_add_link(dentry, inode); if (!err) { - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); return 0; } inode_dec_link_count(inode); @@ -193,8 +192,7 @@ static int ufs_mkdir(struct inode * dir, struct dentry * dentry, umode_t mode) if (err) goto out_fail; - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); return 0; out_fail: diff --git a/include/linux/dcache.h b/include/linux/dcache.h index 94acbde17bb1..66c6e17e61e5 100644 --- a/include/linux/dcache.h +++ b/include/linux/dcache.h @@ -224,6 +224,7 @@ extern seqlock_t rename_lock; * These are the low-level FS interfaces to the dcache.. */ extern void d_instantiate(struct dentry *, struct inode *); +extern void d_instantiate_new(struct dentry *, struct inode *); extern struct dentry * d_instantiate_unique(struct dentry *, struct inode *); extern struct dentry * d_instantiate_anon(struct dentry *, struct inode *); extern int d_instantiate_no_diralias(struct dentry *, struct inode *); From 79f546a696bff2590169fb5684e23d65f4d9f591 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Fri, 11 May 2018 11:20:57 +1000 Subject: [PATCH 061/190] fs: don't scan the inode cache before SB_BORN is set We recently had an oops reported on a 4.14 kernel in xfs_reclaim_inodes_count() where sb->s_fs_info pointed to garbage and so the m_perag_tree lookup walked into lala land. It produces an oops down this path during the failed mount: radix_tree_gang_lookup_tag+0xc4/0x130 xfs_perag_get_tag+0x37/0xf0 xfs_reclaim_inodes_count+0x32/0x40 xfs_fs_nr_cached_objects+0x11/0x20 super_cache_count+0x35/0xc0 shrink_slab.part.66+0xb1/0x370 shrink_node+0x7e/0x1a0 try_to_free_pages+0x199/0x470 __alloc_pages_slowpath+0x3a1/0xd20 __alloc_pages_nodemask+0x1c3/0x200 cache_grow_begin+0x20b/0x2e0 fallback_alloc+0x160/0x200 kmem_cache_alloc+0x111/0x4e0 The problem is that the superblock shrinker is running before the filesystem structures it depends on have been fully set up. i.e. the shrinker is registered in sget(), before ->fill_super() has been called, and the shrinker can call into the filesystem before fill_super() does it's setup work. Essentially we are exposed to both use-after-free and use-before-initialisation bugs here. To fix this, add a check for the SB_BORN flag in super_cache_count. In general, this flag is not set until ->fs_mount() completes successfully, so we know that it is set after the filesystem setup has completed. This matches the trylock_super() behaviour which will not let super_cache_scan() run if SB_BORN is not set, and hence will not allow the superblock shrinker from entering the filesystem while it is being set up or after it has failed setup and is being torn down. Cc: stable@kernel.org Signed-Off-By: Dave Chinner Signed-off-by: Al Viro --- fs/super.c | 30 ++++++++++++++++++++++++------ 1 file changed, 24 insertions(+), 6 deletions(-) diff --git a/fs/super.c b/fs/super.c index 122c402049a2..4b5b562176d0 100644 --- a/fs/super.c +++ b/fs/super.c @@ -121,13 +121,23 @@ static unsigned long super_cache_count(struct shrinker *shrink, sb = container_of(shrink, struct super_block, s_shrink); /* - * Don't call trylock_super as it is a potential - * scalability bottleneck. The counts could get updated - * between super_cache_count and super_cache_scan anyway. - * Call to super_cache_count with shrinker_rwsem held - * ensures the safety of call to list_lru_shrink_count() and - * s_op->nr_cached_objects(). + * We don't call trylock_super() here as it is a scalability bottleneck, + * so we're exposed to partial setup state. The shrinker rwsem does not + * protect filesystem operations backing list_lru_shrink_count() or + * s_op->nr_cached_objects(). Counts can change between + * super_cache_count and super_cache_scan, so we really don't need locks + * here. + * + * However, if we are currently mounting the superblock, the underlying + * filesystem might be in a state of partial construction and hence it + * is dangerous to access it. trylock_super() uses a SB_BORN check to + * avoid this situation, so do the same here. The memory barrier is + * matched with the one in mount_fs() as we don't hold locks here. */ + if (!(sb->s_flags & SB_BORN)) + return 0; + smp_rmb(); + if (sb->s_op && sb->s_op->nr_cached_objects) total_objects = sb->s_op->nr_cached_objects(sb, sc); @@ -1272,6 +1282,14 @@ mount_fs(struct file_system_type *type, int flags, const char *name, void *data) sb = root->d_sb; BUG_ON(!sb); WARN_ON(!sb->s_bdi); + + /* + * Write barrier is for super_cache_count(). We place it before setting + * SB_BORN as the data dependency between the two functions is the + * superblock structure contents that we just set up, not the SB_BORN + * flag. + */ + smp_wmb(); sb->s_flags |= SB_BORN; error = security_sb_kern_mount(sb, flags, secdata); From 361de091a4b97aa9081d304d742f80d486ab7125 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= Date: Tue, 8 May 2018 11:31:04 +0200 Subject: [PATCH 062/190] bcma: fix buffer size caused crash in bcma_core_mips_print_irq() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Used buffer wasn't big enough to hold whole strings. Example output of this function is: [ 0.180892] bcma: bus0: core 0x0800, irq: 2(S)* 3 4 5 6 D I [ 0.180948] bcma: bus0: core 0x0812, irq: 2(S) 3* 4 5 6 D I [ 0.180998] bcma: bus0: core 0x082d, irq: 2(S) 3 4* 5 6 D I [ 0.181046] bcma: bus0: core 0x082c, irq: 2(S) 3 4 5 6 D I* which means we need to store up to 24 chars. Fixes: 758f7e06063a8 ("bcma: Use bcma_debug and not pr_cont in MIPS driver") Signed-off-by: RafaÅ‚ MiÅ‚ecki Cc: stable@vger.kernel.org # v4.15+ Signed-off-by: Kalle Valo --- drivers/bcma/driver_mips.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/bcma/driver_mips.c b/drivers/bcma/driver_mips.c index f040aba48d50..27e9686b6d3a 100644 --- a/drivers/bcma/driver_mips.c +++ b/drivers/bcma/driver_mips.c @@ -184,7 +184,7 @@ static void bcma_core_mips_print_irq(struct bcma_device *dev, unsigned int irq) { int i; static const char *irq_name[] = {"2(S)", "3", "4", "5", "6", "D", "I"}; - char interrupts[20]; + char interrupts[25]; char *ints = interrupts; for (i = 0; i < ARRAY_SIZE(irq_name); i++) From 36910d82a80c1c0c61e505c6d3ecaa901ee13a26 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= Date: Fri, 11 May 2018 11:17:14 +0200 Subject: [PATCH 063/190] Revert "ssb: Prevent build of PCI host features in module" MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This reverts commit 882164a4a928bcaa53280940436ca476e6b1db8e. Above commit added "SSB = y" dependency to the wrong symbol SSB_DRIVER_PCICORE_POSSIBLE and prevented SSB_DRIVER_PCICORE from being selected when needed. PCI core driver for core running in clienthost mode is important for bus initialization. It's perfectly valid scenario to have ssb built as module and use it with buses on PCI cards. This fixes regression that affected all *module* users with PCI cards. Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=1572349 Signed-off-by: RafaÅ‚ MiÅ‚ecki Signed-off-by: Kalle Valo --- drivers/ssb/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/ssb/Kconfig b/drivers/ssb/Kconfig index 9371651d8017..b3f5cae98ea6 100644 --- a/drivers/ssb/Kconfig +++ b/drivers/ssb/Kconfig @@ -117,7 +117,7 @@ config SSB_SERIAL config SSB_DRIVER_PCICORE_POSSIBLE bool - depends on SSB_PCIHOST && SSB = y + depends on SSB_PCIHOST default y config SSB_DRIVER_PCICORE From ebd27d3317c6521a9511f779ea96dc943c4e8003 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= Date: Fri, 11 May 2018 11:17:15 +0200 Subject: [PATCH 064/190] ssb: make SSB_PCICORE_HOSTMODE depend on SSB = y MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit SSB_PCICORE_HOSTMODE protects MIPS specific code that calls not exported symbols pcibios_enable_device and register_pci_controller. This code is supposed to be compiled only with ssb builtin. This fixes: ERROR: "pcibios_enable_device" [drivers/ssb/ssb.ko] undefined! ERROR: "register_pci_controller" [drivers/ssb/ssb.ko] undefined! make[1]: *** [scripts/Makefile.modpost:92: __modpost] Error 1 Reported-by: Matt Redfearn Signed-off-by: RafaÅ‚ MiÅ‚ecki Reviewed-by: Larry Finger Signed-off-by: Kalle Valo --- drivers/ssb/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/ssb/Kconfig b/drivers/ssb/Kconfig index b3f5cae98ea6..c574dd210500 100644 --- a/drivers/ssb/Kconfig +++ b/drivers/ssb/Kconfig @@ -131,7 +131,7 @@ config SSB_DRIVER_PCICORE config SSB_PCICORE_HOSTMODE bool "Hostmode support for SSB PCI core" - depends on SSB_DRIVER_PCICORE && SSB_DRIVER_MIPS + depends on SSB_DRIVER_PCICORE && SSB_DRIVER_MIPS && SSB = y help PCIcore hostmode operation (external PCI bus). From ffed645e3be0e32f8e9ab068d257aee8d0fe8eec Mon Sep 17 00:00:00 2001 From: Konrad Rzeszutek Wilk Date: Fri, 11 May 2018 16:50:35 -0400 Subject: [PATCH 065/190] x86/bugs: Fix the parameters alignment and missing void Fixes: 7bb4d366c ("x86/bugs: Make cpu_show_common() static") Fixes: 24f7fc83b ("x86/bugs: Provide boot parameters for the spec_store_bypass_disable mitigation") Signed-off-by: Konrad Rzeszutek Wilk Signed-off-by: Thomas Gleixner --- arch/x86/kernel/cpu/bugs.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index 784753f0b41e..bab05913f864 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -531,7 +531,7 @@ static enum ssb_mitigation __init __ssb_select_mitigation(void) return mode; } -static void ssb_select_mitigation() +static void ssb_select_mitigation(void) { ssb_mode = __ssb_select_mitigation(); @@ -641,7 +641,7 @@ void x86_spec_ctrl_setup_ap(void) #ifdef CONFIG_SYSFS static ssize_t cpu_show_common(struct device *dev, struct device_attribute *attr, - char *buf, unsigned int bug) + char *buf, unsigned int bug) { if (!boot_cpu_has_bug(bug)) return sprintf(buf, "Not affected\n"); From 7072337e52b3e9d5460500d8dc9cbc1ba2db084c Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Linus=20L=C3=BCssing?= Date: Thu, 10 May 2018 19:44:28 +0200 Subject: [PATCH 066/190] batman-adv: Fix TT sync flags for intermediate TT responses MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The previous TT sync fix so far only fixed TT responses issued by the target node directly. So far, TT responses issued by intermediate nodes still lead to the wrong flags being added, leading to CRC mismatches. This behaviour was observed at Freifunk Hannover in a 800 nodes setup where a considerable amount of nodes were still infected with 'WI' TT flags even with (most) nodes having the previous TT sync fix applied. I was able to reproduce the issue with intermediate TT responses in a four node test setup and this patch fixes this issue by ensuring to use the per originator instead of the summarized, OR'd ones. Fixes: e9c00136a475 ("batman-adv: fix tt_global_entries flags update") Reported-by: Leonardo Mörlein Signed-off-by: Linus Lüssing Signed-off-by: Sven Eckelmann Signed-off-by: Simon Wunderlich --- net/batman-adv/translation-table.c | 61 +++++++++++++++++++++++++----- 1 file changed, 51 insertions(+), 10 deletions(-) diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c index 7fa3a0a0524a..23f9c212ab1e 100644 --- a/net/batman-adv/translation-table.c +++ b/net/batman-adv/translation-table.c @@ -1538,6 +1538,8 @@ batadv_tt_global_orig_entry_find(const struct batadv_tt_global_entry *entry, * handled by a given originator * @entry: the TT global entry to check * @orig_node: the originator to search in the list + * @flags: a pointer to store TT flags for the given @entry received + * from @orig_node * * find out if an orig_node is already in the list of a tt_global_entry. * @@ -1545,7 +1547,8 @@ batadv_tt_global_orig_entry_find(const struct batadv_tt_global_entry *entry, */ static bool batadv_tt_global_entry_has_orig(const struct batadv_tt_global_entry *entry, - const struct batadv_orig_node *orig_node) + const struct batadv_orig_node *orig_node, + u8 *flags) { struct batadv_tt_orig_list_entry *orig_entry; bool found = false; @@ -1553,6 +1556,10 @@ batadv_tt_global_entry_has_orig(const struct batadv_tt_global_entry *entry, orig_entry = batadv_tt_global_orig_entry_find(entry, orig_node); if (orig_entry) { found = true; + + if (flags) + *flags = orig_entry->flags; + batadv_tt_orig_list_entry_put(orig_entry); } @@ -1731,7 +1738,7 @@ static bool batadv_tt_global_add(struct batadv_priv *bat_priv, if (!(common->flags & BATADV_TT_CLIENT_TEMP)) goto out; if (batadv_tt_global_entry_has_orig(tt_global_entry, - orig_node)) + orig_node, NULL)) goto out_remove; batadv_tt_global_del_orig_list(tt_global_entry); goto add_orig_entry; @@ -2880,23 +2887,46 @@ batadv_tt_req_node_new(struct batadv_priv *bat_priv, } /** - * batadv_tt_local_valid() - verify that given tt entry is a valid one + * batadv_tt_local_valid() - verify local tt entry and get flags * @entry_ptr: to be checked local tt entry * @data_ptr: not used but definition required to satisfy the callback prototype + * @flags: a pointer to store TT flags for this client to + * + * Checks the validity of the given local TT entry. If it is, then the provided + * flags pointer is updated. * * Return: true if the entry is a valid, false otherwise. */ -static bool batadv_tt_local_valid(const void *entry_ptr, const void *data_ptr) +static bool batadv_tt_local_valid(const void *entry_ptr, + const void *data_ptr, + u8 *flags) { const struct batadv_tt_common_entry *tt_common_entry = entry_ptr; if (tt_common_entry->flags & BATADV_TT_CLIENT_NEW) return false; + + if (flags) + *flags = tt_common_entry->flags; + return true; } +/** + * batadv_tt_global_valid() - verify global tt entry and get flags + * @entry_ptr: to be checked global tt entry + * @data_ptr: an orig_node object (may be NULL) + * @flags: a pointer to store TT flags for this client to + * + * Checks the validity of the given global TT entry. If it is, then the provided + * flags pointer is updated either with the common (summed) TT flags if data_ptr + * is NULL or the specific, per originator TT flags otherwise. + * + * Return: true if the entry is a valid, false otherwise. + */ static bool batadv_tt_global_valid(const void *entry_ptr, - const void *data_ptr) + const void *data_ptr, + u8 *flags) { const struct batadv_tt_common_entry *tt_common_entry = entry_ptr; const struct batadv_tt_global_entry *tt_global_entry; @@ -2910,7 +2940,8 @@ static bool batadv_tt_global_valid(const void *entry_ptr, struct batadv_tt_global_entry, common); - return batadv_tt_global_entry_has_orig(tt_global_entry, orig_node); + return batadv_tt_global_entry_has_orig(tt_global_entry, orig_node, + flags); } /** @@ -2920,25 +2951,34 @@ static bool batadv_tt_global_valid(const void *entry_ptr, * @hash: hash table containing the tt entries * @tt_len: expected tvlv tt data buffer length in number of bytes * @tvlv_buff: pointer to the buffer to fill with the TT data - * @valid_cb: function to filter tt change entries + * @valid_cb: function to filter tt change entries and to return TT flags * @cb_data: data passed to the filter function as argument + * + * Fills the tvlv buff with the tt entries from the specified hash. If valid_cb + * is not provided then this becomes a no-op. */ static void batadv_tt_tvlv_generate(struct batadv_priv *bat_priv, struct batadv_hashtable *hash, void *tvlv_buff, u16 tt_len, bool (*valid_cb)(const void *, - const void *), + const void *, + u8 *flags), void *cb_data) { struct batadv_tt_common_entry *tt_common_entry; struct batadv_tvlv_tt_change *tt_change; struct hlist_head *head; u16 tt_tot, tt_num_entries = 0; + u8 flags; + bool ret; u32 i; tt_tot = batadv_tt_entries(tt_len); tt_change = (struct batadv_tvlv_tt_change *)tvlv_buff; + if (!valid_cb) + return; + rcu_read_lock(); for (i = 0; i < hash->size; i++) { head = &hash->table[i]; @@ -2948,11 +2988,12 @@ static void batadv_tt_tvlv_generate(struct batadv_priv *bat_priv, if (tt_tot == tt_num_entries) break; - if ((valid_cb) && (!valid_cb(tt_common_entry, cb_data))) + ret = valid_cb(tt_common_entry, cb_data, &flags); + if (!ret) continue; ether_addr_copy(tt_change->addr, tt_common_entry->addr); - tt_change->flags = tt_common_entry->flags; + tt_change->flags = flags; tt_change->vid = htons(tt_common_entry->vid); memset(tt_change->reserved, 0, sizeof(tt_change->reserved)); From 16116dac23396e73c01eeee97b102e4833a4b205 Mon Sep 17 00:00:00 2001 From: Marek Lindner Date: Sat, 12 May 2018 00:23:07 +0800 Subject: [PATCH 067/190] batman-adv: prevent TT request storms by not sending inconsistent TT TLVLs A translation table TVLV changset sent with an OGM consists of a number of headers (one per VLAN) plus the changeset itself (addition and/or deletion of entries). The per-VLAN headers are used by OGM recipients for consistency checks. Said consistency check might determine that a full translation table request is needed to restore consistency. If the TT sender adds per-VLAN headers of empty VLANs into the OGM, recipients are led to believe to have reached an inconsistent state and thus request a full table update. The full table does not contain empty VLANs (due to missing entries) the cycle restarts when the next OGM is issued. Consequently, when the translation table TVLV headers are composed, empty VLANs are to be excluded. Fixes: 21a57f6e7a3b ("batman-adv: make the TT CRC logic VLAN specific") Signed-off-by: Marek Lindner Signed-off-by: Sven Eckelmann Signed-off-by: Simon Wunderlich --- net/batman-adv/translation-table.c | 15 ++++++++++++--- 1 file changed, 12 insertions(+), 3 deletions(-) diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c index 23f9c212ab1e..3986551397ca 100644 --- a/net/batman-adv/translation-table.c +++ b/net/batman-adv/translation-table.c @@ -931,15 +931,20 @@ batadv_tt_prepare_tvlv_local_data(struct batadv_priv *bat_priv, struct batadv_tvlv_tt_vlan_data *tt_vlan; struct batadv_softif_vlan *vlan; u16 num_vlan = 0; - u16 num_entries = 0; + u16 vlan_entries = 0; + u16 total_entries = 0; u16 tvlv_len; u8 *tt_change_ptr; int change_offset; spin_lock_bh(&bat_priv->softif_vlan_list_lock); hlist_for_each_entry_rcu(vlan, &bat_priv->softif_vlan_list, list) { + vlan_entries = atomic_read(&vlan->tt.num_entries); + if (vlan_entries < 1) + continue; + num_vlan++; - num_entries += atomic_read(&vlan->tt.num_entries); + total_entries += vlan_entries; } change_offset = sizeof(**tt_data); @@ -947,7 +952,7 @@ batadv_tt_prepare_tvlv_local_data(struct batadv_priv *bat_priv, /* if tt_len is negative, allocate the space needed by the full table */ if (*tt_len < 0) - *tt_len = batadv_tt_len(num_entries); + *tt_len = batadv_tt_len(total_entries); tvlv_len = *tt_len; tvlv_len += change_offset; @@ -964,6 +969,10 @@ batadv_tt_prepare_tvlv_local_data(struct batadv_priv *bat_priv, tt_vlan = (struct batadv_tvlv_tt_vlan_data *)(*tt_data + 1); hlist_for_each_entry_rcu(vlan, &bat_priv->softif_vlan_list, list) { + vlan_entries = atomic_read(&vlan->tt.num_entries); + if (vlan_entries < 1) + continue; + tt_vlan->vid = htons(vlan->vid); tt_vlan->crc = htonl(vlan->tt.crc); From b127125d9db23e4856156a7c909a3c8e18b69f99 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 25 Apr 2018 10:28:38 -0400 Subject: [PATCH 068/190] fix breakage caused by d_find_alias() semantics change "VFS: don't keep disconnected dentries on d_anon" had a non-trivial side-effect - d_unhashed() now returns true for those dentries, making d_find_alias() skip them altogether. For most of its callers that's fine - we really want a connected alias there. However, there is a codepath where we relied upon picking such aliases if nothing else could be found - selinux delayed initialization of contexts for inodes on already mounted filesystems used to rely upon that. Cc: stable@kernel.org # f1ee616214cb "VFS: don't keep disconnected dentries on d_anon" Signed-off-by: Al Viro --- security/selinux/hooks.c | 20 ++++++++++++++++---- 1 file changed, 16 insertions(+), 4 deletions(-) diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c index 4cafe6a19167..398d165f884e 100644 --- a/security/selinux/hooks.c +++ b/security/selinux/hooks.c @@ -1568,8 +1568,15 @@ static int inode_doinit_with_dentry(struct inode *inode, struct dentry *opt_dent /* Called from d_instantiate or d_splice_alias. */ dentry = dget(opt_dentry); } else { - /* Called from selinux_complete_init, try to find a dentry. */ + /* + * Called from selinux_complete_init, try to find a dentry. + * Some filesystems really want a connected one, so try + * that first. We could split SECURITY_FS_USE_XATTR in + * two, depending upon that... + */ dentry = d_find_alias(inode); + if (!dentry) + dentry = d_find_any_alias(inode); } if (!dentry) { /* @@ -1674,14 +1681,19 @@ static int inode_doinit_with_dentry(struct inode *inode, struct dentry *opt_dent if ((sbsec->flags & SE_SBGENFS) && !S_ISLNK(inode->i_mode)) { /* We must have a dentry to determine the label on * procfs inodes */ - if (opt_dentry) + if (opt_dentry) { /* Called from d_instantiate or * d_splice_alias. */ dentry = dget(opt_dentry); - else + } else { /* Called from selinux_complete_init, try to - * find a dentry. */ + * find a dentry. Some filesystems really want + * a connected one, so try that first. + */ dentry = d_find_alias(inode); + if (!dentry) + dentry = d_find_any_alias(inode); + } /* * This can be hit on boot when a file is accessed * before the policy is loaded. When we load policy we From 5f2b745f5e1304f438f9b2cd03ebc8120b6e0d3b Mon Sep 17 00:00:00 2001 From: Jim Mattson Date: Sun, 13 May 2018 17:33:57 -0400 Subject: [PATCH 069/190] x86/cpu: Make alternative_msr_write work for 32-bit code Cast val and (val >> 32) to (u32), so that they fit in a general-purpose register in both 32-bit and 64-bit code. [ tglx: Made it u32 instead of uintptr_t ] Fixes: c65732e4f721 ("x86/cpu: Restore CPUID_8000_0008_EBX reload") Signed-off-by: Jim Mattson Signed-off-by: Konrad Rzeszutek Wilk Signed-off-by: Thomas Gleixner Acked-by: Linus Torvalds --- arch/x86/include/asm/nospec-branch.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/x86/include/asm/nospec-branch.h b/arch/x86/include/asm/nospec-branch.h index 328ea3cb769f..bc258e644e5e 100644 --- a/arch/x86/include/asm/nospec-branch.h +++ b/arch/x86/include/asm/nospec-branch.h @@ -265,8 +265,8 @@ void alternative_msr_write(unsigned int msr, u64 val, unsigned int feature) { asm volatile(ALTERNATIVE("", "wrmsr", %c[feature]) : : "c" (msr), - "a" (val), - "d" (val >> 32), + "a" ((u32)val), + "d" ((u32)(val >> 32)), [feature] "i" (feature) : "memory"); } From 76936e9a6df17b89481bd2655c8684291afbe656 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Fran=C3=A7ois=20Cami?= Date: Sun, 13 May 2018 20:11:15 +0200 Subject: [PATCH 070/190] libata: Apply NOLPM quirk for SAMSUNG PM830 CXM13D1Q. MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Without this patch the drive errors out regularly: [ 1.090154] ata1.00: ATA-8: SAMSUNG SSD PM830 mSATA 256GB, CXM13D1Q, max UDMA/133 (...) [ 345.154996] ata1.00: exception Emask 0x40 SAct 0x0 SErr 0xc0800 action 0x6 [ 345.155006] ata1.00: irq_stat 0x40000001 [ 345.155013] ata1: SError: { HostInt CommWake 10B8B } [ 345.155018] ata1.00: failed command: SET FEATURES [ 345.155032] ata1.00: cmd ef/05:e1:00:00:00/00:00:00:00:00/40 tag 7 res 51/04:e1:00:00:00/00:00:00:00:00/40 Emask 0x41 (internal error) [ 345.155038] ata1.00: status: { DRDY ERR } [ 345.155042] ata1.00: error: { ABRT } [ 345.155051] ata1: hard resetting link [ 345.465661] ata1: SATA link up 6.0 Gbps (SStatus 133 SControl 300) [ 345.466955] ata1.00: configured for UDMA/133 [ 345.467085] ata1: EH complete Signed-off-by: François Cami Acked-by: Hans de Goede Signed-off-by: Tejun Heo --- drivers/ata/libata-core.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index a2498f5cfb28..7ed2f009911a 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -4553,8 +4553,9 @@ static const struct ata_blacklist_entry ata_device_blacklist [] = { ATA_HORKAGE_ZERO_AFTER_TRIM | ATA_HORKAGE_NOLPM, }, - /* This specific Samsung model/firmware-rev does not handle LPM well */ + /* These specific Samsung models/firmware-revs do not handle LPM well */ { "SAMSUNG MZMPC128HBFU-000MV", "CXM14M1Q", ATA_HORKAGE_NOLPM, }, + { "SAMSUNG SSD PM830 mSATA *", "CXM13D1Q", ATA_HORKAGE_NOLPM, }, /* Sandisk devices which are known to not handle LPM well */ { "SanDisk SD7UB3Q*G1001", NULL, ATA_HORKAGE_NOLPM, }, From 125966db1fea12575ee21b2b3fc95a59032406f0 Mon Sep 17 00:00:00 2001 From: Prasanna Kumar Kalever Date: Thu, 10 May 2018 19:12:18 +0530 Subject: [PATCH 071/190] scsi: target: tcmu: fix error resetting qfull_time_out to default Problem: $ cat /sys/kernel/config/target/core/user_0/block/attrib/qfull_time_out -1 $ echo "-1" > /sys/kernel/config/target/core/user_0/block/attrib/qfull_time_out -bash: echo: write error: Invalid argument Fix: This patch will help reset qfull_time_out to its default i.e. qfull_time_out=-1. Signed-off-by: Prasanna Kumar Kalever Acked-by: Mike Christie Signed-off-by: Martin K. Petersen --- drivers/target/target_core_user.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/target/target_core_user.c b/drivers/target/target_core_user.c index 4ad89ea71a70..4f26bdc3d1dc 100644 --- a/drivers/target/target_core_user.c +++ b/drivers/target/target_core_user.c @@ -2121,6 +2121,8 @@ static ssize_t tcmu_qfull_time_out_store(struct config_item *item, if (val >= 0) { udev->qfull_time_out = val * MSEC_PER_SEC; + } else if (val == -1) { + udev->qfull_time_out = val; } else { printk(KERN_ERR "Invalid qfull timeout value %d\n", val); return -EINVAL; From a406b0a0693eafc6f6b3a633d25749370bf40d8c Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Sun, 13 May 2018 17:10:52 -0700 Subject: [PATCH 072/190] scsi: core: clean up generated file scsi_devinfo_tbl.c "make clean" should remove the generated file "scsi_devinfo_tbl.c", so list it in the clean-files variable so that the file gets cleaned up. Fixes: 345e29608b4b ("scsi: scsi: Export blacklist flags to sysfs") Cc: Hannes Reinecke Signed-off-by: Randy Dunlap Reviewed-by: Hannes Reinecke Signed-off-by: Martin K. Petersen --- drivers/scsi/Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/scsi/Makefile b/drivers/scsi/Makefile index e29f9b8fd66d..56c940394729 100644 --- a/drivers/scsi/Makefile +++ b/drivers/scsi/Makefile @@ -182,7 +182,7 @@ zalon7xx-objs := zalon.o ncr53c8xx.o NCR_Q720_mod-objs := NCR_Q720.o ncr53c8xx.o # Files generated that shall be removed upon make clean -clean-files := 53c700_d.h 53c700_u.h +clean-files := 53c700_d.h 53c700_u.h scsi_devinfo_tbl.c $(obj)/53c700.o $(MODVERDIR)/$(obj)/53c700.ver: $(obj)/53c700_d.h From 8e907ed4882714fd13cfe670681fc6cb5284c780 Mon Sep 17 00:00:00 2001 From: Lidong Chen Date: Tue, 8 May 2018 16:50:16 +0800 Subject: [PATCH 073/190] IB/umem: Use the correct mm during ib_umem_release User-space may invoke ibv_reg_mr and ibv_dereg_mr in different threads. If ibv_dereg_mr is called after the thread which invoked ibv_reg_mr has exited, get_pid_task will return NULL and ib_umem_release will not decrease mm->pinned_vm. Instead of using threads to locate the mm, use the overall tgid from the ib_ucontext struct instead. This matches the behavior of ODP and disassociate in handling the mm of the process that called ibv_reg_mr. Cc: Fixes: 87773dd56d54 ("IB: ib_umem_release() should decrement mm->pinned_vm from ib_umem_get") Signed-off-by: Lidong Chen Signed-off-by: Jason Gunthorpe --- drivers/infiniband/core/umem.c | 7 +------ include/rdma/ib_umem.h | 1 - 2 files changed, 1 insertion(+), 7 deletions(-) diff --git a/drivers/infiniband/core/umem.c b/drivers/infiniband/core/umem.c index 9a4e899d94b3..2b6c9b516070 100644 --- a/drivers/infiniband/core/umem.c +++ b/drivers/infiniband/core/umem.c @@ -119,7 +119,6 @@ struct ib_umem *ib_umem_get(struct ib_ucontext *context, unsigned long addr, umem->length = size; umem->address = addr; umem->page_shift = PAGE_SHIFT; - umem->pid = get_task_pid(current, PIDTYPE_PID); /* * We ask for writable memory if any of the following * access flags are set. "Local write" and "remote write" @@ -132,7 +131,6 @@ struct ib_umem *ib_umem_get(struct ib_ucontext *context, unsigned long addr, IB_ACCESS_REMOTE_ATOMIC | IB_ACCESS_MW_BIND)); if (access & IB_ACCESS_ON_DEMAND) { - put_pid(umem->pid); ret = ib_umem_odp_get(context, umem, access); if (ret) { kfree(umem); @@ -148,7 +146,6 @@ struct ib_umem *ib_umem_get(struct ib_ucontext *context, unsigned long addr, page_list = (struct page **) __get_free_page(GFP_KERNEL); if (!page_list) { - put_pid(umem->pid); kfree(umem); return ERR_PTR(-ENOMEM); } @@ -231,7 +228,6 @@ struct ib_umem *ib_umem_get(struct ib_ucontext *context, unsigned long addr, if (ret < 0) { if (need_release) __ib_umem_release(context->device, umem, 0); - put_pid(umem->pid); kfree(umem); } else current->mm->pinned_vm = locked; @@ -274,8 +270,7 @@ void ib_umem_release(struct ib_umem *umem) __ib_umem_release(umem->context->device, umem, 1); - task = get_pid_task(umem->pid, PIDTYPE_PID); - put_pid(umem->pid); + task = get_pid_task(umem->context->tgid, PIDTYPE_PID); if (!task) goto out; mm = get_task_mm(task); diff --git a/include/rdma/ib_umem.h b/include/rdma/ib_umem.h index 23159dd5be18..a1fd63871d17 100644 --- a/include/rdma/ib_umem.h +++ b/include/rdma/ib_umem.h @@ -48,7 +48,6 @@ struct ib_umem { int writable; int hugetlb; struct work_struct work; - struct pid *pid; struct mm_struct *mm; unsigned long diff; struct ib_umem_odp *odp_data; From 15e6c22fd8e5a42c5ed6d487b7c9fe44c2517765 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Fri, 11 May 2018 15:21:01 +0200 Subject: [PATCH 074/190] KVM: SVM: Move spec control call after restore of GS svm_vcpu_run() invokes x86_spec_ctrl_restore_host() after VMEXIT, but before the host GS is restored. x86_spec_ctrl_restore_host() uses 'current' to determine the host SSBD state of the thread. 'current' is GS based, but host GS is not yet restored and the access causes a triple fault. Move the call after the host GS restore. Fixes: 885f82bfbc6f x86/process: Allow runtime control of Speculative Store Bypass Signed-off-by: Thomas Gleixner Reviewed-by: Borislav Petkov Reviewed-by: Konrad Rzeszutek Wilk Acked-by: Paolo Bonzini --- arch/x86/kvm/svm.c | 24 ++++++++++++------------ 1 file changed, 12 insertions(+), 12 deletions(-) diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 437c1b371129..fa891c8af842 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -5651,6 +5651,18 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu) #endif ); + /* Eliminate branch target predictions from guest mode */ + vmexit_fill_RSB(); + +#ifdef CONFIG_X86_64 + wrmsrl(MSR_GS_BASE, svm->host.gs_base); +#else + loadsegment(fs, svm->host.fs); +#ifndef CONFIG_X86_32_LAZY_GS + loadsegment(gs, svm->host.gs); +#endif +#endif + /* * We do not use IBRS in the kernel. If this vCPU has used the * SPEC_CTRL MSR it may have left it on; save the value and @@ -5671,18 +5683,6 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu) x86_spec_ctrl_restore_host(svm->spec_ctrl); - /* Eliminate branch target predictions from guest mode */ - vmexit_fill_RSB(); - -#ifdef CONFIG_X86_64 - wrmsrl(MSR_GS_BASE, svm->host.gs_base); -#else - loadsegment(fs, svm->host.fs); -#ifndef CONFIG_X86_32_LAZY_GS - loadsegment(gs, svm->host.gs); -#endif -#endif - reload_tss(vcpu); local_irq_disable(); From e7c587da125291db39ddf1f49b18e5970adbac17 Mon Sep 17 00:00:00 2001 From: Borislav Petkov Date: Wed, 2 May 2018 18:15:14 +0200 Subject: [PATCH 075/190] x86/speculation: Use synthetic bits for IBRS/IBPB/STIBP MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Intel and AMD have different CPUID bits hence for those use synthetic bits which get set on the respective vendor's in init_speculation_control(). So that debacles like what the commit message of c65732e4f721 ("x86/cpu: Restore CPUID_8000_0008_EBX reload") talks about don't happen anymore. Signed-off-by: Borislav Petkov Signed-off-by: Thomas Gleixner Reviewed-by: Konrad Rzeszutek Wilk Tested-by: Jörg Otte Cc: Linus Torvalds Cc: "Kirill A. Shutemov" Link: https://lkml.kernel.org/r/20180504161815.GG9257@pd.tnic --- arch/x86/include/asm/cpufeatures.h | 10 ++++++---- arch/x86/kernel/cpu/common.c | 14 ++++++++++---- arch/x86/kvm/cpuid.c | 10 +++++----- arch/x86/kvm/svm.c | 6 +++--- arch/x86/kvm/vmx.c | 9 ++------- 5 files changed, 26 insertions(+), 23 deletions(-) diff --git a/arch/x86/include/asm/cpufeatures.h b/arch/x86/include/asm/cpufeatures.h index 4e1c747acbf8..cb9bca8cecdc 100644 --- a/arch/x86/include/asm/cpufeatures.h +++ b/arch/x86/include/asm/cpufeatures.h @@ -198,7 +198,6 @@ #define X86_FEATURE_CAT_L2 ( 7*32+ 5) /* Cache Allocation Technology L2 */ #define X86_FEATURE_CDP_L3 ( 7*32+ 6) /* Code and Data Prioritization L3 */ #define X86_FEATURE_INVPCID_SINGLE ( 7*32+ 7) /* Effectively INVPCID && CR4.PCIDE=1 */ - #define X86_FEATURE_HW_PSTATE ( 7*32+ 8) /* AMD HW-PState */ #define X86_FEATURE_PROC_FEEDBACK ( 7*32+ 9) /* AMD ProcFeedbackInterface */ #define X86_FEATURE_SME ( 7*32+10) /* AMD Secure Memory Encryption */ @@ -216,6 +215,9 @@ #define X86_FEATURE_USE_IBRS_FW ( 7*32+22) /* "" Use IBRS during runtime firmware calls */ #define X86_FEATURE_SPEC_STORE_BYPASS_DISABLE ( 7*32+23) /* "" Disable Speculative Store Bypass. */ #define X86_FEATURE_AMD_SSBD ( 7*32+24) /* "" AMD SSBD implementation */ +#define X86_FEATURE_IBRS ( 7*32+25) /* Indirect Branch Restricted Speculation */ +#define X86_FEATURE_IBPB ( 7*32+26) /* Indirect Branch Prediction Barrier */ +#define X86_FEATURE_STIBP ( 7*32+27) /* Single Thread Indirect Branch Predictors */ /* Virtualization flags: Linux defined, word 8 */ #define X86_FEATURE_TPR_SHADOW ( 8*32+ 0) /* Intel TPR Shadow */ @@ -276,9 +278,9 @@ #define X86_FEATURE_CLZERO (13*32+ 0) /* CLZERO instruction */ #define X86_FEATURE_IRPERF (13*32+ 1) /* Instructions Retired Count */ #define X86_FEATURE_XSAVEERPTR (13*32+ 2) /* Always save/restore FP error pointers */ -#define X86_FEATURE_IBPB (13*32+12) /* Indirect Branch Prediction Barrier */ -#define X86_FEATURE_IBRS (13*32+14) /* Indirect Branch Restricted Speculation */ -#define X86_FEATURE_STIBP (13*32+15) /* Single Thread Indirect Branch Predictors */ +#define X86_FEATURE_AMD_IBPB (13*32+12) /* "" Indirect Branch Prediction Barrier */ +#define X86_FEATURE_AMD_IBRS (13*32+14) /* "" Indirect Branch Restricted Speculation */ +#define X86_FEATURE_AMD_STIBP (13*32+15) /* "" Single Thread Indirect Branch Predictors */ /* Thermal and Power Management Leaf, CPUID level 0x00000006 (EAX), word 14 */ #define X86_FEATURE_DTHERM (14*32+ 0) /* Digital Thermal Sensor */ diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index 9fbb388fadac..e6cd38b20375 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -757,17 +757,23 @@ static void init_speculation_control(struct cpuinfo_x86 *c) * and they also have a different bit for STIBP support. Also, * a hypervisor might have set the individual AMD bits even on * Intel CPUs, for finer-grained selection of what's available. - * - * We use the AMD bits in 0x8000_0008 EBX as the generic hardware - * features, which are visible in /proc/cpuinfo and used by the - * kernel. So set those accordingly from the Intel bits. */ if (cpu_has(c, X86_FEATURE_SPEC_CTRL)) { set_cpu_cap(c, X86_FEATURE_IBRS); set_cpu_cap(c, X86_FEATURE_IBPB); } + if (cpu_has(c, X86_FEATURE_INTEL_STIBP)) set_cpu_cap(c, X86_FEATURE_STIBP); + + if (cpu_has(c, X86_FEATURE_AMD_IBRS)) + set_cpu_cap(c, X86_FEATURE_IBRS); + + if (cpu_has(c, X86_FEATURE_AMD_IBPB)) + set_cpu_cap(c, X86_FEATURE_IBPB); + + if (cpu_has(c, X86_FEATURE_AMD_STIBP)) + set_cpu_cap(c, X86_FEATURE_STIBP); } void get_cpu_cap(struct cpuinfo_x86 *c) diff --git a/arch/x86/kvm/cpuid.c b/arch/x86/kvm/cpuid.c index 865c9a769864..e5ba48599428 100644 --- a/arch/x86/kvm/cpuid.c +++ b/arch/x86/kvm/cpuid.c @@ -379,7 +379,7 @@ static inline int __do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 function, /* cpuid 0x80000008.ebx */ const u32 kvm_cpuid_8000_0008_ebx_x86_features = - F(IBPB) | F(IBRS); + F(AMD_IBPB) | F(AMD_IBRS); /* cpuid 0xC0000001.edx */ const u32 kvm_cpuid_C000_0001_edx_x86_features = @@ -648,10 +648,10 @@ static inline int __do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 function, entry->eax = g_phys_as | (virt_as << 8); entry->edx = 0; /* IBRS and IBPB aren't necessarily present in hardware cpuid */ - if (boot_cpu_has(X86_FEATURE_IBPB)) - entry->ebx |= F(IBPB); - if (boot_cpu_has(X86_FEATURE_IBRS)) - entry->ebx |= F(IBRS); + if (boot_cpu_has(X86_FEATURE_AMD_IBPB)) + entry->ebx |= F(AMD_IBPB); + if (boot_cpu_has(X86_FEATURE_AMD_IBRS)) + entry->ebx |= F(AMD_IBRS); entry->ebx &= kvm_cpuid_8000_0008_ebx_x86_features; cpuid_mask(&entry->ebx, CPUID_8000_0008_EBX); break; diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index fa891c8af842..1a7a811118a8 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -4108,7 +4108,7 @@ static int svm_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) break; case MSR_IA32_SPEC_CTRL: if (!msr_info->host_initiated && - !guest_cpuid_has(vcpu, X86_FEATURE_IBRS)) + !guest_cpuid_has(vcpu, X86_FEATURE_AMD_IBRS)) return 1; msr_info->data = svm->spec_ctrl; @@ -4203,7 +4203,7 @@ static int svm_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr) break; case MSR_IA32_SPEC_CTRL: if (!msr->host_initiated && - !guest_cpuid_has(vcpu, X86_FEATURE_IBRS)) + !guest_cpuid_has(vcpu, X86_FEATURE_AMD_IBRS)) return 1; /* The STIBP bit doesn't fault even if it's not advertised */ @@ -4230,7 +4230,7 @@ static int svm_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr) break; case MSR_IA32_PRED_CMD: if (!msr->host_initiated && - !guest_cpuid_has(vcpu, X86_FEATURE_IBPB)) + !guest_cpuid_has(vcpu, X86_FEATURE_AMD_IBPB)) return 1; if (data & ~PRED_CMD_IBPB) diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 9b8d80bf3889..210e1b63d4b5 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -3523,9 +3523,7 @@ static int vmx_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) return kvm_get_msr_common(vcpu, msr_info); case MSR_IA32_SPEC_CTRL: if (!msr_info->host_initiated && - !guest_cpuid_has(vcpu, X86_FEATURE_IBRS) && - !guest_cpuid_has(vcpu, X86_FEATURE_SPEC_CTRL) && - !guest_cpuid_has(vcpu, X86_FEATURE_SSBD)) + !guest_cpuid_has(vcpu, X86_FEATURE_SPEC_CTRL)) return 1; msr_info->data = to_vmx(vcpu)->spec_ctrl; @@ -3643,9 +3641,7 @@ static int vmx_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) break; case MSR_IA32_SPEC_CTRL: if (!msr_info->host_initiated && - !guest_cpuid_has(vcpu, X86_FEATURE_IBRS) && - !guest_cpuid_has(vcpu, X86_FEATURE_SPEC_CTRL) && - !guest_cpuid_has(vcpu, X86_FEATURE_SSBD)) + !guest_cpuid_has(vcpu, X86_FEATURE_SPEC_CTRL)) return 1; /* The STIBP bit doesn't fault even if it's not advertised */ @@ -3675,7 +3671,6 @@ static int vmx_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) break; case MSR_IA32_PRED_CMD: if (!msr_info->host_initiated && - !guest_cpuid_has(vcpu, X86_FEATURE_IBPB) && !guest_cpuid_has(vcpu, X86_FEATURE_SPEC_CTRL)) return 1; From 7eb8956a7fec3c1f0abc2a5517dada99ccc8a961 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Thu, 10 May 2018 19:13:18 +0200 Subject: [PATCH 076/190] x86/cpufeatures: Disentangle MSR_SPEC_CTRL enumeration from IBRS The availability of the SPEC_CTRL MSR is enumerated by a CPUID bit on Intel and implied by IBRS or STIBP support on AMD. That's just confusing and in case an AMD CPU has IBRS not supported because the underlying problem has been fixed but has another bit valid in the SPEC_CTRL MSR, the thing falls apart. Add a synthetic feature bit X86_FEATURE_MSR_SPEC_CTRL to denote the availability on both Intel and AMD. While at it replace the boot_cpu_has() checks with static_cpu_has() where possible. This prevents late microcode loading from exposing SPEC_CTRL, but late loading is already very limited as it does not reevaluate the mitigation options and other bits and pieces. Having static_cpu_has() is the simplest and least fragile solution. Signed-off-by: Thomas Gleixner Reviewed-by: Borislav Petkov Reviewed-by: Konrad Rzeszutek Wilk --- arch/x86/include/asm/cpufeatures.h | 1 + arch/x86/kernel/cpu/bugs.c | 18 +++++++++++------- arch/x86/kernel/cpu/common.c | 9 +++++++-- arch/x86/kernel/cpu/intel.c | 1 + 4 files changed, 20 insertions(+), 9 deletions(-) diff --git a/arch/x86/include/asm/cpufeatures.h b/arch/x86/include/asm/cpufeatures.h index cb9bca8cecdc..7d34eb0d3715 100644 --- a/arch/x86/include/asm/cpufeatures.h +++ b/arch/x86/include/asm/cpufeatures.h @@ -206,6 +206,7 @@ #define X86_FEATURE_RETPOLINE_AMD ( 7*32+13) /* "" AMD Retpoline mitigation for Spectre variant 2 */ #define X86_FEATURE_INTEL_PPIN ( 7*32+14) /* Intel Processor Inventory Number */ #define X86_FEATURE_CDP_L2 ( 7*32+15) /* Code and Data Prioritization L2 */ +#define X86_FEATURE_MSR_SPEC_CTRL ( 7*32+16) /* "" MSR SPEC_CTRL is implemented */ #define X86_FEATURE_MBA ( 7*32+18) /* Memory Bandwidth Allocation */ #define X86_FEATURE_RSB_CTXSW ( 7*32+19) /* "" Fill RSB on context switches */ diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index bab05913f864..316cb24092a3 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -64,7 +64,7 @@ void __init check_bugs(void) * have unknown values. AMD64_LS_CFG MSR is cached in the early AMD * init code as it is not enumerated and depends on the family. */ - if (boot_cpu_has(X86_FEATURE_IBRS)) + if (boot_cpu_has(X86_FEATURE_MSR_SPEC_CTRL)) rdmsrl(MSR_IA32_SPEC_CTRL, x86_spec_ctrl_base); /* Select the proper spectre mitigation before patching alternatives */ @@ -145,7 +145,7 @@ u64 x86_spec_ctrl_get_default(void) { u64 msrval = x86_spec_ctrl_base; - if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) + if (static_cpu_has(X86_FEATURE_SPEC_CTRL)) msrval |= ssbd_tif_to_spec_ctrl(current_thread_info()->flags); return msrval; } @@ -155,10 +155,12 @@ void x86_spec_ctrl_set_guest(u64 guest_spec_ctrl) { u64 host = x86_spec_ctrl_base; - if (!boot_cpu_has(X86_FEATURE_IBRS)) + /* Is MSR_SPEC_CTRL implemented ? */ + if (!static_cpu_has(X86_FEATURE_MSR_SPEC_CTRL)) return; - if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) + /* Intel controls SSB in MSR_SPEC_CTRL */ + if (static_cpu_has(X86_FEATURE_SPEC_CTRL)) host |= ssbd_tif_to_spec_ctrl(current_thread_info()->flags); if (host != guest_spec_ctrl) @@ -170,10 +172,12 @@ void x86_spec_ctrl_restore_host(u64 guest_spec_ctrl) { u64 host = x86_spec_ctrl_base; - if (!boot_cpu_has(X86_FEATURE_IBRS)) + /* Is MSR_SPEC_CTRL implemented ? */ + if (!static_cpu_has(X86_FEATURE_MSR_SPEC_CTRL)) return; - if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) + /* Intel controls SSB in MSR_SPEC_CTRL */ + if (static_cpu_has(X86_FEATURE_SPEC_CTRL)) host |= ssbd_tif_to_spec_ctrl(current_thread_info()->flags); if (host != guest_spec_ctrl) @@ -631,7 +635,7 @@ int arch_prctl_spec_ctrl_get(struct task_struct *task, unsigned long which) void x86_spec_ctrl_setup_ap(void) { - if (boot_cpu_has(X86_FEATURE_IBRS)) + if (boot_cpu_has(X86_FEATURE_MSR_SPEC_CTRL)) x86_spec_ctrl_set(x86_spec_ctrl_base & ~x86_spec_ctrl_mask); if (ssb_mode == SPEC_STORE_BYPASS_DISABLE) diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index e6cd38b20375..af54dbe2df9a 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -761,19 +761,24 @@ static void init_speculation_control(struct cpuinfo_x86 *c) if (cpu_has(c, X86_FEATURE_SPEC_CTRL)) { set_cpu_cap(c, X86_FEATURE_IBRS); set_cpu_cap(c, X86_FEATURE_IBPB); + set_cpu_cap(c, X86_FEATURE_MSR_SPEC_CTRL); } if (cpu_has(c, X86_FEATURE_INTEL_STIBP)) set_cpu_cap(c, X86_FEATURE_STIBP); - if (cpu_has(c, X86_FEATURE_AMD_IBRS)) + if (cpu_has(c, X86_FEATURE_AMD_IBRS)) { set_cpu_cap(c, X86_FEATURE_IBRS); + set_cpu_cap(c, X86_FEATURE_MSR_SPEC_CTRL); + } if (cpu_has(c, X86_FEATURE_AMD_IBPB)) set_cpu_cap(c, X86_FEATURE_IBPB); - if (cpu_has(c, X86_FEATURE_AMD_STIBP)) + if (cpu_has(c, X86_FEATURE_AMD_STIBP)) { set_cpu_cap(c, X86_FEATURE_STIBP); + set_cpu_cap(c, X86_FEATURE_MSR_SPEC_CTRL); + } } void get_cpu_cap(struct cpuinfo_x86 *c) diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c index 0eab6c89c8d9..dd37244c587a 100644 --- a/arch/x86/kernel/cpu/intel.c +++ b/arch/x86/kernel/cpu/intel.c @@ -188,6 +188,7 @@ static void early_init_intel(struct cpuinfo_x86 *c) setup_clear_cpu_cap(X86_FEATURE_IBPB); setup_clear_cpu_cap(X86_FEATURE_STIBP); setup_clear_cpu_cap(X86_FEATURE_SPEC_CTRL); + setup_clear_cpu_cap(X86_FEATURE_MSR_SPEC_CTRL); setup_clear_cpu_cap(X86_FEATURE_INTEL_STIBP); setup_clear_cpu_cap(X86_FEATURE_SSBD); } From 52817587e706686fcdb27f14c1b000c92f266c96 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Thu, 10 May 2018 20:21:36 +0200 Subject: [PATCH 077/190] x86/cpufeatures: Disentangle SSBD enumeration The SSBD enumeration is similarly to the other bits magically shared between Intel and AMD though the mechanisms are different. Make X86_FEATURE_SSBD synthetic and set it depending on the vendor specific features or family dependent setup. Change the Intel bit to X86_FEATURE_SPEC_CTRL_SSBD to denote that SSBD is controlled via MSR_SPEC_CTRL and fix up the usage sites. Signed-off-by: Thomas Gleixner Reviewed-by: Borislav Petkov Reviewed-by: Konrad Rzeszutek Wilk --- arch/x86/include/asm/cpufeatures.h | 7 +++---- arch/x86/kernel/cpu/amd.c | 7 +------ arch/x86/kernel/cpu/bugs.c | 10 +++++----- arch/x86/kernel/cpu/common.c | 3 +++ arch/x86/kernel/cpu/intel.c | 1 + arch/x86/kernel/process.c | 2 +- 6 files changed, 14 insertions(+), 16 deletions(-) diff --git a/arch/x86/include/asm/cpufeatures.h b/arch/x86/include/asm/cpufeatures.h index 7d34eb0d3715..61c34c1a525c 100644 --- a/arch/x86/include/asm/cpufeatures.h +++ b/arch/x86/include/asm/cpufeatures.h @@ -207,15 +207,14 @@ #define X86_FEATURE_INTEL_PPIN ( 7*32+14) /* Intel Processor Inventory Number */ #define X86_FEATURE_CDP_L2 ( 7*32+15) /* Code and Data Prioritization L2 */ #define X86_FEATURE_MSR_SPEC_CTRL ( 7*32+16) /* "" MSR SPEC_CTRL is implemented */ - +#define X86_FEATURE_SSBD ( 7*32+17) /* Speculative Store Bypass Disable */ #define X86_FEATURE_MBA ( 7*32+18) /* Memory Bandwidth Allocation */ #define X86_FEATURE_RSB_CTXSW ( 7*32+19) /* "" Fill RSB on context switches */ #define X86_FEATURE_SEV ( 7*32+20) /* AMD Secure Encrypted Virtualization */ - #define X86_FEATURE_USE_IBPB ( 7*32+21) /* "" Indirect Branch Prediction Barrier enabled */ #define X86_FEATURE_USE_IBRS_FW ( 7*32+22) /* "" Use IBRS during runtime firmware calls */ #define X86_FEATURE_SPEC_STORE_BYPASS_DISABLE ( 7*32+23) /* "" Disable Speculative Store Bypass. */ -#define X86_FEATURE_AMD_SSBD ( 7*32+24) /* "" AMD SSBD implementation */ +#define X86_FEATURE_LS_CFG_SSBD ( 7*32+24) /* "" AMD SSBD implementation via LS_CFG MSR */ #define X86_FEATURE_IBRS ( 7*32+25) /* Indirect Branch Restricted Speculation */ #define X86_FEATURE_IBPB ( 7*32+26) /* Indirect Branch Prediction Barrier */ #define X86_FEATURE_STIBP ( 7*32+27) /* Single Thread Indirect Branch Predictors */ @@ -339,7 +338,7 @@ #define X86_FEATURE_SPEC_CTRL (18*32+26) /* "" Speculation Control (IBRS + IBPB) */ #define X86_FEATURE_INTEL_STIBP (18*32+27) /* "" Single Thread Indirect Branch Predictors */ #define X86_FEATURE_ARCH_CAPABILITIES (18*32+29) /* IA32_ARCH_CAPABILITIES MSR (Intel) */ -#define X86_FEATURE_SSBD (18*32+31) /* Speculative Store Bypass Disable */ +#define X86_FEATURE_SPEC_CTRL_SSBD (18*32+31) /* "" Speculative Store Bypass Disable */ /* * BUG word(s) diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index 7bde990b0385..2d2d8985654b 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -570,8 +570,8 @@ static void bsp_init_amd(struct cpuinfo_x86 *c) * avoid RMW. If that faults, do not enable SSBD. */ if (!rdmsrl_safe(MSR_AMD64_LS_CFG, &x86_amd_ls_cfg_base)) { + setup_force_cpu_cap(X86_FEATURE_LS_CFG_SSBD); setup_force_cpu_cap(X86_FEATURE_SSBD); - setup_force_cpu_cap(X86_FEATURE_AMD_SSBD); x86_amd_ls_cfg_ssbd_mask = 1ULL << bit; } } @@ -919,11 +919,6 @@ static void init_amd(struct cpuinfo_x86 *c) /* AMD CPUs don't reset SS attributes on SYSRET, Xen does. */ if (!cpu_has(c, X86_FEATURE_XENPV)) set_cpu_bug(c, X86_BUG_SYSRET_SS_ATTRS); - - if (boot_cpu_has(X86_FEATURE_AMD_SSBD)) { - set_cpu_cap(c, X86_FEATURE_SSBD); - set_cpu_cap(c, X86_FEATURE_AMD_SSBD); - } } #ifdef CONFIG_X86_32 diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index 316cb24092a3..7ebd6373fc31 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -159,8 +159,8 @@ void x86_spec_ctrl_set_guest(u64 guest_spec_ctrl) if (!static_cpu_has(X86_FEATURE_MSR_SPEC_CTRL)) return; - /* Intel controls SSB in MSR_SPEC_CTRL */ - if (static_cpu_has(X86_FEATURE_SPEC_CTRL)) + /* SSBD controlled in MSR_SPEC_CTRL */ + if (static_cpu_has(X86_FEATURE_SPEC_CTRL_SSBD)) host |= ssbd_tif_to_spec_ctrl(current_thread_info()->flags); if (host != guest_spec_ctrl) @@ -176,8 +176,8 @@ void x86_spec_ctrl_restore_host(u64 guest_spec_ctrl) if (!static_cpu_has(X86_FEATURE_MSR_SPEC_CTRL)) return; - /* Intel controls SSB in MSR_SPEC_CTRL */ - if (static_cpu_has(X86_FEATURE_SPEC_CTRL)) + /* SSBD controlled in MSR_SPEC_CTRL */ + if (static_cpu_has(X86_FEATURE_SPEC_CTRL_SSBD)) host |= ssbd_tif_to_spec_ctrl(current_thread_info()->flags); if (host != guest_spec_ctrl) @@ -189,7 +189,7 @@ static void x86_amd_ssb_disable(void) { u64 msrval = x86_amd_ls_cfg_base | x86_amd_ls_cfg_ssbd_mask; - if (boot_cpu_has(X86_FEATURE_AMD_SSBD)) + if (boot_cpu_has(X86_FEATURE_LS_CFG_SSBD)) wrmsrl(MSR_AMD64_LS_CFG, msrval); } diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index af54dbe2df9a..68282514c025 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -767,6 +767,9 @@ static void init_speculation_control(struct cpuinfo_x86 *c) if (cpu_has(c, X86_FEATURE_INTEL_STIBP)) set_cpu_cap(c, X86_FEATURE_STIBP); + if (cpu_has(c, X86_FEATURE_SPEC_CTRL_SSBD)) + set_cpu_cap(c, X86_FEATURE_SSBD); + if (cpu_has(c, X86_FEATURE_AMD_IBRS)) { set_cpu_cap(c, X86_FEATURE_IBRS); set_cpu_cap(c, X86_FEATURE_MSR_SPEC_CTRL); diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c index dd37244c587a..577e7f7ae273 100644 --- a/arch/x86/kernel/cpu/intel.c +++ b/arch/x86/kernel/cpu/intel.c @@ -191,6 +191,7 @@ static void early_init_intel(struct cpuinfo_x86 *c) setup_clear_cpu_cap(X86_FEATURE_MSR_SPEC_CTRL); setup_clear_cpu_cap(X86_FEATURE_INTEL_STIBP); setup_clear_cpu_cap(X86_FEATURE_SSBD); + setup_clear_cpu_cap(X86_FEATURE_SPEC_CTRL_SSBD); } /* diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index b77a091bf3b8..d71ef7eaa7ef 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -283,7 +283,7 @@ static __always_inline void __speculative_store_bypass_update(unsigned long tifn { u64 msr; - if (static_cpu_has(X86_FEATURE_AMD_SSBD)) { + if (static_cpu_has(X86_FEATURE_LS_CFG_SSBD)) { msr = x86_amd_ls_cfg_base | ssbd_tif_to_amd_ls_cfg(tifn); wrmsrl(MSR_AMD64_LS_CFG, msr); } else { From d1035d971829dcf80e8686ccde26f94b0a069472 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Thu, 10 May 2018 16:26:00 +0200 Subject: [PATCH 078/190] x86/cpufeatures: Add FEATURE_ZEN Add a ZEN feature bit so family-dependent static_cpu_has() optimizations can be built for ZEN. Signed-off-by: Thomas Gleixner Reviewed-by: Borislav Petkov Reviewed-by: Konrad Rzeszutek Wilk --- arch/x86/include/asm/cpufeatures.h | 1 + arch/x86/kernel/cpu/amd.c | 1 + 2 files changed, 2 insertions(+) diff --git a/arch/x86/include/asm/cpufeatures.h b/arch/x86/include/asm/cpufeatures.h index 61c34c1a525c..8099be4fc3e1 100644 --- a/arch/x86/include/asm/cpufeatures.h +++ b/arch/x86/include/asm/cpufeatures.h @@ -218,6 +218,7 @@ #define X86_FEATURE_IBRS ( 7*32+25) /* Indirect Branch Restricted Speculation */ #define X86_FEATURE_IBPB ( 7*32+26) /* Indirect Branch Prediction Barrier */ #define X86_FEATURE_STIBP ( 7*32+27) /* Single Thread Indirect Branch Predictors */ +#define X86_FEATURE_ZEN ( 7*32+28) /* "" CPU is AMD family 0x17 (Zen) */ /* Virtualization flags: Linux defined, word 8 */ #define X86_FEATURE_TPR_SHADOW ( 8*32+ 0) /* Intel TPR Shadow */ diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index 2d2d8985654b..1b18be3f35a8 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -812,6 +812,7 @@ static void init_amd_bd(struct cpuinfo_x86 *c) static void init_amd_zn(struct cpuinfo_x86 *c) { + set_cpu_cap(c, X86_FEATURE_ZEN); /* * Fix erratum 1076: CPB feature bit not being set in CPUID. It affects * all up to and including B1. From 1f50ddb4f4189243c05926b842dc1a0332195f31 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Wed, 9 May 2018 21:53:09 +0200 Subject: [PATCH 079/190] x86/speculation: Handle HT correctly on AMD The AMD64_LS_CFG MSR is a per core MSR on Family 17H CPUs. That means when hyperthreading is enabled the SSBD bit toggle needs to take both cores into account. Otherwise the following situation can happen: CPU0 CPU1 disable SSB disable SSB enable SSB <- Enables it for the Core, i.e. for CPU0 as well So after the SSB enable on CPU1 the task on CPU0 runs with SSB enabled again. On Intel the SSBD control is per core as well, but the synchronization logic is implemented behind the per thread SPEC_CTRL MSR. It works like this: CORE_SPEC_CTRL = THREAD0_SPEC_CTRL | THREAD1_SPEC_CTRL i.e. if one of the threads enables a mitigation then this affects both and the mitigation is only disabled in the core when both threads disabled it. Add the necessary synchronization logic for AMD family 17H. Unfortunately that requires a spinlock to serialize the access to the MSR, but the locks are only shared between siblings. Signed-off-by: Thomas Gleixner Reviewed-by: Borislav Petkov Reviewed-by: Konrad Rzeszutek Wilk --- arch/x86/include/asm/spec-ctrl.h | 6 ++ arch/x86/kernel/process.c | 131 ++++++++++++++++++++++++++++--- arch/x86/kernel/smpboot.c | 5 ++ 3 files changed, 133 insertions(+), 9 deletions(-) diff --git a/arch/x86/include/asm/spec-ctrl.h b/arch/x86/include/asm/spec-ctrl.h index dc21209790bf..0cb49c4564b0 100644 --- a/arch/x86/include/asm/spec-ctrl.h +++ b/arch/x86/include/asm/spec-ctrl.h @@ -33,6 +33,12 @@ static inline u64 ssbd_tif_to_amd_ls_cfg(u64 tifn) return (tifn & _TIF_SSBD) ? x86_amd_ls_cfg_ssbd_mask : 0ULL; } +#ifdef CONFIG_SMP +extern void speculative_store_bypass_ht_init(void); +#else +static inline void speculative_store_bypass_ht_init(void) { } +#endif + extern void speculative_store_bypass_update(void); #endif diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index d71ef7eaa7ef..d6fe8648d3f6 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -279,22 +279,135 @@ static inline void switch_to_bitmap(struct tss_struct *tss, } } +#ifdef CONFIG_SMP + +struct ssb_state { + struct ssb_state *shared_state; + raw_spinlock_t lock; + unsigned int disable_state; + unsigned long local_state; +}; + +#define LSTATE_SSB 0 + +static DEFINE_PER_CPU(struct ssb_state, ssb_state); + +void speculative_store_bypass_ht_init(void) +{ + struct ssb_state *st = this_cpu_ptr(&ssb_state); + unsigned int this_cpu = smp_processor_id(); + unsigned int cpu; + + st->local_state = 0; + + /* + * Shared state setup happens once on the first bringup + * of the CPU. It's not destroyed on CPU hotunplug. + */ + if (st->shared_state) + return; + + raw_spin_lock_init(&st->lock); + + /* + * Go over HT siblings and check whether one of them has set up the + * shared state pointer already. + */ + for_each_cpu(cpu, topology_sibling_cpumask(this_cpu)) { + if (cpu == this_cpu) + continue; + + if (!per_cpu(ssb_state, cpu).shared_state) + continue; + + /* Link it to the state of the sibling: */ + st->shared_state = per_cpu(ssb_state, cpu).shared_state; + return; + } + + /* + * First HT sibling to come up on the core. Link shared state of + * the first HT sibling to itself. The siblings on the same core + * which come up later will see the shared state pointer and link + * themself to the state of this CPU. + */ + st->shared_state = st; +} + +/* + * Logic is: First HT sibling enables SSBD for both siblings in the core + * and last sibling to disable it, disables it for the whole core. This how + * MSR_SPEC_CTRL works in "hardware": + * + * CORE_SPEC_CTRL = THREAD0_SPEC_CTRL | THREAD1_SPEC_CTRL + */ +static __always_inline void amd_set_core_ssb_state(unsigned long tifn) +{ + struct ssb_state *st = this_cpu_ptr(&ssb_state); + u64 msr = x86_amd_ls_cfg_base; + + if (!static_cpu_has(X86_FEATURE_ZEN)) { + msr |= ssbd_tif_to_amd_ls_cfg(tifn); + wrmsrl(MSR_AMD64_LS_CFG, msr); + return; + } + + if (tifn & _TIF_SSBD) { + /* + * Since this can race with prctl(), block reentry on the + * same CPU. + */ + if (__test_and_set_bit(LSTATE_SSB, &st->local_state)) + return; + + msr |= x86_amd_ls_cfg_ssbd_mask; + + raw_spin_lock(&st->shared_state->lock); + /* First sibling enables SSBD: */ + if (!st->shared_state->disable_state) + wrmsrl(MSR_AMD64_LS_CFG, msr); + st->shared_state->disable_state++; + raw_spin_unlock(&st->shared_state->lock); + } else { + if (!__test_and_clear_bit(LSTATE_SSB, &st->local_state)) + return; + + raw_spin_lock(&st->shared_state->lock); + st->shared_state->disable_state--; + if (!st->shared_state->disable_state) + wrmsrl(MSR_AMD64_LS_CFG, msr); + raw_spin_unlock(&st->shared_state->lock); + } +} +#else +static __always_inline void amd_set_core_ssb_state(unsigned long tifn) +{ + u64 msr = x86_amd_ls_cfg_base | ssbd_tif_to_amd_ls_cfg(tifn); + + wrmsrl(MSR_AMD64_LS_CFG, msr); +} +#endif + +static __always_inline void intel_set_ssb_state(unsigned long tifn) +{ + u64 msr = x86_spec_ctrl_base | ssbd_tif_to_spec_ctrl(tifn); + + wrmsrl(MSR_IA32_SPEC_CTRL, msr); +} + static __always_inline void __speculative_store_bypass_update(unsigned long tifn) { - u64 msr; - - if (static_cpu_has(X86_FEATURE_LS_CFG_SSBD)) { - msr = x86_amd_ls_cfg_base | ssbd_tif_to_amd_ls_cfg(tifn); - wrmsrl(MSR_AMD64_LS_CFG, msr); - } else { - msr = x86_spec_ctrl_base | ssbd_tif_to_spec_ctrl(tifn); - wrmsrl(MSR_IA32_SPEC_CTRL, msr); - } + if (static_cpu_has(X86_FEATURE_LS_CFG_SSBD)) + amd_set_core_ssb_state(tifn); + else + intel_set_ssb_state(tifn); } void speculative_store_bypass_update(void) { + preempt_disable(); __speculative_store_bypass_update(current_thread_info()->flags); + preempt_enable(); } void __switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p, diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c index 0f1cbb042f49..9dd324ae4832 100644 --- a/arch/x86/kernel/smpboot.c +++ b/arch/x86/kernel/smpboot.c @@ -79,6 +79,7 @@ #include #include #include +#include /* Number of siblings per CPU package */ int smp_num_siblings = 1; @@ -244,6 +245,8 @@ static void notrace start_secondary(void *unused) */ check_tsc_sync_target(); + speculative_store_bypass_ht_init(); + /* * Lock vector_lock, set CPU online and bring the vector * allocator online. Online must be set with vector_lock held @@ -1292,6 +1295,8 @@ void __init native_smp_prepare_cpus(unsigned int max_cpus) set_mtrr_aps_delayed_init(); smp_quirk_init_udelay(); + + speculative_store_bypass_ht_init(); } void arch_enable_nonboot_cpus_begin(void) From ccbcd2674472a978b48c91c1fbfb66c0ff959f24 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Wed, 9 May 2018 23:01:01 +0200 Subject: [PATCH 080/190] x86/bugs, KVM: Extend speculation control for VIRT_SPEC_CTRL AMD is proposing a VIRT_SPEC_CTRL MSR to handle the Speculative Store Bypass Disable via MSR_AMD64_LS_CFG so that guests do not have to care about the bit position of the SSBD bit and thus facilitate migration. Also, the sibling coordination on Family 17H CPUs can only be done on the host. Extend x86_spec_ctrl_set_guest() and x86_spec_ctrl_restore_host() with an extra argument for the VIRT_SPEC_CTRL MSR. Hand in 0 from VMX and in SVM add a new virt_spec_ctrl member to the CPU data structure which is going to be used in later patches for the actual implementation. Signed-off-by: Thomas Gleixner Reviewed-by: Borislav Petkov Reviewed-by: Konrad Rzeszutek Wilk --- arch/x86/include/asm/spec-ctrl.h | 9 ++++++--- arch/x86/kernel/cpu/bugs.c | 20 ++++++++++++++++++-- arch/x86/kvm/svm.c | 11 +++++++++-- arch/x86/kvm/vmx.c | 4 ++-- 4 files changed, 35 insertions(+), 9 deletions(-) diff --git a/arch/x86/include/asm/spec-ctrl.h b/arch/x86/include/asm/spec-ctrl.h index 0cb49c4564b0..6e2874049afd 100644 --- a/arch/x86/include/asm/spec-ctrl.h +++ b/arch/x86/include/asm/spec-ctrl.h @@ -10,10 +10,13 @@ * the guest has, while on VMEXIT we restore the host view. This * would be easier if SPEC_CTRL were architecturally maskable or * shadowable for guests but this is not (currently) the case. - * Takes the guest view of SPEC_CTRL MSR as a parameter. + * Takes the guest view of SPEC_CTRL MSR as a parameter and also + * the guest's version of VIRT_SPEC_CTRL, if emulated. */ -extern void x86_spec_ctrl_set_guest(u64); -extern void x86_spec_ctrl_restore_host(u64); +extern void x86_spec_ctrl_set_guest(u64 guest_spec_ctrl, + u64 guest_virt_spec_ctrl); +extern void x86_spec_ctrl_restore_host(u64 guest_spec_ctrl, + u64 guest_virt_spec_ctrl); /* AMD specific Speculative Store Bypass MSR data */ extern u64 x86_amd_ls_cfg_base; diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index 7ebd6373fc31..d3afd38f30d1 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -151,7 +151,15 @@ u64 x86_spec_ctrl_get_default(void) } EXPORT_SYMBOL_GPL(x86_spec_ctrl_get_default); -void x86_spec_ctrl_set_guest(u64 guest_spec_ctrl) +/** + * x86_spec_ctrl_set_guest - Set speculation control registers for the guest + * @guest_spec_ctrl: The guest content of MSR_SPEC_CTRL + * @guest_virt_spec_ctrl: The guest controlled bits of MSR_VIRT_SPEC_CTRL + * (may get translated to MSR_AMD64_LS_CFG bits) + * + * Avoids writing to the MSR if the content/bits are the same + */ +void x86_spec_ctrl_set_guest(u64 guest_spec_ctrl, u64 guest_virt_spec_ctrl) { u64 host = x86_spec_ctrl_base; @@ -168,7 +176,15 @@ void x86_spec_ctrl_set_guest(u64 guest_spec_ctrl) } EXPORT_SYMBOL_GPL(x86_spec_ctrl_set_guest); -void x86_spec_ctrl_restore_host(u64 guest_spec_ctrl) +/** + * x86_spec_ctrl_restore_host - Restore host speculation control registers + * @guest_spec_ctrl: The guest content of MSR_SPEC_CTRL + * @guest_virt_spec_ctrl: The guest controlled bits of MSR_VIRT_SPEC_CTRL + * (may get translated to MSR_AMD64_LS_CFG bits) + * + * Avoids writing to the MSR if the content/bits are the same + */ +void x86_spec_ctrl_restore_host(u64 guest_spec_ctrl, u64 guest_virt_spec_ctrl) { u64 host = x86_spec_ctrl_base; diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 1a7a811118a8..c07dbcc6d449 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -213,6 +213,12 @@ struct vcpu_svm { } host; u64 spec_ctrl; + /* + * Contains guest-controlled bits of VIRT_SPEC_CTRL, which will be + * translated into the appropriate L2_CFG bits on the host to + * perform speculative control. + */ + u64 virt_spec_ctrl; u32 *msrpm; @@ -2060,6 +2066,7 @@ static void svm_vcpu_reset(struct kvm_vcpu *vcpu, bool init_event) vcpu->arch.microcode_version = 0x01000065; svm->spec_ctrl = 0; + svm->virt_spec_ctrl = 0; if (!init_event) { svm->vcpu.arch.apic_base = APIC_DEFAULT_PHYS_BASE | @@ -5557,7 +5564,7 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu) * is no need to worry about the conditional branch over the wrmsr * being speculatively taken. */ - x86_spec_ctrl_set_guest(svm->spec_ctrl); + x86_spec_ctrl_set_guest(svm->spec_ctrl, svm->virt_spec_ctrl); asm volatile ( "push %%" _ASM_BP "; \n\t" @@ -5681,7 +5688,7 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu) if (unlikely(!msr_write_intercepted(vcpu, MSR_IA32_SPEC_CTRL))) svm->spec_ctrl = native_read_msr(MSR_IA32_SPEC_CTRL); - x86_spec_ctrl_restore_host(svm->spec_ctrl); + x86_spec_ctrl_restore_host(svm->spec_ctrl, svm->virt_spec_ctrl); reload_tss(vcpu); diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 210e1b63d4b5..5d733a03a6fa 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -9717,7 +9717,7 @@ static void __noclone vmx_vcpu_run(struct kvm_vcpu *vcpu) * is no need to worry about the conditional branch over the wrmsr * being speculatively taken. */ - x86_spec_ctrl_set_guest(vmx->spec_ctrl); + x86_spec_ctrl_set_guest(vmx->spec_ctrl, 0); vmx->__launched = vmx->loaded_vmcs->launched; @@ -9865,7 +9865,7 @@ static void __noclone vmx_vcpu_run(struct kvm_vcpu *vcpu) if (unlikely(!msr_write_intercepted(vcpu, MSR_IA32_SPEC_CTRL))) vmx->spec_ctrl = native_read_msr(MSR_IA32_SPEC_CTRL); - x86_spec_ctrl_restore_host(vmx->spec_ctrl); + x86_spec_ctrl_restore_host(vmx->spec_ctrl, 0); /* Eliminate branch target predictions from guest mode */ vmexit_fill_RSB(); From 11fb0683493b2da112cd64c9dada221b52463bf7 Mon Sep 17 00:00:00 2001 From: Tom Lendacky Date: Thu, 17 May 2018 17:09:18 +0200 Subject: [PATCH 081/190] x86/speculation: Add virtualized speculative store bypass disable support Some AMD processors only support a non-architectural means of enabling speculative store bypass disable (SSBD). To allow a simplified view of this to a guest, an architectural definition has been created through a new CPUID bit, 0x80000008_EBX[25], and a new MSR, 0xc001011f. With this, a hypervisor can virtualize the existence of this definition and provide an architectural method for using SSBD to a guest. Add the new CPUID feature, the new MSR and update the existing SSBD support to use this MSR when present. Signed-off-by: Tom Lendacky Signed-off-by: Thomas Gleixner Reviewed-by: Borislav Petkov --- arch/x86/include/asm/cpufeatures.h | 1 + arch/x86/include/asm/msr-index.h | 2 ++ arch/x86/kernel/cpu/bugs.c | 4 +++- arch/x86/kernel/process.c | 13 ++++++++++++- 4 files changed, 18 insertions(+), 2 deletions(-) diff --git a/arch/x86/include/asm/cpufeatures.h b/arch/x86/include/asm/cpufeatures.h index 8099be4fc3e1..fb00a2fca990 100644 --- a/arch/x86/include/asm/cpufeatures.h +++ b/arch/x86/include/asm/cpufeatures.h @@ -282,6 +282,7 @@ #define X86_FEATURE_AMD_IBPB (13*32+12) /* "" Indirect Branch Prediction Barrier */ #define X86_FEATURE_AMD_IBRS (13*32+14) /* "" Indirect Branch Restricted Speculation */ #define X86_FEATURE_AMD_STIBP (13*32+15) /* "" Single Thread Indirect Branch Predictors */ +#define X86_FEATURE_VIRT_SSBD (13*32+25) /* Virtualized Speculative Store Bypass Disable */ /* Thermal and Power Management Leaf, CPUID level 0x00000006 (EAX), word 14 */ #define X86_FEATURE_DTHERM (14*32+ 0) /* Digital Thermal Sensor */ diff --git a/arch/x86/include/asm/msr-index.h b/arch/x86/include/asm/msr-index.h index 0da3ca260b06..562414d5b834 100644 --- a/arch/x86/include/asm/msr-index.h +++ b/arch/x86/include/asm/msr-index.h @@ -347,6 +347,8 @@ #define MSR_AMD64_SEV_ENABLED_BIT 0 #define MSR_AMD64_SEV_ENABLED BIT_ULL(MSR_AMD64_SEV_ENABLED_BIT) +#define MSR_AMD64_VIRT_SPEC_CTRL 0xc001011f + /* Fam 17h MSRs */ #define MSR_F17H_IRPERF 0xc00000e9 diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index d3afd38f30d1..82422a04b506 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -205,7 +205,9 @@ static void x86_amd_ssb_disable(void) { u64 msrval = x86_amd_ls_cfg_base | x86_amd_ls_cfg_ssbd_mask; - if (boot_cpu_has(X86_FEATURE_LS_CFG_SSBD)) + if (boot_cpu_has(X86_FEATURE_VIRT_SSBD)) + wrmsrl(MSR_AMD64_VIRT_SPEC_CTRL, SPEC_CTRL_SSBD); + else if (boot_cpu_has(X86_FEATURE_LS_CFG_SSBD)) wrmsrl(MSR_AMD64_LS_CFG, msrval); } diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index d6fe8648d3f6..91c3398286d8 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -388,6 +388,15 @@ static __always_inline void amd_set_core_ssb_state(unsigned long tifn) } #endif +static __always_inline void amd_set_ssb_virt_state(unsigned long tifn) +{ + /* + * SSBD has the same definition in SPEC_CTRL and VIRT_SPEC_CTRL, + * so ssbd_tif_to_spec_ctrl() just works. + */ + wrmsrl(MSR_AMD64_VIRT_SPEC_CTRL, ssbd_tif_to_spec_ctrl(tifn)); +} + static __always_inline void intel_set_ssb_state(unsigned long tifn) { u64 msr = x86_spec_ctrl_base | ssbd_tif_to_spec_ctrl(tifn); @@ -397,7 +406,9 @@ static __always_inline void intel_set_ssb_state(unsigned long tifn) static __always_inline void __speculative_store_bypass_update(unsigned long tifn) { - if (static_cpu_has(X86_FEATURE_LS_CFG_SSBD)) + if (static_cpu_has(X86_FEATURE_VIRT_SSBD)) + amd_set_ssb_virt_state(tifn); + else if (static_cpu_has(X86_FEATURE_LS_CFG_SSBD)) amd_set_core_ssb_state(tifn); else intel_set_ssb_state(tifn); From 0270be3e34efb05a88bc4c422572ece038ef3608 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Thu, 10 May 2018 20:31:44 +0200 Subject: [PATCH 082/190] x86/speculation: Rework speculative_store_bypass_update() The upcoming support for the virtual SPEC_CTRL MSR on AMD needs to reuse speculative_store_bypass_update() to avoid code duplication. Add an argument for supplying a thread info (TIF) value and create a wrapper speculative_store_bypass_update_current() which is used at the existing call site. Signed-off-by: Thomas Gleixner Reviewed-by: Borislav Petkov Reviewed-by: Konrad Rzeszutek Wilk --- arch/x86/include/asm/spec-ctrl.h | 7 ++++++- arch/x86/kernel/cpu/bugs.c | 2 +- arch/x86/kernel/process.c | 4 ++-- 3 files changed, 9 insertions(+), 4 deletions(-) diff --git a/arch/x86/include/asm/spec-ctrl.h b/arch/x86/include/asm/spec-ctrl.h index 6e2874049afd..82b6c5a0d61e 100644 --- a/arch/x86/include/asm/spec-ctrl.h +++ b/arch/x86/include/asm/spec-ctrl.h @@ -42,6 +42,11 @@ extern void speculative_store_bypass_ht_init(void); static inline void speculative_store_bypass_ht_init(void) { } #endif -extern void speculative_store_bypass_update(void); +extern void speculative_store_bypass_update(unsigned long tif); + +static inline void speculative_store_bypass_update_current(void) +{ + speculative_store_bypass_update(current_thread_info()->flags); +} #endif diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index 82422a04b506..f2f0c1b3bf50 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -598,7 +598,7 @@ static int ssb_prctl_set(struct task_struct *task, unsigned long ctrl) * mitigation until it is next scheduled. */ if (task == current && update) - speculative_store_bypass_update(); + speculative_store_bypass_update_current(); return 0; } diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index 91c3398286d8..30ca2d1a9231 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -414,10 +414,10 @@ static __always_inline void __speculative_store_bypass_update(unsigned long tifn intel_set_ssb_state(tifn); } -void speculative_store_bypass_update(void) +void speculative_store_bypass_update(unsigned long tif) { preempt_disable(); - __speculative_store_bypass_update(current_thread_info()->flags); + __speculative_store_bypass_update(tif); preempt_enable(); } From cc69b34989210f067b2c51d5539b5f96ebcc3a01 Mon Sep 17 00:00:00 2001 From: Borislav Petkov Date: Sat, 12 May 2018 00:14:51 +0200 Subject: [PATCH 083/190] x86/bugs: Unify x86_spec_ctrl_{set_guest,restore_host} Function bodies are very similar and are going to grow more almost identical code. Add a bool arg to determine whether SPEC_CTRL is being set for the guest or restored to the host. No functional changes. Signed-off-by: Borislav Petkov Signed-off-by: Thomas Gleixner Reviewed-by: Konrad Rzeszutek Wilk --- arch/x86/include/asm/spec-ctrl.h | 33 +++++++++++++++--- arch/x86/kernel/cpu/bugs.c | 58 ++++++++------------------------ 2 files changed, 43 insertions(+), 48 deletions(-) diff --git a/arch/x86/include/asm/spec-ctrl.h b/arch/x86/include/asm/spec-ctrl.h index 82b6c5a0d61e..9cecbe5e57ee 100644 --- a/arch/x86/include/asm/spec-ctrl.h +++ b/arch/x86/include/asm/spec-ctrl.h @@ -13,10 +13,35 @@ * Takes the guest view of SPEC_CTRL MSR as a parameter and also * the guest's version of VIRT_SPEC_CTRL, if emulated. */ -extern void x86_spec_ctrl_set_guest(u64 guest_spec_ctrl, - u64 guest_virt_spec_ctrl); -extern void x86_spec_ctrl_restore_host(u64 guest_spec_ctrl, - u64 guest_virt_spec_ctrl); +extern void x86_virt_spec_ctrl(u64 guest_spec_ctrl, u64 guest_virt_spec_ctrl, bool guest); + +/** + * x86_spec_ctrl_set_guest - Set speculation control registers for the guest + * @guest_spec_ctrl: The guest content of MSR_SPEC_CTRL + * @guest_virt_spec_ctrl: The guest controlled bits of MSR_VIRT_SPEC_CTRL + * (may get translated to MSR_AMD64_LS_CFG bits) + * + * Avoids writing to the MSR if the content/bits are the same + */ +static inline +void x86_spec_ctrl_set_guest(u64 guest_spec_ctrl, u64 guest_virt_spec_ctrl) +{ + x86_virt_spec_ctrl(guest_spec_ctrl, guest_virt_spec_ctrl, true); +} + +/** + * x86_spec_ctrl_restore_host - Restore host speculation control registers + * @guest_spec_ctrl: The guest content of MSR_SPEC_CTRL + * @guest_virt_spec_ctrl: The guest controlled bits of MSR_VIRT_SPEC_CTRL + * (may get translated to MSR_AMD64_LS_CFG bits) + * + * Avoids writing to the MSR if the content/bits are the same + */ +static inline +void x86_spec_ctrl_restore_host(u64 guest_spec_ctrl, u64 guest_virt_spec_ctrl) +{ + x86_virt_spec_ctrl(guest_spec_ctrl, guest_virt_spec_ctrl, false); +} /* AMD specific Speculative Store Bypass MSR data */ extern u64 x86_amd_ls_cfg_base; diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index f2f0c1b3bf50..feb7d597c265 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -151,55 +151,25 @@ u64 x86_spec_ctrl_get_default(void) } EXPORT_SYMBOL_GPL(x86_spec_ctrl_get_default); -/** - * x86_spec_ctrl_set_guest - Set speculation control registers for the guest - * @guest_spec_ctrl: The guest content of MSR_SPEC_CTRL - * @guest_virt_spec_ctrl: The guest controlled bits of MSR_VIRT_SPEC_CTRL - * (may get translated to MSR_AMD64_LS_CFG bits) - * - * Avoids writing to the MSR if the content/bits are the same - */ -void x86_spec_ctrl_set_guest(u64 guest_spec_ctrl, u64 guest_virt_spec_ctrl) +void +x86_virt_spec_ctrl(u64 guest_spec_ctrl, u64 guest_virt_spec_ctrl, bool setguest) { - u64 host = x86_spec_ctrl_base; + struct thread_info *ti = current_thread_info(); + u64 msr, host = x86_spec_ctrl_base; /* Is MSR_SPEC_CTRL implemented ? */ - if (!static_cpu_has(X86_FEATURE_MSR_SPEC_CTRL)) - return; + if (static_cpu_has(X86_FEATURE_MSR_SPEC_CTRL)) { + /* SSBD controlled in MSR_SPEC_CTRL */ + if (static_cpu_has(X86_FEATURE_SPEC_CTRL_SSBD)) + host |= ssbd_tif_to_spec_ctrl(ti->flags); - /* SSBD controlled in MSR_SPEC_CTRL */ - if (static_cpu_has(X86_FEATURE_SPEC_CTRL_SSBD)) - host |= ssbd_tif_to_spec_ctrl(current_thread_info()->flags); - - if (host != guest_spec_ctrl) - wrmsrl(MSR_IA32_SPEC_CTRL, guest_spec_ctrl); + if (host != guest_spec_ctrl) { + msr = setguest ? guest_spec_ctrl : host; + wrmsrl(MSR_IA32_SPEC_CTRL, msr); + } + } } -EXPORT_SYMBOL_GPL(x86_spec_ctrl_set_guest); - -/** - * x86_spec_ctrl_restore_host - Restore host speculation control registers - * @guest_spec_ctrl: The guest content of MSR_SPEC_CTRL - * @guest_virt_spec_ctrl: The guest controlled bits of MSR_VIRT_SPEC_CTRL - * (may get translated to MSR_AMD64_LS_CFG bits) - * - * Avoids writing to the MSR if the content/bits are the same - */ -void x86_spec_ctrl_restore_host(u64 guest_spec_ctrl, u64 guest_virt_spec_ctrl) -{ - u64 host = x86_spec_ctrl_base; - - /* Is MSR_SPEC_CTRL implemented ? */ - if (!static_cpu_has(X86_FEATURE_MSR_SPEC_CTRL)) - return; - - /* SSBD controlled in MSR_SPEC_CTRL */ - if (static_cpu_has(X86_FEATURE_SPEC_CTRL_SSBD)) - host |= ssbd_tif_to_spec_ctrl(current_thread_info()->flags); - - if (host != guest_spec_ctrl) - wrmsrl(MSR_IA32_SPEC_CTRL, host); -} -EXPORT_SYMBOL_GPL(x86_spec_ctrl_restore_host); +EXPORT_SYMBOL_GPL(x86_virt_spec_ctrl); static void x86_amd_ssb_disable(void) { From fa8ac4988249c38476f6ad678a4848a736373403 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Sat, 12 May 2018 20:49:16 +0200 Subject: [PATCH 084/190] x86/bugs: Expose x86_spec_ctrl_base directly x86_spec_ctrl_base is the system wide default value for the SPEC_CTRL MSR. x86_spec_ctrl_get_default() returns x86_spec_ctrl_base and was intended to prevent modification to that variable. Though the variable is read only after init and globaly visible already. Remove the function and export the variable instead. Signed-off-by: Thomas Gleixner Reviewed-by: Borislav Petkov Reviewed-by: Konrad Rzeszutek Wilk --- arch/x86/include/asm/nospec-branch.h | 16 +++++----------- arch/x86/include/asm/spec-ctrl.h | 3 --- arch/x86/kernel/cpu/bugs.c | 11 +---------- 3 files changed, 6 insertions(+), 24 deletions(-) diff --git a/arch/x86/include/asm/nospec-branch.h b/arch/x86/include/asm/nospec-branch.h index bc258e644e5e..8d9deec00de9 100644 --- a/arch/x86/include/asm/nospec-branch.h +++ b/arch/x86/include/asm/nospec-branch.h @@ -217,16 +217,7 @@ enum spectre_v2_mitigation { SPECTRE_V2_IBRS, }; -/* - * The Intel specification for the SPEC_CTRL MSR requires that we - * preserve any already set reserved bits at boot time (e.g. for - * future additions that this kernel is not currently aware of). - * We then set any additional mitigation bits that we want - * ourselves and always use this as the base for SPEC_CTRL. - * We also use this when handling guest entry/exit as below. - */ extern void x86_spec_ctrl_set(u64); -extern u64 x86_spec_ctrl_get_default(void); /* The Speculative Store Bypass disable variants */ enum ssb_mitigation { @@ -278,6 +269,9 @@ static inline void indirect_branch_prediction_barrier(void) alternative_msr_write(MSR_IA32_PRED_CMD, val, X86_FEATURE_USE_IBPB); } +/* The Intel SPEC CTRL MSR base value cache */ +extern u64 x86_spec_ctrl_base; + /* * With retpoline, we must use IBRS to restrict branch prediction * before calling into firmware. @@ -286,7 +280,7 @@ static inline void indirect_branch_prediction_barrier(void) */ #define firmware_restrict_branch_speculation_start() \ do { \ - u64 val = x86_spec_ctrl_get_default() | SPEC_CTRL_IBRS; \ + u64 val = x86_spec_ctrl_base | SPEC_CTRL_IBRS; \ \ preempt_disable(); \ alternative_msr_write(MSR_IA32_SPEC_CTRL, val, \ @@ -295,7 +289,7 @@ do { \ #define firmware_restrict_branch_speculation_end() \ do { \ - u64 val = x86_spec_ctrl_get_default(); \ + u64 val = x86_spec_ctrl_base; \ \ alternative_msr_write(MSR_IA32_SPEC_CTRL, val, \ X86_FEATURE_USE_IBRS_FW); \ diff --git a/arch/x86/include/asm/spec-ctrl.h b/arch/x86/include/asm/spec-ctrl.h index 9cecbe5e57ee..763d49710329 100644 --- a/arch/x86/include/asm/spec-ctrl.h +++ b/arch/x86/include/asm/spec-ctrl.h @@ -47,9 +47,6 @@ void x86_spec_ctrl_restore_host(u64 guest_spec_ctrl, u64 guest_virt_spec_ctrl) extern u64 x86_amd_ls_cfg_base; extern u64 x86_amd_ls_cfg_ssbd_mask; -/* The Intel SPEC CTRL MSR base value cache */ -extern u64 x86_spec_ctrl_base; - static inline u64 ssbd_tif_to_spec_ctrl(u64 tifn) { BUILD_BUG_ON(TIF_SSBD < SPEC_CTRL_SSBD_SHIFT); diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index feb7d597c265..00f51deba493 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -36,6 +36,7 @@ static void __init ssb_select_mitigation(void); * writes to SPEC_CTRL contain whatever reserved bits have been set. */ u64 __ro_after_init x86_spec_ctrl_base; +EXPORT_SYMBOL_GPL(x86_spec_ctrl_base); /* * The vendor and possibly platform specific bits which can be modified in @@ -141,16 +142,6 @@ void x86_spec_ctrl_set(u64 val) } EXPORT_SYMBOL_GPL(x86_spec_ctrl_set); -u64 x86_spec_ctrl_get_default(void) -{ - u64 msrval = x86_spec_ctrl_base; - - if (static_cpu_has(X86_FEATURE_SPEC_CTRL)) - msrval |= ssbd_tif_to_spec_ctrl(current_thread_info()->flags); - return msrval; -} -EXPORT_SYMBOL_GPL(x86_spec_ctrl_get_default); - void x86_virt_spec_ctrl(u64 guest_spec_ctrl, u64 guest_virt_spec_ctrl, bool setguest) { From 4b59bdb569453a60b752b274ca61f009e37f4dae Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Sat, 12 May 2018 20:53:14 +0200 Subject: [PATCH 085/190] x86/bugs: Remove x86_spec_ctrl_set() x86_spec_ctrl_set() is only used in bugs.c and the extra mask checks there provide no real value as both call sites can just write x86_spec_ctrl_base to MSR_SPEC_CTRL. x86_spec_ctrl_base is valid and does not need any extra masking or checking. Signed-off-by: Thomas Gleixner Reviewed-by: Borislav Petkov Reviewed-by: Konrad Rzeszutek Wilk --- arch/x86/include/asm/nospec-branch.h | 2 -- arch/x86/kernel/cpu/bugs.c | 13 ++----------- 2 files changed, 2 insertions(+), 13 deletions(-) diff --git a/arch/x86/include/asm/nospec-branch.h b/arch/x86/include/asm/nospec-branch.h index 8d9deec00de9..8b38df98548e 100644 --- a/arch/x86/include/asm/nospec-branch.h +++ b/arch/x86/include/asm/nospec-branch.h @@ -217,8 +217,6 @@ enum spectre_v2_mitigation { SPECTRE_V2_IBRS, }; -extern void x86_spec_ctrl_set(u64); - /* The Speculative Store Bypass disable variants */ enum ssb_mitigation { SPEC_STORE_BYPASS_NONE, diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index 00f51deba493..e0b2e3b3301e 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -133,15 +133,6 @@ static const char *spectre_v2_strings[] = { static enum spectre_v2_mitigation spectre_v2_enabled __ro_after_init = SPECTRE_V2_NONE; -void x86_spec_ctrl_set(u64 val) -{ - if (val & x86_spec_ctrl_mask) - WARN_ONCE(1, "SPEC_CTRL MSR value 0x%16llx is unknown.\n", val); - else - wrmsrl(MSR_IA32_SPEC_CTRL, x86_spec_ctrl_base | val); -} -EXPORT_SYMBOL_GPL(x86_spec_ctrl_set); - void x86_virt_spec_ctrl(u64 guest_spec_ctrl, u64 guest_virt_spec_ctrl, bool setguest) { @@ -503,7 +494,7 @@ static enum ssb_mitigation __init __ssb_select_mitigation(void) case X86_VENDOR_INTEL: x86_spec_ctrl_base |= SPEC_CTRL_SSBD; x86_spec_ctrl_mask &= ~SPEC_CTRL_SSBD; - x86_spec_ctrl_set(SPEC_CTRL_SSBD); + wrmsrl(MSR_IA32_SPEC_CTRL, x86_spec_ctrl_base); break; case X86_VENDOR_AMD: x86_amd_ssb_disable(); @@ -615,7 +606,7 @@ int arch_prctl_spec_ctrl_get(struct task_struct *task, unsigned long which) void x86_spec_ctrl_setup_ap(void) { if (boot_cpu_has(X86_FEATURE_MSR_SPEC_CTRL)) - x86_spec_ctrl_set(x86_spec_ctrl_base & ~x86_spec_ctrl_mask); + wrmsrl(MSR_IA32_SPEC_CTRL, x86_spec_ctrl_base); if (ssb_mode == SPEC_STORE_BYPASS_DISABLE) x86_amd_ssb_disable(); From be6fcb5478e95bb1c91f489121238deb3abca46a Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Sat, 12 May 2018 20:10:00 +0200 Subject: [PATCH 086/190] x86/bugs: Rework spec_ctrl base and mask logic x86_spec_ctrL_mask is intended to mask out bits from a MSR_SPEC_CTRL value which are not to be modified. However the implementation is not really used and the bitmask was inverted to make a check easier, which was removed in "x86/bugs: Remove x86_spec_ctrl_set()" Aside of that it is missing the STIBP bit if it is supported by the platform, so if the mask would be used in x86_virt_spec_ctrl() then it would prevent a guest from setting STIBP. Add the STIBP bit if supported and use the mask in x86_virt_spec_ctrl() to sanitize the value which is supplied by the guest. Signed-off-by: Thomas Gleixner Reviewed-by: Borislav Petkov --- arch/x86/kernel/cpu/bugs.c | 26 +++++++++++++++++++------- 1 file changed, 19 insertions(+), 7 deletions(-) diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index e0b2e3b3301e..8e327bfec513 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -42,7 +42,7 @@ EXPORT_SYMBOL_GPL(x86_spec_ctrl_base); * The vendor and possibly platform specific bits which can be modified in * x86_spec_ctrl_base. */ -static u64 __ro_after_init x86_spec_ctrl_mask = ~SPEC_CTRL_IBRS; +static u64 __ro_after_init x86_spec_ctrl_mask = SPEC_CTRL_IBRS; /* * AMD specific MSR info for Speculative Store Bypass control. @@ -68,6 +68,10 @@ void __init check_bugs(void) if (boot_cpu_has(X86_FEATURE_MSR_SPEC_CTRL)) rdmsrl(MSR_IA32_SPEC_CTRL, x86_spec_ctrl_base); + /* Allow STIBP in MSR_SPEC_CTRL if supported */ + if (boot_cpu_has(X86_FEATURE_STIBP)) + x86_spec_ctrl_mask |= SPEC_CTRL_STIBP; + /* Select the proper spectre mitigation before patching alternatives */ spectre_v2_select_mitigation(); @@ -136,18 +140,26 @@ static enum spectre_v2_mitigation spectre_v2_enabled __ro_after_init = void x86_virt_spec_ctrl(u64 guest_spec_ctrl, u64 guest_virt_spec_ctrl, bool setguest) { + u64 msrval, guestval, hostval = x86_spec_ctrl_base; struct thread_info *ti = current_thread_info(); - u64 msr, host = x86_spec_ctrl_base; /* Is MSR_SPEC_CTRL implemented ? */ if (static_cpu_has(X86_FEATURE_MSR_SPEC_CTRL)) { + /* + * Restrict guest_spec_ctrl to supported values. Clear the + * modifiable bits in the host base value and or the + * modifiable bits from the guest value. + */ + guestval = hostval & ~x86_spec_ctrl_mask; + guestval |= guest_spec_ctrl & x86_spec_ctrl_mask; + /* SSBD controlled in MSR_SPEC_CTRL */ if (static_cpu_has(X86_FEATURE_SPEC_CTRL_SSBD)) - host |= ssbd_tif_to_spec_ctrl(ti->flags); + hostval |= ssbd_tif_to_spec_ctrl(ti->flags); - if (host != guest_spec_ctrl) { - msr = setguest ? guest_spec_ctrl : host; - wrmsrl(MSR_IA32_SPEC_CTRL, msr); + if (hostval != guestval) { + msrval = setguest ? guestval : hostval; + wrmsrl(MSR_IA32_SPEC_CTRL, msrval); } } } @@ -493,7 +505,7 @@ static enum ssb_mitigation __init __ssb_select_mitigation(void) switch (boot_cpu_data.x86_vendor) { case X86_VENDOR_INTEL: x86_spec_ctrl_base |= SPEC_CTRL_SSBD; - x86_spec_ctrl_mask &= ~SPEC_CTRL_SSBD; + x86_spec_ctrl_mask |= SPEC_CTRL_SSBD; wrmsrl(MSR_IA32_SPEC_CTRL, x86_spec_ctrl_base); break; case X86_VENDOR_AMD: From 47c61b3955cf712cadfc25635bf9bc174af030ea Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Thu, 10 May 2018 20:42:48 +0200 Subject: [PATCH 087/190] x86/speculation, KVM: Implement support for VIRT_SPEC_CTRL/LS_CFG Add the necessary logic for supporting the emulated VIRT_SPEC_CTRL MSR to x86_virt_spec_ctrl(). If either X86_FEATURE_LS_CFG_SSBD or X86_FEATURE_VIRT_SPEC_CTRL is set then use the new guest_virt_spec_ctrl argument to check whether the state must be modified on the host. The update reuses speculative_store_bypass_update() so the ZEN-specific sibling coordination can be reused. Signed-off-by: Thomas Gleixner --- arch/x86/include/asm/spec-ctrl.h | 6 ++++++ arch/x86/kernel/cpu/bugs.c | 30 ++++++++++++++++++++++++++++++ 2 files changed, 36 insertions(+) diff --git a/arch/x86/include/asm/spec-ctrl.h b/arch/x86/include/asm/spec-ctrl.h index 763d49710329..ae7c2c5cd7f0 100644 --- a/arch/x86/include/asm/spec-ctrl.h +++ b/arch/x86/include/asm/spec-ctrl.h @@ -53,6 +53,12 @@ static inline u64 ssbd_tif_to_spec_ctrl(u64 tifn) return (tifn & _TIF_SSBD) >> (TIF_SSBD - SPEC_CTRL_SSBD_SHIFT); } +static inline unsigned long ssbd_spec_ctrl_to_tif(u64 spec_ctrl) +{ + BUILD_BUG_ON(TIF_SSBD < SPEC_CTRL_SSBD_SHIFT); + return (spec_ctrl & SPEC_CTRL_SSBD) << (TIF_SSBD - SPEC_CTRL_SSBD_SHIFT); +} + static inline u64 ssbd_tif_to_amd_ls_cfg(u64 tifn) { return (tifn & _TIF_SSBD) ? x86_amd_ls_cfg_ssbd_mask : 0ULL; diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index 8e327bfec513..7416fc206b4a 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -162,6 +162,36 @@ x86_virt_spec_ctrl(u64 guest_spec_ctrl, u64 guest_virt_spec_ctrl, bool setguest) wrmsrl(MSR_IA32_SPEC_CTRL, msrval); } } + + /* + * If SSBD is not handled in MSR_SPEC_CTRL on AMD, update + * MSR_AMD64_L2_CFG or MSR_VIRT_SPEC_CTRL if supported. + */ + if (!static_cpu_has(X86_FEATURE_LS_CFG_SSBD) && + !static_cpu_has(X86_FEATURE_VIRT_SSBD)) + return; + + /* + * If the host has SSBD mitigation enabled, force it in the host's + * virtual MSR value. If its not permanently enabled, evaluate + * current's TIF_SSBD thread flag. + */ + if (static_cpu_has(X86_FEATURE_SPEC_STORE_BYPASS_DISABLE)) + hostval = SPEC_CTRL_SSBD; + else + hostval = ssbd_tif_to_spec_ctrl(ti->flags); + + /* Sanitize the guest value */ + guestval = guest_virt_spec_ctrl & SPEC_CTRL_SSBD; + + if (hostval != guestval) { + unsigned long tif; + + tif = setguest ? ssbd_spec_ctrl_to_tif(guestval) : + ssbd_spec_ctrl_to_tif(hostval); + + speculative_store_bypass_update(tif); + } } EXPORT_SYMBOL_GPL(x86_virt_spec_ctrl); From bc226f07dcd3c9ef0b7f6236fe356ea4a9cb4769 Mon Sep 17 00:00:00 2001 From: Tom Lendacky Date: Thu, 10 May 2018 22:06:39 +0200 Subject: [PATCH 088/190] KVM: SVM: Implement VIRT_SPEC_CTRL support for SSBD Expose the new virtualized architectural mechanism, VIRT_SSBD, for using speculative store bypass disable (SSBD) under SVM. This will allow guests to use SSBD on hardware that uses non-architectural mechanisms for enabling SSBD. [ tglx: Folded the migration fixup from Paolo Bonzini ] Signed-off-by: Tom Lendacky Signed-off-by: Thomas Gleixner --- arch/x86/include/asm/kvm_host.h | 2 +- arch/x86/kernel/cpu/common.c | 3 ++- arch/x86/kvm/cpuid.c | 11 +++++++++-- arch/x86/kvm/svm.c | 21 +++++++++++++++++++-- arch/x86/kvm/vmx.c | 18 +++++++++++++++--- arch/x86/kvm/x86.c | 13 ++++--------- 6 files changed, 50 insertions(+), 18 deletions(-) diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index c25775fad4ed..f4b2588865e9 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -924,7 +924,7 @@ struct kvm_x86_ops { int (*hardware_setup)(void); /* __init */ void (*hardware_unsetup)(void); /* __exit */ bool (*cpu_has_accelerated_tpr)(void); - bool (*cpu_has_high_real_mode_segbase)(void); + bool (*has_emulated_msr)(int index); void (*cpuid_update)(struct kvm_vcpu *vcpu); struct kvm *(*vm_alloc)(void); diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index 68282514c025..b4247ed0c81e 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -767,7 +767,8 @@ static void init_speculation_control(struct cpuinfo_x86 *c) if (cpu_has(c, X86_FEATURE_INTEL_STIBP)) set_cpu_cap(c, X86_FEATURE_STIBP); - if (cpu_has(c, X86_FEATURE_SPEC_CTRL_SSBD)) + if (cpu_has(c, X86_FEATURE_SPEC_CTRL_SSBD) || + cpu_has(c, X86_FEATURE_VIRT_SSBD)) set_cpu_cap(c, X86_FEATURE_SSBD); if (cpu_has(c, X86_FEATURE_AMD_IBRS)) { diff --git a/arch/x86/kvm/cpuid.c b/arch/x86/kvm/cpuid.c index e5ba48599428..ced851169730 100644 --- a/arch/x86/kvm/cpuid.c +++ b/arch/x86/kvm/cpuid.c @@ -379,7 +379,7 @@ static inline int __do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 function, /* cpuid 0x80000008.ebx */ const u32 kvm_cpuid_8000_0008_ebx_x86_features = - F(AMD_IBPB) | F(AMD_IBRS); + F(AMD_IBPB) | F(AMD_IBRS) | F(VIRT_SSBD); /* cpuid 0xC0000001.edx */ const u32 kvm_cpuid_C000_0001_edx_x86_features = @@ -647,13 +647,20 @@ static inline int __do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 function, g_phys_as = phys_as; entry->eax = g_phys_as | (virt_as << 8); entry->edx = 0; - /* IBRS and IBPB aren't necessarily present in hardware cpuid */ + /* + * IBRS, IBPB and VIRT_SSBD aren't necessarily present in + * hardware cpuid + */ if (boot_cpu_has(X86_FEATURE_AMD_IBPB)) entry->ebx |= F(AMD_IBPB); if (boot_cpu_has(X86_FEATURE_AMD_IBRS)) entry->ebx |= F(AMD_IBRS); + if (boot_cpu_has(X86_FEATURE_VIRT_SSBD)) + entry->ebx |= F(VIRT_SSBD); entry->ebx &= kvm_cpuid_8000_0008_ebx_x86_features; cpuid_mask(&entry->ebx, CPUID_8000_0008_EBX); + if (boot_cpu_has(X86_FEATURE_LS_CFG_SSBD)) + entry->ebx |= F(VIRT_SSBD); break; } case 0x80000019: diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index c07dbcc6d449..26110c202b19 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -4120,6 +4120,13 @@ static int svm_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) msr_info->data = svm->spec_ctrl; break; + case MSR_AMD64_VIRT_SPEC_CTRL: + if (!msr_info->host_initiated && + !guest_cpuid_has(vcpu, X86_FEATURE_VIRT_SSBD)) + return 1; + + msr_info->data = svm->virt_spec_ctrl; + break; case MSR_F15H_IC_CFG: { int family, model; @@ -4251,6 +4258,16 @@ static int svm_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr) break; set_msr_interception(svm->msrpm, MSR_IA32_PRED_CMD, 0, 1); break; + case MSR_AMD64_VIRT_SPEC_CTRL: + if (!msr->host_initiated && + !guest_cpuid_has(vcpu, X86_FEATURE_VIRT_SSBD)) + return 1; + + if (data & ~SPEC_CTRL_SSBD) + return 1; + + svm->virt_spec_ctrl = data; + break; case MSR_STAR: svm->vmcb->save.star = data; break; @@ -5791,7 +5808,7 @@ static bool svm_cpu_has_accelerated_tpr(void) return false; } -static bool svm_has_high_real_mode_segbase(void) +static bool svm_has_emulated_msr(int index) { return true; } @@ -7017,7 +7034,7 @@ static struct kvm_x86_ops svm_x86_ops __ro_after_init = { .hardware_enable = svm_hardware_enable, .hardware_disable = svm_hardware_disable, .cpu_has_accelerated_tpr = svm_cpu_has_accelerated_tpr, - .cpu_has_high_real_mode_segbase = svm_has_high_real_mode_segbase, + .has_emulated_msr = svm_has_emulated_msr, .vcpu_create = svm_create_vcpu, .vcpu_free = svm_free_vcpu, diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 5d733a03a6fa..0c57fb4df2d1 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -9477,9 +9477,21 @@ static void vmx_handle_external_intr(struct kvm_vcpu *vcpu) } STACK_FRAME_NON_STANDARD(vmx_handle_external_intr); -static bool vmx_has_high_real_mode_segbase(void) +static bool vmx_has_emulated_msr(int index) { - return enable_unrestricted_guest || emulate_invalid_guest_state; + switch (index) { + case MSR_IA32_SMBASE: + /* + * We cannot do SMM unless we can run the guest in big + * real mode. + */ + return enable_unrestricted_guest || emulate_invalid_guest_state; + case MSR_AMD64_VIRT_SPEC_CTRL: + /* This is AMD only. */ + return false; + default: + return true; + } } static bool vmx_mpx_supported(void) @@ -12625,7 +12637,7 @@ static struct kvm_x86_ops vmx_x86_ops __ro_after_init = { .hardware_enable = hardware_enable, .hardware_disable = hardware_disable, .cpu_has_accelerated_tpr = report_flexpriority, - .cpu_has_high_real_mode_segbase = vmx_has_high_real_mode_segbase, + .has_emulated_msr = vmx_has_emulated_msr, .vm_init = vmx_vm_init, .vm_alloc = vmx_vm_alloc, diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 51ecd381793b..421a39e40d5e 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -1058,6 +1058,7 @@ static u32 emulated_msrs[] = { MSR_SMI_COUNT, MSR_PLATFORM_INFO, MSR_MISC_FEATURES_ENABLES, + MSR_AMD64_VIRT_SPEC_CTRL, }; static unsigned num_emulated_msrs; @@ -2903,7 +2904,7 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext) * fringe case that is not enabled except via specific settings * of the module parameters. */ - r = kvm_x86_ops->cpu_has_high_real_mode_segbase(); + r = kvm_x86_ops->has_emulated_msr(MSR_IA32_SMBASE); break; case KVM_CAP_VAPIC: r = !kvm_x86_ops->cpu_has_accelerated_tpr(); @@ -4603,14 +4604,8 @@ static void kvm_init_msr_list(void) num_msrs_to_save = j; for (i = j = 0; i < ARRAY_SIZE(emulated_msrs); i++) { - switch (emulated_msrs[i]) { - case MSR_IA32_SMBASE: - if (!kvm_x86_ops->cpu_has_high_real_mode_segbase()) - continue; - break; - default: - break; - } + if (!kvm_x86_ops->has_emulated_msr(emulated_msrs[i])) + continue; if (j < i) emulated_msrs[j] = emulated_msrs[i]; From faf37c44a105f3608115785f17cbbf3500f8bc71 Mon Sep 17 00:00:00 2001 From: Michael Neuling Date: Fri, 18 May 2018 11:37:42 +1000 Subject: [PATCH 089/190] powerpc/64s: Clear PCR on boot Clear the PCR (Processor Compatibility Register) on boot to ensure we are not running in a compatibility mode. We've seen this cause problems when a crash (and kdump) occurs while running compat mode guests. The kdump kernel then runs with the PCR set and causes problems. The symptom in the kdump kernel (also seen in petitboot after fast-reboot) is early userspace programs taking sigills on newer instructions (seen in libc). Signed-off-by: Michael Neuling Cc: stable@vger.kernel.org Signed-off-by: Michael Ellerman --- arch/powerpc/kernel/cpu_setup_power.S | 6 ++++++ arch/powerpc/kernel/dt_cpu_ftrs.c | 1 + 2 files changed, 7 insertions(+) diff --git a/arch/powerpc/kernel/cpu_setup_power.S b/arch/powerpc/kernel/cpu_setup_power.S index 3f30c994e931..458b928dbd84 100644 --- a/arch/powerpc/kernel/cpu_setup_power.S +++ b/arch/powerpc/kernel/cpu_setup_power.S @@ -28,6 +28,7 @@ _GLOBAL(__setup_cpu_power7) beqlr li r0,0 mtspr SPRN_LPID,r0 + mtspr SPRN_PCR,r0 mfspr r3,SPRN_LPCR li r4,(LPCR_LPES1 >> LPCR_LPES_SH) bl __init_LPCR_ISA206 @@ -41,6 +42,7 @@ _GLOBAL(__restore_cpu_power7) beqlr li r0,0 mtspr SPRN_LPID,r0 + mtspr SPRN_PCR,r0 mfspr r3,SPRN_LPCR li r4,(LPCR_LPES1 >> LPCR_LPES_SH) bl __init_LPCR_ISA206 @@ -57,6 +59,7 @@ _GLOBAL(__setup_cpu_power8) beqlr li r0,0 mtspr SPRN_LPID,r0 + mtspr SPRN_PCR,r0 mfspr r3,SPRN_LPCR ori r3, r3, LPCR_PECEDH li r4,0 /* LPES = 0 */ @@ -78,6 +81,7 @@ _GLOBAL(__restore_cpu_power8) beqlr li r0,0 mtspr SPRN_LPID,r0 + mtspr SPRN_PCR,r0 mfspr r3,SPRN_LPCR ori r3, r3, LPCR_PECEDH li r4,0 /* LPES = 0 */ @@ -99,6 +103,7 @@ _GLOBAL(__setup_cpu_power9) mtspr SPRN_PSSCR,r0 mtspr SPRN_LPID,r0 mtspr SPRN_PID,r0 + mtspr SPRN_PCR,r0 mfspr r3,SPRN_LPCR LOAD_REG_IMMEDIATE(r4, LPCR_PECEDH | LPCR_PECE_HVEE | LPCR_HVICE | LPCR_HEIC) or r3, r3, r4 @@ -123,6 +128,7 @@ _GLOBAL(__restore_cpu_power9) mtspr SPRN_PSSCR,r0 mtspr SPRN_LPID,r0 mtspr SPRN_PID,r0 + mtspr SPRN_PCR,r0 mfspr r3,SPRN_LPCR LOAD_REG_IMMEDIATE(r4, LPCR_PECEDH | LPCR_PECE_HVEE | LPCR_HVICE | LPCR_HEIC) or r3, r3, r4 diff --git a/arch/powerpc/kernel/dt_cpu_ftrs.c b/arch/powerpc/kernel/dt_cpu_ftrs.c index 8ab51f6ca03a..c904477abaf3 100644 --- a/arch/powerpc/kernel/dt_cpu_ftrs.c +++ b/arch/powerpc/kernel/dt_cpu_ftrs.c @@ -101,6 +101,7 @@ static void __restore_cpu_cpufeatures(void) if (hv_mode) { mtspr(SPRN_LPID, 0); mtspr(SPRN_HFSCR, system_registers.hfscr); + mtspr(SPRN_PCR, 0); } mtspr(SPRN_FSCR, system_registers.fscr); From 3ae180972564846e6d794e3615e1ab0a1e6c4ef9 Mon Sep 17 00:00:00 2001 From: Ben Hutchings Date: Thu, 17 May 2018 22:34:39 +0100 Subject: [PATCH 090/190] ALSA: timer: Fix pause event notification Commit f65e0d299807 ("ALSA: timer: Call notifier in the same spinlock") combined the start/continue and stop/pause functions, and in doing so changed the event code for the pause case to SNDRV_TIMER_EVENT_CONTINUE. Change it back to SNDRV_TIMER_EVENT_PAUSE. Fixes: f65e0d299807 ("ALSA: timer: Call notifier in the same spinlock") Signed-off-by: Ben Hutchings Cc: stable@vger.kernel.org Signed-off-by: Takashi Iwai --- sound/core/timer.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/sound/core/timer.c b/sound/core/timer.c index dc87728c5b74..0ddcae495838 100644 --- a/sound/core/timer.c +++ b/sound/core/timer.c @@ -592,7 +592,7 @@ static int snd_timer_stop1(struct snd_timer_instance *timeri, bool stop) else timeri->flags |= SNDRV_TIMER_IFLG_PAUSED; snd_timer_notify1(timeri, stop ? SNDRV_TIMER_EVENT_STOP : - SNDRV_TIMER_EVENT_CONTINUE); + SNDRV_TIMER_EVENT_PAUSE); unlock: spin_unlock_irqrestore(&timer->lock, flags); return result; @@ -614,7 +614,7 @@ static int snd_timer_stop_slave(struct snd_timer_instance *timeri, bool stop) list_del_init(&timeri->ack_list); list_del_init(&timeri->active_list); snd_timer_notify1(timeri, stop ? SNDRV_TIMER_EVENT_STOP : - SNDRV_TIMER_EVENT_CONTINUE); + SNDRV_TIMER_EVENT_PAUSE); spin_unlock(&timeri->timer->lock); } spin_unlock_irqrestore(&slave_active_lock, flags); From 814596495dd2b9d4aab92d8f89cf19060d25d2ea Mon Sep 17 00:00:00 2001 From: Eric Biggers Date: Mon, 14 May 2018 20:09:24 -0700 Subject: [PATCH 091/190] cfg80211: further limit wiphy names to 64 bytes wiphy names were recently limited to 128 bytes by commit a7cfebcb7594 ("cfg80211: limit wiphy names to 128 bytes"). As it turns out though, this isn't sufficient because dev_vprintk_emit() needs the syslog header string "SUBSYSTEM=ieee80211\0DEVICE=+ieee80211:$devname" to fit into 128 bytes. This triggered the "device/subsystem name too long" WARN when the device name was >= 90 bytes. As before, this was reproduced by syzbot by sending an HWSIM_CMD_NEW_RADIO command to the MAC80211_HWSIM generic netlink family. Fix it by further limiting wiphy names to 64 bytes. Reported-by: syzbot+e64565577af34b3768dc@syzkaller.appspotmail.com Fixes: a7cfebcb7594 ("cfg80211: limit wiphy names to 128 bytes") Signed-off-by: Eric Biggers Signed-off-by: Johannes Berg --- include/uapi/linux/nl80211.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/uapi/linux/nl80211.h b/include/uapi/linux/nl80211.h index 9c3630146cec..271b93783d28 100644 --- a/include/uapi/linux/nl80211.h +++ b/include/uapi/linux/nl80211.h @@ -2698,7 +2698,7 @@ enum nl80211_attrs { #define NL80211_ATTR_KEYS NL80211_ATTR_KEYS #define NL80211_ATTR_FEATURE_FLAGS NL80211_ATTR_FEATURE_FLAGS -#define NL80211_WIPHY_NAME_MAXLEN 128 +#define NL80211_WIPHY_NAME_MAXLEN 64 #define NL80211_MAX_SUPP_RATES 32 #define NL80211_MAX_SUPP_HT_RATES 77 From 1039d08100e58f016a814347baadb77bf8dd21ab Mon Sep 17 00:00:00 2001 From: Dedy Lansky Date: Thu, 17 May 2018 16:25:03 +0300 Subject: [PATCH 092/190] nl80211: fix nlmsg allocation in cfg80211_ft_event Allocation size of nlmsg in cfg80211_ft_event is based on ric_ies_len and doesn't take into account ies_len. This leads to NL80211_CMD_FT_EVENT message construction failure in case ft_event contains large enough ies buffer. Add ies_len to the nlmsg allocation size. Signed-off-by: Dedy Lansky Signed-off-by: Johannes Berg --- net/wireless/nl80211.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index a052693c2e85..7c5135a92d76 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -15555,7 +15555,8 @@ void cfg80211_ft_event(struct net_device *netdev, if (!ft_event->target_ap) return; - msg = nlmsg_new(100 + ft_event->ric_ies_len, GFP_KERNEL); + msg = nlmsg_new(100 + ft_event->ies_len + ft_event->ric_ies_len, + GFP_KERNEL); if (!msg) return; From 1d6741d86429a294f51f2773c751c8f7662e7ca2 Mon Sep 17 00:00:00 2001 From: Bob Copeland Date: Thu, 17 May 2018 09:09:28 -0400 Subject: [PATCH 093/190] mac80211: mesh: fix premature update of rc stats The mesh_neighbour_update() function, queued via beacon rx, can race with userspace creating the same station. If the station already exists by the time mesh_neighbour_update() is called, the function wrongly assumes rate control has been initialized and calls rate_control_rate_update(), which in turn calls into the driver. Updating the rate control before it has been initialized can cause a crash in some drivers, for example this firmware crash in ath10k due to sta->rx_nss being 0: [ 3078.088247] mesh0: Inserted STA 5c:e2:8c:f1:ab:ba [ 3078.258407] ath10k_pci 0000:0d:00.0: firmware crashed! (uuid d6ed5961-93cc-4d61-803f-5eda55bb8643) [ 3078.258421] ath10k_pci 0000:0d:00.0: qca988x hw2.0 target 0x4100016c chip_id 0x043202ff sub 0000:0000 [ 3078.258426] ath10k_pci 0000:0d:00.0: kconfig debug 1 debugfs 1 tracing 1 dfs 0 testmode 0 [ 3078.258608] ath10k_pci 0000:0d:00.0: firmware ver 10.2.4.70.59-2 api 5 features no-p2p,raw-mode,mfp crc32 4159f498 [ 3078.258613] ath10k_pci 0000:0d:00.0: board_file api 1 bmi_id N/A crc32 bebc7c08 [ 3078.258617] ath10k_pci 0000:0d:00.0: htt-ver 2.1 wmi-op 5 htt-op 2 cal otp max-sta 128 raw 0 hwcrypto 1 [ 3078.260627] ath10k_pci 0000:0d:00.0: firmware register dump: [ 3078.260640] ath10k_pci 0000:0d:00.0: [00]: 0x4100016C 0x000015B3 0x009A31BB 0x00955B31 [ 3078.260647] ath10k_pci 0000:0d:00.0: [04]: 0x009A31BB 0x00060130 0x00000008 0x00000007 [ 3078.260652] ath10k_pci 0000:0d:00.0: [08]: 0x00000000 0x00955B31 0x00000000 0x0040F89E [ 3078.260656] ath10k_pci 0000:0d:00.0: [12]: 0x00000009 0xFFFFFFFF 0x009580F5 0x00958117 [ 3078.260660] ath10k_pci 0000:0d:00.0: [16]: 0x00958080 0x0094085D 0x00000000 0x00000000 [ 3078.260664] ath10k_pci 0000:0d:00.0: [20]: 0x409A31BB 0x0040AA84 0x00000002 0x00000001 [ 3078.260669] ath10k_pci 0000:0d:00.0: [24]: 0x809A2B8D 0x0040AAE4 0x00000088 0xC09A31BB [ 3078.260673] ath10k_pci 0000:0d:00.0: [28]: 0x809898C8 0x0040AB04 0x0043F91C 0x009C6458 [ 3078.260677] ath10k_pci 0000:0d:00.0: [32]: 0x809B66AC 0x0040AB34 0x009C6458 0x0043F91C [ 3078.260686] ath10k_pci 0000:0d:00.0: [36]: 0x809B2824 0x0040ADA4 0x00400000 0x00416EB4 [ 3078.260692] ath10k_pci 0000:0d:00.0: [40]: 0x809C07D9 0x0040ADE4 0x0040AE08 0x00412028 [ 3078.260696] ath10k_pci 0000:0d:00.0: [44]: 0x809486FA 0x0040AE04 0x00000001 0x00000000 [ 3078.260700] ath10k_pci 0000:0d:00.0: [48]: 0x80948E2C 0x0040AEA4 0x0041F4F0 0x00412634 [ 3078.260704] ath10k_pci 0000:0d:00.0: [52]: 0x809BFC39 0x0040AEC4 0x0041F4F0 0x00000001 [ 3078.260709] ath10k_pci 0000:0d:00.0: [56]: 0x80940F18 0x0040AF14 0x00000010 0x00403AC0 [ 3078.284130] ath10k_pci 0000:0d:00.0: failed to to request monitor vdev 1 stop: -108 Fix this by checking whether the sta has already initialized rate control using the flag for that purpose. We can also drop the unnecessary insert parameter here. Signed-off-by: Bob Copeland Signed-off-by: Johannes Berg --- net/mac80211/mesh_plink.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/net/mac80211/mesh_plink.c b/net/mac80211/mesh_plink.c index 0f6c9ca59062..5b5b0f95ffd1 100644 --- a/net/mac80211/mesh_plink.c +++ b/net/mac80211/mesh_plink.c @@ -401,7 +401,7 @@ u32 mesh_plink_deactivate(struct sta_info *sta) static void mesh_sta_info_init(struct ieee80211_sub_if_data *sdata, struct sta_info *sta, - struct ieee802_11_elems *elems, bool insert) + struct ieee802_11_elems *elems) { struct ieee80211_local *local = sdata->local; struct ieee80211_supported_band *sband; @@ -447,7 +447,7 @@ static void mesh_sta_info_init(struct ieee80211_sub_if_data *sdata, sta->sta.bandwidth = IEEE80211_STA_RX_BW_20; } - if (insert) + if (!test_sta_flag(sta, WLAN_STA_RATE_CONTROL)) rate_control_rate_init(sta); else rate_control_rate_update(local, sband, sta, changed); @@ -551,7 +551,7 @@ mesh_sta_info_get(struct ieee80211_sub_if_data *sdata, rcu_read_lock(); sta = sta_info_get(sdata, addr); if (sta) { - mesh_sta_info_init(sdata, sta, elems, false); + mesh_sta_info_init(sdata, sta, elems); } else { rcu_read_unlock(); /* can't run atomic */ @@ -561,7 +561,7 @@ mesh_sta_info_get(struct ieee80211_sub_if_data *sdata, return NULL; } - mesh_sta_info_init(sdata, sta, elems, true); + mesh_sta_info_init(sdata, sta, elems); if (sta_info_insert_rcu(sta)) return NULL; From 240da953fcc6a9008c92fae5b1f727ee5ed167ab Mon Sep 17 00:00:00 2001 From: Konrad Rzeszutek Wilk Date: Wed, 16 May 2018 23:18:09 -0400 Subject: [PATCH 094/190] x86/bugs: Rename SSBD_NO to SSB_NO The "336996 Speculative Execution Side Channel Mitigations" from May defines this as SSB_NO, hence lets sync-up. Signed-off-by: Konrad Rzeszutek Wilk Signed-off-by: Thomas Gleixner --- arch/x86/include/asm/msr-index.h | 2 +- arch/x86/kernel/cpu/common.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/x86/include/asm/msr-index.h b/arch/x86/include/asm/msr-index.h index 562414d5b834..fda2114197b3 100644 --- a/arch/x86/include/asm/msr-index.h +++ b/arch/x86/include/asm/msr-index.h @@ -70,7 +70,7 @@ #define MSR_IA32_ARCH_CAPABILITIES 0x0000010a #define ARCH_CAP_RDCL_NO (1 << 0) /* Not susceptible to Meltdown */ #define ARCH_CAP_IBRS_ALL (1 << 1) /* Enhanced IBRS support */ -#define ARCH_CAP_SSBD_NO (1 << 4) /* +#define ARCH_CAP_SSB_NO (1 << 4) /* * Not susceptible to Speculative Store Bypass * attack, so no Speculative Store Bypass * control required. diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index b4247ed0c81e..78decc3e3067 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -974,7 +974,7 @@ static void __init cpu_set_bug_bits(struct cpuinfo_x86 *c) rdmsrl(MSR_IA32_ARCH_CAPABILITIES, ia32_cap); if (!x86_match_cpu(cpu_no_spec_store_bypass) && - !(ia32_cap & ARCH_CAP_SSBD_NO)) + !(ia32_cap & ARCH_CAP_SSB_NO)) setup_force_cpu_bug(X86_BUG_SPEC_STORE_BYPASS); if (x86_match_cpu(cpu_no_speculation)) From a45b599ad808c3c982fdcdc12b0b8611c2f92824 Mon Sep 17 00:00:00 2001 From: Alexander Potapenko Date: Fri, 18 May 2018 16:23:18 +0200 Subject: [PATCH 095/190] scsi: sg: allocate with __GFP_ZERO in sg_build_indirect() This shall help avoid copying uninitialized memory to the userspace when calling ioctl(fd, SG_IO) with an empty command. Reported-by: syzbot+7d26fc1eea198488deab@syzkaller.appspotmail.com Cc: stable@vger.kernel.org Signed-off-by: Alexander Potapenko Acked-by: Douglas Gilbert Reviewed-by: Johannes Thumshirn Signed-off-by: Martin K. Petersen --- drivers/scsi/sg.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/scsi/sg.c b/drivers/scsi/sg.c index c198b96368dd..5c40d809830f 100644 --- a/drivers/scsi/sg.c +++ b/drivers/scsi/sg.c @@ -1894,7 +1894,7 @@ sg_build_indirect(Sg_scatter_hold * schp, Sg_fd * sfp, int buff_size) num = (rem_sz > scatter_elem_sz_prev) ? scatter_elem_sz_prev : rem_sz; - schp->pages[k] = alloc_pages(gfp_mask, order); + schp->pages[k] = alloc_pages(gfp_mask | __GFP_ZERO, order); if (!schp->pages[k]) goto out; From 4855c92dbb7b3b85c23e88ab7ca04f99b9677b41 Mon Sep 17 00:00:00 2001 From: Joe Jin Date: Thu, 17 May 2018 12:33:28 -0700 Subject: [PATCH 096/190] xen-swiotlb: fix the check condition for xen_swiotlb_free_coherent When run raidconfig from Dom0 we found that the Xen DMA heap is reduced, but Dom Heap is increased by the same size. Tracing raidconfig we found that the related ioctl() in megaraid_sas will call dma_alloc_coherent() to apply memory. If the memory allocated by Dom0 is not in the DMA area, it will exchange memory with Xen to meet the requiment. Later drivers call dma_free_coherent() to free the memory, on xen_swiotlb_free_coherent() the check condition (dev_addr + size - 1 <= dma_mask) is always false, it prevents calling xen_destroy_contiguous_region() to return the memory to the Xen DMA heap. This issue introduced by commit 6810df88dcfc2 "xen-swiotlb: When doing coherent alloc/dealloc check before swizzling the MFNs.". Signed-off-by: Joe Jin Tested-by: John Sobecki Reviewed-by: Rzeszutek Wilk Cc: stable@vger.kernel.org Signed-off-by: Konrad Rzeszutek Wilk --- drivers/xen/swiotlb-xen.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/xen/swiotlb-xen.c b/drivers/xen/swiotlb-xen.c index e1c60899fdbc..a6f9ba85dc4b 100644 --- a/drivers/xen/swiotlb-xen.c +++ b/drivers/xen/swiotlb-xen.c @@ -351,7 +351,7 @@ xen_swiotlb_free_coherent(struct device *hwdev, size_t size, void *vaddr, * physical address */ phys = xen_bus_to_phys(dev_addr); - if (((dev_addr + size - 1 > dma_mask)) || + if (((dev_addr + size - 1 <= dma_mask)) || range_straddles_page_boundary(phys, size)) xen_destroy_contiguous_region(phys, order); From a6837d26674afd2ccb789fe666b4aff7095dc3d2 Mon Sep 17 00:00:00 2001 From: Anders Roxell Date: Fri, 18 May 2018 20:23:24 +0200 Subject: [PATCH 097/190] selftests: bpf: config: enable NET_SCH_INGRESS for xdp_meta.sh When running bpf's selftest test_xdp_meta.sh it fails: ./test_xdp_meta.sh Error: Specified qdisc not found. selftests: test_xdp_meta [FAILED] Need to enable CONFIG_NET_SCH_INGRESS and CONFIG_NET_CLS_ACT to get the test to pass. Fixes: 22c8852624fc ("bpf: improve selftests and add tests for meta pointer") Signed-off-by: Anders Roxell Signed-off-by: Daniel Borkmann --- tools/testing/selftests/bpf/config | 2 ++ 1 file changed, 2 insertions(+) diff --git a/tools/testing/selftests/bpf/config b/tools/testing/selftests/bpf/config index 983dd25d49f4..1eefe211a4a8 100644 --- a/tools/testing/selftests/bpf/config +++ b/tools/testing/selftests/bpf/config @@ -5,3 +5,5 @@ CONFIG_BPF_EVENTS=y CONFIG_TEST_BPF=m CONFIG_CGROUP_BPF=y CONFIG_NETDEVSIM=m +CONFIG_NET_CLS_ACT=y +CONFIG_NET_SCH_INGRESS=y From af86ca4e3088fe5eacf2f7e58c01fa68ca067672 Mon Sep 17 00:00:00 2001 From: Alexei Starovoitov Date: Tue, 15 May 2018 09:27:05 -0700 Subject: [PATCH 098/190] bpf: Prevent memory disambiguation attack Detect code patterns where malicious 'speculative store bypass' can be used and sanitize such patterns. 39: (bf) r3 = r10 40: (07) r3 += -216 41: (79) r8 = *(u64 *)(r7 +0) // slow read 42: (7a) *(u64 *)(r10 -72) = 0 // verifier inserts this instruction 43: (7b) *(u64 *)(r8 +0) = r3 // this store becomes slow due to r8 44: (79) r1 = *(u64 *)(r6 +0) // cpu speculatively executes this load 45: (71) r2 = *(u8 *)(r1 +0) // speculatively arbitrary 'load byte' // is now sanitized Above code after x86 JIT becomes: e5: mov %rbp,%rdx e8: add $0xffffffffffffff28,%rdx ef: mov 0x0(%r13),%r14 f3: movq $0x0,-0x48(%rbp) fb: mov %rdx,0x0(%r14) ff: mov 0x0(%rbx),%rdi 103: movzbq 0x0(%rdi),%rsi Signed-off-by: Alexei Starovoitov Signed-off-by: Thomas Gleixner --- include/linux/bpf_verifier.h | 1 + kernel/bpf/verifier.c | 59 ++++++++++++++++++++++++++++++++++-- 2 files changed, 57 insertions(+), 3 deletions(-) diff --git a/include/linux/bpf_verifier.h b/include/linux/bpf_verifier.h index 7e61c395fddf..65cfc2f59db9 100644 --- a/include/linux/bpf_verifier.h +++ b/include/linux/bpf_verifier.h @@ -146,6 +146,7 @@ struct bpf_insn_aux_data { s32 call_imm; /* saved imm field of call insn */ }; int ctx_field_size; /* the ctx field size for load insn, maybe 0 */ + int sanitize_stack_off; /* stack slot to be cleared */ bool seen; /* this insn was processed by the verifier */ }; diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index 5dd1dcb902bf..2ce967a63ede 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -978,7 +978,7 @@ static bool register_is_null(struct bpf_reg_state *reg) */ static int check_stack_write(struct bpf_verifier_env *env, struct bpf_func_state *state, /* func where register points to */ - int off, int size, int value_regno) + int off, int size, int value_regno, int insn_idx) { struct bpf_func_state *cur; /* state of the current function */ int i, slot = -off - 1, spi = slot / BPF_REG_SIZE, err; @@ -1017,8 +1017,33 @@ static int check_stack_write(struct bpf_verifier_env *env, state->stack[spi].spilled_ptr = cur->regs[value_regno]; state->stack[spi].spilled_ptr.live |= REG_LIVE_WRITTEN; - for (i = 0; i < BPF_REG_SIZE; i++) + for (i = 0; i < BPF_REG_SIZE; i++) { + if (state->stack[spi].slot_type[i] == STACK_MISC && + !env->allow_ptr_leaks) { + int *poff = &env->insn_aux_data[insn_idx].sanitize_stack_off; + int soff = (-spi - 1) * BPF_REG_SIZE; + + /* detected reuse of integer stack slot with a pointer + * which means either llvm is reusing stack slot or + * an attacker is trying to exploit CVE-2018-3639 + * (speculative store bypass) + * Have to sanitize that slot with preemptive + * store of zero. + */ + if (*poff && *poff != soff) { + /* disallow programs where single insn stores + * into two different stack slots, since verifier + * cannot sanitize them + */ + verbose(env, + "insn %d cannot access two stack slots fp%d and fp%d", + insn_idx, *poff, soff); + return -EINVAL; + } + *poff = soff; + } state->stack[spi].slot_type[i] = STACK_SPILL; + } } else { u8 type = STACK_MISC; @@ -1694,7 +1719,7 @@ static int check_mem_access(struct bpf_verifier_env *env, int insn_idx, u32 regn if (t == BPF_WRITE) err = check_stack_write(env, state, off, size, - value_regno); + value_regno, insn_idx); else err = check_stack_read(env, state, off, size, value_regno); @@ -5169,6 +5194,34 @@ static int convert_ctx_accesses(struct bpf_verifier_env *env) else continue; + if (type == BPF_WRITE && + env->insn_aux_data[i + delta].sanitize_stack_off) { + struct bpf_insn patch[] = { + /* Sanitize suspicious stack slot with zero. + * There are no memory dependencies for this store, + * since it's only using frame pointer and immediate + * constant of zero + */ + BPF_ST_MEM(BPF_DW, BPF_REG_FP, + env->insn_aux_data[i + delta].sanitize_stack_off, + 0), + /* the original STX instruction will immediately + * overwrite the same stack slot with appropriate value + */ + *insn, + }; + + cnt = ARRAY_SIZE(patch); + new_prog = bpf_patch_insn_data(env, i + delta, patch, cnt); + if (!new_prog) + return -ENOMEM; + + delta += cnt - 1; + env->prog = new_prog; + insn = new_prog->insnsi + i + delta; + continue; + } + if (env->insn_aux_data[i + delta].ptr_type != PTR_TO_CTX) continue; From 643ca198aacc671f32ef7c0c2783f0b539070a36 Mon Sep 17 00:00:00 2001 From: Laurent Pinchart Date: Fri, 27 Apr 2018 22:40:21 +0300 Subject: [PATCH 099/190] drm: rcar-du: lvds: Fix crash in .atomic_check when disabling connector The connector .atomic_check() handler can be called with a NULL crtc pointer in the connector state when the connector gets disabled explicitly (through performing a legacy mode set or setting the connector's CRTC_ID property to 0). This causes a crash as the crtc pointer is dereferenced without any check. Fix it by returning from the .atomic_check() handler when then crtc pointer is NULL, as there is no check to be performed when the connector gets disabled. Fixes: c6a27fa41fab ("drm: rcar-du: Convert LVDS encoder code to bridge driver") Signed-off-by: Laurent Pinchart Reviewed-by: Kieran Bingham --- drivers/gpu/drm/rcar-du/rcar_lvds.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/gpu/drm/rcar-du/rcar_lvds.c b/drivers/gpu/drm/rcar-du/rcar_lvds.c index 3d2d3bbd1342..155ad840f3c5 100644 --- a/drivers/gpu/drm/rcar-du/rcar_lvds.c +++ b/drivers/gpu/drm/rcar-du/rcar_lvds.c @@ -88,6 +88,9 @@ static int rcar_lvds_connector_atomic_check(struct drm_connector *connector, const struct drm_display_mode *panel_mode; struct drm_crtc_state *crtc_state; + if (!state->crtc) + return 0; + if (list_empty(&connector->modes)) { dev_dbg(lvds->dev, "connector: empty modes list\n"); return -EINVAL; From b25b750df99bcba29317d3f9d9f93c4ec58890e6 Mon Sep 17 00:00:00 2001 From: Mathieu Malaterre Date: Wed, 16 May 2018 21:20:20 +0200 Subject: [PATCH 100/190] mmc: block: propagate correct returned value in mmc_rpmb_ioctl MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit In commit 97548575bef3 ("mmc: block: Convert RPMB to a character device") a new function `mmc_rpmb_ioctl` was added. The final return is simply returning a value of `0` instead of propagating the correct return code. Discovered during a compilation with W=1, silence the following gcc warning drivers/mmc/core/block.c:2470:6: warning: variable ‘ret’ set but not used [-Wunused-but-set-variable] Signed-off-by: Mathieu Malaterre Reviewed-by: Shawn Lin Fixes: 97548575bef3 ("mmc: block: Convert RPMB to a character device") Cc: stable@vger.kernel.org # v4.15+ Signed-off-by: Ulf Hansson --- drivers/mmc/core/block.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/mmc/core/block.c b/drivers/mmc/core/block.c index 9e923cd1d80e..38a7586b00cc 100644 --- a/drivers/mmc/core/block.c +++ b/drivers/mmc/core/block.c @@ -2485,7 +2485,7 @@ static long mmc_rpmb_ioctl(struct file *filp, unsigned int cmd, break; } - return 0; + return ret; } #ifdef CONFIG_COMPAT From 4c94238f37af87a2165c3fb491b4a8b50e90649c Mon Sep 17 00:00:00 2001 From: Srinath Mannam Date: Fri, 18 May 2018 15:03:55 -0700 Subject: [PATCH 101/190] mmc: sdhci-iproc: remove hard coded mmc cap 1.8v Remove hard coded mmc cap 1.8v from platform data as it is board specific. The 1.8v DDR mmc caps can be enabled using DTS property for those boards that support it. Fixes: b17b4ab8ce38 ("mmc: sdhci-iproc: define MMC caps in platform data") Signed-off-by: Srinath Mannam Signed-off-by: Scott Branden Reviewed-by: Ray Jui Cc: stable@vger.kernel.org # v4.8+ Signed-off-by: Ulf Hansson --- drivers/mmc/host/sdhci-iproc.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/mmc/host/sdhci-iproc.c b/drivers/mmc/host/sdhci-iproc.c index 0ef741bc515d..6f430da9ae2f 100644 --- a/drivers/mmc/host/sdhci-iproc.c +++ b/drivers/mmc/host/sdhci-iproc.c @@ -206,7 +206,6 @@ static const struct sdhci_iproc_data iproc_data = { .caps1 = SDHCI_DRIVER_TYPE_C | SDHCI_DRIVER_TYPE_D | SDHCI_SUPPORT_DDR50, - .mmc_caps = MMC_CAP_1_8V_DDR, }; static const struct sdhci_pltfm_data sdhci_bcm2835_pltfm_data = { From 5f651b870485ee60f5abbbd85195a6852978894a Mon Sep 17 00:00:00 2001 From: Corneliu Doban Date: Fri, 18 May 2018 15:03:56 -0700 Subject: [PATCH 102/190] mmc: sdhci-iproc: fix 32bit writes for TRANSFER_MODE register When the host controller accepts only 32bit writes, the value of the 16bit TRANSFER_MODE register, that has the same 32bit address as the 16bit COMMAND register, needs to be saved and it will be written in a 32bit write together with the command as this will trigger the host to send the command on the SD interface. When sending the tuning command, TRANSFER_MODE is written and then sdhci_set_transfer_mode reads it back to clear AUTO_CMD12 bit and write it again resulting in wrong value to be written because the initial write value was saved in a shadow and the read-back returned a wrong value, from the register. Fix sdhci_iproc_readw to return the saved value of TRANSFER_MODE when a saved value exist. Same fix for read of BLOCK_SIZE and BLOCK_COUNT registers, that are saved for a different reason, although a scenario that will cause the mentioned problem on this registers is not probable. Fixes: b580c52d58d9 ("mmc: sdhci-iproc: add IPROC SDHCI driver") Signed-off-by: Corneliu Doban Signed-off-by: Scott Branden Cc: stable@vger.kernel.org # v4.1+ Signed-off-by: Ulf Hansson --- drivers/mmc/host/sdhci-iproc.c | 30 +++++++++++++++++++++++++----- 1 file changed, 25 insertions(+), 5 deletions(-) diff --git a/drivers/mmc/host/sdhci-iproc.c b/drivers/mmc/host/sdhci-iproc.c index 6f430da9ae2f..1f0ab08e8adf 100644 --- a/drivers/mmc/host/sdhci-iproc.c +++ b/drivers/mmc/host/sdhci-iproc.c @@ -33,6 +33,8 @@ struct sdhci_iproc_host { const struct sdhci_iproc_data *data; u32 shadow_cmd; u32 shadow_blk; + bool is_cmd_shadowed; + bool is_blk_shadowed; }; #define REG_OFFSET_IN_BITS(reg) ((reg) << 3 & 0x18) @@ -48,8 +50,22 @@ static inline u32 sdhci_iproc_readl(struct sdhci_host *host, int reg) static u16 sdhci_iproc_readw(struct sdhci_host *host, int reg) { - u32 val = sdhci_iproc_readl(host, (reg & ~3)); - u16 word = val >> REG_OFFSET_IN_BITS(reg) & 0xffff; + struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host); + struct sdhci_iproc_host *iproc_host = sdhci_pltfm_priv(pltfm_host); + u32 val; + u16 word; + + if ((reg == SDHCI_TRANSFER_MODE) && iproc_host->is_cmd_shadowed) { + /* Get the saved transfer mode */ + val = iproc_host->shadow_cmd; + } else if ((reg == SDHCI_BLOCK_SIZE || reg == SDHCI_BLOCK_COUNT) && + iproc_host->is_blk_shadowed) { + /* Get the saved block info */ + val = iproc_host->shadow_blk; + } else { + val = sdhci_iproc_readl(host, (reg & ~3)); + } + word = val >> REG_OFFSET_IN_BITS(reg) & 0xffff; return word; } @@ -105,13 +121,15 @@ static void sdhci_iproc_writew(struct sdhci_host *host, u16 val, int reg) if (reg == SDHCI_COMMAND) { /* Write the block now as we are issuing a command */ - if (iproc_host->shadow_blk != 0) { + if (iproc_host->is_blk_shadowed) { sdhci_iproc_writel(host, iproc_host->shadow_blk, SDHCI_BLOCK_SIZE); - iproc_host->shadow_blk = 0; + iproc_host->is_blk_shadowed = false; } oldval = iproc_host->shadow_cmd; - } else if (reg == SDHCI_BLOCK_SIZE || reg == SDHCI_BLOCK_COUNT) { + iproc_host->is_cmd_shadowed = false; + } else if ((reg == SDHCI_BLOCK_SIZE || reg == SDHCI_BLOCK_COUNT) && + iproc_host->is_blk_shadowed) { /* Block size and count are stored in shadow reg */ oldval = iproc_host->shadow_blk; } else { @@ -123,9 +141,11 @@ static void sdhci_iproc_writew(struct sdhci_host *host, u16 val, int reg) if (reg == SDHCI_TRANSFER_MODE) { /* Save the transfer mode until the command is issued */ iproc_host->shadow_cmd = newval; + iproc_host->is_cmd_shadowed = true; } else if (reg == SDHCI_BLOCK_SIZE || reg == SDHCI_BLOCK_COUNT) { /* Save the block info until the command is issued */ iproc_host->shadow_blk = newval; + iproc_host->is_blk_shadowed = true; } else { /* Command or other regular 32-bit write */ sdhci_iproc_writel(host, newval, reg & ~3); From 3de06d5a1f05c11c94cbb68af14dbfa7fb81d78b Mon Sep 17 00:00:00 2001 From: Corneliu Doban Date: Fri, 18 May 2018 15:03:57 -0700 Subject: [PATCH 103/190] mmc: sdhci-iproc: add SDHCI_QUIRK2_HOST_OFF_CARD_ON for cygnus The SDHCI_QUIRK2_HOST_OFF_CARD_ON is needed for the driver to properly reset the host controller (reset all) on initialization after exiting deep sleep. Signed-off-by: Corneliu Doban Signed-off-by: Scott Branden Reviewed-by: Ray Jui Reviewed-by: Srinath Mannam Fixes: c833e92bbb60 ("mmc: sdhci-iproc: support standard byte register accesses") Cc: stable@vger.kernel.org # v4.10+ Signed-off-by: Ulf Hansson --- drivers/mmc/host/sdhci-iproc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/mmc/host/sdhci-iproc.c b/drivers/mmc/host/sdhci-iproc.c index 1f0ab08e8adf..d0e83db42ae5 100644 --- a/drivers/mmc/host/sdhci-iproc.c +++ b/drivers/mmc/host/sdhci-iproc.c @@ -186,7 +186,7 @@ static const struct sdhci_ops sdhci_iproc_32only_ops = { static const struct sdhci_pltfm_data sdhci_iproc_cygnus_pltfm_data = { .quirks = SDHCI_QUIRK_DATA_TIMEOUT_USES_SDCLK, - .quirks2 = SDHCI_QUIRK2_ACMD23_BROKEN, + .quirks2 = SDHCI_QUIRK2_ACMD23_BROKEN | SDHCI_QUIRK2_HOST_OFF_CARD_ON, .ops = &sdhci_iproc_32only_ops, }; From 136d769e0b3475d71350aa3648a116a6ee7a8f6c Mon Sep 17 00:00:00 2001 From: Sudip Mukherjee Date: Sat, 19 May 2018 22:29:36 +0100 Subject: [PATCH 104/190] libata: blacklist Micron 500IT SSD with MU01 firmware While whitelisting Micron M500DC drives, the tweaked blacklist entry enabled queued TRIM from M500IT variants also. But these do not support queued TRIM. And while using those SSDs with the latest kernel we have seen errors and even the partition table getting corrupted. Some part from the dmesg: [ 6.727384] ata1.00: ATA-9: Micron_M500IT_MTFDDAK060MBD, MU01, max UDMA/133 [ 6.727390] ata1.00: 117231408 sectors, multi 16: LBA48 NCQ (depth 31/32), AA [ 6.741026] ata1.00: supports DRM functions and may not be fully accessible [ 6.759887] ata1.00: configured for UDMA/133 [ 6.762256] scsi 0:0:0:0: Direct-Access ATA Micron_M500IT_MT MU01 PQ: 0 ANSI: 5 and then for the error: [ 120.860334] ata1.00: exception Emask 0x1 SAct 0x7ffc0007 SErr 0x0 action 0x6 frozen [ 120.860338] ata1.00: irq_stat 0x40000008 [ 120.860342] ata1.00: failed command: SEND FPDMA QUEUED [ 120.860351] ata1.00: cmd 64/01:00:00:00:00/00:00:00:00:00/a0 tag 0 ncq dma 512 out res 40/00:00:00:00:00/00:00:00:00:00/00 Emask 0x5 (timeout) [ 120.860353] ata1.00: status: { DRDY } [ 120.860543] ata1: hard resetting link [ 121.166128] ata1: SATA link up 3.0 Gbps (SStatus 123 SControl 300) [ 121.166376] ata1.00: supports DRM functions and may not be fully accessible [ 121.186238] ata1.00: supports DRM functions and may not be fully accessible [ 121.204445] ata1.00: configured for UDMA/133 [ 121.204454] ata1.00: device reported invalid CHS sector 0 [ 121.204541] sd 0:0:0:0: [sda] tag#18 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=0x08 [ 121.204546] sd 0:0:0:0: [sda] tag#18 Sense Key : 0x5 [current] [ 121.204550] sd 0:0:0:0: [sda] tag#18 ASC=0x21 ASCQ=0x4 [ 121.204555] sd 0:0:0:0: [sda] tag#18 CDB: opcode=0x93 93 08 00 00 00 00 00 04 28 80 00 00 00 30 00 00 [ 121.204559] print_req_error: I/O error, dev sda, sector 272512 After few reboots with these errors, and the SSD is corrupted. After blacklisting it, the errors are not seen and the SSD does not get corrupted any more. Fixes: 243918be6393 ("libata: Do not blacklist Micron M500DC") Cc: Martin K. Petersen Cc: stable@vger.kernel.org Signed-off-by: Sudip Mukherjee Signed-off-by: Tejun Heo --- drivers/ata/libata-core.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index 7ed2f009911a..346b163f6e89 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -4561,6 +4561,8 @@ static const struct ata_blacklist_entry ata_device_blacklist [] = { { "SanDisk SD7UB3Q*G1001", NULL, ATA_HORKAGE_NOLPM, }, /* devices that don't properly handle queued TRIM commands */ + { "Micron_M500IT_*", "MU01", ATA_HORKAGE_NO_NCQ_TRIM | + ATA_HORKAGE_ZERO_AFTER_TRIM, }, { "Micron_M500_*", NULL, ATA_HORKAGE_NO_NCQ_TRIM | ATA_HORKAGE_ZERO_AFTER_TRIM, }, { "Crucial_CT*M500*", NULL, ATA_HORKAGE_NO_NCQ_TRIM | From 32c3fa7cdf0c4a3eb8405fc3e13398de019e828b Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Mon, 21 May 2018 17:44:57 +0100 Subject: [PATCH 105/190] arm64: lse: Add early clobbers to some input/output asm operands For LSE atomics that read and write a register operand, we need to ensure that these operands are annotated as "early clobber" if the register is written before all of the input operands have been consumed. Failure to do so can result in the compiler allocating the same register to both operands, leading to splats such as: Unable to handle kernel paging request at virtual address 11111122222221 [...] x1 : 1111111122222222 x0 : 1111111122222221 Process swapper/0 (pid: 1, stack limit = 0x000000008209f908) Call trace: test_atomic64+0x1360/0x155c where x0 has been allocated as both the value to be stored and also the atomic_t pointer. This patch adds the missing clobbers. Cc: Cc: Dave Martin Cc: Robin Murphy Reported-by: Mark Salter Signed-off-by: Will Deacon --- arch/arm64/include/asm/atomic_lse.h | 24 ++++++++++++------------ 1 file changed, 12 insertions(+), 12 deletions(-) diff --git a/arch/arm64/include/asm/atomic_lse.h b/arch/arm64/include/asm/atomic_lse.h index 9ef0797380cb..f9b0b09153e0 100644 --- a/arch/arm64/include/asm/atomic_lse.h +++ b/arch/arm64/include/asm/atomic_lse.h @@ -117,7 +117,7 @@ static inline void atomic_and(int i, atomic_t *v) /* LSE atomics */ " mvn %w[i], %w[i]\n" " stclr %w[i], %[v]") - : [i] "+r" (w0), [v] "+Q" (v->counter) + : [i] "+&r" (w0), [v] "+Q" (v->counter) : "r" (x1) : __LL_SC_CLOBBERS); } @@ -135,7 +135,7 @@ static inline int atomic_fetch_and##name(int i, atomic_t *v) \ /* LSE atomics */ \ " mvn %w[i], %w[i]\n" \ " ldclr" #mb " %w[i], %w[i], %[v]") \ - : [i] "+r" (w0), [v] "+Q" (v->counter) \ + : [i] "+&r" (w0), [v] "+Q" (v->counter) \ : "r" (x1) \ : __LL_SC_CLOBBERS, ##cl); \ \ @@ -161,7 +161,7 @@ static inline void atomic_sub(int i, atomic_t *v) /* LSE atomics */ " neg %w[i], %w[i]\n" " stadd %w[i], %[v]") - : [i] "+r" (w0), [v] "+Q" (v->counter) + : [i] "+&r" (w0), [v] "+Q" (v->counter) : "r" (x1) : __LL_SC_CLOBBERS); } @@ -180,7 +180,7 @@ static inline int atomic_sub_return##name(int i, atomic_t *v) \ " neg %w[i], %w[i]\n" \ " ldadd" #mb " %w[i], w30, %[v]\n" \ " add %w[i], %w[i], w30") \ - : [i] "+r" (w0), [v] "+Q" (v->counter) \ + : [i] "+&r" (w0), [v] "+Q" (v->counter) \ : "r" (x1) \ : __LL_SC_CLOBBERS , ##cl); \ \ @@ -207,7 +207,7 @@ static inline int atomic_fetch_sub##name(int i, atomic_t *v) \ /* LSE atomics */ \ " neg %w[i], %w[i]\n" \ " ldadd" #mb " %w[i], %w[i], %[v]") \ - : [i] "+r" (w0), [v] "+Q" (v->counter) \ + : [i] "+&r" (w0), [v] "+Q" (v->counter) \ : "r" (x1) \ : __LL_SC_CLOBBERS, ##cl); \ \ @@ -314,7 +314,7 @@ static inline void atomic64_and(long i, atomic64_t *v) /* LSE atomics */ " mvn %[i], %[i]\n" " stclr %[i], %[v]") - : [i] "+r" (x0), [v] "+Q" (v->counter) + : [i] "+&r" (x0), [v] "+Q" (v->counter) : "r" (x1) : __LL_SC_CLOBBERS); } @@ -332,7 +332,7 @@ static inline long atomic64_fetch_and##name(long i, atomic64_t *v) \ /* LSE atomics */ \ " mvn %[i], %[i]\n" \ " ldclr" #mb " %[i], %[i], %[v]") \ - : [i] "+r" (x0), [v] "+Q" (v->counter) \ + : [i] "+&r" (x0), [v] "+Q" (v->counter) \ : "r" (x1) \ : __LL_SC_CLOBBERS, ##cl); \ \ @@ -358,7 +358,7 @@ static inline void atomic64_sub(long i, atomic64_t *v) /* LSE atomics */ " neg %[i], %[i]\n" " stadd %[i], %[v]") - : [i] "+r" (x0), [v] "+Q" (v->counter) + : [i] "+&r" (x0), [v] "+Q" (v->counter) : "r" (x1) : __LL_SC_CLOBBERS); } @@ -377,7 +377,7 @@ static inline long atomic64_sub_return##name(long i, atomic64_t *v) \ " neg %[i], %[i]\n" \ " ldadd" #mb " %[i], x30, %[v]\n" \ " add %[i], %[i], x30") \ - : [i] "+r" (x0), [v] "+Q" (v->counter) \ + : [i] "+&r" (x0), [v] "+Q" (v->counter) \ : "r" (x1) \ : __LL_SC_CLOBBERS, ##cl); \ \ @@ -404,7 +404,7 @@ static inline long atomic64_fetch_sub##name(long i, atomic64_t *v) \ /* LSE atomics */ \ " neg %[i], %[i]\n" \ " ldadd" #mb " %[i], %[i], %[v]") \ - : [i] "+r" (x0), [v] "+Q" (v->counter) \ + : [i] "+&r" (x0), [v] "+Q" (v->counter) \ : "r" (x1) \ : __LL_SC_CLOBBERS, ##cl); \ \ @@ -435,7 +435,7 @@ static inline long atomic64_dec_if_positive(atomic64_t *v) " sub x30, x30, %[ret]\n" " cbnz x30, 1b\n" "2:") - : [ret] "+r" (x0), [v] "+Q" (v->counter) + : [ret] "+&r" (x0), [v] "+Q" (v->counter) : : __LL_SC_CLOBBERS, "cc", "memory"); @@ -516,7 +516,7 @@ static inline long __cmpxchg_double##name(unsigned long old1, \ " eor %[old1], %[old1], %[oldval1]\n" \ " eor %[old2], %[old2], %[oldval2]\n" \ " orr %[old1], %[old1], %[old2]") \ - : [old1] "+r" (x0), [old2] "+r" (x1), \ + : [old1] "+&r" (x0), [old2] "+&r" (x1), \ [v] "+Q" (*(unsigned long *)ptr) \ : [new1] "r" (x2), [new2] "r" (x3), [ptr] "r" (x4), \ [oldval1] "r" (oldval1), [oldval2] "r" (oldval2) \ From 255845fc43a3aaf806852a1d3bc89bff1411ebe3 Mon Sep 17 00:00:00 2001 From: "Jason A. Donenfeld" Date: Sat, 28 Apr 2018 00:42:52 +0200 Subject: [PATCH 106/190] arm64: export tishift functions to modules Otherwise modules that use these arithmetic operations will fail to link. We accomplish this with the usual EXPORT_SYMBOL, which on most architectures goes in the .S file but the ARM64 maintainers prefer that insead it goes into arm64ksyms. While we're at it, we also fix this up to use SPDX, and I personally choose to relicense this as GPL2||BSD so that these symbols don't need to be export_symbol_gpl, so all modules can use the routines, since these are important general purpose compiler-generated function calls. Signed-off-by: Jason A. Donenfeld Reported-by: PaX Team Cc: stable@vger.kernel.org Signed-off-by: Will Deacon --- arch/arm64/kernel/arm64ksyms.c | 8 ++++++++ arch/arm64/lib/tishift.S | 15 ++------------- 2 files changed, 10 insertions(+), 13 deletions(-) diff --git a/arch/arm64/kernel/arm64ksyms.c b/arch/arm64/kernel/arm64ksyms.c index 66be504edb6c..d894a20b70b2 100644 --- a/arch/arm64/kernel/arm64ksyms.c +++ b/arch/arm64/kernel/arm64ksyms.c @@ -75,3 +75,11 @@ NOKPROBE_SYMBOL(_mcount); /* arm-smccc */ EXPORT_SYMBOL(__arm_smccc_smc); EXPORT_SYMBOL(__arm_smccc_hvc); + + /* tishift.S */ +extern long long __ashlti3(long long a, int b); +EXPORT_SYMBOL(__ashlti3); +extern long long __ashrti3(long long a, int b); +EXPORT_SYMBOL(__ashrti3); +extern long long __lshrti3(long long a, int b); +EXPORT_SYMBOL(__lshrti3); diff --git a/arch/arm64/lib/tishift.S b/arch/arm64/lib/tishift.S index d3db9b2cd479..0fdff97794de 100644 --- a/arch/arm64/lib/tishift.S +++ b/arch/arm64/lib/tishift.S @@ -1,17 +1,6 @@ -/* - * Copyright (C) 2017 Jason A. Donenfeld . All Rights Reserved. +/* SPDX-License-Identifier: (GPL-2.0 OR BSD-3-Clause) * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program. If not, see . + * Copyright (C) 2017-2018 Jason A. Donenfeld . All Rights Reserved. */ #include From f7068114d45ec55996b9040e98111afa56e010fe Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Mon, 21 May 2018 12:21:14 -0600 Subject: [PATCH 107/190] sr: pass down correctly sized SCSI sense buffer We're casting the CDROM layer request_sense to the SCSI sense buffer, but the former is 64 bytes and the latter is 96 bytes. As we generally allocate these on the stack, we end up blowing up the stack. Fix this by wrapping the scsi_execute() call with a properly sized sense buffer, and copying back the bits for the CDROM layer. Cc: stable@vger.kernel.org Reported-by: Piotr Gabriel Kosinski Reported-by: Daniel Shapira Tested-by: Kees Cook Fixes: 82ed4db499b8 ("block: split scsi_request out of struct request") Signed-off-by: Jens Axboe --- drivers/scsi/sr_ioctl.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/drivers/scsi/sr_ioctl.c b/drivers/scsi/sr_ioctl.c index 2a21f2d48592..35fab1e18adc 100644 --- a/drivers/scsi/sr_ioctl.c +++ b/drivers/scsi/sr_ioctl.c @@ -188,9 +188,13 @@ int sr_do_ioctl(Scsi_CD *cd, struct packet_command *cgc) struct scsi_device *SDev; struct scsi_sense_hdr sshdr; int result, err = 0, retries = 0; + unsigned char sense_buffer[SCSI_SENSE_BUFFERSIZE], *senseptr = NULL; SDev = cd->device; + if (cgc->sense) + senseptr = sense_buffer; + retry: if (!scsi_block_when_processing_errors(SDev)) { err = -ENODEV; @@ -198,10 +202,12 @@ int sr_do_ioctl(Scsi_CD *cd, struct packet_command *cgc) } result = scsi_execute(SDev, cgc->cmd, cgc->data_direction, - cgc->buffer, cgc->buflen, - (unsigned char *)cgc->sense, &sshdr, + cgc->buffer, cgc->buflen, senseptr, &sshdr, cgc->timeout, IOCTL_RETRIES, 0, 0, NULL); + if (cgc->sense) + memcpy(cgc->sense, sense_buffer, sizeof(*cgc->sense)); + /* Minimal error checking. Ignore cases we know about, and report the rest. */ if (driver_byte(result) != 0) { switch (sshdr.sense_key) { From 30da870ce4a4e007c901858a96e9e394a1daa74a Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 6 May 2018 12:15:20 -0400 Subject: [PATCH 108/190] affs_lookup(): close a race with affs_remove_link() we unlock the directory hash too early - if we are looking at secondary link and primary (in another directory) gets removed just as we unlock, we could have the old primary moved in place of the secondary, leaving us to look into freed entry (and leaving our dentry with ->d_fsdata pointing to a freed entry). Cc: stable@vger.kernel.org # 2.4.4+ Acked-by: David Sterba Signed-off-by: Al Viro --- fs/affs/namei.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/fs/affs/namei.c b/fs/affs/namei.c index d8aa0ae3d037..1ed0fa4c4d48 100644 --- a/fs/affs/namei.c +++ b/fs/affs/namei.c @@ -206,9 +206,10 @@ affs_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) affs_lock_dir(dir); bh = affs_find_entry(dir, dentry); - affs_unlock_dir(dir); - if (IS_ERR(bh)) + if (IS_ERR(bh)) { + affs_unlock_dir(dir); return ERR_CAST(bh); + } if (bh) { u32 ino = bh->b_blocknr; @@ -222,10 +223,13 @@ affs_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) } affs_brelse(bh); inode = affs_iget(sb, ino); - if (IS_ERR(inode)) + if (IS_ERR(inode)) { + affs_unlock_dir(dir); return ERR_CAST(inode); + } } d_add(dentry, inode); + affs_unlock_dir(dir); return NULL; } From 87fbd639c02ec96d67738e40b6521fb070ed7168 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 6 May 2018 12:20:40 -0400 Subject: [PATCH 109/190] affs_lookup: switch to d_splice_alias() Making something exportable takes more than providing ->s_export_ops. In particular, ->lookup() *MUST* use d_splice_alias() instead of d_add(). Reading Documentation/filesystems/nfs/Exporting would've been a good idea; as it is, exporting AFFS is badly (and exploitably) broken. Partially-Fixes: ed4433d72394 "fs/affs: make affs exportable" Acked-by: David Sterba Signed-off-by: Al Viro --- fs/affs/namei.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) diff --git a/fs/affs/namei.c b/fs/affs/namei.c index 1ed0fa4c4d48..41c5749f4db7 100644 --- a/fs/affs/namei.c +++ b/fs/affs/namei.c @@ -201,6 +201,7 @@ affs_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) struct super_block *sb = dir->i_sb; struct buffer_head *bh; struct inode *inode = NULL; + struct dentry *res; pr_debug("%s(\"%pd\")\n", __func__, dentry); @@ -223,14 +224,12 @@ affs_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) } affs_brelse(bh); inode = affs_iget(sb, ino); - if (IS_ERR(inode)) { - affs_unlock_dir(dir); - return ERR_CAST(inode); - } } - d_add(dentry, inode); + res = d_splice_alias(inode, dentry); + if (!IS_ERR_OR_NULL(res)) + res->d_fsdata = dentry->d_fsdata; affs_unlock_dir(dir); - return NULL; + return res; } int From f4e4d434fe3f5eceea470bf821683677dabe39c4 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 30 Apr 2018 19:02:02 -0400 Subject: [PATCH 110/190] befs_lookup(): use d_splice_alias() RTFS(Documentation/filesystems/nfs/Exporting) if you try to make something exportable. Fixes: ac632f5b6301 "befs: add NFS export support" Signed-off-by: Al Viro --- fs/befs/linuxvfs.c | 17 +++++------------ 1 file changed, 5 insertions(+), 12 deletions(-) diff --git a/fs/befs/linuxvfs.c b/fs/befs/linuxvfs.c index af2832aaeec5..4700b4534439 100644 --- a/fs/befs/linuxvfs.c +++ b/fs/befs/linuxvfs.c @@ -198,23 +198,16 @@ befs_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) if (ret == BEFS_BT_NOT_FOUND) { befs_debug(sb, "<--- %s %pd not found", __func__, dentry); - d_add(dentry, NULL); - return ERR_PTR(-ENOENT); - + inode = NULL; } else if (ret != BEFS_OK || offset == 0) { befs_error(sb, "<--- %s Error", __func__); - return ERR_PTR(-ENODATA); + inode = ERR_PTR(-ENODATA); + } else { + inode = befs_iget(dir->i_sb, (ino_t) offset); } - - inode = befs_iget(dir->i_sb, (ino_t) offset); - if (IS_ERR(inode)) - return ERR_CAST(inode); - - d_add(dentry, inode); - befs_debug(sb, "<--- %s", __func__); - return NULL; + return d_splice_alias(inode, dentry); } static int From 08a8f3086880325433d66b2dc9cdfb3f095adddf Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Sun, 13 May 2018 15:05:47 -0700 Subject: [PATCH 111/190] cramfs: Fix IS_ENABLED typo There's an extra C here... Fixes: 99c18ce580c6 ("cramfs: direct memory access support") Acked-by: Nicolas Pitre Signed-off-by: Joe Perches Signed-off-by: Al Viro --- fs/cramfs/inode.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/cramfs/inode.c b/fs/cramfs/inode.c index 017b0ab19bc4..124b093d14e5 100644 --- a/fs/cramfs/inode.c +++ b/fs/cramfs/inode.c @@ -492,7 +492,7 @@ static void cramfs_kill_sb(struct super_block *sb) { struct cramfs_sb_info *sbi = CRAMFS_SB(sb); - if (IS_ENABLED(CCONFIG_CRAMFS_MTD) && sb->s_mtd) { + if (IS_ENABLED(CONFIG_CRAMFS_MTD) && sb->s_mtd) { if (sbi && sbi->mtd_point_size) mtd_unpoint(sb->s_mtd, 0, sbi->mtd_point_size); kill_mtd_super(sb); From 82382acec0c97b91830fff7130d0acce4ac4f3f3 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 3 Apr 2018 00:22:29 -0400 Subject: [PATCH 112/190] kernfs: deal with kernfs_fill_super() failures make sure that info->node is initialized early, so that kernfs_kill_sb() can list_del() it safely. Signed-off-by: Al Viro --- fs/kernfs/mount.c | 1 + 1 file changed, 1 insertion(+) diff --git a/fs/kernfs/mount.c b/fs/kernfs/mount.c index 26dd9a50f383..ff2716f9322e 100644 --- a/fs/kernfs/mount.c +++ b/fs/kernfs/mount.c @@ -316,6 +316,7 @@ struct dentry *kernfs_mount_ns(struct file_system_type *fs_type, int flags, info->root = root; info->ns = ns; + INIT_LIST_HEAD(&info->node); sb = sget_userns(fs_type, kernfs_test_super, kernfs_set_super, flags, &init_user_ns, info); From 7b745a4e4051e1bbce40e0b1c2cf636c70583aa4 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 14 May 2018 00:03:34 -0400 Subject: [PATCH 113/190] unfuck sysfs_mount() new_sb is left uninitialized in case of early failures in kernfs_mount_ns(), and while IS_ERR(root) is true in all such cases, using IS_ERR(root) || !new_sb is not a solution - IS_ERR(root) is true in some cases when new_sb is true. Make sure new_sb is initialized (and matches the reality) in all cases and fix the condition for dropping kobj reference - we want it done precisely in those situations where the reference has not been transferred into a new super_block instance. Signed-off-by: Al Viro --- fs/sysfs/mount.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/fs/sysfs/mount.c b/fs/sysfs/mount.c index b428d317ae92..92682fcc41f6 100644 --- a/fs/sysfs/mount.c +++ b/fs/sysfs/mount.c @@ -25,7 +25,7 @@ static struct dentry *sysfs_mount(struct file_system_type *fs_type, { struct dentry *root; void *ns; - bool new_sb; + bool new_sb = false; if (!(flags & SB_KERNMOUNT)) { if (!kobj_ns_current_may_mount(KOBJ_NS_TYPE_NET)) @@ -35,9 +35,9 @@ static struct dentry *sysfs_mount(struct file_system_type *fs_type, ns = kobj_ns_grab_current(KOBJ_NS_TYPE_NET); root = kernfs_mount_ns(fs_type, flags, sysfs_root, SYSFS_MAGIC, &new_sb, ns); - if (IS_ERR(root) || !new_sb) + if (!new_sb) kobj_ns_drop(KOBJ_NS_TYPE_NET, ns); - else if (new_sb) + else if (!IS_ERR(root)) root->d_sb->s_iflags |= SB_I_USERNS_VISIBLE; return root; From 9c3e9025a3f7ed25c99a0add8af65431c8043800 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 10 May 2018 22:59:45 -0400 Subject: [PATCH 114/190] cachefiles: vfs_mkdir() might succeed leaving dentry negative unhashed That can (and does, on some filesystems) happen - ->mkdir() (and thus vfs_mkdir()) can legitimately leave its argument negative and just unhash it, counting upon the lookup to pick the object we'd created next time we try to look at that name. Some vfs_mkdir() callers forget about that possibility... Signed-off-by: Al Viro --- fs/cachefiles/namei.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/fs/cachefiles/namei.c b/fs/cachefiles/namei.c index 0daa1e3fe0df..ab0bbe93b398 100644 --- a/fs/cachefiles/namei.c +++ b/fs/cachefiles/namei.c @@ -572,6 +572,11 @@ int cachefiles_walk_to_object(struct cachefiles_object *parent, if (ret < 0) goto create_error; + if (unlikely(d_unhashed(next))) { + dput(next); + inode_unlock(d_inode(dir)); + goto lookup_again; + } ASSERT(d_backing_inode(next)); _debug("mkdir -> %p{%p{ino=%lu}}", @@ -764,6 +769,7 @@ struct dentry *cachefiles_get_directory(struct cachefiles_cache *cache, /* search the current directory for the element name */ inode_lock(d_inode(dir)); +retry: start = jiffies; subdir = lookup_one_len(dirname, dir, strlen(dirname)); cachefiles_hist(cachefiles_lookup_histogram, start); @@ -793,6 +799,10 @@ struct dentry *cachefiles_get_directory(struct cachefiles_cache *cache, if (ret < 0) goto mkdir_error; + if (unlikely(d_unhashed(subdir))) { + dput(subdir); + goto retry; + } ASSERT(d_backing_inode(subdir)); _debug("mkdir -> %p{%p{ino=%lu}}", From 3819bb0d79f50b05910db5bdc6d9ef512184e3b1 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 11 May 2018 17:03:19 -0400 Subject: [PATCH 115/190] nfsd: vfs_mkdir() might succeed leaving dentry negative unhashed That can (and does, on some filesystems) happen - ->mkdir() (and thus vfs_mkdir()) can legitimately leave its argument negative and just unhash it, counting upon the lookup to pick the object we'd created next time we try to look at that name. Some vfs_mkdir() callers forget about that possibility... Acked-by: J. Bruce Fields Signed-off-by: Al Viro --- fs/nfsd/vfs.c | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+) diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index 2410b093a2e6..b0555d7d8200 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -1201,6 +1201,28 @@ nfsd_create_locked(struct svc_rqst *rqstp, struct svc_fh *fhp, break; case S_IFDIR: host_err = vfs_mkdir(dirp, dchild, iap->ia_mode); + if (!host_err && unlikely(d_unhashed(dchild))) { + struct dentry *d; + d = lookup_one_len(dchild->d_name.name, + dchild->d_parent, + dchild->d_name.len); + if (IS_ERR(d)) { + host_err = PTR_ERR(d); + break; + } + if (unlikely(d_is_negative(d))) { + dput(d); + err = nfserr_serverfault; + goto out; + } + dput(resfhp->fh_dentry); + resfhp->fh_dentry = dget(d); + err = fh_update(resfhp); + dput(dchild); + dchild = d; + if (err) + goto out; + } break; case S_IFCHR: case S_IFBLK: From 5aa1437d2d9a068c0334bd7c9dafa8ec4f97f13b Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 17 May 2018 17:18:30 -0400 Subject: [PATCH 116/190] ext2: fix a block leak open file, unlink it, then use ioctl(2) to make it immutable or append only. Now close it and watch the blocks *not* freed... Immutable/append-only checks belong in ->setattr(). Note: the bug is old and backport to anything prior to 737f2e93b972 ("ext2: convert to use the new truncate convention") will need these checks lifted into ext2_setattr(). Cc: stable@kernel.org Signed-off-by: Al Viro --- fs/ext2/inode.c | 10 ---------- 1 file changed, 10 deletions(-) diff --git a/fs/ext2/inode.c b/fs/ext2/inode.c index 1e01fabef130..71635909df3b 100644 --- a/fs/ext2/inode.c +++ b/fs/ext2/inode.c @@ -1264,21 +1264,11 @@ static void __ext2_truncate_blocks(struct inode *inode, loff_t offset) static void ext2_truncate_blocks(struct inode *inode, loff_t offset) { - /* - * XXX: it seems like a bug here that we don't allow - * IS_APPEND inode to have blocks-past-i_size trimmed off. - * review and fix this. - * - * Also would be nice to be able to handle IO errors and such, - * but that's probably too much to ask. - */ if (!(S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode))) return; if (ext2_inode_is_fast_symlink(inode)) return; - if (IS_APPEND(inode) || IS_IMMUTABLE(inode)) - return; dax_sem_down_write(EXT2_I(inode)); __ext2_truncate_blocks(inode, offset); From baf10564fbb66ea222cae66fbff11c444590ffd9 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 20 May 2018 16:46:23 -0400 Subject: [PATCH 117/190] aio: fix io_destroy(2) vs. lookup_ioctx() race kill_ioctx() used to have an explicit RCU delay between removing the reference from ->ioctx_table and percpu_ref_kill() dropping the refcount. At some point that delay had been removed, on the theory that percpu_ref_kill() itself contained an RCU delay. Unfortunately, that was the wrong kind of RCU delay and it didn't care about rcu_read_lock() used by lookup_ioctx(). As the result, we could get ctx freed right under lookup_ioctx(). Tejun has fixed that in a6d7cff472e ("fs/aio: Add explicit RCU grace period when freeing kioctx"); however, that fix is not enough. Suppose io_destroy() from one thread races with e.g. io_setup() from another; CPU1 removes the reference from current->mm->ioctx_table[...] just as CPU2 has picked it (under rcu_read_lock()). Then CPU1 proceeds to drop the refcount, getting it to 0 and triggering a call of free_ioctx_users(), which proceeds to drop the secondary refcount and once that reaches zero calls free_ioctx_reqs(). That does INIT_RCU_WORK(&ctx->free_rwork, free_ioctx); queue_rcu_work(system_wq, &ctx->free_rwork); and schedules freeing the whole thing after RCU delay. In the meanwhile CPU2 has gotten around to percpu_ref_get(), bumping the refcount from 0 to 1 and returned the reference to io_setup(). Tejun's fix (that queue_rcu_work() in there) guarantees that ctx won't get freed until after percpu_ref_get(). Sure, we'd increment the counter before ctx can be freed. Now we are out of rcu_read_lock() and there's nothing to stop freeing of the whole thing. Unfortunately, CPU2 assumes that since it has grabbed the reference, ctx is *NOT* going away until it gets around to dropping that reference. The fix is obvious - use percpu_ref_tryget_live() and treat failure as miss. It's not costlier than what we currently do in normal case, it's safe to call since freeing *is* delayed and it closes the race window - either lookup_ioctx() comes before percpu_ref_kill() (in which case ctx->users won't reach 0 until the caller of lookup_ioctx() drops it) or lookup_ioctx() fails, ctx->users is unaffected and caller of lookup_ioctx() doesn't see the object in question at all. Cc: stable@kernel.org Fixes: a6d7cff472e "fs/aio: Add explicit RCU grace period when freeing kioctx" Signed-off-by: Al Viro --- fs/aio.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/fs/aio.c b/fs/aio.c index 88d7927ffbc6..8061d9787e54 100644 --- a/fs/aio.c +++ b/fs/aio.c @@ -1078,8 +1078,8 @@ static struct kioctx *lookup_ioctx(unsigned long ctx_id) ctx = rcu_dereference(table->table[id]); if (ctx && ctx->user_id == ctx_id) { - percpu_ref_get(&ctx->users); - ret = ctx; + if (percpu_ref_tryget_live(&ctx->users)) + ret = ctx; } out: rcu_read_unlock(); From eedffa28c9b00ca2dcb4d541b5a530f4c917052d Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Mon, 21 May 2018 14:35:03 -0400 Subject: [PATCH 118/190] loop: clear wb_err in bd_inode when detaching backing file When a loop block device encounters a writeback error, that error will get propagated to the bd_inode's wb_err field. If we then detach the backing file from it, attach another and fsync it, we'll get back the writeback error that we had from the previous backing file. This is a bit of a grey area as POSIX doesn't cover loop devices, but it is somewhat counterintuitive. If we detach a backing file from the loopdev while there are still unreported errors, take it as a sign that we're no longer interested in the previous file, and clear out the wb_err in the loop blockdev. Reported-and-Tested-by: Theodore Y. Ts'o Signed-off-by: Jeff Layton Signed-off-by: Jens Axboe --- drivers/block/loop.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/block/loop.c b/drivers/block/loop.c index 5d4e31655d96..55cf554bc914 100644 --- a/drivers/block/loop.c +++ b/drivers/block/loop.c @@ -1068,6 +1068,7 @@ static int loop_clr_fd(struct loop_device *lo) if (bdev) { bdput(bdev); invalidate_bdev(bdev); + bdev->bd_inode->i_mapping->wb_err = 0; } set_capacity(lo->lo_disk, 0); loop_sysfs_exit(lo); From a048a07d7f4535baa4cbad6bc024f175317ab938 Mon Sep 17 00:00:00 2001 From: Nicholas Piggin Date: Tue, 22 May 2018 09:00:00 +1000 Subject: [PATCH 119/190] powerpc/64s: Add support for a store forwarding barrier at kernel entry/exit MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit On some CPUs we can prevent a vulnerability related to store-to-load forwarding by preventing store forwarding between privilege domains, by inserting a barrier in kernel entry and exit paths. This is known to be the case on at least Power7, Power8 and Power9 powerpc CPUs. Barriers must be inserted generally before the first load after moving to a higher privilege, and after the last store before moving to a lower privilege, HV and PR privilege transitions must be protected. Barriers are added as patch sections, with all kernel/hypervisor entry points patched, and the exit points to lower privilge levels patched similarly to the RFI flush patching. Firmware advertisement is not implemented yet, so CPU flush types are hard coded. Thanks to Michal Suchánek for bug fixes and review. Signed-off-by: Nicholas Piggin Signed-off-by: Mauricio Faria de Oliveira Signed-off-by: Michael Neuling Signed-off-by: Michal Suchánek Signed-off-by: Michael Ellerman Signed-off-by: Linus Torvalds --- arch/powerpc/include/asm/exception-64s.h | 29 ++++ arch/powerpc/include/asm/feature-fixups.h | 19 +++ arch/powerpc/include/asm/security_features.h | 11 ++ arch/powerpc/kernel/exceptions-64s.S | 19 ++- arch/powerpc/kernel/security.c | 149 +++++++++++++++++++ arch/powerpc/kernel/vmlinux.lds.S | 14 ++ arch/powerpc/lib/feature-fixups.c | 115 ++++++++++++++ arch/powerpc/platforms/powernv/setup.c | 1 + arch/powerpc/platforms/pseries/setup.c | 1 + 9 files changed, 356 insertions(+), 2 deletions(-) diff --git a/arch/powerpc/include/asm/exception-64s.h b/arch/powerpc/include/asm/exception-64s.h index 471b2274fbeb..c40b4380951c 100644 --- a/arch/powerpc/include/asm/exception-64s.h +++ b/arch/powerpc/include/asm/exception-64s.h @@ -74,6 +74,27 @@ */ #define EX_R3 EX_DAR +#define STF_ENTRY_BARRIER_SLOT \ + STF_ENTRY_BARRIER_FIXUP_SECTION; \ + nop; \ + nop; \ + nop + +#define STF_EXIT_BARRIER_SLOT \ + STF_EXIT_BARRIER_FIXUP_SECTION; \ + nop; \ + nop; \ + nop; \ + nop; \ + nop; \ + nop + +/* + * r10 must be free to use, r13 must be paca + */ +#define INTERRUPT_TO_KERNEL \ + STF_ENTRY_BARRIER_SLOT + /* * Macros for annotating the expected destination of (h)rfid * @@ -90,16 +111,19 @@ rfid #define RFI_TO_USER \ + STF_EXIT_BARRIER_SLOT; \ RFI_FLUSH_SLOT; \ rfid; \ b rfi_flush_fallback #define RFI_TO_USER_OR_KERNEL \ + STF_EXIT_BARRIER_SLOT; \ RFI_FLUSH_SLOT; \ rfid; \ b rfi_flush_fallback #define RFI_TO_GUEST \ + STF_EXIT_BARRIER_SLOT; \ RFI_FLUSH_SLOT; \ rfid; \ b rfi_flush_fallback @@ -108,21 +132,25 @@ hrfid #define HRFI_TO_USER \ + STF_EXIT_BARRIER_SLOT; \ RFI_FLUSH_SLOT; \ hrfid; \ b hrfi_flush_fallback #define HRFI_TO_USER_OR_KERNEL \ + STF_EXIT_BARRIER_SLOT; \ RFI_FLUSH_SLOT; \ hrfid; \ b hrfi_flush_fallback #define HRFI_TO_GUEST \ + STF_EXIT_BARRIER_SLOT; \ RFI_FLUSH_SLOT; \ hrfid; \ b hrfi_flush_fallback #define HRFI_TO_UNKNOWN \ + STF_EXIT_BARRIER_SLOT; \ RFI_FLUSH_SLOT; \ hrfid; \ b hrfi_flush_fallback @@ -254,6 +282,7 @@ END_FTR_SECTION_NESTED(ftr,ftr,943) #define __EXCEPTION_PROLOG_1_PRE(area) \ OPT_SAVE_REG_TO_PACA(area+EX_PPR, r9, CPU_FTR_HAS_PPR); \ OPT_SAVE_REG_TO_PACA(area+EX_CFAR, r10, CPU_FTR_CFAR); \ + INTERRUPT_TO_KERNEL; \ SAVE_CTR(r10, area); \ mfcr r9; diff --git a/arch/powerpc/include/asm/feature-fixups.h b/arch/powerpc/include/asm/feature-fixups.h index 1e82eb3caabd..a9b64df34e2a 100644 --- a/arch/powerpc/include/asm/feature-fixups.h +++ b/arch/powerpc/include/asm/feature-fixups.h @@ -187,6 +187,22 @@ label##3: \ FTR_ENTRY_OFFSET label##1b-label##3b; \ .popsection; +#define STF_ENTRY_BARRIER_FIXUP_SECTION \ +953: \ + .pushsection __stf_entry_barrier_fixup,"a"; \ + .align 2; \ +954: \ + FTR_ENTRY_OFFSET 953b-954b; \ + .popsection; + +#define STF_EXIT_BARRIER_FIXUP_SECTION \ +955: \ + .pushsection __stf_exit_barrier_fixup,"a"; \ + .align 2; \ +956: \ + FTR_ENTRY_OFFSET 955b-956b; \ + .popsection; + #define RFI_FLUSH_FIXUP_SECTION \ 951: \ .pushsection __rfi_flush_fixup,"a"; \ @@ -199,6 +215,9 @@ label##3: \ #ifndef __ASSEMBLY__ #include +extern long stf_barrier_fallback; +extern long __start___stf_entry_barrier_fixup, __stop___stf_entry_barrier_fixup; +extern long __start___stf_exit_barrier_fixup, __stop___stf_exit_barrier_fixup; extern long __start___rfi_flush_fixup, __stop___rfi_flush_fixup; void apply_feature_fixups(void); diff --git a/arch/powerpc/include/asm/security_features.h b/arch/powerpc/include/asm/security_features.h index fa4d2e1cf772..44989b22383c 100644 --- a/arch/powerpc/include/asm/security_features.h +++ b/arch/powerpc/include/asm/security_features.h @@ -12,6 +12,17 @@ extern unsigned long powerpc_security_features; extern bool rfi_flush; +/* These are bit flags */ +enum stf_barrier_type { + STF_BARRIER_NONE = 0x1, + STF_BARRIER_FALLBACK = 0x2, + STF_BARRIER_EIEIO = 0x4, + STF_BARRIER_SYNC_ORI = 0x8, +}; + +void setup_stf_barrier(void); +void do_stf_barrier_fixups(enum stf_barrier_type types); + static inline void security_ftr_set(unsigned long feature) { powerpc_security_features |= feature; diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S index ae6a849db60b..f283958129f2 100644 --- a/arch/powerpc/kernel/exceptions-64s.S +++ b/arch/powerpc/kernel/exceptions-64s.S @@ -885,7 +885,7 @@ END_FTR_SECTION_IFSET(CPU_FTR_TM) #endif -EXC_REAL_MASKABLE(decrementer, 0x900, 0x80, IRQS_DISABLED) +EXC_REAL_OOL_MASKABLE(decrementer, 0x900, 0x80, IRQS_DISABLED) EXC_VIRT_MASKABLE(decrementer, 0x4900, 0x80, 0x900, IRQS_DISABLED) TRAMP_KVM(PACA_EXGEN, 0x900) EXC_COMMON_ASYNC(decrementer_common, 0x900, timer_interrupt) @@ -961,6 +961,7 @@ EXC_COMMON(trap_0b_common, 0xb00, unknown_exception) mtctr r13; \ GET_PACA(r13); \ std r10,PACA_EXGEN+EX_R10(r13); \ + INTERRUPT_TO_KERNEL; \ KVMTEST_PR(0xc00); /* uses r10, branch to do_kvm_0xc00_system_call */ \ HMT_MEDIUM; \ mfctr r9; @@ -969,7 +970,8 @@ EXC_COMMON(trap_0b_common, 0xb00, unknown_exception) #define SYSCALL_KVMTEST \ HMT_MEDIUM; \ mr r9,r13; \ - GET_PACA(r13); + GET_PACA(r13); \ + INTERRUPT_TO_KERNEL; #endif #define LOAD_SYSCALL_HANDLER(reg) \ @@ -1507,6 +1509,19 @@ masked_##_H##interrupt: \ b .; \ MASKED_DEC_HANDLER(_H) +TRAMP_REAL_BEGIN(stf_barrier_fallback) + std r9,PACA_EXRFI+EX_R9(r13) + std r10,PACA_EXRFI+EX_R10(r13) + sync + ld r9,PACA_EXRFI+EX_R9(r13) + ld r10,PACA_EXRFI+EX_R10(r13) + ori 31,31,0 + .rept 14 + b 1f +1: + .endr + blr + TRAMP_REAL_BEGIN(rfi_flush_fallback) SET_SCRATCH0(r13); GET_PACA(r13); diff --git a/arch/powerpc/kernel/security.c b/arch/powerpc/kernel/security.c index bab5a27ea805..b98a722da915 100644 --- a/arch/powerpc/kernel/security.c +++ b/arch/powerpc/kernel/security.c @@ -8,6 +8,7 @@ #include #include +#include #include @@ -86,3 +87,151 @@ ssize_t cpu_show_spectre_v2(struct device *dev, struct device_attribute *attr, c return s.len; } + +/* + * Store-forwarding barrier support. + */ + +static enum stf_barrier_type stf_enabled_flush_types; +static bool no_stf_barrier; +bool stf_barrier; + +static int __init handle_no_stf_barrier(char *p) +{ + pr_info("stf-barrier: disabled on command line."); + no_stf_barrier = true; + return 0; +} + +early_param("no_stf_barrier", handle_no_stf_barrier); + +/* This is the generic flag used by other architectures */ +static int __init handle_ssbd(char *p) +{ + if (!p || strncmp(p, "auto", 5) == 0 || strncmp(p, "on", 2) == 0 ) { + /* Until firmware tells us, we have the barrier with auto */ + return 0; + } else if (strncmp(p, "off", 3) == 0) { + handle_no_stf_barrier(NULL); + return 0; + } else + return 1; + + return 0; +} +early_param("spec_store_bypass_disable", handle_ssbd); + +/* This is the generic flag used by other architectures */ +static int __init handle_no_ssbd(char *p) +{ + handle_no_stf_barrier(NULL); + return 0; +} +early_param("nospec_store_bypass_disable", handle_no_ssbd); + +static void stf_barrier_enable(bool enable) +{ + if (enable) + do_stf_barrier_fixups(stf_enabled_flush_types); + else + do_stf_barrier_fixups(STF_BARRIER_NONE); + + stf_barrier = enable; +} + +void setup_stf_barrier(void) +{ + enum stf_barrier_type type; + bool enable, hv; + + hv = cpu_has_feature(CPU_FTR_HVMODE); + + /* Default to fallback in case fw-features are not available */ + if (cpu_has_feature(CPU_FTR_ARCH_300)) + type = STF_BARRIER_EIEIO; + else if (cpu_has_feature(CPU_FTR_ARCH_207S)) + type = STF_BARRIER_SYNC_ORI; + else if (cpu_has_feature(CPU_FTR_ARCH_206)) + type = STF_BARRIER_FALLBACK; + else + type = STF_BARRIER_NONE; + + enable = security_ftr_enabled(SEC_FTR_FAVOUR_SECURITY) && + (security_ftr_enabled(SEC_FTR_L1D_FLUSH_PR) || + (security_ftr_enabled(SEC_FTR_L1D_FLUSH_HV) && hv)); + + if (type == STF_BARRIER_FALLBACK) { + pr_info("stf-barrier: fallback barrier available\n"); + } else if (type == STF_BARRIER_SYNC_ORI) { + pr_info("stf-barrier: hwsync barrier available\n"); + } else if (type == STF_BARRIER_EIEIO) { + pr_info("stf-barrier: eieio barrier available\n"); + } + + stf_enabled_flush_types = type; + + if (!no_stf_barrier) + stf_barrier_enable(enable); +} + +ssize_t cpu_show_spec_store_bypass(struct device *dev, struct device_attribute *attr, char *buf) +{ + if (stf_barrier && stf_enabled_flush_types != STF_BARRIER_NONE) { + const char *type; + switch (stf_enabled_flush_types) { + case STF_BARRIER_EIEIO: + type = "eieio"; + break; + case STF_BARRIER_SYNC_ORI: + type = "hwsync"; + break; + case STF_BARRIER_FALLBACK: + type = "fallback"; + break; + default: + type = "unknown"; + } + return sprintf(buf, "Mitigation: Kernel entry/exit barrier (%s)\n", type); + } + + if (!security_ftr_enabled(SEC_FTR_L1D_FLUSH_HV) && + !security_ftr_enabled(SEC_FTR_L1D_FLUSH_PR)) + return sprintf(buf, "Not affected\n"); + + return sprintf(buf, "Vulnerable\n"); +} + +#ifdef CONFIG_DEBUG_FS +static int stf_barrier_set(void *data, u64 val) +{ + bool enable; + + if (val == 1) + enable = true; + else if (val == 0) + enable = false; + else + return -EINVAL; + + /* Only do anything if we're changing state */ + if (enable != stf_barrier) + stf_barrier_enable(enable); + + return 0; +} + +static int stf_barrier_get(void *data, u64 *val) +{ + *val = stf_barrier ? 1 : 0; + return 0; +} + +DEFINE_SIMPLE_ATTRIBUTE(fops_stf_barrier, stf_barrier_get, stf_barrier_set, "%llu\n"); + +static __init int stf_barrier_debugfs_init(void) +{ + debugfs_create_file("stf_barrier", 0600, powerpc_debugfs_root, NULL, &fops_stf_barrier); + return 0; +} +device_initcall(stf_barrier_debugfs_init); +#endif /* CONFIG_DEBUG_FS */ diff --git a/arch/powerpc/kernel/vmlinux.lds.S b/arch/powerpc/kernel/vmlinux.lds.S index c8af90ff49f0..b8d82678f8b4 100644 --- a/arch/powerpc/kernel/vmlinux.lds.S +++ b/arch/powerpc/kernel/vmlinux.lds.S @@ -133,6 +133,20 @@ SECTIONS RO_DATA(PAGE_SIZE) #ifdef CONFIG_PPC64 + . = ALIGN(8); + __stf_entry_barrier_fixup : AT(ADDR(__stf_entry_barrier_fixup) - LOAD_OFFSET) { + __start___stf_entry_barrier_fixup = .; + *(__stf_entry_barrier_fixup) + __stop___stf_entry_barrier_fixup = .; + } + + . = ALIGN(8); + __stf_exit_barrier_fixup : AT(ADDR(__stf_exit_barrier_fixup) - LOAD_OFFSET) { + __start___stf_exit_barrier_fixup = .; + *(__stf_exit_barrier_fixup) + __stop___stf_exit_barrier_fixup = .; + } + . = ALIGN(8); __rfi_flush_fixup : AT(ADDR(__rfi_flush_fixup) - LOAD_OFFSET) { __start___rfi_flush_fixup = .; diff --git a/arch/powerpc/lib/feature-fixups.c b/arch/powerpc/lib/feature-fixups.c index 288fe4f0db4e..e1bcdc32a851 100644 --- a/arch/powerpc/lib/feature-fixups.c +++ b/arch/powerpc/lib/feature-fixups.c @@ -23,6 +23,7 @@ #include #include #include +#include #include struct fixup_entry { @@ -117,6 +118,120 @@ void do_feature_fixups(unsigned long value, void *fixup_start, void *fixup_end) } #ifdef CONFIG_PPC_BOOK3S_64 +void do_stf_entry_barrier_fixups(enum stf_barrier_type types) +{ + unsigned int instrs[3], *dest; + long *start, *end; + int i; + + start = PTRRELOC(&__start___stf_entry_barrier_fixup), + end = PTRRELOC(&__stop___stf_entry_barrier_fixup); + + instrs[0] = 0x60000000; /* nop */ + instrs[1] = 0x60000000; /* nop */ + instrs[2] = 0x60000000; /* nop */ + + i = 0; + if (types & STF_BARRIER_FALLBACK) { + instrs[i++] = 0x7d4802a6; /* mflr r10 */ + instrs[i++] = 0x60000000; /* branch patched below */ + instrs[i++] = 0x7d4803a6; /* mtlr r10 */ + } else if (types & STF_BARRIER_EIEIO) { + instrs[i++] = 0x7e0006ac; /* eieio + bit 6 hint */ + } else if (types & STF_BARRIER_SYNC_ORI) { + instrs[i++] = 0x7c0004ac; /* hwsync */ + instrs[i++] = 0xe94d0000; /* ld r10,0(r13) */ + instrs[i++] = 0x63ff0000; /* ori 31,31,0 speculation barrier */ + } + + for (i = 0; start < end; start++, i++) { + dest = (void *)start + *start; + + pr_devel("patching dest %lx\n", (unsigned long)dest); + + patch_instruction(dest, instrs[0]); + + if (types & STF_BARRIER_FALLBACK) + patch_branch(dest + 1, (unsigned long)&stf_barrier_fallback, + BRANCH_SET_LINK); + else + patch_instruction(dest + 1, instrs[1]); + + patch_instruction(dest + 2, instrs[2]); + } + + printk(KERN_DEBUG "stf-barrier: patched %d entry locations (%s barrier)\n", i, + (types == STF_BARRIER_NONE) ? "no" : + (types == STF_BARRIER_FALLBACK) ? "fallback" : + (types == STF_BARRIER_EIEIO) ? "eieio" : + (types == (STF_BARRIER_SYNC_ORI)) ? "hwsync" + : "unknown"); +} + +void do_stf_exit_barrier_fixups(enum stf_barrier_type types) +{ + unsigned int instrs[6], *dest; + long *start, *end; + int i; + + start = PTRRELOC(&__start___stf_exit_barrier_fixup), + end = PTRRELOC(&__stop___stf_exit_barrier_fixup); + + instrs[0] = 0x60000000; /* nop */ + instrs[1] = 0x60000000; /* nop */ + instrs[2] = 0x60000000; /* nop */ + instrs[3] = 0x60000000; /* nop */ + instrs[4] = 0x60000000; /* nop */ + instrs[5] = 0x60000000; /* nop */ + + i = 0; + if (types & STF_BARRIER_FALLBACK || types & STF_BARRIER_SYNC_ORI) { + if (cpu_has_feature(CPU_FTR_HVMODE)) { + instrs[i++] = 0x7db14ba6; /* mtspr 0x131, r13 (HSPRG1) */ + instrs[i++] = 0x7db04aa6; /* mfspr r13, 0x130 (HSPRG0) */ + } else { + instrs[i++] = 0x7db243a6; /* mtsprg 2,r13 */ + instrs[i++] = 0x7db142a6; /* mfsprg r13,1 */ + } + instrs[i++] = 0x7c0004ac; /* hwsync */ + instrs[i++] = 0xe9ad0000; /* ld r13,0(r13) */ + instrs[i++] = 0x63ff0000; /* ori 31,31,0 speculation barrier */ + if (cpu_has_feature(CPU_FTR_HVMODE)) { + instrs[i++] = 0x7db14aa6; /* mfspr r13, 0x131 (HSPRG1) */ + } else { + instrs[i++] = 0x7db242a6; /* mfsprg r13,2 */ + } + } else if (types & STF_BARRIER_EIEIO) { + instrs[i++] = 0x7e0006ac; /* eieio + bit 6 hint */ + } + + for (i = 0; start < end; start++, i++) { + dest = (void *)start + *start; + + pr_devel("patching dest %lx\n", (unsigned long)dest); + + patch_instruction(dest, instrs[0]); + patch_instruction(dest + 1, instrs[1]); + patch_instruction(dest + 2, instrs[2]); + patch_instruction(dest + 3, instrs[3]); + patch_instruction(dest + 4, instrs[4]); + patch_instruction(dest + 5, instrs[5]); + } + printk(KERN_DEBUG "stf-barrier: patched %d exit locations (%s barrier)\n", i, + (types == STF_BARRIER_NONE) ? "no" : + (types == STF_BARRIER_FALLBACK) ? "fallback" : + (types == STF_BARRIER_EIEIO) ? "eieio" : + (types == (STF_BARRIER_SYNC_ORI)) ? "hwsync" + : "unknown"); +} + + +void do_stf_barrier_fixups(enum stf_barrier_type types) +{ + do_stf_entry_barrier_fixups(types); + do_stf_exit_barrier_fixups(types); +} + void do_rfi_flush_fixups(enum l1d_flush_type types) { unsigned int instrs[3], *dest; diff --git a/arch/powerpc/platforms/powernv/setup.c b/arch/powerpc/platforms/powernv/setup.c index ef8c9ce53a61..a6648ec99ca7 100644 --- a/arch/powerpc/platforms/powernv/setup.c +++ b/arch/powerpc/platforms/powernv/setup.c @@ -131,6 +131,7 @@ static void __init pnv_setup_arch(void) set_arch_panic_timeout(10, ARCH_PANIC_TIMEOUT); pnv_setup_rfi_flush(); + setup_stf_barrier(); /* Initialize SMP */ pnv_smp_init(); diff --git a/arch/powerpc/platforms/pseries/setup.c b/arch/powerpc/platforms/pseries/setup.c index b55ad4286dc7..fdb32e056ef4 100644 --- a/arch/powerpc/platforms/pseries/setup.c +++ b/arch/powerpc/platforms/pseries/setup.c @@ -710,6 +710,7 @@ static void __init pSeries_setup_arch(void) fwnmi_init(); pseries_setup_rfi_flush(); + setup_stf_barrier(); /* By default, only probe PCI (can be overridden by rtas_pci) */ pci_add_flags(PCI_PROBE_ONLY); From 5247a77ced2d3396a32679c7480e8bcd5b26e471 Mon Sep 17 00:00:00 2001 From: Haim Dreyfuss Date: Mon, 21 May 2018 23:34:57 +0300 Subject: [PATCH 120/190] cfg80211: fix NULL pointer derference when querying regdb Some drivers may call this function when regdb is not initialized yet, so we need to make sure regdb is valid before trying to access it. Make sure regdb is initialized before trying to access it in reg_query_regdb_wmm() and query_regdb(). Reported-by: Eric Biggers Signed-off-by: Haim Dreyfuss Signed-off-by: Luca Coelho Signed-off-by: Johannes Berg --- net/wireless/reg.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/wireless/reg.c b/net/wireless/reg.c index ac3e12c32aa3..5fcec5c94eb7 100644 --- a/net/wireless/reg.c +++ b/net/wireless/reg.c @@ -916,6 +916,9 @@ int reg_query_regdb_wmm(char *alpha2, int freq, u32 *dbptr, const struct fwdb_header *hdr = regdb; const struct fwdb_country *country; + if (!regdb) + return -ENODATA; + if (IS_ERR(regdb)) return PTR_ERR(regdb); From fed4825096cfbbfd654cb292ab6eb193911aef01 Mon Sep 17 00:00:00 2001 From: Andrew Zaborowski Date: Tue, 22 May 2018 02:43:56 +0200 Subject: [PATCH 121/190] mac80211_hwsim: Fix radio dump for radio idx 0 Since 6335698e24ec11e1324b916177da6721df724dd8 the radio with idx of 0 will not get dumped in HWSIM_CMD_GET_RADIO because of the last_idx checks. Offset cb->args[0] by 1 similarly to what is done in nl80211.c. Fixes: 6335698e24ec ("mac80211_hwsim: add generation count for netlink dump operation") Signed-off-by: Andrew Zaborowski Signed-off-by: Johannes Berg --- drivers/net/wireless/mac80211_hwsim.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/wireless/mac80211_hwsim.c b/drivers/net/wireless/mac80211_hwsim.c index 4a017a0d71ea..920c23e542a5 100644 --- a/drivers/net/wireless/mac80211_hwsim.c +++ b/drivers/net/wireless/mac80211_hwsim.c @@ -3340,7 +3340,7 @@ static int hwsim_get_radio_nl(struct sk_buff *msg, struct genl_info *info) static int hwsim_dump_radio_nl(struct sk_buff *skb, struct netlink_callback *cb) { - int last_idx = cb->args[0]; + int last_idx = cb->args[0] - 1; struct mac80211_hwsim_data *data = NULL; int res = 0; void *hdr; @@ -3368,7 +3368,7 @@ static int hwsim_dump_radio_nl(struct sk_buff *skb, last_idx = data->idx; } - cb->args[0] = last_idx; + cb->args[0] = last_idx + 1; /* list changed, but no new element sent, set interrupted flag */ if (skb->len == 0 && cb->prev_seq && cb->seq != cb->prev_seq) { From c76700ab337a022cbd620d6e36ac6bd1342331dd Mon Sep 17 00:00:00 2001 From: Kalle Valo Date: Fri, 18 May 2018 19:08:11 +0300 Subject: [PATCH 122/190] MAINTAINERS: update Kalle's email address I switched to use my codeaurora.org address. Signed-off-by: Kalle Valo --- MAINTAINERS | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/MAINTAINERS b/MAINTAINERS index 6d21f2f74578..4be9ff892ec0 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2345,7 +2345,7 @@ S: Maintained F: drivers/net/wireless/ath/ath5k/ ATHEROS ATH6KL WIRELESS DRIVER -M: Kalle Valo +M: Kalle Valo L: linux-wireless@vger.kernel.org W: http://wireless.kernel.org/en/users/Drivers/ath6kl T: git git://git.kernel.org/pub/scm/linux/kernel/git/kvalo/ath.git @@ -11651,7 +11651,7 @@ S: Maintained F: drivers/media/tuners/qt1010* QUALCOMM ATHEROS ATH10K WIRELESS DRIVER -M: Kalle Valo +M: Kalle Valo L: ath10k@lists.infradead.org W: http://wireless.kernel.org/en/users/Drivers/ath10k T: git git://git.kernel.org/pub/scm/linux/kernel/git/kvalo/ath.git From b561290995b2704cfc92bea6996818a3be36c592 Mon Sep 17 00:00:00 2001 From: Kalle Valo Date: Fri, 18 May 2018 19:08:12 +0300 Subject: [PATCH 123/190] MAINTAINERS: change Kalle as ath.ko maintainer Luis hasn't worked on ath.ko for some time now. Acked-by: Luis R. Rodriguez Signed-off-by: Kalle Valo --- MAINTAINERS | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index 4be9ff892ec0..5dace1734086 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2330,7 +2330,7 @@ F: drivers/gpio/gpio-ath79.c F: Documentation/devicetree/bindings/gpio/gpio-ath79.txt ATHEROS ATH GENERIC UTILITIES -M: "Luis R. Rodriguez" +M: Kalle Valo L: linux-wireless@vger.kernel.org S: Supported F: drivers/net/wireless/ath/* From 813477aa49aac5deba04eb4956360dde58a0e807 Mon Sep 17 00:00:00 2001 From: Kalle Valo Date: Fri, 18 May 2018 19:08:13 +0300 Subject: [PATCH 124/190] MAINTAINERS: change Kalle as wcn36xx maintainer Eugene hasn't worked on wcn36xx for some time now. Signed-off-by: Kalle Valo --- MAINTAINERS | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index 5dace1734086..c1b15dde5a46 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -11702,7 +11702,7 @@ S: Maintained F: drivers/media/platform/qcom/venus/ QUALCOMM WCN36XX WIRELESS DRIVER -M: Eugene Krasnikov +M: Kalle Valo L: wcn36xx@lists.infradead.org W: http://wireless.kernel.org/en/users/Drivers/wcn36xx T: git git://github.com/KrasnikovEugene/wcn36xx.git From c62ec4610c40bcc44f2d3d5ed1c312737279e2f3 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Tue, 22 May 2018 13:02:17 +0200 Subject: [PATCH 125/190] PM / core: Fix direct_complete handling for devices with no callbacks Commit 08810a4119aa (PM / core: Add NEVER_SKIP and SMART_PREPARE driver flags) inadvertently prevented the power.direct_complete flag from being set for devices without PM callbacks and with disabled runtime PM which also prevents power.direct_complete from being set for their parents. That led to problems including a resume crash on HP ZBook 14u. Restore the previous behavior by causing power.direct_complete to be set for those devices again, but do that in a more direct way to avoid overlooking that case in the future. Link: https://bugzilla.kernel.org/show_bug.cgi?id=199693 Fixes: 08810a4119aa (PM / core: Add NEVER_SKIP and SMART_PREPARE driver flags) Reported-by: Thomas Martitz Tested-by: Thomas Martitz Cc: 4.15+ # 4.15+ Signed-off-by: Rafael J. Wysocki Reviewed-by: Ulf Hansson Reviewed-by: Johan Hovold --- drivers/base/power/main.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c index 02a497e7c785..e5e067091572 100644 --- a/drivers/base/power/main.c +++ b/drivers/base/power/main.c @@ -1923,10 +1923,8 @@ static int device_prepare(struct device *dev, pm_message_t state) dev->power.wakeup_path = false; - if (dev->power.no_pm_callbacks) { - ret = 1; /* Let device go direct_complete */ + if (dev->power.no_pm_callbacks) goto unlock; - } if (dev->pm_domain) callback = dev->pm_domain->ops.prepare; @@ -1960,7 +1958,8 @@ static int device_prepare(struct device *dev, pm_message_t state) */ spin_lock_irq(&dev->power.lock); dev->power.direct_complete = state.event == PM_EVENT_SUSPEND && - pm_runtime_suspended(dev) && ret > 0 && + ((pm_runtime_suspended(dev) && ret > 0) || + dev->power.no_pm_callbacks) && !dev_pm_test_driver_flags(dev, DPM_FLAG_NEVER_SKIP); spin_unlock_irq(&dev->power.lock); return 0; From cc19846079a70abcfd91b5a0791a5f17d69458a5 Mon Sep 17 00:00:00 2001 From: Peter Maydell Date: Tue, 22 May 2018 17:11:20 +0100 Subject: [PATCH 126/190] arm64: fault: Don't leak data in ESR context for user fault on kernel VA If userspace faults on a kernel address, handing them the raw ESR value on the sigframe as part of the delivered signal can leak data useful to attackers who are using information about the underlying hardware fault type (e.g. translation vs permission) as a mechanism to defeat KASLR. However there are also legitimate uses for the information provided in the ESR -- notably the GCC and LLVM sanitizers use this to report whether wild pointer accesses by the application are reads or writes (since a wild write is a more serious bug than a wild read), so we don't want to drop the ESR information entirely. For faulting addresses in the kernel, sanitize the ESR. We choose to present userspace with the illusion that there is nothing mapped in the kernel's part of the address space at all, by reporting all faults as level 0 translation faults taken to EL1. These fields are safe to pass through to userspace as they depend only on the instruction that userspace used to provoke the fault: EC IL (always) ISV CM WNR (for all data aborts) All the other fields in ESR except DFSC are architecturally RES0 for an L0 translation fault taken to EL1, so can be zeroed out without confusing userspace. The illusion is not entirely perfect, as there is a tiny wrinkle where we will report an alignment fault that was not due to the memory type (for instance a LDREX to an unaligned address) as a translation fault, whereas if you do this on real unmapped memory the alignment fault takes precedence. This is not likely to trip anybody up in practice, as the only users we know of for the ESR information who care about the behaviour for kernel addresses only really want to know about the WnR bit. Signed-off-by: Peter Maydell Signed-off-by: Will Deacon --- arch/arm64/mm/fault.c | 51 +++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 51 insertions(+) diff --git a/arch/arm64/mm/fault.c b/arch/arm64/mm/fault.c index 4165485e8b6e..2af3dd89bcdb 100644 --- a/arch/arm64/mm/fault.c +++ b/arch/arm64/mm/fault.c @@ -293,6 +293,57 @@ static void __do_kernel_fault(unsigned long addr, unsigned int esr, static void __do_user_fault(struct siginfo *info, unsigned int esr) { current->thread.fault_address = (unsigned long)info->si_addr; + + /* + * If the faulting address is in the kernel, we must sanitize the ESR. + * From userspace's point of view, kernel-only mappings don't exist + * at all, so we report them as level 0 translation faults. + * (This is not quite the way that "no mapping there at all" behaves: + * an alignment fault not caused by the memory type would take + * precedence over translation fault for a real access to empty + * space. Unfortunately we can't easily distinguish "alignment fault + * not caused by memory type" from "alignment fault caused by memory + * type", so we ignore this wrinkle and just return the translation + * fault.) + */ + if (current->thread.fault_address >= TASK_SIZE) { + switch (ESR_ELx_EC(esr)) { + case ESR_ELx_EC_DABT_LOW: + /* + * These bits provide only information about the + * faulting instruction, which userspace knows already. + * We explicitly clear bits which are architecturally + * RES0 in case they are given meanings in future. + * We always report the ESR as if the fault was taken + * to EL1 and so ISV and the bits in ISS[23:14] are + * clear. (In fact it always will be a fault to EL1.) + */ + esr &= ESR_ELx_EC_MASK | ESR_ELx_IL | + ESR_ELx_CM | ESR_ELx_WNR; + esr |= ESR_ELx_FSC_FAULT; + break; + case ESR_ELx_EC_IABT_LOW: + /* + * Claim a level 0 translation fault. + * All other bits are architecturally RES0 for faults + * reported with that DFSC value, so we clear them. + */ + esr &= ESR_ELx_EC_MASK | ESR_ELx_IL; + esr |= ESR_ELx_FSC_FAULT; + break; + default: + /* + * This should never happen (entry.S only brings us + * into this code for insn and data aborts from a lower + * exception level). Fail safe by not providing an ESR + * context record at all. + */ + WARN(1, "ESR 0x%x is not DABT or IABT from EL0\n", esr); + esr = 0; + break; + } + } + current->thread.fault_code = esr; arm64_force_sig_info(info, esr_to_fault_info(esr)->name, current); } From 644fbdeacf1d3edd366e44b8ba214de9d1dd66a9 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Sun, 20 May 2018 16:39:10 +0800 Subject: [PATCH 127/190] sctp: fix the issue that flags are ignored when using kernel_connect Now sctp uses inet_dgram_connect as its proto_ops .connect, and the flags param can't be passed into its proto .connect where this flags is really needed. sctp works around it by getting flags from socket file in __sctp_connect. It works for connecting from userspace, as inherently the user sock has socket file and it passes f_flags as the flags param into the proto_ops .connect. However, the sock created by sock_create_kern doesn't have a socket file, and it passes the flags (like O_NONBLOCK) by using the flags param in kernel_connect, which calls proto_ops .connect later. So to fix it, this patch defines a new proto_ops .connect for sctp, sctp_inet_connect, which calls __sctp_connect() directly with this flags param. After this, the sctp's proto .connect can be removed. Note that sctp_inet_connect doesn't need to do some checks that are not needed for sctp, which makes thing better than with inet_dgram_connect. Suggested-by: Marcelo Ricardo Leitner Signed-off-by: Xin Long Acked-by: Neil Horman Acked-by: Marcelo Ricardo Leitner Reviewed-by: Michal Kubecek Signed-off-by: David S. Miller --- include/net/sctp/sctp.h | 2 ++ net/sctp/ipv6.c | 2 +- net/sctp/protocol.c | 2 +- net/sctp/socket.c | 51 ++++++++++++++++++++++++++++------------- 4 files changed, 39 insertions(+), 18 deletions(-) diff --git a/include/net/sctp/sctp.h b/include/net/sctp/sctp.h index 28b996d63490..35498e613ff5 100644 --- a/include/net/sctp/sctp.h +++ b/include/net/sctp/sctp.h @@ -103,6 +103,8 @@ void sctp_addr_wq_mgmt(struct net *, struct sctp_sockaddr_entry *, int); /* * sctp/socket.c */ +int sctp_inet_connect(struct socket *sock, struct sockaddr *uaddr, + int addr_len, int flags); int sctp_backlog_rcv(struct sock *sk, struct sk_buff *skb); int sctp_inet_listen(struct socket *sock, int backlog); void sctp_write_space(struct sock *sk); diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c index 42247110d842..0cd2e764f47f 100644 --- a/net/sctp/ipv6.c +++ b/net/sctp/ipv6.c @@ -1006,7 +1006,7 @@ static const struct proto_ops inet6_seqpacket_ops = { .owner = THIS_MODULE, .release = inet6_release, .bind = inet6_bind, - .connect = inet_dgram_connect, + .connect = sctp_inet_connect, .socketpair = sock_no_socketpair, .accept = inet_accept, .getname = sctp_getname, diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index d685f8456762..6bf0a9971888 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -1012,7 +1012,7 @@ static const struct proto_ops inet_seqpacket_ops = { .owner = THIS_MODULE, .release = inet_release, /* Needs to be wrapped... */ .bind = inet_bind, - .connect = inet_dgram_connect, + .connect = sctp_inet_connect, .socketpair = sock_no_socketpair, .accept = inet_accept, .getname = inet_getname, /* Semantics are different. */ diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 80835ac26d2c..ae7e7c606f72 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -1086,7 +1086,7 @@ static int sctp_setsockopt_bindx(struct sock *sk, */ static int __sctp_connect(struct sock *sk, struct sockaddr *kaddrs, - int addrs_size, + int addrs_size, int flags, sctp_assoc_t *assoc_id) { struct net *net = sock_net(sk); @@ -1104,7 +1104,6 @@ static int __sctp_connect(struct sock *sk, union sctp_addr *sa_addr = NULL; void *addr_buf; unsigned short port; - unsigned int f_flags = 0; sp = sctp_sk(sk); ep = sp->ep; @@ -1254,13 +1253,7 @@ static int __sctp_connect(struct sock *sk, sp->pf->to_sk_daddr(sa_addr, sk); sk->sk_err = 0; - /* in-kernel sockets don't generally have a file allocated to them - * if all they do is call sock_create_kern(). - */ - if (sk->sk_socket->file) - f_flags = sk->sk_socket->file->f_flags; - - timeo = sock_sndtimeo(sk, f_flags & O_NONBLOCK); + timeo = sock_sndtimeo(sk, flags & O_NONBLOCK); if (assoc_id) *assoc_id = asoc->assoc_id; @@ -1348,7 +1341,7 @@ static int __sctp_setsockopt_connectx(struct sock *sk, sctp_assoc_t *assoc_id) { struct sockaddr *kaddrs; - int err = 0; + int err = 0, flags = 0; pr_debug("%s: sk:%p addrs:%p addrs_size:%d\n", __func__, sk, addrs, addrs_size); @@ -1367,7 +1360,13 @@ static int __sctp_setsockopt_connectx(struct sock *sk, if (err) goto out_free; - err = __sctp_connect(sk, kaddrs, addrs_size, assoc_id); + /* in-kernel sockets don't generally have a file allocated to them + * if all they do is call sock_create_kern(). + */ + if (sk->sk_socket->file) + flags = sk->sk_socket->file->f_flags; + + err = __sctp_connect(sk, kaddrs, addrs_size, flags, assoc_id); out_free: kvfree(kaddrs); @@ -4397,16 +4396,26 @@ static int sctp_setsockopt(struct sock *sk, int level, int optname, * len: the size of the address. */ static int sctp_connect(struct sock *sk, struct sockaddr *addr, - int addr_len) + int addr_len, int flags) { - int err = 0; + struct inet_sock *inet = inet_sk(sk); struct sctp_af *af; + int err = 0; lock_sock(sk); pr_debug("%s: sk:%p, sockaddr:%p, addr_len:%d\n", __func__, sk, addr, addr_len); + /* We may need to bind the socket. */ + if (!inet->inet_num) { + if (sk->sk_prot->get_port(sk, 0)) { + release_sock(sk); + return -EAGAIN; + } + inet->inet_sport = htons(inet->inet_num); + } + /* Validate addr_len before calling common connect/connectx routine. */ af = sctp_get_af_specific(addr->sa_family); if (!af || addr_len < af->sockaddr_len) { @@ -4415,13 +4424,25 @@ static int sctp_connect(struct sock *sk, struct sockaddr *addr, /* Pass correct addr len to common routine (so it knows there * is only one address being passed. */ - err = __sctp_connect(sk, addr, af->sockaddr_len, NULL); + err = __sctp_connect(sk, addr, af->sockaddr_len, flags, NULL); } release_sock(sk); return err; } +int sctp_inet_connect(struct socket *sock, struct sockaddr *uaddr, + int addr_len, int flags) +{ + if (addr_len < sizeof(uaddr->sa_family)) + return -EINVAL; + + if (uaddr->sa_family == AF_UNSPEC) + return -EOPNOTSUPP; + + return sctp_connect(sock->sk, uaddr, addr_len, flags); +} + /* FIXME: Write comments. */ static int sctp_disconnect(struct sock *sk, int flags) { @@ -8724,7 +8745,6 @@ struct proto sctp_prot = { .name = "SCTP", .owner = THIS_MODULE, .close = sctp_close, - .connect = sctp_connect, .disconnect = sctp_disconnect, .accept = sctp_accept, .ioctl = sctp_ioctl, @@ -8767,7 +8787,6 @@ struct proto sctpv6_prot = { .name = "SCTPv6", .owner = THIS_MODULE, .close = sctp_close, - .connect = sctp_connect, .disconnect = sctp_disconnect, .accept = sctp_accept, .ioctl = sctp_ioctl, From 9fcca5effcc02088a956f367b3c8d564c4b45450 Mon Sep 17 00:00:00 2001 From: Fabio Estevam Date: Sun, 20 May 2018 13:55:18 -0300 Subject: [PATCH 128/190] net: fec: ptp: Switch to SPDX identifier Adopt the SPDX license identifier headers to ease license compliance management. Signed-off-by: Fabio Estevam Acked-by: Fugang Duan Signed-off-by: David S. Miller --- drivers/net/ethernet/freescale/fec_ptp.c | 14 +------------- 1 file changed, 1 insertion(+), 13 deletions(-) diff --git a/drivers/net/ethernet/freescale/fec_ptp.c b/drivers/net/ethernet/freescale/fec_ptp.c index f81439796ac7..43d973215040 100644 --- a/drivers/net/ethernet/freescale/fec_ptp.c +++ b/drivers/net/ethernet/freescale/fec_ptp.c @@ -1,20 +1,8 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Fast Ethernet Controller (ENET) PTP driver for MX6x. * * Copyright (C) 2012 Freescale Semiconductor, Inc. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms and conditions of the GNU General Public License, - * version 2, as published by the Free Software Foundation. - * - * This program is distributed in the hope it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License along with - * this program; if not, write to the Free Software Foundation, Inc., - * 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. */ #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt From 1f508124e987fac97ac1765e3fbb83da36f47107 Mon Sep 17 00:00:00 2001 From: Fabio Estevam Date: Sun, 20 May 2018 13:55:19 -0300 Subject: [PATCH 129/190] net: fec: Add a SPDX identifier Currently there is no license information in the header of this file. The MODULE_LICENSE field contains ("GPL"), which means GNU Public License v2 or later, so add a corresponding SPDX license identifier. Signed-off-by: Fabio Estevam Acked-by: Fugang Duan Signed-off-by: David S. Miller --- drivers/net/ethernet/freescale/fec_main.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c index d4604bc8eb5b..9d3eed46830d 100644 --- a/drivers/net/ethernet/freescale/fec_main.c +++ b/drivers/net/ethernet/freescale/fec_main.c @@ -1,3 +1,4 @@ +// SPDX-License-Identifier: GPL-2.0+ /* * Fast Ethernet Controller (FEC) driver for Motorola MPC8xx. * Copyright (c) 1997 Dan Malek (dmalek@jlc.net) From 6009d1fe6ba3bb2dab55921da60465329cc1cd89 Mon Sep 17 00:00:00 2001 From: Wenwen Wang Date: Mon, 21 May 2018 01:58:07 -0500 Subject: [PATCH 130/190] isdn: eicon: fix a missing-check bug In divasmain.c, the function divas_write() firstly invokes the function diva_xdi_open_adapter() to open the adapter that matches with the adapter number provided by the user, and then invokes the function diva_xdi_write() to perform the write operation using the matched adapter. The two functions diva_xdi_open_adapter() and diva_xdi_write() are located in diva.c. In diva_xdi_open_adapter(), the user command is copied to the object 'msg' from the userspace pointer 'src' through the function pointer 'cp_fn', which eventually calls copy_from_user() to do the copy. Then, the adapter number 'msg.adapter' is used to find out a matched adapter from the 'adapter_queue'. A matched adapter will be returned if it is found. Otherwise, NULL is returned to indicate the failure of the verification on the adapter number. As mentioned above, if a matched adapter is returned, the function diva_xdi_write() is invoked to perform the write operation. In this function, the user command is copied once again from the userspace pointer 'src', which is the same as the 'src' pointer in diva_xdi_open_adapter() as both of them are from the 'buf' pointer in divas_write(). Similarly, the copy is achieved through the function pointer 'cp_fn', which finally calls copy_from_user(). After the successful copy, the corresponding command processing handler of the matched adapter is invoked to perform the write operation. It is obvious that there are two copies here from userspace, one is in diva_xdi_open_adapter(), and one is in diva_xdi_write(). Plus, both of these two copies share the same source userspace pointer, i.e., the 'buf' pointer in divas_write(). Given that a malicious userspace process can race to change the content pointed by the 'buf' pointer, this can pose potential security issues. For example, in the first copy, the user provides a valid adapter number to pass the verification process and a valid adapter can be found. Then the user can modify the adapter number to an invalid number. This way, the user can bypass the verification process of the adapter number and inject inconsistent data. This patch reuses the data copied in diva_xdi_open_adapter() and passes it to diva_xdi_write(). This way, the above issues can be avoided. Signed-off-by: Wenwen Wang Signed-off-by: David S. Miller --- drivers/isdn/hardware/eicon/diva.c | 22 +++++++++++++++------- drivers/isdn/hardware/eicon/diva.h | 5 +++-- drivers/isdn/hardware/eicon/divasmain.c | 18 +++++++++++------- 3 files changed, 29 insertions(+), 16 deletions(-) diff --git a/drivers/isdn/hardware/eicon/diva.c b/drivers/isdn/hardware/eicon/diva.c index 944a7f338099..1b25d8bc153a 100644 --- a/drivers/isdn/hardware/eicon/diva.c +++ b/drivers/isdn/hardware/eicon/diva.c @@ -388,10 +388,10 @@ void divasa_xdi_driver_unload(void) ** Receive and process command from user mode utility */ void *diva_xdi_open_adapter(void *os_handle, const void __user *src, - int length, + int length, void *mptr, divas_xdi_copy_from_user_fn_t cp_fn) { - diva_xdi_um_cfg_cmd_t msg; + diva_xdi_um_cfg_cmd_t *msg = (diva_xdi_um_cfg_cmd_t *)mptr; diva_os_xdi_adapter_t *a = NULL; diva_os_spin_lock_magic_t old_irql; struct list_head *tmp; @@ -401,21 +401,21 @@ void *diva_xdi_open_adapter(void *os_handle, const void __user *src, length, sizeof(diva_xdi_um_cfg_cmd_t))) return NULL; } - if ((*cp_fn) (os_handle, &msg, src, sizeof(msg)) <= 0) { + if ((*cp_fn) (os_handle, msg, src, sizeof(*msg)) <= 0) { DBG_ERR(("A: A(?) open, write error")) return NULL; } diva_os_enter_spin_lock(&adapter_lock, &old_irql, "open_adapter"); list_for_each(tmp, &adapter_queue) { a = list_entry(tmp, diva_os_xdi_adapter_t, link); - if (a->controller == (int)msg.adapter) + if (a->controller == (int)msg->adapter) break; a = NULL; } diva_os_leave_spin_lock(&adapter_lock, &old_irql, "open_adapter"); if (!a) { - DBG_ERR(("A: A(%d) open, adapter not found", msg.adapter)) + DBG_ERR(("A: A(%d) open, adapter not found", msg->adapter)) } return (a); @@ -437,8 +437,10 @@ void diva_xdi_close_adapter(void *adapter, void *os_handle) int diva_xdi_write(void *adapter, void *os_handle, const void __user *src, - int length, divas_xdi_copy_from_user_fn_t cp_fn) + int length, void *mptr, + divas_xdi_copy_from_user_fn_t cp_fn) { + diva_xdi_um_cfg_cmd_t *msg = (diva_xdi_um_cfg_cmd_t *)mptr; diva_os_xdi_adapter_t *a = (diva_os_xdi_adapter_t *) adapter; void *data; @@ -459,7 +461,13 @@ diva_xdi_write(void *adapter, void *os_handle, const void __user *src, return (-2); } - length = (*cp_fn) (os_handle, data, src, length); + if (msg) { + *(diva_xdi_um_cfg_cmd_t *)data = *msg; + length = (*cp_fn) (os_handle, (char *)data + sizeof(*msg), + src + sizeof(*msg), length - sizeof(*msg)); + } else { + length = (*cp_fn) (os_handle, data, src, length); + } if (length > 0) { if ((*(a->interface.cmd_proc)) (a, (diva_xdi_um_cfg_cmd_t *) data, length)) { diff --git a/drivers/isdn/hardware/eicon/diva.h b/drivers/isdn/hardware/eicon/diva.h index b067032093a8..1ad76650fbf9 100644 --- a/drivers/isdn/hardware/eicon/diva.h +++ b/drivers/isdn/hardware/eicon/diva.h @@ -20,10 +20,11 @@ int diva_xdi_read(void *adapter, void *os_handle, void __user *dst, int max_length, divas_xdi_copy_to_user_fn_t cp_fn); int diva_xdi_write(void *adapter, void *os_handle, const void __user *src, - int length, divas_xdi_copy_from_user_fn_t cp_fn); + int length, void *msg, + divas_xdi_copy_from_user_fn_t cp_fn); void *diva_xdi_open_adapter(void *os_handle, const void __user *src, - int length, + int length, void *msg, divas_xdi_copy_from_user_fn_t cp_fn); void diva_xdi_close_adapter(void *adapter, void *os_handle); diff --git a/drivers/isdn/hardware/eicon/divasmain.c b/drivers/isdn/hardware/eicon/divasmain.c index b9980e84f9db..b6a3950b2564 100644 --- a/drivers/isdn/hardware/eicon/divasmain.c +++ b/drivers/isdn/hardware/eicon/divasmain.c @@ -591,19 +591,22 @@ static int divas_release(struct inode *inode, struct file *file) static ssize_t divas_write(struct file *file, const char __user *buf, size_t count, loff_t *ppos) { + diva_xdi_um_cfg_cmd_t msg; int ret = -EINVAL; if (!file->private_data) { file->private_data = diva_xdi_open_adapter(file, buf, - count, + count, &msg, xdi_copy_from_user); - } - if (!file->private_data) { - return (-ENODEV); + if (!file->private_data) + return (-ENODEV); + ret = diva_xdi_write(file->private_data, file, + buf, count, &msg, xdi_copy_from_user); + } else { + ret = diva_xdi_write(file->private_data, file, + buf, count, NULL, xdi_copy_from_user); } - ret = diva_xdi_write(file->private_data, file, - buf, count, xdi_copy_from_user); switch (ret) { case -1: /* Message should be removed from rx mailbox first */ ret = -EBUSY; @@ -622,11 +625,12 @@ static ssize_t divas_write(struct file *file, const char __user *buf, static ssize_t divas_read(struct file *file, char __user *buf, size_t count, loff_t *ppos) { + diva_xdi_um_cfg_cmd_t msg; int ret = -EINVAL; if (!file->private_data) { file->private_data = diva_xdi_open_adapter(file, buf, - count, + count, &msg, xdi_copy_from_user); } if (!file->private_data) { From 2677d20677314101293e6da0094ede7b5526d2b1 Mon Sep 17 00:00:00 2001 From: Alexey Kodanev Date: Mon, 21 May 2018 19:28:44 +0300 Subject: [PATCH 131/190] dccp: don't free ccid2_hc_tx_sock struct in dccp_disconnect() Syzbot reported the use-after-free in timer_is_static_object() [1]. This can happen because the structure for the rto timer (ccid2_hc_tx_sock) is removed in dccp_disconnect(), and ccid2_hc_tx_rto_expire() can be called after that. The report [1] is similar to the one in commit 120e9dabaf55 ("dccp: defer ccid_hc_tx_delete() at dismantle time"). And the fix is the same, delay freeing ccid2_hc_tx_sock structure, so that it is freed in dccp_sk_destruct(). [1] ================================================================== BUG: KASAN: use-after-free in timer_is_static_object+0x80/0x90 kernel/time/timer.c:607 Read of size 8 at addr ffff8801bebb5118 by task syz-executor2/25299 CPU: 1 PID: 25299 Comm: syz-executor2 Not tainted 4.17.0-rc5+ #54 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Call Trace: __dump_stack lib/dump_stack.c:77 [inline] dump_stack+0x1b9/0x294 lib/dump_stack.c:113 print_address_description+0x6c/0x20b mm/kasan/report.c:256 kasan_report_error mm/kasan/report.c:354 [inline] kasan_report.cold.7+0x242/0x2fe mm/kasan/report.c:412 __asan_report_load8_noabort+0x14/0x20 mm/kasan/report.c:433 timer_is_static_object+0x80/0x90 kernel/time/timer.c:607 debug_object_activate+0x2d9/0x670 lib/debugobjects.c:508 debug_timer_activate kernel/time/timer.c:709 [inline] debug_activate kernel/time/timer.c:764 [inline] __mod_timer kernel/time/timer.c:1041 [inline] mod_timer+0x4d3/0x13b0 kernel/time/timer.c:1102 sk_reset_timer+0x22/0x60 net/core/sock.c:2742 ccid2_hc_tx_rto_expire+0x587/0x680 net/dccp/ccids/ccid2.c:147 call_timer_fn+0x230/0x940 kernel/time/timer.c:1326 expire_timers kernel/time/timer.c:1363 [inline] __run_timers+0x79e/0xc50 kernel/time/timer.c:1666 run_timer_softirq+0x4c/0x70 kernel/time/timer.c:1692 __do_softirq+0x2e0/0xaf5 kernel/softirq.c:285 invoke_softirq kernel/softirq.c:365 [inline] irq_exit+0x1d1/0x200 kernel/softirq.c:405 exiting_irq arch/x86/include/asm/apic.h:525 [inline] smp_apic_timer_interrupt+0x17e/0x710 arch/x86/kernel/apic/apic.c:1052 apic_timer_interrupt+0xf/0x20 arch/x86/entry/entry_64.S:863 ... Allocated by task 25374: save_stack+0x43/0xd0 mm/kasan/kasan.c:448 set_track mm/kasan/kasan.c:460 [inline] kasan_kmalloc+0xc4/0xe0 mm/kasan/kasan.c:553 kasan_slab_alloc+0x12/0x20 mm/kasan/kasan.c:490 kmem_cache_alloc+0x12e/0x760 mm/slab.c:3554 ccid_new+0x25b/0x3e0 net/dccp/ccid.c:151 dccp_hdlr_ccid+0x27/0x150 net/dccp/feat.c:44 __dccp_feat_activate+0x184/0x270 net/dccp/feat.c:344 dccp_feat_activate_values+0x3a7/0x819 net/dccp/feat.c:1538 dccp_create_openreq_child+0x472/0x610 net/dccp/minisocks.c:128 dccp_v4_request_recv_sock+0x12c/0xca0 net/dccp/ipv4.c:408 dccp_v6_request_recv_sock+0x125d/0x1f10 net/dccp/ipv6.c:415 dccp_check_req+0x455/0x6a0 net/dccp/minisocks.c:197 dccp_v4_rcv+0x7b8/0x1f3f net/dccp/ipv4.c:841 ip_local_deliver_finish+0x2e3/0xd80 net/ipv4/ip_input.c:215 NF_HOOK include/linux/netfilter.h:288 [inline] ip_local_deliver+0x1e1/0x720 net/ipv4/ip_input.c:256 dst_input include/net/dst.h:450 [inline] ip_rcv_finish+0x81b/0x2200 net/ipv4/ip_input.c:396 NF_HOOK include/linux/netfilter.h:288 [inline] ip_rcv+0xb70/0x143d net/ipv4/ip_input.c:492 __netif_receive_skb_core+0x26f5/0x3630 net/core/dev.c:4592 __netif_receive_skb+0x2c/0x1e0 net/core/dev.c:4657 process_backlog+0x219/0x760 net/core/dev.c:5337 napi_poll net/core/dev.c:5735 [inline] net_rx_action+0x7b7/0x1930 net/core/dev.c:5801 __do_softirq+0x2e0/0xaf5 kernel/softirq.c:285 Freed by task 25374: save_stack+0x43/0xd0 mm/kasan/kasan.c:448 set_track mm/kasan/kasan.c:460 [inline] __kasan_slab_free+0x11a/0x170 mm/kasan/kasan.c:521 kasan_slab_free+0xe/0x10 mm/kasan/kasan.c:528 __cache_free mm/slab.c:3498 [inline] kmem_cache_free+0x86/0x2d0 mm/slab.c:3756 ccid_hc_tx_delete+0xc3/0x100 net/dccp/ccid.c:190 dccp_disconnect+0x130/0xc66 net/dccp/proto.c:286 dccp_close+0x3bc/0xe60 net/dccp/proto.c:1045 inet_release+0x104/0x1f0 net/ipv4/af_inet.c:427 inet6_release+0x50/0x70 net/ipv6/af_inet6.c:460 sock_release+0x96/0x1b0 net/socket.c:594 sock_close+0x16/0x20 net/socket.c:1149 __fput+0x34d/0x890 fs/file_table.c:209 ____fput+0x15/0x20 fs/file_table.c:243 task_work_run+0x1e4/0x290 kernel/task_work.c:113 tracehook_notify_resume include/linux/tracehook.h:191 [inline] exit_to_usermode_loop+0x2bd/0x310 arch/x86/entry/common.c:166 prepare_exit_to_usermode arch/x86/entry/common.c:196 [inline] syscall_return_slowpath arch/x86/entry/common.c:265 [inline] do_syscall_64+0x6ac/0x800 arch/x86/entry/common.c:290 entry_SYSCALL_64_after_hwframe+0x49/0xbe The buggy address belongs to the object at ffff8801bebb4cc0 which belongs to the cache ccid2_hc_tx_sock of size 1240 The buggy address is located 1112 bytes inside of 1240-byte region [ffff8801bebb4cc0, ffff8801bebb5198) The buggy address belongs to the page: page:ffffea0006faed00 count:1 mapcount:0 mapping:ffff8801bebb41c0 index:0xffff8801bebb5240 compound_mapcount: 0 flags: 0x2fffc0000008100(slab|head) raw: 02fffc0000008100 ffff8801bebb41c0 ffff8801bebb5240 0000000100000003 raw: ffff8801cdba3138 ffffea0007634120 ffff8801cdbaab40 0000000000000000 page dumped because: kasan: bad access detected ... ================================================================== Reported-by: syzbot+5d47e9ec91a6f15dbd6f@syzkaller.appspotmail.com Signed-off-by: Alexey Kodanev Signed-off-by: David S. Miller --- net/dccp/proto.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/net/dccp/proto.c b/net/dccp/proto.c index 84cd4e3fd01b..0d56e36a6db7 100644 --- a/net/dccp/proto.c +++ b/net/dccp/proto.c @@ -283,9 +283,7 @@ int dccp_disconnect(struct sock *sk, int flags) dccp_clear_xmit_timers(sk); ccid_hc_rx_delete(dp->dccps_hc_rx_ccid, sk); - ccid_hc_tx_delete(dp->dccps_hc_tx_ccid, sk); dp->dccps_hc_rx_ccid = NULL; - dp->dccps_hc_tx_ccid = NULL; __skb_queue_purge(&sk->sk_receive_queue); __skb_queue_purge(&sk->sk_write_queue); From 66fb33254f45df4b049f487aff1cbde1ef919390 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 21 May 2018 10:51:53 -0700 Subject: [PATCH 132/190] ipmr: properly check rhltable_init() return value commit 8fb472c09b9d ("ipmr: improve hash scalability") added a call to rhltable_init() without checking its return value. This problem was then later copied to IPv6 and factorized in commit 0bbbf0e7d0e7 ("ipmr, ip6mr: Unite creation of new mr_table") kasan: CONFIG_KASAN_INLINE enabled kasan: GPF could be caused by NULL-ptr deref or user memory access general protection fault: 0000 [#1] SMP KASAN Dumping ftrace buffer: (ftrace buffer empty) Modules linked in: CPU: 1 PID: 31552 Comm: syz-executor7 Not tainted 4.17.0-rc5+ #60 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 RIP: 0010:rht_key_hashfn include/linux/rhashtable.h:277 [inline] RIP: 0010:__rhashtable_lookup include/linux/rhashtable.h:630 [inline] RIP: 0010:rhltable_lookup include/linux/rhashtable.h:716 [inline] RIP: 0010:mr_mfc_find_parent+0x2ad/0xbb0 net/ipv4/ipmr_base.c:63 RSP: 0018:ffff8801826aef70 EFLAGS: 00010203 RAX: 0000000000000001 RBX: 0000000000000001 RCX: ffffc90001ea0000 RDX: 0000000000000079 RSI: ffffffff8661e859 RDI: 000000000000000c RBP: ffff8801826af1c0 R08: ffff8801b2212000 R09: ffffed003b5e46c2 R10: ffffed003b5e46c2 R11: ffff8801daf23613 R12: dffffc0000000000 R13: ffff8801826af198 R14: ffff8801cf8225c0 R15: ffff8801826af658 FS: 00007ff7fa732700(0000) GS:ffff8801daf00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00000003ffffff9c CR3: 00000001b0210000 CR4: 00000000001406e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: ip6mr_cache_find_parent net/ipv6/ip6mr.c:981 [inline] ip6mr_mfc_delete+0x1fe/0x6b0 net/ipv6/ip6mr.c:1221 ip6_mroute_setsockopt+0x15c6/0x1d70 net/ipv6/ip6mr.c:1698 do_ipv6_setsockopt.isra.9+0x422/0x4660 net/ipv6/ipv6_sockglue.c:163 ipv6_setsockopt+0xbd/0x170 net/ipv6/ipv6_sockglue.c:922 rawv6_setsockopt+0x59/0x140 net/ipv6/raw.c:1060 sock_common_setsockopt+0x9a/0xe0 net/core/sock.c:3039 __sys_setsockopt+0x1bd/0x390 net/socket.c:1903 __do_sys_setsockopt net/socket.c:1914 [inline] __se_sys_setsockopt net/socket.c:1911 [inline] __x64_sys_setsockopt+0xbe/0x150 net/socket.c:1911 do_syscall_64+0x1b1/0x800 arch/x86/entry/common.c:287 entry_SYSCALL_64_after_hwframe+0x49/0xbe Fixes: 8fb472c09b9d ("ipmr: improve hash scalability") Fixes: 0bbbf0e7d0e7 ("ipmr, ip6mr: Unite creation of new mr_table") Signed-off-by: Eric Dumazet Cc: Nikolay Aleksandrov Cc: Yuval Mintz Reported-by: syzbot Acked-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- net/ipv4/ipmr_base.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/net/ipv4/ipmr_base.c b/net/ipv4/ipmr_base.c index 4fe97723b53f..30221701614c 100644 --- a/net/ipv4/ipmr_base.c +++ b/net/ipv4/ipmr_base.c @@ -43,7 +43,10 @@ mr_table_alloc(struct net *net, u32 id, write_pnet(&mrt->net, net); mrt->ops = *ops; - rhltable_init(&mrt->mfc_hash, mrt->ops.rht_params); + if (rhltable_init(&mrt->mfc_hash, mrt->ops.rht_params)) { + kfree(mrt); + return NULL; + } INIT_LIST_HEAD(&mrt->mfc_cache_list); INIT_LIST_HEAD(&mrt->mfc_unres_queue); From fdd13dd350dda1826579eb5c333d76b14513b812 Mon Sep 17 00:00:00 2001 From: Shahed Shaikh Date: Mon, 21 May 2018 12:31:47 -0700 Subject: [PATCH 133/190] qed: Fix mask for physical address in ILT entry ILT entry requires 12 bit right shifted physical address. Existing mask for ILT entry of physical address i.e. ILT_ENTRY_PHY_ADDR_MASK is not sufficient to handle 64bit address because upper 8 bits of 64 bit address were getting masked which resulted in completer abort error on PCIe bus due to invalid address. Fix that mask to handle 64bit physical address. Fixes: fe56b9e6a8d9 ("qed: Add module with basic common support") Signed-off-by: Shahed Shaikh Signed-off-by: Ariel Elior Signed-off-by: David S. Miller --- drivers/net/ethernet/qlogic/qed/qed_cxt.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/qlogic/qed/qed_cxt.c b/drivers/net/ethernet/qlogic/qed/qed_cxt.c index 00f41c145d4d..820b226d6ff8 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_cxt.c +++ b/drivers/net/ethernet/qlogic/qed/qed_cxt.c @@ -77,7 +77,7 @@ #define ILT_CFG_REG(cli, reg) PSWRQ2_REG_ ## cli ## _ ## reg ## _RT_OFFSET /* ILT entry structure */ -#define ILT_ENTRY_PHY_ADDR_MASK 0x000FFFFFFFFFFFULL +#define ILT_ENTRY_PHY_ADDR_MASK (~0ULL >> 12) #define ILT_ENTRY_PHY_ADDR_SHIFT 0 #define ILT_ENTRY_VALID_MASK 0x1ULL #define ILT_ENTRY_VALID_SHIFT 52 From d7db318651d5430eb79eceef1e6aa9b084ae78e6 Mon Sep 17 00:00:00 2001 From: Bo Chen Date: Mon, 21 May 2018 14:44:49 -0700 Subject: [PATCH 134/190] pcnet32: add an error handling path in pcnet32_probe_pci() Make sure to invoke pci_disable_device() when errors occur in pcnet32_probe_pci(). Signed-off-by: Bo Chen Signed-off-by: David S. Miller --- drivers/net/ethernet/amd/pcnet32.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/amd/pcnet32.c b/drivers/net/ethernet/amd/pcnet32.c index a561705f232c..be198cc0b10c 100644 --- a/drivers/net/ethernet/amd/pcnet32.c +++ b/drivers/net/ethernet/amd/pcnet32.c @@ -1552,22 +1552,26 @@ pcnet32_probe_pci(struct pci_dev *pdev, const struct pci_device_id *ent) if (!ioaddr) { if (pcnet32_debug & NETIF_MSG_PROBE) pr_err("card has no PCI IO resources, aborting\n"); - return -ENODEV; + err = -ENODEV; + goto err_disable_dev; } err = pci_set_dma_mask(pdev, PCNET32_DMA_MASK); if (err) { if (pcnet32_debug & NETIF_MSG_PROBE) pr_err("architecture does not support 32bit PCI busmaster DMA\n"); - return err; + goto err_disable_dev; } if (!request_region(ioaddr, PCNET32_TOTAL_SIZE, "pcnet32_probe_pci")) { if (pcnet32_debug & NETIF_MSG_PROBE) pr_err("io address range already allocated\n"); - return -EBUSY; + err = -EBUSY; + goto err_disable_dev; } err = pcnet32_probe1(ioaddr, 1, pdev); + +err_disable_dev: if (err < 0) pci_disable_device(pdev); From 6db615431a21b6057f68ed87583a663ee69f7601 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Wed, 9 May 2018 16:04:51 +0200 Subject: [PATCH 135/190] alpha: use dma_direct_ops for jensen The generic dma_direct implementation does the same thing as the alpha pci-noop implementation, just with more bells and whistles. And unlike the current code it at least has a theoretical chance to actually compile. Signed-off-by: Christoph Hellwig Signed-off-by: Matt Turner --- arch/alpha/Kconfig | 1 + arch/alpha/include/asm/dma-mapping.h | 4 ++++ arch/alpha/kernel/pci-noop.c | 33 ---------------------------- 3 files changed, 5 insertions(+), 33 deletions(-) diff --git a/arch/alpha/Kconfig b/arch/alpha/Kconfig index b2022885ced8..f19dc31288c8 100644 --- a/arch/alpha/Kconfig +++ b/arch/alpha/Kconfig @@ -211,6 +211,7 @@ config ALPHA_EIGER config ALPHA_JENSEN bool "Jensen" depends on BROKEN + select DMA_DIRECT_OPS help DEC PC 150 AXP (aka Jensen): This is a very old Digital system - one of the first-generation Alpha systems. A number of these systems diff --git a/arch/alpha/include/asm/dma-mapping.h b/arch/alpha/include/asm/dma-mapping.h index b78f61f20796..76ce923ecca1 100644 --- a/arch/alpha/include/asm/dma-mapping.h +++ b/arch/alpha/include/asm/dma-mapping.h @@ -6,7 +6,11 @@ extern const struct dma_map_ops *dma_ops; static inline const struct dma_map_ops *get_arch_dma_ops(struct bus_type *bus) { +#ifdef CONFIG_ALPHA_JENSEN + return &dma_direct_ops; +#else return dma_ops; +#endif } #endif /* _ALPHA_DMA_MAPPING_H */ diff --git a/arch/alpha/kernel/pci-noop.c b/arch/alpha/kernel/pci-noop.c index b6ebb65127a8..c7c5879869d3 100644 --- a/arch/alpha/kernel/pci-noop.c +++ b/arch/alpha/kernel/pci-noop.c @@ -102,36 +102,3 @@ SYSCALL_DEFINE5(pciconfig_write, unsigned long, bus, unsigned long, dfn, else return -ENODEV; } - -static void *alpha_noop_alloc_coherent(struct device *dev, size_t size, - dma_addr_t *dma_handle, gfp_t gfp, - unsigned long attrs) -{ - void *ret; - - if (!dev || *dev->dma_mask >= 0xffffffffUL) - gfp &= ~GFP_DMA; - ret = (void *)__get_free_pages(gfp, get_order(size)); - if (ret) { - memset(ret, 0, size); - *dma_handle = virt_to_phys(ret); - } - return ret; -} - -static int alpha_noop_supported(struct device *dev, u64 mask) -{ - return mask < 0x00ffffffUL ? 0 : 1; -} - -const struct dma_map_ops alpha_noop_ops = { - .alloc = alpha_noop_alloc_coherent, - .free = dma_noop_free_coherent, - .map_page = dma_noop_map_page, - .map_sg = dma_noop_map_sg, - .mapping_error = dma_noop_mapping_error, - .dma_supported = alpha_noop_supported, -}; - -const struct dma_map_ops *dma_ops = &alpha_noop_ops; -EXPORT_SYMBOL(dma_ops); From f5e82fa26063e6fad10624ff600457d878fa6e41 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Wed, 9 May 2018 16:04:52 +0200 Subject: [PATCH 136/190] alpha: simplify get_arch_dma_ops Remove the dma_ops indirection. Signed-off-by: Christoph Hellwig Signed-off-by: Matt Turner --- arch/alpha/include/asm/dma-mapping.h | 4 ++-- arch/alpha/kernel/pci_iommu.c | 4 +--- 2 files changed, 3 insertions(+), 5 deletions(-) diff --git a/arch/alpha/include/asm/dma-mapping.h b/arch/alpha/include/asm/dma-mapping.h index 76ce923ecca1..8beeafd4f68e 100644 --- a/arch/alpha/include/asm/dma-mapping.h +++ b/arch/alpha/include/asm/dma-mapping.h @@ -2,14 +2,14 @@ #ifndef _ALPHA_DMA_MAPPING_H #define _ALPHA_DMA_MAPPING_H -extern const struct dma_map_ops *dma_ops; +extern const struct dma_map_ops alpha_pci_ops; static inline const struct dma_map_ops *get_arch_dma_ops(struct bus_type *bus) { #ifdef CONFIG_ALPHA_JENSEN return &dma_direct_ops; #else - return dma_ops; + return &alpha_pci_ops; #endif } diff --git a/arch/alpha/kernel/pci_iommu.c b/arch/alpha/kernel/pci_iommu.c index 83b34b9188ea..6923b0d9c1e1 100644 --- a/arch/alpha/kernel/pci_iommu.c +++ b/arch/alpha/kernel/pci_iommu.c @@ -950,6 +950,4 @@ const struct dma_map_ops alpha_pci_ops = { .mapping_error = alpha_pci_mapping_error, .dma_supported = alpha_pci_supported, }; - -const struct dma_map_ops *dma_ops = &alpha_pci_ops; -EXPORT_SYMBOL(dma_ops); +EXPORT_SYMBOL(alpha_pci_ops); From 92d7223a74235054f2aa7227d207d9c57f84dca0 Mon Sep 17 00:00:00 2001 From: Sinan Kaya Date: Mon, 16 Apr 2018 18:16:56 -0400 Subject: [PATCH 137/190] alpha: io: reorder barriers to guarantee writeX() and iowriteX() ordering #2 memory-barriers.txt has been updated with the following requirement. "When using writel(), a prior wmb() is not needed to guarantee that the cache coherent memory writes have completed before writing to the MMIO region." Current writeX() and iowriteX() implementations on alpha are not satisfying this requirement as the barrier is after the register write. Move mb() in writeX() and iowriteX() functions to guarantee that HW observes memory changes before performing register operations. Signed-off-by: Sinan Kaya Reported-by: Arnd Bergmann Signed-off-by: Matt Turner --- arch/alpha/kernel/io.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/arch/alpha/kernel/io.c b/arch/alpha/kernel/io.c index 3e3d49c254c5..c025a3e5e357 100644 --- a/arch/alpha/kernel/io.c +++ b/arch/alpha/kernel/io.c @@ -37,20 +37,20 @@ unsigned int ioread32(void __iomem *addr) void iowrite8(u8 b, void __iomem *addr) { - IO_CONCAT(__IO_PREFIX,iowrite8)(b, addr); mb(); + IO_CONCAT(__IO_PREFIX,iowrite8)(b, addr); } void iowrite16(u16 b, void __iomem *addr) { - IO_CONCAT(__IO_PREFIX,iowrite16)(b, addr); mb(); + IO_CONCAT(__IO_PREFIX,iowrite16)(b, addr); } void iowrite32(u32 b, void __iomem *addr) { - IO_CONCAT(__IO_PREFIX,iowrite32)(b, addr); mb(); + IO_CONCAT(__IO_PREFIX,iowrite32)(b, addr); } EXPORT_SYMBOL(ioread8); @@ -176,26 +176,26 @@ u64 readq(const volatile void __iomem *addr) void writeb(u8 b, volatile void __iomem *addr) { - __raw_writeb(b, addr); mb(); + __raw_writeb(b, addr); } void writew(u16 b, volatile void __iomem *addr) { - __raw_writew(b, addr); mb(); + __raw_writew(b, addr); } void writel(u32 b, volatile void __iomem *addr) { - __raw_writel(b, addr); mb(); + __raw_writel(b, addr); } void writeq(u64 b, volatile void __iomem *addr) { - __raw_writeq(b, addr); mb(); + __raw_writeq(b, addr); } EXPORT_SYMBOL(readb); From 11799564fc7eedff50801950090773928f867996 Mon Sep 17 00:00:00 2001 From: Brian Norris Date: Tue, 22 May 2018 17:23:10 -0700 Subject: [PATCH 138/190] mfd: cros_ec: Retry commands when EC is known to be busy Commit 001dde9400d5 ("mfd: cros ec: spi: Fix "in progress" error signaling") pointed out some bad code, but its analysis and conclusion was not 100% correct. It *is* correct that we should not propagate result==EC_RES_IN_PROGRESS for transport errors, because this has a special meaning -- that we should follow up with EC_CMD_GET_COMMS_STATUS until the EC is no longer busy. This is definitely the wrong thing for many commands, because among other problems, EC_CMD_GET_COMMS_STATUS doesn't actually retrieve any RX data from the EC, so commands that expected some data back will instead start processing junk. For such commands, the right answer is to either propagate the error (and return that error to the caller) or resend the original command (*not* EC_CMD_GET_COMMS_STATUS). Unfortunately, commit 001dde9400d5 forgets a crucial point: that for some long-running operations, the EC physically cannot respond to commands any more. For example, with EC_CMD_FLASH_ERASE, the EC may be re-flashing its own code regions, so it can't respond to SPI interrupts. Instead, the EC prepares us ahead of time for being busy for a "long" time, and fills its hardware buffer with EC_SPI_PAST_END. Thus, we expect to see several "transport" errors (or, messages filled with EC_SPI_PAST_END). So we should really translate that to a retryable error (-EAGAIN) and continue sending EC_CMD_GET_COMMS_STATUS until we get a ready status. IOW, it is actually important to treat some of these "junk" values as retryable errors. Together with commit 001dde9400d5, this resolves bugs like the following: 1. EC_CMD_FLASH_ERASE now works again (with commit 001dde9400d5, we would abort the first time we saw EC_SPI_PAST_END) 2. Before commit 001dde9400d5, transport errors (e.g., EC_SPI_RX_BAD_DATA) seen in other commands (e.g., EC_CMD_RTC_GET_VALUE) used to yield junk data in the RX buffer; they will now yield -EAGAIN return values, and tools like 'hwclock' will simply fail instead of retrieving and re-programming undefined time values Fixes: 001dde9400d5 ("mfd: cros ec: spi: Fix "in progress" error signaling") Signed-off-by: Brian Norris Signed-off-by: Lee Jones --- drivers/mfd/cros_ec_spi.c | 24 ++++++++++++++++++++---- drivers/platform/chrome/cros_ec_proto.c | 2 ++ 2 files changed, 22 insertions(+), 4 deletions(-) diff --git a/drivers/mfd/cros_ec_spi.c b/drivers/mfd/cros_ec_spi.c index 1b52b8557034..2060d1483043 100644 --- a/drivers/mfd/cros_ec_spi.c +++ b/drivers/mfd/cros_ec_spi.c @@ -419,10 +419,25 @@ static int cros_ec_pkt_xfer_spi(struct cros_ec_device *ec_dev, /* Verify that EC can process command */ for (i = 0; i < len; i++) { rx_byte = rx_buf[i]; + /* + * Seeing the PAST_END, RX_BAD_DATA, or NOT_READY + * markers are all signs that the EC didn't fully + * receive our command. e.g., if the EC is flashing + * itself, it can't respond to any commands and instead + * clocks out EC_SPI_PAST_END from its SPI hardware + * buffer. Similar occurrences can happen if the AP is + * too slow to clock out data after asserting CS -- the + * EC will abort and fill its buffer with + * EC_SPI_RX_BAD_DATA. + * + * In all cases, these errors should be safe to retry. + * Report -EAGAIN and let the caller decide what to do + * about that. + */ if (rx_byte == EC_SPI_PAST_END || rx_byte == EC_SPI_RX_BAD_DATA || rx_byte == EC_SPI_NOT_READY) { - ret = -EREMOTEIO; + ret = -EAGAIN; break; } } @@ -431,7 +446,7 @@ static int cros_ec_pkt_xfer_spi(struct cros_ec_device *ec_dev, if (!ret) ret = cros_ec_spi_receive_packet(ec_dev, ec_msg->insize + sizeof(*response)); - else + else if (ret != -EAGAIN) dev_err(ec_dev->dev, "spi transfer failed: %d\n", ret); final_ret = terminate_request(ec_dev); @@ -537,10 +552,11 @@ static int cros_ec_cmd_xfer_spi(struct cros_ec_device *ec_dev, /* Verify that EC can process command */ for (i = 0; i < len; i++) { rx_byte = rx_buf[i]; + /* See comments in cros_ec_pkt_xfer_spi() */ if (rx_byte == EC_SPI_PAST_END || rx_byte == EC_SPI_RX_BAD_DATA || rx_byte == EC_SPI_NOT_READY) { - ret = -EREMOTEIO; + ret = -EAGAIN; break; } } @@ -549,7 +565,7 @@ static int cros_ec_cmd_xfer_spi(struct cros_ec_device *ec_dev, if (!ret) ret = cros_ec_spi_receive_response(ec_dev, ec_msg->insize + EC_MSG_TX_PROTO_BYTES); - else + else if (ret != -EAGAIN) dev_err(ec_dev->dev, "spi transfer failed: %d\n", ret); final_ret = terminate_request(ec_dev); diff --git a/drivers/platform/chrome/cros_ec_proto.c b/drivers/platform/chrome/cros_ec_proto.c index e7bbdf947bbc..8350ca2311c7 100644 --- a/drivers/platform/chrome/cros_ec_proto.c +++ b/drivers/platform/chrome/cros_ec_proto.c @@ -91,6 +91,8 @@ static int send_command(struct cros_ec_device *ec_dev, usleep_range(10000, 11000); ret = (*xfer_fxn)(ec_dev, status_msg); + if (ret == -EAGAIN) + continue; if (ret < 0) break; From 1a2b80ecc7ad374e9ef6a3de6fdd032d94be2270 Mon Sep 17 00:00:00 2001 From: Anders Roxell Date: Sat, 19 May 2018 00:27:37 +0200 Subject: [PATCH 139/190] selftests: net: reuseport_bpf_numa: don't fail if no numa support The reuseport_bpf_numa test case fails there's no numa support. The test shouldn't fail if there's no support it should be skipped. Fixes: 3c2c3c16aaf6 ("reuseport, bpf: add test case for bpf_get_numa_node_id") Signed-off-by: Anders Roxell Signed-off-by: Daniel Borkmann --- tools/testing/selftests/net/reuseport_bpf_numa.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/tools/testing/selftests/net/reuseport_bpf_numa.c b/tools/testing/selftests/net/reuseport_bpf_numa.c index 365c32e84189..c9f478b40996 100644 --- a/tools/testing/selftests/net/reuseport_bpf_numa.c +++ b/tools/testing/selftests/net/reuseport_bpf_numa.c @@ -23,6 +23,8 @@ #include #include +#include "../kselftest.h" + static const int PORT = 8888; static void build_rcv_group(int *rcv_fd, size_t len, int family, int proto) @@ -229,7 +231,7 @@ int main(void) int *rcv_fd, nodes; if (numa_available() < 0) - error(1, errno, "no numa api support"); + ksft_exit_skip("no numa api support\n"); nodes = numa_max_node() + 1; From 938ae7259c908ad031da35d551da297640bb640c Mon Sep 17 00:00:00 2001 From: Thomas Hellstrom Date: Wed, 23 May 2018 16:11:24 +0200 Subject: [PATCH 140/190] drm/vmwgfx: Fix 32-bit VMW_PORT_HB_[IN|OUT] macros Depending on whether the kernel is compiled with frame-pointer or not, the temporary memory location used for the bp parameter in these macros is referenced relative to the stack pointer or the frame pointer. Hence we can never reference that parameter when we've modified either the stack pointer or the frame pointer, because then the compiler would generate an incorrect stack reference. Fix this by pushing the temporary memory parameter on a known location on the stack before modifying the stack- and frame pointers. Cc: Signed-off-by: Thomas Hellstrom Reviewed-by: Brian Paul Reviewed-by: Sinclair Yeh --- drivers/gpu/drm/vmwgfx/vmwgfx_msg.h | 25 +++++++++++++++++-------- 1 file changed, 17 insertions(+), 8 deletions(-) diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_msg.h b/drivers/gpu/drm/vmwgfx/vmwgfx_msg.h index 557a033fb610..8545488aa0cf 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_msg.h +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_msg.h @@ -135,17 +135,24 @@ #else -/* In the 32-bit version of this macro, we use "m" because there is no - * more register left for bp +/* + * In the 32-bit version of this macro, we store bp in a memory location + * because we've ran out of registers. + * Now we can't reference that memory location while we've modified + * %esp or %ebp, so we first push it on the stack, just before we push + * %ebp, and then when we need it we read it from the stack where we + * just pushed it. */ #define VMW_PORT_HB_OUT(cmd, in_ecx, in_si, in_di, \ port_num, magic, bp, \ eax, ebx, ecx, edx, si, di) \ ({ \ - asm volatile ("push %%ebp;" \ - "mov %12, %%ebp;" \ + asm volatile ("push %12;" \ + "push %%ebp;" \ + "mov 0x04(%%esp), %%ebp;" \ "rep outsb;" \ - "pop %%ebp;" : \ + "pop %%ebp;" \ + "add $0x04, %%esp;" : \ "=a"(eax), \ "=b"(ebx), \ "=c"(ecx), \ @@ -167,10 +174,12 @@ port_num, magic, bp, \ eax, ebx, ecx, edx, si, di) \ ({ \ - asm volatile ("push %%ebp;" \ - "mov %12, %%ebp;" \ + asm volatile ("push %12;" \ + "push %%ebp;" \ + "mov 0x04(%%esp), %%ebp;" \ "rep insb;" \ - "pop %%ebp" : \ + "pop %%ebp;" \ + "add $0x04, %%esp;" : \ "=a"(eax), \ "=b"(ebx), \ "=c"(ecx), \ From f37230c0ad481091bc136788ff8b37dc86300c6d Mon Sep 17 00:00:00 2001 From: Thomas Hellstrom Date: Wed, 23 May 2018 16:13:20 +0200 Subject: [PATCH 141/190] drm/vmwgfx: Fix host logging / guestinfo reading error paths The error paths were leaking opened channels. Fix by using dedicated error paths. Cc: Signed-off-by: Thomas Hellstrom Reviewed-by: Brian Paul Reviewed-by: Sinclair Yeh --- drivers/gpu/drm/vmwgfx/vmwgfx_msg.c | 48 +++++++++++++++++++---------- 1 file changed, 31 insertions(+), 17 deletions(-) diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_msg.c b/drivers/gpu/drm/vmwgfx/vmwgfx_msg.c index cdff99211602..21d746bdc922 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_msg.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_msg.c @@ -329,8 +329,6 @@ int vmw_host_get_guestinfo(const char *guest_info_param, struct rpc_channel channel; char *msg, *reply = NULL; size_t reply_len = 0; - int ret = 0; - if (!vmw_msg_enabled) return -ENODEV; @@ -344,15 +342,14 @@ int vmw_host_get_guestinfo(const char *guest_info_param, return -ENOMEM; } - if (vmw_open_channel(&channel, RPCI_PROTOCOL_NUM) || - vmw_send_msg(&channel, msg) || - vmw_recv_msg(&channel, (void *) &reply, &reply_len) || - vmw_close_channel(&channel)) { - DRM_ERROR("Failed to get %s", guest_info_param); + if (vmw_open_channel(&channel, RPCI_PROTOCOL_NUM)) + goto out_open; - ret = -EINVAL; - } + if (vmw_send_msg(&channel, msg) || + vmw_recv_msg(&channel, (void *) &reply, &reply_len)) + goto out_msg; + vmw_close_channel(&channel); if (buffer && reply && reply_len > 0) { /* Remove reply code, which are the first 2 characters of * the reply @@ -369,7 +366,17 @@ int vmw_host_get_guestinfo(const char *guest_info_param, kfree(reply); kfree(msg); - return ret; + return 0; + +out_msg: + vmw_close_channel(&channel); + kfree(reply); +out_open: + *length = 0; + kfree(msg); + DRM_ERROR("Failed to get %s", guest_info_param); + + return -EINVAL; } @@ -400,15 +407,22 @@ int vmw_host_log(const char *log) return -ENOMEM; } - if (vmw_open_channel(&channel, RPCI_PROTOCOL_NUM) || - vmw_send_msg(&channel, msg) || - vmw_close_channel(&channel)) { - DRM_ERROR("Failed to send log\n"); + if (vmw_open_channel(&channel, RPCI_PROTOCOL_NUM)) + goto out_open; - ret = -EINVAL; - } + if (vmw_send_msg(&channel, msg)) + goto out_msg; + vmw_close_channel(&channel); kfree(msg); - return ret; + return 0; + +out_msg: + vmw_close_channel(&channel); +out_open: + kfree(msg); + DRM_ERROR("Failed to send log\n"); + + return -EINVAL; } From 6a93cea15ed38e2dba4a0552483d28b7a87a03bd Mon Sep 17 00:00:00 2001 From: Thomas Hellstrom Date: Wed, 23 May 2018 16:14:54 +0200 Subject: [PATCH 142/190] drm/vmwgfx: Schedule an fb dirty update after resume We have had problems displaying fbdev after a resume and as a workaround we have had to call vmw_fb_refresh(). This has had a number of unwanted side-effects. The root of the problem was, however that the coalesced fbdev dirty region was not empty on the first dirty_mark() after a resume, so a flush was never scheduled. Fix this by force scheduling an fbdev flush after resume, and remove the workaround. Signed-off-by: Thomas Hellstrom Reviewed-by: Brian Paul Reviewed-by: Deepak Rawat --- drivers/gpu/drm/vmwgfx/vmwgfx_drv.c | 5 ----- drivers/gpu/drm/vmwgfx/vmwgfx_drv.h | 1 - drivers/gpu/drm/vmwgfx/vmwgfx_fb.c | 24 ++++++++---------------- 3 files changed, 8 insertions(+), 22 deletions(-) diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c index 70e1a8820a7c..8b770a8e02cd 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c @@ -1278,8 +1278,6 @@ static void vmw_master_drop(struct drm_device *dev, dev_priv->active_master = &dev_priv->fbdev_master; ttm_lock_set_kill(&dev_priv->fbdev_master.lock, false, SIGTERM); ttm_vt_unlock(&dev_priv->fbdev_master.lock); - - vmw_fb_refresh(dev_priv); } /** @@ -1483,7 +1481,6 @@ static int vmw_pm_freeze(struct device *kdev) vmw_kms_resume(dev); if (dev_priv->enable_fb) vmw_fb_on(dev_priv); - vmw_fb_refresh(dev_priv); return -EBUSY; } @@ -1523,8 +1520,6 @@ static int vmw_pm_restore(struct device *kdev) if (dev_priv->enable_fb) vmw_fb_on(dev_priv); - vmw_fb_refresh(dev_priv); - return 0; } diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h index f34f368c1a2e..5fcbe1620d50 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h @@ -910,7 +910,6 @@ int vmw_fb_init(struct vmw_private *vmw_priv); int vmw_fb_close(struct vmw_private *dev_priv); int vmw_fb_off(struct vmw_private *vmw_priv); int vmw_fb_on(struct vmw_private *vmw_priv); -void vmw_fb_refresh(struct vmw_private *vmw_priv); /** * Kernel modesetting - vmwgfx_kms.c diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_fb.c b/drivers/gpu/drm/vmwgfx/vmwgfx_fb.c index ba0cdb743c3e..54e300365a5c 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_fb.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_fb.c @@ -866,21 +866,13 @@ int vmw_fb_on(struct vmw_private *vmw_priv) spin_lock_irqsave(&par->dirty.lock, flags); par->dirty.active = true; spin_unlock_irqrestore(&par->dirty.lock, flags); - + + /* + * Need to reschedule a dirty update, because otherwise that's + * only done in dirty_mark() if the previous coalesced + * dirty region was empty. + */ + schedule_delayed_work(&par->local_work, 0); + return 0; } - -/** - * vmw_fb_refresh - Refresh fb display - * - * @vmw_priv: Pointer to device private - * - * Call into kms to show the fbdev display(s). - */ -void vmw_fb_refresh(struct vmw_private *vmw_priv) -{ - if (!vmw_priv->fb_info) - return; - - vmw_fb_set_par(vmw_priv->fb_info); -} From 6890418bbb780f0ee9cf124055afa79777f1b4f1 Mon Sep 17 00:00:00 2001 From: Jason Wang Date: Tue, 22 May 2018 11:44:28 +0800 Subject: [PATCH 143/190] virtio-net: correctly redirect linearized packet After a linearized packet was redirected by XDP, we should not go for the err path which will try to pop buffers for the next packet and increase the drop counter. Fixing this by just drop the page refcnt for the original page. Fixes: 186b3c998c50 ("virtio-net: support XDP_REDIRECT") Reported-by: David Ahern Tested-by: David Ahern Acked-by: Michael S. Tsirkin Signed-off-by: Jason Wang Signed-off-by: David S. Miller --- drivers/net/virtio_net.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index 770422e953f7..c15d240f643f 100644 --- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -787,7 +787,7 @@ static struct sk_buff *receive_mergeable(struct net_device *dev, } *xdp_xmit = true; if (unlikely(xdp_page != page)) - goto err_xdp; + put_page(page); rcu_read_unlock(); goto xdp_xmit; default: From 5d458a13dd59d04b4d6658a6d5b94d42732b15ae Mon Sep 17 00:00:00 2001 From: Jason Wang Date: Tue, 22 May 2018 11:44:29 +0800 Subject: [PATCH 144/190] virtio-net: correctly transmit XDP buff after linearizing We should not go for the error path after successfully transmitting a XDP buffer after linearizing. Since the error path may try to pop and drop next packet and increase the drop counters. Fixing this by simply drop the refcnt of original page and go for xmit path. Fixes: 72979a6c3590 ("virtio_net: xdp, add slowpath case for non contiguous buffers") Cc: John Fastabend Acked-by: Michael S. Tsirkin Signed-off-by: Jason Wang Signed-off-by: David S. Miller --- drivers/net/virtio_net.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index c15d240f643f..6260d6549ee8 100644 --- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -775,7 +775,7 @@ static struct sk_buff *receive_mergeable(struct net_device *dev, } *xdp_xmit = true; if (unlikely(xdp_page != page)) - goto err_xdp; + put_page(page); rcu_read_unlock(); goto xdp_xmit; case XDP_REDIRECT: From 850e088d5bbb333342fd4def08d0a4035f2b7126 Mon Sep 17 00:00:00 2001 From: Jason Wang Date: Tue, 22 May 2018 11:44:30 +0800 Subject: [PATCH 145/190] virtio-net: correctly check num_buf during err path If we successfully linearize the packet, num_buf will be set to zero which may confuse error handling path which assumes num_buf is at least 1 and this can lead the code tries to pop the descriptor of next buffer. Fixing this by checking num_buf against 1 before decreasing. Fixes: 4941d472bf95 ("virtio-net: do not reset during XDP set") Signed-off-by: Jason Wang Signed-off-by: David S. Miller --- drivers/net/virtio_net.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index 6260d6549ee8..326e2476153d 100644 --- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -875,7 +875,7 @@ static struct sk_buff *receive_mergeable(struct net_device *dev, rcu_read_unlock(); err_skb: put_page(page); - while (--num_buf) { + while (num_buf-- > 1) { buf = virtqueue_get_buf(rq->vq, &len); if (unlikely(!buf)) { pr_debug("%s: rx error: %d buffers missing\n", From 3d62b2a0db505bbf9ed0755f254e45d775f9807f Mon Sep 17 00:00:00 2001 From: Jason Wang Date: Tue, 22 May 2018 11:44:31 +0800 Subject: [PATCH 146/190] virtio-net: fix leaking page for gso packet during mergeable XDP We need to drop refcnt to xdp_page if we see a gso packet. Otherwise it will be leaked. Fixing this by moving the check of gso packet above the linearizing logic. While at it, remove useless comment as well. Cc: John Fastabend Fixes: 72979a6c3590 ("virtio_net: xdp, add slowpath case for non contiguous buffers") Signed-off-by: Jason Wang Signed-off-by: David S. Miller --- drivers/net/virtio_net.c | 15 +++++++-------- 1 file changed, 7 insertions(+), 8 deletions(-) diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index 326e2476153d..032e1ac10a30 100644 --- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -707,6 +707,13 @@ static struct sk_buff *receive_mergeable(struct net_device *dev, void *data; u32 act; + /* Transient failure which in theory could occur if + * in-flight packets from before XDP was enabled reach + * the receive path after XDP is loaded. + */ + if (unlikely(hdr->hdr.gso_type)) + goto err_xdp; + /* This happens when rx buffer size is underestimated * or headroom is not enough because of the buffer * was refilled before XDP is set. This should only @@ -727,14 +734,6 @@ static struct sk_buff *receive_mergeable(struct net_device *dev, xdp_page = page; } - /* Transient failure which in theory could occur if - * in-flight packets from before XDP was enabled reach - * the receive path after XDP is loaded. In practice I - * was not able to create this condition. - */ - if (unlikely(hdr->hdr.gso_type)) - goto err_xdp; - /* Allow consuming headroom but reserve enough space to push * the descriptor on if we get an XDP_TX return code. */ From 2f3ab6221e4c87960347d65c7cab9bd917d1f637 Mon Sep 17 00:00:00 2001 From: Jason Wang Date: Tue, 22 May 2018 14:21:04 +0800 Subject: [PATCH 147/190] tuntap: correctly set SOCKWQ_ASYNC_NOSPACE When link is down, writes to the device might fail with -EIO. Userspace needs an indication when the status is resolved. As a fix, tun_net_open() attempts to wake up writers - but that is only effective if SOCKWQ_ASYNC_NOSPACE has been set in the past. This is not the case of vhost_net which only poll for EPOLLOUT after it meets errors during sendmsg(). This patch fixes this by making sure SOCKWQ_ASYNC_NOSPACE is set when socket is not writable or device is down to guarantee EPOLLOUT will be raised in either tun_chr_poll() or tun_sock_write_space() after device is up. Cc: Hannes Frederic Sowa Cc: Eric Dumazet Fixes: 1bd4978a88ac2 ("tun: honor IFF_UP in tun_get_user()") Signed-off-by: Jason Wang Signed-off-by: David S. Miller --- drivers/net/tun.c | 19 +++++++++++++++---- 1 file changed, 15 insertions(+), 4 deletions(-) diff --git a/drivers/net/tun.c b/drivers/net/tun.c index d45ac37e1287..45d807796a18 100644 --- a/drivers/net/tun.c +++ b/drivers/net/tun.c @@ -1423,6 +1423,13 @@ static void tun_net_init(struct net_device *dev) dev->max_mtu = MAX_MTU - dev->hard_header_len; } +static bool tun_sock_writeable(struct tun_struct *tun, struct tun_file *tfile) +{ + struct sock *sk = tfile->socket.sk; + + return (tun->dev->flags & IFF_UP) && sock_writeable(sk); +} + /* Character device part */ /* Poll */ @@ -1445,10 +1452,14 @@ static __poll_t tun_chr_poll(struct file *file, poll_table *wait) if (!ptr_ring_empty(&tfile->tx_ring)) mask |= EPOLLIN | EPOLLRDNORM; - if (tun->dev->flags & IFF_UP && - (sock_writeable(sk) || - (!test_and_set_bit(SOCKWQ_ASYNC_NOSPACE, &sk->sk_socket->flags) && - sock_writeable(sk)))) + /* Make sure SOCKWQ_ASYNC_NOSPACE is set if not writable to + * guarantee EPOLLOUT to be raised by either here or + * tun_sock_write_space(). Then process could get notification + * after it writes to a down device and meets -EIO. + */ + if (tun_sock_writeable(tun, tfile) || + (!test_and_set_bit(SOCKWQ_ASYNC_NOSPACE, &sk->sk_socket->flags) && + tun_sock_writeable(tun, tfile))) mask |= EPOLLOUT | EPOLLWRNORM; if (tun->dev->reg_state != NETREG_REGISTERED) From 73f9d36440a3bbaa034013e17d2032a905552ab2 Mon Sep 17 00:00:00 2001 From: Nathan Fontenot Date: Tue, 22 May 2018 11:21:10 -0500 Subject: [PATCH 148/190] ibmvnic: Only do H_EOI for mobility events When enabling the sub-CRQ IRQ a previous update sent a H_EOI prior to the enablement to clear any pending interrupts that may be present across a partition migration. This fixed a firmware bug where a migration could erroneously indicate that a H_EOI was pending. The H_EOI should only be sent when enabling during a mobility event though. Doing so at other time could wrong and can produce extra driver output when IRQs are enabled when doing TX completion. Signed-off-by: Nathan Fontenot Signed-off-by: David S. Miller --- drivers/net/ethernet/ibm/ibmvnic.c | 15 +++++++++------ 1 file changed, 9 insertions(+), 6 deletions(-) diff --git a/drivers/net/ethernet/ibm/ibmvnic.c b/drivers/net/ethernet/ibm/ibmvnic.c index 4bb4646a5f92..62cd3602c633 100644 --- a/drivers/net/ethernet/ibm/ibmvnic.c +++ b/drivers/net/ethernet/ibm/ibmvnic.c @@ -2617,18 +2617,21 @@ static int enable_scrq_irq(struct ibmvnic_adapter *adapter, { struct device *dev = &adapter->vdev->dev; unsigned long rc; - u64 val; if (scrq->hw_irq > 0x100000000ULL) { dev_err(dev, "bad hw_irq = %lx\n", scrq->hw_irq); return 1; } - val = (0xff000000) | scrq->hw_irq; - rc = plpar_hcall_norets(H_EOI, val); - if (rc) - dev_err(dev, "H_EOI FAILED irq 0x%llx. rc=%ld\n", - val, rc); + if (adapter->resetting && + adapter->reset_reason == VNIC_RESET_MOBILITY) { + u64 val = (0xff000000) | scrq->hw_irq; + + rc = plpar_hcall_norets(H_EOI, val); + if (rc) + dev_err(dev, "H_EOI FAILED irq 0x%llx. rc=%ld\n", + val, rc); + } rc = plpar_hcall_norets(H_VIOCTL, adapter->vdev->unit_address, H_ENABLE_VIO_INTERRUPT, scrq->hw_irq, 0, 0); From 4f7f56b6b11296909897696c0ad09d420b34208a Mon Sep 17 00:00:00 2001 From: Colin Ian King Date: Tue, 22 May 2018 16:42:51 +0100 Subject: [PATCH 149/190] net/mlx4: fix spelling mistake: "Inrerface" -> "Interface" and rephrase message Trivial fix to spelling mistake in mlx4_dbg debug message and also change the phrasing of the message so that is is more readable Signed-off-by: Colin Ian King Reviewed-by: Tariq Toukan Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx4/intf.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/mellanox/mlx4/intf.c b/drivers/net/ethernet/mellanox/mlx4/intf.c index 2edcce98ab2d..65482f004e50 100644 --- a/drivers/net/ethernet/mellanox/mlx4/intf.c +++ b/drivers/net/ethernet/mellanox/mlx4/intf.c @@ -172,7 +172,7 @@ int mlx4_do_bond(struct mlx4_dev *dev, bool enable) list_add_tail(&dev_ctx->list, &priv->ctx_list); spin_unlock_irqrestore(&priv->ctx_lock, flags); - mlx4_dbg(dev, "Inrerface for protocol %d restarted with when bonded mode is %s\n", + mlx4_dbg(dev, "Interface for protocol %d restarted with bonded mode %s\n", dev_ctx->intf->protocol, enable ? "enabled" : "disabled"); } From 2eabd764cb5512f1338d06ffc054c8bc9fbe9104 Mon Sep 17 00:00:00 2001 From: Roopa Prabhu Date: Tue, 22 May 2018 13:44:51 -0700 Subject: [PATCH 150/190] net: ipv4: add missing RTA_TABLE to rtm_ipv4_policy Signed-off-by: Roopa Prabhu Signed-off-by: David S. Miller --- net/ipv4/fib_frontend.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c index 4d622112bf95..e66172aaf241 100644 --- a/net/ipv4/fib_frontend.c +++ b/net/ipv4/fib_frontend.c @@ -649,6 +649,7 @@ const struct nla_policy rtm_ipv4_policy[RTA_MAX + 1] = { [RTA_ENCAP] = { .type = NLA_NESTED }, [RTA_UID] = { .type = NLA_U32 }, [RTA_MARK] = { .type = NLA_U32 }, + [RTA_TABLE] = { .type = NLA_U32 }, }; static int rtm_to_fib_config(struct net *net, struct sk_buff *skb, From 733a969a7ed14fc5786bcc59c1bdda83c7ddb46e Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Tue, 22 May 2018 16:22:26 -0700 Subject: [PATCH 151/190] net: phy: broadcom: Fix auxiliary control register reads We are currently doing auxiliary control register reads with the shadow register value 0b111 (0x7) which incidentally is also the selector value that should be present in bits [2:0]. Fix this by using the appropriate selector mask which is defined (MII_BCM54XX_AUXCTL_SHDWSEL_MASK). This does not have a functional impact yet because we always access the MII_BCM54XX_AUXCTL_SHDWSEL_MISC (0x7) register in the current code. This might change at some point though. Fixes: 5b4e29005123 ("net: phy: broadcom: add bcm54xx_auxctl_read") Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/phy/bcm-phy-lib.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/phy/bcm-phy-lib.c b/drivers/net/phy/bcm-phy-lib.c index 5ad130c3da43..d5e0833d69b9 100644 --- a/drivers/net/phy/bcm-phy-lib.c +++ b/drivers/net/phy/bcm-phy-lib.c @@ -56,7 +56,7 @@ int bcm54xx_auxctl_read(struct phy_device *phydev, u16 regnum) /* The register must be written to both the Shadow Register Select and * the Shadow Read Register Selector */ - phy_write(phydev, MII_BCM54XX_AUX_CTL, regnum | + phy_write(phydev, MII_BCM54XX_AUX_CTL, MII_BCM54XX_AUXCTL_SHDWSEL_MASK | regnum << MII_BCM54XX_AUXCTL_SHDWSEL_READ_SHIFT); return phy_read(phydev, MII_BCM54XX_AUX_CTL); } From 79fb218d97980d4fee9a64f4c8ff05289364ba25 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Tue, 22 May 2018 17:04:49 -0700 Subject: [PATCH 152/190] net: phy: broadcom: Fix bcm_write_exp() On newer PHYs, we need to select the expansion register to write with setting bits [11:8] to 0xf. This was done correctly by bcm7xxx.c prior to being migrated to generic code under bcm-phy-lib.c which unfortunately used the older implementation from the BCM54xx days. Fix this by creating an inline stub: bcm_write_exp_sel() which adds the correct value (MII_BCM54XX_EXP_SEL_ER) and update both the Cygnus PHY and BCM7xxx PHY drivers which require setting these bits. broadcom.c is unchanged because some PHYs even use a different selector method, so let them specify it directly (e.g: SerDes secondary selector). Fixes: a1cba5613edf ("net: phy: Add Broadcom phy library for common interfaces") Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/phy/bcm-cygnus.c | 6 +++--- drivers/net/phy/bcm-phy-lib.h | 7 +++++++ drivers/net/phy/bcm7xxx.c | 4 ++-- 3 files changed, 12 insertions(+), 5 deletions(-) diff --git a/drivers/net/phy/bcm-cygnus.c b/drivers/net/phy/bcm-cygnus.c index 6838129839ca..e757b09f1889 100644 --- a/drivers/net/phy/bcm-cygnus.c +++ b/drivers/net/phy/bcm-cygnus.c @@ -61,17 +61,17 @@ static int bcm_cygnus_afe_config(struct phy_device *phydev) return rc; /* make rcal=100, since rdb default is 000 */ - rc = bcm_phy_write_exp(phydev, MII_BRCM_CORE_EXPB1, 0x10); + rc = bcm_phy_write_exp_sel(phydev, MII_BRCM_CORE_EXPB1, 0x10); if (rc < 0) return rc; /* CORE_EXPB0, Reset R_CAL/RC_CAL Engine */ - rc = bcm_phy_write_exp(phydev, MII_BRCM_CORE_EXPB0, 0x10); + rc = bcm_phy_write_exp_sel(phydev, MII_BRCM_CORE_EXPB0, 0x10); if (rc < 0) return rc; /* CORE_EXPB0, Disable Reset R_CAL/RC_CAL Engine */ - rc = bcm_phy_write_exp(phydev, MII_BRCM_CORE_EXPB0, 0x00); + rc = bcm_phy_write_exp_sel(phydev, MII_BRCM_CORE_EXPB0, 0x00); return 0; } diff --git a/drivers/net/phy/bcm-phy-lib.h b/drivers/net/phy/bcm-phy-lib.h index 7c73808cbbde..81cceaa412fe 100644 --- a/drivers/net/phy/bcm-phy-lib.h +++ b/drivers/net/phy/bcm-phy-lib.h @@ -14,11 +14,18 @@ #ifndef _LINUX_BCM_PHY_LIB_H #define _LINUX_BCM_PHY_LIB_H +#include #include int bcm_phy_write_exp(struct phy_device *phydev, u16 reg, u16 val); int bcm_phy_read_exp(struct phy_device *phydev, u16 reg); +static inline int bcm_phy_write_exp_sel(struct phy_device *phydev, + u16 reg, u16 val) +{ + return bcm_phy_write_exp(phydev, reg | MII_BCM54XX_EXP_SEL_ER, val); +} + int bcm54xx_auxctl_write(struct phy_device *phydev, u16 regnum, u16 val); int bcm54xx_auxctl_read(struct phy_device *phydev, u16 regnum); diff --git a/drivers/net/phy/bcm7xxx.c b/drivers/net/phy/bcm7xxx.c index 29b1c88b55cc..01d2ff2f6241 100644 --- a/drivers/net/phy/bcm7xxx.c +++ b/drivers/net/phy/bcm7xxx.c @@ -65,10 +65,10 @@ struct bcm7xxx_phy_priv { static void r_rc_cal_reset(struct phy_device *phydev) { /* Reset R_CAL/RC_CAL Engine */ - bcm_phy_write_exp(phydev, 0x00b0, 0x0010); + bcm_phy_write_exp_sel(phydev, 0x00b0, 0x0010); /* Disable Reset R_AL/RC_CAL Engine */ - bcm_phy_write_exp(phydev, 0x00b0, 0x0000); + bcm_phy_write_exp_sel(phydev, 0x00b0, 0x0000); } static int bcm7xxx_28nm_b0_afe_config_init(struct phy_device *phydev) From d546b67cda015fb92bfee93d5dc0ceadb91deaee Mon Sep 17 00:00:00 2001 From: Jack Morgenstein Date: Wed, 23 May 2018 10:41:59 +0300 Subject: [PATCH 153/190] net/mlx4: Fix irq-unsafe spinlock usage spin_lock/unlock was used instead of spin_un/lock_irq in a procedure used in process space, on a spinlock which can be grabbed in an interrupt. This caused the stack trace below to be displayed (on kernel 4.17.0-rc1 compiled with Lock Debugging enabled): [ 154.661474] WARNING: SOFTIRQ-safe -> SOFTIRQ-unsafe lock order detected [ 154.668909] 4.17.0-rc1-rdma_rc_mlx+ #3 Tainted: G I [ 154.675856] ----------------------------------------------------- [ 154.682706] modprobe/10159 [HC0[0]:SC0[0]:HE0:SE1] is trying to acquire: [ 154.690254] 00000000f3b0e495 (&(&qp_table->lock)->rlock){+.+.}, at: mlx4_qp_remove+0x20/0x50 [mlx4_core] [ 154.700927] and this task is already holding: [ 154.707461] 0000000094373b5d (&(&cq->lock)->rlock/1){....}, at: destroy_qp_common+0x111/0x560 [mlx4_ib] [ 154.718028] which would create a new lock dependency: [ 154.723705] (&(&cq->lock)->rlock/1){....} -> (&(&qp_table->lock)->rlock){+.+.} [ 154.731922] but this new dependency connects a SOFTIRQ-irq-safe lock: [ 154.740798] (&(&cq->lock)->rlock){..-.} [ 154.740800] ... which became SOFTIRQ-irq-safe at: [ 154.752163] _raw_spin_lock_irqsave+0x3e/0x50 [ 154.757163] mlx4_ib_poll_cq+0x36/0x900 [mlx4_ib] [ 154.762554] ipoib_tx_poll+0x4a/0xf0 [ib_ipoib] ... to a SOFTIRQ-irq-unsafe lock: [ 154.815603] (&(&qp_table->lock)->rlock){+.+.} [ 154.815604] ... which became SOFTIRQ-irq-unsafe at: [ 154.827718] ... [ 154.827720] _raw_spin_lock+0x35/0x50 [ 154.833912] mlx4_qp_lookup+0x1e/0x50 [mlx4_core] [ 154.839302] mlx4_flow_attach+0x3f/0x3d0 [mlx4_core] Since mlx4_qp_lookup() is called only in process space, we can simply replace the spin_un/lock calls with spin_un/lock_irq calls. Fixes: 6dc06c08bef1 ("net/mlx4: Fix the check in attaching steering rules") Signed-off-by: Jack Morgenstein Signed-off-by: Tariq Toukan Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx4/qp.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx4/qp.c b/drivers/net/ethernet/mellanox/mlx4/qp.c index 3aaf4bad6c5a..427e7a31862c 100644 --- a/drivers/net/ethernet/mellanox/mlx4/qp.c +++ b/drivers/net/ethernet/mellanox/mlx4/qp.c @@ -393,11 +393,11 @@ struct mlx4_qp *mlx4_qp_lookup(struct mlx4_dev *dev, u32 qpn) struct mlx4_qp_table *qp_table = &mlx4_priv(dev)->qp_table; struct mlx4_qp *qp; - spin_lock(&qp_table->lock); + spin_lock_irq(&qp_table->lock); qp = __mlx4_qp_lookup(dev, qpn); - spin_unlock(&qp_table->lock); + spin_unlock_irq(&qp_table->lock); return qp; } From 30bf066cd9989fef34aeeef9080368867fe42be7 Mon Sep 17 00:00:00 2001 From: "Kalderon, Michal" Date: Tue, 15 May 2018 15:13:33 +0300 Subject: [PATCH 154/190] RDMA/qedr: Fix doorbell bar mapping for dpi > 1 Each user_context receives a separate dpi value and thus a different address on the doorbell bar. The qedr_mmap function needs to validate the address and map the doorbell bar accordingly. The current implementation always checked against dpi=0 doorbell range leading to a wrong mapping for doorbell bar. (It entered an else case that mapped the address differently). qedr_mmap should only be used for doorbells, so the else was actually wrong in the first place. This only has an affect on arm architecture and not an issue on a x86 based architecture. This lead to doorbells not occurring on arm based systems and left applications that use more than one dpi (or several applications run simultaneously ) to hang. Fixes: ac1b36e55a51 ("qedr: Add support for user context verbs") Signed-off-by: Ariel Elior Signed-off-by: Michal Kalderon Reviewed-by: Leon Romanovsky Signed-off-by: Jason Gunthorpe --- drivers/infiniband/hw/qedr/verbs.c | 62 +++++++++++++++--------------- 1 file changed, 30 insertions(+), 32 deletions(-) diff --git a/drivers/infiniband/hw/qedr/verbs.c b/drivers/infiniband/hw/qedr/verbs.c index 7d3763b2e01c..3f9afc02d166 100644 --- a/drivers/infiniband/hw/qedr/verbs.c +++ b/drivers/infiniband/hw/qedr/verbs.c @@ -401,49 +401,47 @@ int qedr_mmap(struct ib_ucontext *context, struct vm_area_struct *vma) { struct qedr_ucontext *ucontext = get_qedr_ucontext(context); struct qedr_dev *dev = get_qedr_dev(context->device); - unsigned long vm_page = vma->vm_pgoff << PAGE_SHIFT; - u64 unmapped_db = dev->db_phys_addr; + unsigned long phys_addr = vma->vm_pgoff << PAGE_SHIFT; unsigned long len = (vma->vm_end - vma->vm_start); - int rc = 0; - bool found; + unsigned long dpi_start; + + dpi_start = dev->db_phys_addr + (ucontext->dpi * ucontext->dpi_size); DP_DEBUG(dev, QEDR_MSG_INIT, - "qedr_mmap called vm_page=0x%lx vm_pgoff=0x%lx unmapped_db=0x%llx db_size=%x, len=%lx\n", - vm_page, vma->vm_pgoff, unmapped_db, dev->db_size, len); - if (vma->vm_start & (PAGE_SIZE - 1)) { - DP_ERR(dev, "Vma_start not page aligned = %ld\n", - vma->vm_start); + "mmap invoked with vm_start=0x%pK, vm_end=0x%pK,vm_pgoff=0x%pK; dpi_start=0x%pK dpi_size=0x%x\n", + (void *)vma->vm_start, (void *)vma->vm_end, + (void *)vma->vm_pgoff, (void *)dpi_start, ucontext->dpi_size); + + if ((vma->vm_start & (PAGE_SIZE - 1)) || (len & (PAGE_SIZE - 1))) { + DP_ERR(dev, + "failed mmap, adrresses must be page aligned: start=0x%pK, end=0x%pK\n", + (void *)vma->vm_start, (void *)vma->vm_end); return -EINVAL; } - found = qedr_search_mmap(ucontext, vm_page, len); - if (!found) { - DP_ERR(dev, "Vma_pgoff not found in mapped array = %ld\n", + if (!qedr_search_mmap(ucontext, phys_addr, len)) { + DP_ERR(dev, "failed mmap, vm_pgoff=0x%lx is not authorized\n", vma->vm_pgoff); return -EINVAL; } - DP_DEBUG(dev, QEDR_MSG_INIT, "Mapping doorbell bar\n"); - - if ((vm_page >= unmapped_db) && (vm_page <= (unmapped_db + - dev->db_size))) { - DP_DEBUG(dev, QEDR_MSG_INIT, "Mapping doorbell bar\n"); - if (vma->vm_flags & VM_READ) { - DP_ERR(dev, "Trying to map doorbell bar for read\n"); - return -EPERM; - } - - vma->vm_page_prot = pgprot_writecombine(vma->vm_page_prot); - - rc = io_remap_pfn_range(vma, vma->vm_start, vma->vm_pgoff, - PAGE_SIZE, vma->vm_page_prot); - } else { - DP_DEBUG(dev, QEDR_MSG_INIT, "Mapping chains\n"); - rc = remap_pfn_range(vma, vma->vm_start, - vma->vm_pgoff, len, vma->vm_page_prot); + if (phys_addr < dpi_start || + ((phys_addr + len) > (dpi_start + ucontext->dpi_size))) { + DP_ERR(dev, + "failed mmap, pages are outside of dpi; page address=0x%pK, dpi_start=0x%pK, dpi_size=0x%x\n", + (void *)phys_addr, (void *)dpi_start, + ucontext->dpi_size); + return -EINVAL; } - DP_DEBUG(dev, QEDR_MSG_INIT, "qedr_mmap return code: %d\n", rc); - return rc; + + if (vma->vm_flags & VM_READ) { + DP_ERR(dev, "failed mmap, cannot map doorbell bar for read\n"); + return -EINVAL; + } + + vma->vm_page_prot = pgprot_writecombine(vma->vm_page_prot); + return io_remap_pfn_range(vma, vma->vm_start, vma->vm_pgoff, len, + vma->vm_page_prot); } struct ib_pd *qedr_alloc_pd(struct ib_device *ibdev, From f4602cbb0a2478dda8238a4f382867da425daa8e Mon Sep 17 00:00:00 2001 From: Jason Gunthorpe Date: Tue, 22 May 2018 15:56:51 -0600 Subject: [PATCH 155/190] IB/uverbs: Fix uverbs_attr_get_obj The err pointer comes from uverbs_attr_get, not from the uobject member, which does not store an ERR_PTR. Fixes: be934cca9e98 ("IB/uverbs: Add device memory registration ioctl support") Signed-off-by: Jason Gunthorpe Reviewed-by: Leon Romanovsky --- include/rdma/uverbs_ioctl.h | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/include/rdma/uverbs_ioctl.h b/include/rdma/uverbs_ioctl.h index 4a4201d997a7..095383a4bd1a 100644 --- a/include/rdma/uverbs_ioctl.h +++ b/include/rdma/uverbs_ioctl.h @@ -411,13 +411,13 @@ static inline int uverbs_attr_get_enum_id(const struct uverbs_attr_bundle *attrs static inline void *uverbs_attr_get_obj(const struct uverbs_attr_bundle *attrs_bundle, u16 idx) { - struct ib_uobject *uobj = - uverbs_attr_get(attrs_bundle, idx)->obj_attr.uobject; + const struct uverbs_attr *attr; - if (IS_ERR(uobj)) - return uobj; + attr = uverbs_attr_get(attrs_bundle, idx); + if (IS_ERR(attr)) + return ERR_CAST(attr); - return uobj->object; + return attr->obj_attr.uobject->object; } static inline int uverbs_copy_to(const struct uverbs_attr_bundle *attrs_bundle, From 05d6a4ddb654ef6f2fbbcf9dcb3b263184baa8e4 Mon Sep 17 00:00:00 2001 From: oulijun Date: Tue, 22 May 2018 20:47:14 +0800 Subject: [PATCH 156/190] RDMA/hns: Bugfix for cq record db for kernel When use cq record db for kernel, it needs to set the hr_cq->db_en to 1 and configure the dma address of record cq db of qp context. Fixes: 86188a8810ed ("RDMA/hns: Support cq record doorbell for kernel space") Signed-off-by: Lijun Ou Signed-off-by: Jason Gunthorpe --- drivers/infiniband/hw/hns/hns_roce_cq.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/infiniband/hw/hns/hns_roce_cq.c b/drivers/infiniband/hw/hns/hns_roce_cq.c index 14734d0d0b76..3a485f50fede 100644 --- a/drivers/infiniband/hw/hns/hns_roce_cq.c +++ b/drivers/infiniband/hw/hns/hns_roce_cq.c @@ -377,6 +377,7 @@ struct ib_cq *hns_roce_ib_create_cq(struct ib_device *ib_dev, hr_cq->set_ci_db = hr_cq->db.db_record; *hr_cq->set_ci_db = 0; + hr_cq->db_en = 1; } /* Init mmt table and write buff address to mtt table */ From 55ba49cbcef37053d973f9a45bc58818c333fe13 Mon Sep 17 00:00:00 2001 From: oulijun Date: Tue, 22 May 2018 20:47:15 +0800 Subject: [PATCH 157/190] RDMA/hns: Move the location for initializing tmp_len When posted work request, it need to compute the length of all sges of every wr and fill it into the msg_len field of send wqe. Thus, While posting multiple wr, tmp_len should be reinitialized to zero. Fixes: 8b9b8d143b46 ("RDMA/hns: Fix the endian problem for hns") Signed-off-by: Lijun Ou Signed-off-by: Jason Gunthorpe --- drivers/infiniband/hw/hns/hns_roce_hw_v2.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c index 7ecf0c911808..1f0965bb64ee 100644 --- a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c +++ b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c @@ -142,8 +142,8 @@ static int hns_roce_v2_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, unsigned long flags; unsigned int ind; void *wqe = NULL; - u32 tmp_len = 0; bool loopback; + u32 tmp_len; int ret = 0; u8 *smac; int nreq; @@ -189,6 +189,7 @@ static int hns_roce_v2_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, owner_bit = ~(((qp->sq.head + nreq) >> ilog2(qp->sq.wqe_cnt)) & 0x1); + tmp_len = 0; /* Corresponding to the QP type, wqe process separately */ if (ibqp->qp_type == IB_QPT_GSI) { From d50147381aa0c9725d63a677c138c47f55d6d3bc Mon Sep 17 00:00:00 2001 From: Omar Sandoval Date: Tue, 22 May 2018 09:47:58 -0700 Subject: [PATCH 158/190] Btrfs: fix error handling in btrfs_truncate() Jun Wu at Facebook reported that an internal service was seeing a return value of 1 from ftruncate() on Btrfs in some cases. This is coming from the NEED_TRUNCATE_BLOCK return value from btrfs_truncate_inode_items(). btrfs_truncate() uses two variables for error handling, ret and err. When btrfs_truncate_inode_items() returns non-zero, we set err to the return value. However, NEED_TRUNCATE_BLOCK is not an error. Make sure we only set err if ret is an error (i.e., negative). To reproduce the issue: mount a filesystem with -o compress-force=zstd and the following program will encounter return value of 1 from ftruncate: int main(void) { char buf[256] = { 0 }; int ret; int fd; fd = open("test", O_CREAT | O_WRONLY | O_TRUNC, 0666); if (fd == -1) { perror("open"); return EXIT_FAILURE; } if (write(fd, buf, sizeof(buf)) != sizeof(buf)) { perror("write"); close(fd); return EXIT_FAILURE; } if (fsync(fd) == -1) { perror("fsync"); close(fd); return EXIT_FAILURE; } ret = ftruncate(fd, 128); if (ret) { printf("ftruncate() returned %d\n", ret); close(fd); return EXIT_FAILURE; } close(fd); return EXIT_SUCCESS; } Fixes: ddfae63cc8e0 ("btrfs: move btrfs_truncate_block out of trans handle") CC: stable@vger.kernel.org # 4.15+ Reported-by: Jun Wu Signed-off-by: Omar Sandoval Signed-off-by: David Sterba --- fs/btrfs/inode.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 8e604e7071f1..d82afca0c05f 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -9124,7 +9124,8 @@ static int btrfs_truncate(struct inode *inode, bool skip_writeback) BTRFS_EXTENT_DATA_KEY); trans->block_rsv = &fs_info->trans_block_rsv; if (ret != -ENOSPC && ret != -EAGAIN) { - err = ret; + if (ret < 0) + err = ret; break; } From 82034c23fcbc2389c73d97737f61fa2dd6526413 Mon Sep 17 00:00:00 2001 From: Laura Abbott Date: Wed, 23 May 2018 11:43:46 -0700 Subject: [PATCH 159/190] arm64: Make sure permission updates happen for pmd/pud Commit 15122ee2c515 ("arm64: Enforce BBM for huge IO/VMAP mappings") disallowed block mappings for ioremap since that code does not honor break-before-make. The same APIs are also used for permission updating though and the extra checks prevent the permission updates from happening, even though this should be permitted. This results in read-only permissions not being fully applied. Visibly, this can occasionaly be seen as a failure on the built in rodata test when the test data ends up in a section or as an odd RW gap on the page table dump. Fix this by using pgattr_change_is_safe instead of p*d_present for determining if the change is permitted. Reviewed-by: Kees Cook Tested-by: Peter Robinson Reported-by: Peter Robinson Fixes: 15122ee2c515 ("arm64: Enforce BBM for huge IO/VMAP mappings") Signed-off-by: Laura Abbott Signed-off-by: Will Deacon --- arch/arm64/mm/mmu.c | 16 ++++++++++------ 1 file changed, 10 insertions(+), 6 deletions(-) diff --git a/arch/arm64/mm/mmu.c b/arch/arm64/mm/mmu.c index 2dbb2c9f1ec1..493ff75670ff 100644 --- a/arch/arm64/mm/mmu.c +++ b/arch/arm64/mm/mmu.c @@ -933,13 +933,15 @@ int pud_set_huge(pud_t *pudp, phys_addr_t phys, pgprot_t prot) { pgprot_t sect_prot = __pgprot(PUD_TYPE_SECT | pgprot_val(mk_sect_prot(prot))); + pud_t new_pud = pfn_pud(__phys_to_pfn(phys), sect_prot); - /* ioremap_page_range doesn't honour BBM */ - if (pud_present(READ_ONCE(*pudp))) + /* Only allow permission changes for now */ + if (!pgattr_change_is_safe(READ_ONCE(pud_val(*pudp)), + pud_val(new_pud))) return 0; BUG_ON(phys & ~PUD_MASK); - set_pud(pudp, pfn_pud(__phys_to_pfn(phys), sect_prot)); + set_pud(pudp, new_pud); return 1; } @@ -947,13 +949,15 @@ int pmd_set_huge(pmd_t *pmdp, phys_addr_t phys, pgprot_t prot) { pgprot_t sect_prot = __pgprot(PMD_TYPE_SECT | pgprot_val(mk_sect_prot(prot))); + pmd_t new_pmd = pfn_pmd(__phys_to_pfn(phys), sect_prot); - /* ioremap_page_range doesn't honour BBM */ - if (pmd_present(READ_ONCE(*pmdp))) + /* Only allow permission changes for now */ + if (!pgattr_change_is_safe(READ_ONCE(pmd_val(*pmdp)), + pmd_val(new_pmd))) return 0; BUG_ON(phys & ~PMD_MASK); - set_pmd(pmdp, pfn_pmd(__phys_to_pfn(phys), sect_prot)); + set_pmd(pmdp, new_pmd); return 1; } From 4544e403eb25552aed7f0ee181a7a506b8800403 Mon Sep 17 00:00:00 2001 From: Mika Westerberg Date: Thu, 24 May 2018 11:12:16 +0300 Subject: [PATCH 160/190] ahci: Add PCI ID for Cannon Lake PCH-LP AHCI This one should be using the default LPM policy for mobile chipsets so add the PCI ID to the driver list of supported revices. Signed-off-by: Mika Westerberg Signed-off-by: Tejun Heo Cc: stable@vger.kernel.org --- drivers/ata/ahci.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c index 6389c88b3500..738fb22978dd 100644 --- a/drivers/ata/ahci.c +++ b/drivers/ata/ahci.c @@ -334,6 +334,7 @@ static const struct pci_device_id ahci_pci_tbl[] = { { PCI_VDEVICE(INTEL, 0x9c07), board_ahci_mobile }, /* Lynx LP RAID */ { PCI_VDEVICE(INTEL, 0x9c0e), board_ahci_mobile }, /* Lynx LP RAID */ { PCI_VDEVICE(INTEL, 0x9c0f), board_ahci_mobile }, /* Lynx LP RAID */ + { PCI_VDEVICE(INTEL, 0x9dd3), board_ahci_mobile }, /* Cannon Lake PCH-LP AHCI */ { PCI_VDEVICE(INTEL, 0x1f22), board_ahci }, /* Avoton AHCI */ { PCI_VDEVICE(INTEL, 0x1f23), board_ahci }, /* Avoton AHCI */ { PCI_VDEVICE(INTEL, 0x1f24), board_ahci }, /* Avoton RAID */ From c93552c443ebc63b14e26e46d2e76941c88e0d71 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Thu, 24 May 2018 02:32:53 +0200 Subject: [PATCH 161/190] bpf: properly enforce index mask to prevent out-of-bounds speculation While reviewing the verifier code, I recently noticed that the following two program variants in relation to tail calls can be loaded. Variant 1: # bpftool p d x i 15 0: (15) if r1 == 0x0 goto pc+3 1: (18) r2 = map[id:5] 3: (05) goto pc+2 4: (18) r2 = map[id:6] 6: (b7) r3 = 7 7: (35) if r3 >= 0xa0 goto pc+2 8: (54) (u32) r3 &= (u32) 255 9: (85) call bpf_tail_call#12 10: (b7) r0 = 1 11: (95) exit # bpftool m s i 5 5: prog_array flags 0x0 key 4B value 4B max_entries 4 memlock 4096B # bpftool m s i 6 6: prog_array flags 0x0 key 4B value 4B max_entries 160 memlock 4096B Variant 2: # bpftool p d x i 20 0: (15) if r1 == 0x0 goto pc+3 1: (18) r2 = map[id:8] 3: (05) goto pc+2 4: (18) r2 = map[id:7] 6: (b7) r3 = 7 7: (35) if r3 >= 0x4 goto pc+2 8: (54) (u32) r3 &= (u32) 3 9: (85) call bpf_tail_call#12 10: (b7) r0 = 1 11: (95) exit # bpftool m s i 8 8: prog_array flags 0x0 key 4B value 4B max_entries 160 memlock 4096B # bpftool m s i 7 7: prog_array flags 0x0 key 4B value 4B max_entries 4 memlock 4096B In both cases the index masking inserted by the verifier in order to control out of bounds speculation from a CPU via b2157399cc98 ("bpf: prevent out-of-bounds speculation") seems to be incorrect in what it is enforcing. In the 1st variant, the mask is applied from the map with the significantly larger number of entries where we would allow to a certain degree out of bounds speculation for the smaller map, and in the 2nd variant where the mask is applied from the map with the smaller number of entries, we get buggy behavior since we truncate the index of the larger map. The original intent from commit b2157399cc98 is to reject such occasions where two or more different tail call maps are used in the same tail call helper invocation. However, the check on the BPF_MAP_PTR_POISON is never hit since we never poisoned the saved pointer in the first place! We do this explicitly for map lookups but in case of tail calls we basically used the tail call map in insn_aux_data that was processed in the most recent path which the verifier walked. Thus any prior path that stored a pointer in insn_aux_data at the helper location was always overridden. Fix it by moving the map pointer poison logic into a small helper that covers both BPF helpers with the same logic. After that in fixup_bpf_calls() the poison check is then hit for tail calls and the program rejected. Latter only happens in unprivileged case since this is the *only* occasion where a rewrite needs to happen, and where such rewrite is specific to the map (max_entries, index_mask). In the privileged case the rewrite is generic for the insn->imm / insn->code update so multiple maps from different paths can be handled just fine since all the remaining logic happens in the instruction processing itself. This is similar to the case of map lookups: in case there is a collision of maps in fixup_bpf_calls() we must skip the inlined rewrite since this will turn the generic instruction sequence into a non- generic one. Thus the patch_call_imm will simply update the insn->imm location where the bpf_map_lookup_elem() will later take care of the dispatch. Given we need this 'poison' state as a check, the information of whether a map is an unpriv_array gets lost, so enforcing it prior to that needs an additional state. In general this check is needed since there are some complex and tail call intensive BPF programs out there where LLVM tends to generate such code occasionally. We therefore convert the map_ptr rather into map_state to store all this w/o extra memory overhead, and the bit whether one of the maps involved in the collision was from an unpriv_array thus needs to be retained as well there. Fixes: b2157399cc98 ("bpf: prevent out-of-bounds speculation") Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: Alexei Starovoitov --- include/linux/bpf_verifier.h | 2 +- kernel/bpf/verifier.c | 86 +++++++++++++++++++++++++++--------- 2 files changed, 65 insertions(+), 23 deletions(-) diff --git a/include/linux/bpf_verifier.h b/include/linux/bpf_verifier.h index 7e61c395fddf..52fb077d3c45 100644 --- a/include/linux/bpf_verifier.h +++ b/include/linux/bpf_verifier.h @@ -142,7 +142,7 @@ struct bpf_verifier_state_list { struct bpf_insn_aux_data { union { enum bpf_reg_type ptr_type; /* pointer type for load/store insns */ - struct bpf_map *map_ptr; /* pointer for call insn into lookup_elem */ + unsigned long map_state; /* pointer/poison value for maps */ s32 call_imm; /* saved imm field of call insn */ }; int ctx_field_size; /* the ctx field size for load insn, maybe 0 */ diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index 5dd1dcb902bf..dcebf3f7365c 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -156,7 +156,29 @@ struct bpf_verifier_stack_elem { #define BPF_COMPLEXITY_LIMIT_INSNS 131072 #define BPF_COMPLEXITY_LIMIT_STACK 1024 -#define BPF_MAP_PTR_POISON ((void *)0xeB9F + POISON_POINTER_DELTA) +#define BPF_MAP_PTR_UNPRIV 1UL +#define BPF_MAP_PTR_POISON ((void *)((0xeB9FUL << 1) + \ + POISON_POINTER_DELTA)) +#define BPF_MAP_PTR(X) ((struct bpf_map *)((X) & ~BPF_MAP_PTR_UNPRIV)) + +static bool bpf_map_ptr_poisoned(const struct bpf_insn_aux_data *aux) +{ + return BPF_MAP_PTR(aux->map_state) == BPF_MAP_PTR_POISON; +} + +static bool bpf_map_ptr_unpriv(const struct bpf_insn_aux_data *aux) +{ + return aux->map_state & BPF_MAP_PTR_UNPRIV; +} + +static void bpf_map_ptr_store(struct bpf_insn_aux_data *aux, + const struct bpf_map *map, bool unpriv) +{ + BUILD_BUG_ON((unsigned long)BPF_MAP_PTR_POISON & BPF_MAP_PTR_UNPRIV); + unpriv |= bpf_map_ptr_unpriv(aux); + aux->map_state = (unsigned long)map | + (unpriv ? BPF_MAP_PTR_UNPRIV : 0UL); +} struct bpf_call_arg_meta { struct bpf_map *map_ptr; @@ -2333,6 +2355,29 @@ static int prepare_func_exit(struct bpf_verifier_env *env, int *insn_idx) return 0; } +static int +record_func_map(struct bpf_verifier_env *env, struct bpf_call_arg_meta *meta, + int func_id, int insn_idx) +{ + struct bpf_insn_aux_data *aux = &env->insn_aux_data[insn_idx]; + + if (func_id != BPF_FUNC_tail_call && + func_id != BPF_FUNC_map_lookup_elem) + return 0; + if (meta->map_ptr == NULL) { + verbose(env, "kernel subsystem misconfigured verifier\n"); + return -EINVAL; + } + + if (!BPF_MAP_PTR(aux->map_state)) + bpf_map_ptr_store(aux, meta->map_ptr, + meta->map_ptr->unpriv_array); + else if (BPF_MAP_PTR(aux->map_state) != meta->map_ptr) + bpf_map_ptr_store(aux, BPF_MAP_PTR_POISON, + meta->map_ptr->unpriv_array); + return 0; +} + static int check_helper_call(struct bpf_verifier_env *env, int func_id, int insn_idx) { const struct bpf_func_proto *fn = NULL; @@ -2387,13 +2432,6 @@ static int check_helper_call(struct bpf_verifier_env *env, int func_id, int insn err = check_func_arg(env, BPF_REG_2, fn->arg2_type, &meta); if (err) return err; - if (func_id == BPF_FUNC_tail_call) { - if (meta.map_ptr == NULL) { - verbose(env, "verifier bug\n"); - return -EINVAL; - } - env->insn_aux_data[insn_idx].map_ptr = meta.map_ptr; - } err = check_func_arg(env, BPF_REG_3, fn->arg3_type, &meta); if (err) return err; @@ -2404,6 +2442,10 @@ static int check_helper_call(struct bpf_verifier_env *env, int func_id, int insn if (err) return err; + err = record_func_map(env, &meta, func_id, insn_idx); + if (err) + return err; + /* Mark slots with STACK_MISC in case of raw mode, stack offset * is inferred from register state. */ @@ -2428,8 +2470,6 @@ static int check_helper_call(struct bpf_verifier_env *env, int func_id, int insn } else if (fn->ret_type == RET_VOID) { regs[BPF_REG_0].type = NOT_INIT; } else if (fn->ret_type == RET_PTR_TO_MAP_VALUE_OR_NULL) { - struct bpf_insn_aux_data *insn_aux; - regs[BPF_REG_0].type = PTR_TO_MAP_VALUE_OR_NULL; /* There is no offset yet applied, variable or fixed */ mark_reg_known_zero(env, regs, BPF_REG_0); @@ -2445,11 +2485,6 @@ static int check_helper_call(struct bpf_verifier_env *env, int func_id, int insn } regs[BPF_REG_0].map_ptr = meta.map_ptr; regs[BPF_REG_0].id = ++env->id_gen; - insn_aux = &env->insn_aux_data[insn_idx]; - if (!insn_aux->map_ptr) - insn_aux->map_ptr = meta.map_ptr; - else if (insn_aux->map_ptr != meta.map_ptr) - insn_aux->map_ptr = BPF_MAP_PTR_POISON; } else { verbose(env, "unknown return type %d of func %s#%d\n", fn->ret_type, func_id_name(func_id), func_id); @@ -5417,6 +5452,7 @@ static int fixup_bpf_calls(struct bpf_verifier_env *env) struct bpf_insn *insn = prog->insnsi; const struct bpf_func_proto *fn; const int insn_cnt = prog->len; + struct bpf_insn_aux_data *aux; struct bpf_insn insn_buf[16]; struct bpf_prog *new_prog; struct bpf_map *map_ptr; @@ -5491,19 +5527,22 @@ static int fixup_bpf_calls(struct bpf_verifier_env *env) insn->imm = 0; insn->code = BPF_JMP | BPF_TAIL_CALL; + aux = &env->insn_aux_data[i + delta]; + if (!bpf_map_ptr_unpriv(aux)) + continue; + /* instead of changing every JIT dealing with tail_call * emit two extra insns: * if (index >= max_entries) goto out; * index &= array->index_mask; * to avoid out-of-bounds cpu speculation */ - map_ptr = env->insn_aux_data[i + delta].map_ptr; - if (map_ptr == BPF_MAP_PTR_POISON) { + if (bpf_map_ptr_poisoned(aux)) { verbose(env, "tail_call abusing map_ptr\n"); return -EINVAL; } - if (!map_ptr->unpriv_array) - continue; + + map_ptr = BPF_MAP_PTR(aux->map_state); insn_buf[0] = BPF_JMP_IMM(BPF_JGE, BPF_REG_3, map_ptr->max_entries, 2); insn_buf[1] = BPF_ALU32_IMM(BPF_AND, BPF_REG_3, @@ -5527,9 +5566,12 @@ static int fixup_bpf_calls(struct bpf_verifier_env *env) */ if (prog->jit_requested && BITS_PER_LONG == 64 && insn->imm == BPF_FUNC_map_lookup_elem) { - map_ptr = env->insn_aux_data[i + delta].map_ptr; - if (map_ptr == BPF_MAP_PTR_POISON || - !map_ptr->ops->map_gen_lookup) + aux = &env->insn_aux_data[i + delta]; + if (bpf_map_ptr_poisoned(aux)) + goto patch_call_imm; + + map_ptr = BPF_MAP_PTR(aux->map_state); + if (!map_ptr->ops->map_gen_lookup) goto patch_call_imm; cnt = map_ptr->ops->map_gen_lookup(map_ptr, insn_buf); From d883c6cf3b39f1f42506e82ad2779fb88004acf3 Mon Sep 17 00:00:00 2001 From: Joonsoo Kim Date: Wed, 23 May 2018 10:18:21 +0900 Subject: [PATCH 162/190] Revert "mm/cma: manage the memory of the CMA area by using the ZONE_MOVABLE" MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This reverts the following commits that change CMA design in MM. 3d2054ad8c2d ("ARM: CMA: avoid double mapping to the CMA area if CONFIG_HIGHMEM=y") 1d47a3ec09b5 ("mm/cma: remove ALLOC_CMA") bad8c6c0b114 ("mm/cma: manage the memory of the CMA area by using the ZONE_MOVABLE") Ville reported a following error on i386. Inode-cache hash table entries: 65536 (order: 6, 262144 bytes) microcode: microcode updated early to revision 0x4, date = 2013-06-28 Initializing CPU#0 Initializing HighMem for node 0 (000377fe:00118000) Initializing Movable for node 0 (00000001:00118000) BUG: Bad page state in process swapper pfn:377fe page:f53effc0 count:0 mapcount:-127 mapping:00000000 index:0x0 flags: 0x80000000() raw: 80000000 00000000 00000000 ffffff80 00000000 00000100 00000200 00000001 page dumped because: nonzero mapcount Modules linked in: CPU: 0 PID: 0 Comm: swapper Not tainted 4.17.0-rc5-elk+ #145 Hardware name: Dell Inc. Latitude E5410/03VXMC, BIOS A15 07/11/2013 Call Trace: dump_stack+0x60/0x96 bad_page+0x9a/0x100 free_pages_check_bad+0x3f/0x60 free_pcppages_bulk+0x29d/0x5b0 free_unref_page_commit+0x84/0xb0 free_unref_page+0x3e/0x70 __free_pages+0x1d/0x20 free_highmem_page+0x19/0x40 add_highpages_with_active_regions+0xab/0xeb set_highmem_pages_init+0x66/0x73 mem_init+0x1b/0x1d7 start_kernel+0x17a/0x363 i386_start_kernel+0x95/0x99 startup_32_smp+0x164/0x168 The reason for this error is that the span of MOVABLE_ZONE is extended to whole node span for future CMA initialization, and, normal memory is wrongly freed here. I submitted the fix and it seems to work, but, another problem happened. It's so late time to fix the later problem so I decide to reverting the series. Reported-by: Ville Syrjälä Acked-by: Laura Abbott Acked-by: Michal Hocko Cc: Andrew Morton Signed-off-by: Joonsoo Kim Signed-off-by: Linus Torvalds --- arch/arm/mm/dma-mapping.c | 16 +------ include/linux/memory_hotplug.h | 3 ++ include/linux/mm.h | 1 - mm/cma.c | 83 +++++----------------------------- mm/compaction.c | 4 +- mm/internal.h | 4 +- mm/page_alloc.c | 83 ++++++++++++---------------------- 7 files changed, 49 insertions(+), 145 deletions(-) diff --git a/arch/arm/mm/dma-mapping.c b/arch/arm/mm/dma-mapping.c index 8c398fedbbb6..ada8eb206a90 100644 --- a/arch/arm/mm/dma-mapping.c +++ b/arch/arm/mm/dma-mapping.c @@ -466,12 +466,6 @@ void __init dma_contiguous_early_fixup(phys_addr_t base, unsigned long size) void __init dma_contiguous_remap(void) { int i; - - if (!dma_mmu_remap_num) - return; - - /* call flush_cache_all() since CMA area would be large enough */ - flush_cache_all(); for (i = 0; i < dma_mmu_remap_num; i++) { phys_addr_t start = dma_mmu_remap[i].base; phys_addr_t end = start + dma_mmu_remap[i].size; @@ -504,15 +498,7 @@ void __init dma_contiguous_remap(void) flush_tlb_kernel_range(__phys_to_virt(start), __phys_to_virt(end)); - /* - * All the memory in CMA region will be on ZONE_MOVABLE. - * If that zone is considered as highmem, the memory in CMA - * region is also considered as highmem even if it's - * physical address belong to lowmem. In this case, - * re-mapping isn't required. - */ - if (!is_highmem_idx(ZONE_MOVABLE)) - iotable_init(&map, 1); + iotable_init(&map, 1); } } diff --git a/include/linux/memory_hotplug.h b/include/linux/memory_hotplug.h index e0e49b5b1ee1..2b0265265c28 100644 --- a/include/linux/memory_hotplug.h +++ b/include/linux/memory_hotplug.h @@ -216,6 +216,9 @@ void put_online_mems(void); void mem_hotplug_begin(void); void mem_hotplug_done(void); +extern void set_zone_contiguous(struct zone *zone); +extern void clear_zone_contiguous(struct zone *zone); + #else /* ! CONFIG_MEMORY_HOTPLUG */ #define pfn_to_online_page(pfn) \ ({ \ diff --git a/include/linux/mm.h b/include/linux/mm.h index c6fa9a255dbf..02a616e2f17d 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -2109,7 +2109,6 @@ extern void setup_per_cpu_pageset(void); extern void zone_pcp_update(struct zone *zone); extern void zone_pcp_reset(struct zone *zone); -extern void setup_zone_pageset(struct zone *zone); /* page_alloc.c */ extern int min_free_kbytes; diff --git a/mm/cma.c b/mm/cma.c index aa40e6c7b042..5809bbe360d7 100644 --- a/mm/cma.c +++ b/mm/cma.c @@ -39,7 +39,6 @@ #include #include "cma.h" -#include "internal.h" struct cma cma_areas[MAX_CMA_AREAS]; unsigned cma_area_count; @@ -110,25 +109,23 @@ static int __init cma_activate_area(struct cma *cma) if (!cma->bitmap) return -ENOMEM; + WARN_ON_ONCE(!pfn_valid(pfn)); + zone = page_zone(pfn_to_page(pfn)); + do { unsigned j; base_pfn = pfn; - if (!pfn_valid(base_pfn)) - goto err; - - zone = page_zone(pfn_to_page(base_pfn)); for (j = pageblock_nr_pages; j; --j, pfn++) { - if (!pfn_valid(pfn)) - goto err; - + WARN_ON_ONCE(!pfn_valid(pfn)); /* - * In init_cma_reserved_pageblock(), present_pages - * is adjusted with assumption that all pages in - * the pageblock come from a single zone. + * alloc_contig_range requires the pfn range + * specified to be in the same zone. Make this + * simple by forcing the entire CMA resv range + * to be in the same zone. */ if (page_zone(pfn_to_page(pfn)) != zone) - goto err; + goto not_in_zone; } init_cma_reserved_pageblock(pfn_to_page(base_pfn)); } while (--i); @@ -142,7 +139,7 @@ static int __init cma_activate_area(struct cma *cma) return 0; -err: +not_in_zone: pr_err("CMA area %s could not be activated\n", cma->name); kfree(cma->bitmap); cma->count = 0; @@ -152,41 +149,6 @@ static int __init cma_activate_area(struct cma *cma) static int __init cma_init_reserved_areas(void) { int i; - struct zone *zone; - pg_data_t *pgdat; - - if (!cma_area_count) - return 0; - - for_each_online_pgdat(pgdat) { - unsigned long start_pfn = UINT_MAX, end_pfn = 0; - - zone = &pgdat->node_zones[ZONE_MOVABLE]; - - /* - * In this case, we cannot adjust the zone range - * since it is now maximum node span and we don't - * know original zone range. - */ - if (populated_zone(zone)) - continue; - - for (i = 0; i < cma_area_count; i++) { - if (pfn_to_nid(cma_areas[i].base_pfn) != - pgdat->node_id) - continue; - - start_pfn = min(start_pfn, cma_areas[i].base_pfn); - end_pfn = max(end_pfn, cma_areas[i].base_pfn + - cma_areas[i].count); - } - - if (!end_pfn) - continue; - - zone->zone_start_pfn = start_pfn; - zone->spanned_pages = end_pfn - start_pfn; - } for (i = 0; i < cma_area_count; i++) { int ret = cma_activate_area(&cma_areas[i]); @@ -195,32 +157,9 @@ static int __init cma_init_reserved_areas(void) return ret; } - /* - * Reserved pages for ZONE_MOVABLE are now activated and - * this would change ZONE_MOVABLE's managed page counter and - * the other zones' present counter. We need to re-calculate - * various zone information that depends on this initialization. - */ - build_all_zonelists(NULL); - for_each_populated_zone(zone) { - if (zone_idx(zone) == ZONE_MOVABLE) { - zone_pcp_reset(zone); - setup_zone_pageset(zone); - } else - zone_pcp_update(zone); - - set_zone_contiguous(zone); - } - - /* - * We need to re-init per zone wmark by calling - * init_per_zone_wmark_min() but doesn't call here because it is - * registered on core_initcall and it will be called later than us. - */ - return 0; } -pure_initcall(cma_init_reserved_areas); +core_initcall(cma_init_reserved_areas); /** * cma_init_reserved_mem() - create custom contiguous area from reserved memory diff --git a/mm/compaction.c b/mm/compaction.c index 028b7210a669..29bd1df18b98 100644 --- a/mm/compaction.c +++ b/mm/compaction.c @@ -1450,12 +1450,14 @@ static enum compact_result __compaction_suitable(struct zone *zone, int order, * if compaction succeeds. * For costly orders, we require low watermark instead of min for * compaction to proceed to increase its chances. + * ALLOC_CMA is used, as pages in CMA pageblocks are considered + * suitable migration targets */ watermark = (order > PAGE_ALLOC_COSTLY_ORDER) ? low_wmark_pages(zone) : min_wmark_pages(zone); watermark += compact_gap(order); if (!__zone_watermark_ok(zone, 0, watermark, classzone_idx, - 0, wmark_target)) + ALLOC_CMA, wmark_target)) return COMPACT_SKIPPED; return COMPACT_CONTINUE; diff --git a/mm/internal.h b/mm/internal.h index 62d8c34e63d5..502d14189794 100644 --- a/mm/internal.h +++ b/mm/internal.h @@ -168,9 +168,6 @@ extern void post_alloc_hook(struct page *page, unsigned int order, gfp_t gfp_flags); extern int user_min_free_kbytes; -extern void set_zone_contiguous(struct zone *zone); -extern void clear_zone_contiguous(struct zone *zone); - #if defined CONFIG_COMPACTION || defined CONFIG_CMA /* @@ -498,6 +495,7 @@ unsigned long reclaim_clean_pages_from_list(struct zone *zone, #define ALLOC_HARDER 0x10 /* try to alloc harder */ #define ALLOC_HIGH 0x20 /* __GFP_HIGH set */ #define ALLOC_CPUSET 0x40 /* check for correct cpuset */ +#define ALLOC_CMA 0x80 /* allow allocations from CMA areas */ enum ttu_flags; struct tlbflush_unmap_batch; diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 905db9d7962f..511a7124d7f9 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1743,38 +1743,16 @@ void __init page_alloc_init_late(void) } #ifdef CONFIG_CMA -static void __init adjust_present_page_count(struct page *page, long count) -{ - struct zone *zone = page_zone(page); - - /* We don't need to hold a lock since it is boot-up process */ - zone->present_pages += count; -} - /* Free whole pageblock and set its migration type to MIGRATE_CMA. */ void __init init_cma_reserved_pageblock(struct page *page) { unsigned i = pageblock_nr_pages; - unsigned long pfn = page_to_pfn(page); struct page *p = page; - int nid = page_to_nid(page); - - /* - * ZONE_MOVABLE will steal present pages from other zones by - * changing page links so page_zone() is changed. Before that, - * we need to adjust previous zone's page count first. - */ - adjust_present_page_count(page, -pageblock_nr_pages); do { __ClearPageReserved(p); set_page_count(p, 0); - - /* Steal pages from other zones */ - set_page_links(p, ZONE_MOVABLE, nid, pfn); - } while (++p, ++pfn, --i); - - adjust_present_page_count(page, pageblock_nr_pages); + } while (++p, --i); set_pageblock_migratetype(page, MIGRATE_CMA); @@ -2889,7 +2867,7 @@ int __isolate_free_page(struct page *page, unsigned int order) * exists. */ watermark = min_wmark_pages(zone) + (1UL << order); - if (!zone_watermark_ok(zone, 0, watermark, 0, 0)) + if (!zone_watermark_ok(zone, 0, watermark, 0, ALLOC_CMA)) return 0; __mod_zone_freepage_state(zone, -(1UL << order), mt); @@ -3165,6 +3143,12 @@ bool __zone_watermark_ok(struct zone *z, unsigned int order, unsigned long mark, } +#ifdef CONFIG_CMA + /* If allocation can't use CMA areas don't use free CMA pages */ + if (!(alloc_flags & ALLOC_CMA)) + free_pages -= zone_page_state(z, NR_FREE_CMA_PAGES); +#endif + /* * Check watermarks for an order-0 allocation request. If these * are not met, then a high-order request also cannot go ahead @@ -3191,8 +3175,10 @@ bool __zone_watermark_ok(struct zone *z, unsigned int order, unsigned long mark, } #ifdef CONFIG_CMA - if (!list_empty(&area->free_list[MIGRATE_CMA])) + if ((alloc_flags & ALLOC_CMA) && + !list_empty(&area->free_list[MIGRATE_CMA])) { return true; + } #endif if (alloc_harder && !list_empty(&area->free_list[MIGRATE_HIGHATOMIC])) @@ -3212,6 +3198,13 @@ static inline bool zone_watermark_fast(struct zone *z, unsigned int order, unsigned long mark, int classzone_idx, unsigned int alloc_flags) { long free_pages = zone_page_state(z, NR_FREE_PAGES); + long cma_pages = 0; + +#ifdef CONFIG_CMA + /* If allocation can't use CMA areas don't use free CMA pages */ + if (!(alloc_flags & ALLOC_CMA)) + cma_pages = zone_page_state(z, NR_FREE_CMA_PAGES); +#endif /* * Fast check for order-0 only. If this fails then the reserves @@ -3220,7 +3213,7 @@ static inline bool zone_watermark_fast(struct zone *z, unsigned int order, * the caller is !atomic then it'll uselessly search the free * list. That corner case is then slower but it is harmless. */ - if (!order && free_pages > mark + z->lowmem_reserve[classzone_idx]) + if (!order && (free_pages - cma_pages) > mark + z->lowmem_reserve[classzone_idx]) return true; return __zone_watermark_ok(z, order, mark, classzone_idx, alloc_flags, @@ -3856,6 +3849,10 @@ gfp_to_alloc_flags(gfp_t gfp_mask) } else if (unlikely(rt_task(current)) && !in_interrupt()) alloc_flags |= ALLOC_HARDER; +#ifdef CONFIG_CMA + if (gfpflags_to_migratetype(gfp_mask) == MIGRATE_MOVABLE) + alloc_flags |= ALLOC_CMA; +#endif return alloc_flags; } @@ -4322,6 +4319,9 @@ static inline bool prepare_alloc_pages(gfp_t gfp_mask, unsigned int order, if (should_fail_alloc_page(gfp_mask, order)) return false; + if (IS_ENABLED(CONFIG_CMA) && ac->migratetype == MIGRATE_MOVABLE) + *alloc_flags |= ALLOC_CMA; + return true; } @@ -6204,7 +6204,6 @@ static void __paginginit free_area_init_core(struct pglist_data *pgdat) { enum zone_type j; int nid = pgdat->node_id; - unsigned long node_end_pfn = 0; pgdat_resize_init(pgdat); #ifdef CONFIG_NUMA_BALANCING @@ -6232,13 +6231,9 @@ static void __paginginit free_area_init_core(struct pglist_data *pgdat) struct zone *zone = pgdat->node_zones + j; unsigned long size, realsize, freesize, memmap_pages; unsigned long zone_start_pfn = zone->zone_start_pfn; - unsigned long movable_size = 0; size = zone->spanned_pages; realsize = freesize = zone->present_pages; - if (zone_end_pfn(zone) > node_end_pfn) - node_end_pfn = zone_end_pfn(zone); - /* * Adjust freesize so that it accounts for how much memory @@ -6287,30 +6282,12 @@ static void __paginginit free_area_init_core(struct pglist_data *pgdat) zone_seqlock_init(zone); zone_pcp_init(zone); - /* - * The size of the CMA area is unknown now so we need to - * prepare the memory for the usemap at maximum. - */ - if (IS_ENABLED(CONFIG_CMA) && j == ZONE_MOVABLE && - pgdat->node_spanned_pages) { - movable_size = node_end_pfn - pgdat->node_start_pfn; - } - - if (!size && !movable_size) + if (!size) continue; set_pageblock_order(); - if (movable_size) { - zone->zone_start_pfn = pgdat->node_start_pfn; - zone->spanned_pages = movable_size; - setup_usemap(pgdat, zone, - pgdat->node_start_pfn, movable_size); - init_currently_empty_zone(zone, - pgdat->node_start_pfn, movable_size); - } else { - setup_usemap(pgdat, zone, zone_start_pfn, size); - init_currently_empty_zone(zone, zone_start_pfn, size); - } + setup_usemap(pgdat, zone, zone_start_pfn, size); + init_currently_empty_zone(zone, zone_start_pfn, size); memmap_init(size, nid, j, zone_start_pfn); } } @@ -7951,7 +7928,7 @@ void free_contig_range(unsigned long pfn, unsigned nr_pages) } #endif -#if defined CONFIG_MEMORY_HOTPLUG || defined CONFIG_CMA +#ifdef CONFIG_MEMORY_HOTPLUG /* * The zone indicated has a new number of managed_pages; batch sizes and percpu * page high values need to be recalulated. From 009f8c90f571d87855914dbc20e6c0ea2a3b19ae Mon Sep 17 00:00:00 2001 From: Lukas Wunner Date: Thu, 24 May 2018 19:01:07 +0200 Subject: [PATCH 163/190] ALSA: hda - Fix runtime PM MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Before commit 3b5b899ca67d ("ALSA: hda: Make use of core codec functions to sync power state"), hda_set_power_state() returned the response to the Get Power State verb, a 32-bit unsigned integer whose expected value is 0x233 after transitioning a codec to D3, and 0x0 after transitioning it to D0. The response value is significant because hda_codec_runtime_suspend() does not clear the codec's bit in the codec_powered bitmask unless the AC_PWRST_CLK_STOP_OK bit (0x200) is set in the response value. That in turn prevents the HDA controller from runtime suspending because azx_runtime_idle() checks that the codec_powered bitmask is zero. Since commit 3b5b899ca67d, hda_set_power_state() only returns 0x0 or 0x1, thereby breaking runtime PM for any HDA controller. That's because an inline function introduced by the commit returns a bool instead of a 32-bit unsigned int. The change was likely erroneous and resulted from copying and pasting snd_hda_check_power_state(), which is immediately preceding the newly introduced inline function. Fix it. Link: https://bugs.freedesktop.org/show_bug.cgi?id=106597 Fixes: 3b5b899ca67d ("ALSA: hda: Make use of core codec functions to sync power state") Cc: Alex Deucher Cc: Abhijeet Kumar Reported-and-tested-by: Gunnar Krüger Signed-off-by: Lukas Wunner Acked-by: Alex Deucher Signed-off-by: Takashi Iwai --- sound/pci/hda/hda_local.h | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/sound/pci/hda/hda_local.h b/sound/pci/hda/hda_local.h index 321e78baa63c..9bd935216c18 100644 --- a/sound/pci/hda/hda_local.h +++ b/sound/pci/hda/hda_local.h @@ -622,8 +622,10 @@ snd_hda_check_power_state(struct hda_codec *codec, hda_nid_t nid, { return snd_hdac_check_power_state(&codec->core, nid, target_state); } -static inline bool snd_hda_sync_power_state(struct hda_codec *codec, - hda_nid_t nid, unsigned int target_state) + +static inline unsigned int snd_hda_sync_power_state(struct hda_codec *codec, + hda_nid_t nid, + unsigned int target_state) { return snd_hdac_sync_power_state(&codec->core, nid, target_state); } From 902a545904c71d719ed144234d67df75f31db63b Mon Sep 17 00:00:00 2001 From: Eran Ben Elisha Date: Tue, 1 May 2018 16:25:07 +0300 Subject: [PATCH 164/190] net/mlx5e: When RXFCS is set, add FCS data into checksum calculation When RXFCS feature is enabled, the HW do not strip the FCS data, however it is not present in the checksum calculated by the HW. Fix that by manually calculating the FCS checksum and adding it to the SKB checksum field. Add helper function to find the FCS data for all SKB forms (linear, one fragment or more). Fixes: 102722fc6832 ("net/mlx5e: Add support for RXFCS feature flag") Signed-off-by: Eran Ben Elisha Signed-off-by: Saeed Mahameed --- .../net/ethernet/mellanox/mlx5/core/en_rx.c | 42 +++++++++++++++++++ 1 file changed, 42 insertions(+) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c index 176645762e49..1ff0b0e93804 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c @@ -615,6 +615,45 @@ static inline bool is_last_ethertype_ip(struct sk_buff *skb, int *network_depth) return (ethertype == htons(ETH_P_IP) || ethertype == htons(ETH_P_IPV6)); } +static __be32 mlx5e_get_fcs(struct sk_buff *skb) +{ + int last_frag_sz, bytes_in_prev, nr_frags; + u8 *fcs_p1, *fcs_p2; + skb_frag_t *last_frag; + __be32 fcs_bytes; + + if (!skb_is_nonlinear(skb)) + return *(__be32 *)(skb->data + skb->len - ETH_FCS_LEN); + + nr_frags = skb_shinfo(skb)->nr_frags; + last_frag = &skb_shinfo(skb)->frags[nr_frags - 1]; + last_frag_sz = skb_frag_size(last_frag); + + /* If all FCS data is in last frag */ + if (last_frag_sz >= ETH_FCS_LEN) + return *(__be32 *)(skb_frag_address(last_frag) + + last_frag_sz - ETH_FCS_LEN); + + fcs_p2 = (u8 *)skb_frag_address(last_frag); + bytes_in_prev = ETH_FCS_LEN - last_frag_sz; + + /* Find where the other part of the FCS is - Linear or another frag */ + if (nr_frags == 1) { + fcs_p1 = skb_tail_pointer(skb); + } else { + skb_frag_t *prev_frag = &skb_shinfo(skb)->frags[nr_frags - 2]; + + fcs_p1 = skb_frag_address(prev_frag) + + skb_frag_size(prev_frag); + } + fcs_p1 -= bytes_in_prev; + + memcpy(&fcs_bytes, fcs_p1, bytes_in_prev); + memcpy(((u8 *)&fcs_bytes) + bytes_in_prev, fcs_p2, last_frag_sz); + + return fcs_bytes; +} + static inline void mlx5e_handle_csum(struct net_device *netdev, struct mlx5_cqe64 *cqe, struct mlx5e_rq *rq, @@ -643,6 +682,9 @@ static inline void mlx5e_handle_csum(struct net_device *netdev, skb->csum = csum_partial(skb->data + ETH_HLEN, network_depth - ETH_HLEN, skb->csum); + if (unlikely(netdev->features & NETIF_F_RXFCS)) + skb->csum = csum_add(skb->csum, + (__force __wsum)mlx5e_get_fcs(skb)); rq->stats.csum_complete++; return; } From 1dcbc01f73f9abc4779f71eae5e6dc61bee37229 Mon Sep 17 00:00:00 2001 From: Yossi Kuperman Date: Tue, 17 Oct 2017 20:39:17 +0300 Subject: [PATCH 165/190] net/mlx5: IPSec, Fix a race between concurrent sandbox QP commands Sandbox QP Commands are retired in the order they are sent. Outstanding commands are stored in a linked-list in the order they appear. Once a response is received and the callback gets called, we pull the first element off the pending list, assuming they correspond. Sending a message and adding it to the pending list is not done atomically, hence there is an opportunity for a race between concurrent requests. Bind both send and add under a critical section. Fixes: bebb23e6cb02 ("net/mlx5: Accel, Add IPSec acceleration interface") Signed-off-by: Yossi Kuperman Signed-off-by: Adi Nissim Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.c | 12 +++++------- 1 file changed, 5 insertions(+), 7 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.c b/drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.c index 0f5da499a223..fad8c2e3804e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.c @@ -237,19 +237,17 @@ static void *mlx5_fpga_ipsec_cmd_exec(struct mlx5_core_dev *mdev, context->buf.sg[0].data = &context->command; spin_lock_irqsave(&fdev->ipsec->pending_cmds_lock, flags); - list_add_tail(&context->list, &fdev->ipsec->pending_cmds); + res = mlx5_fpga_sbu_conn_sendmsg(fdev->ipsec->conn, &context->buf); + if (!res) + list_add_tail(&context->list, &fdev->ipsec->pending_cmds); spin_unlock_irqrestore(&fdev->ipsec->pending_cmds_lock, flags); - res = mlx5_fpga_sbu_conn_sendmsg(fdev->ipsec->conn, &context->buf); if (res) { - mlx5_fpga_warn(fdev, "Failure sending IPSec command: %d\n", - res); - spin_lock_irqsave(&fdev->ipsec->pending_cmds_lock, flags); - list_del(&context->list); - spin_unlock_irqrestore(&fdev->ipsec->pending_cmds_lock, flags); + mlx5_fpga_warn(fdev, "Failed to send IPSec command: %d\n", res); kfree(context); return ERR_PTR(res); } + /* Context will be freed by wait func after completion */ return context; } From 9aad13b087ab0a588cd68259de618f100053360e Mon Sep 17 00:00:00 2001 From: Willem de Bruijn Date: Thu, 24 May 2018 18:10:30 -0400 Subject: [PATCH 166/190] packet: fix reserve calculation Commit b84bbaf7a6c8 ("packet: in packet_snd start writing at link layer allocation") ensures that packet_snd always starts writing the link layer header in reserved headroom allocated for this purpose. This is needed because packets may be shorter than hard_header_len, in which case the space up to hard_header_len may be zeroed. But that necessary padding is not accounted for in skb->len. The fix, however, is buggy. It calls skb_push, which grows skb->len when moving skb->data back. But in this case packet length should not change. Instead, call skb_reserve, which moves both skb->data and skb->tail back, without changing length. Fixes: b84bbaf7a6c8 ("packet: in packet_snd start writing at link layer allocation") Reported-by: Tariq Toukan Signed-off-by: Willem de Bruijn Acked-by: Soheil Hassas Yeganeh Signed-off-by: David S. Miller --- net/packet/af_packet.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index e9422fe45179..acb7b86574cd 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -2911,7 +2911,7 @@ static int packet_snd(struct socket *sock, struct msghdr *msg, size_t len) if (unlikely(offset < 0)) goto out_free; } else if (reserve) { - skb_push(skb, reserve); + skb_reserve(skb, -reserve); } /* Returns -EFAULT on error */ From 1b15ad683ab42a203f98b67045b40720e99d0e9a Mon Sep 17 00:00:00 2001 From: Jason Wang Date: Tue, 22 May 2018 19:58:57 +0800 Subject: [PATCH 167/190] vhost: synchronize IOTLB message with dev cleanup DaeRyong Jeong reports a race between vhost_dev_cleanup() and vhost_process_iotlb_msg(): Thread interleaving: CPU0 (vhost_process_iotlb_msg) CPU1 (vhost_dev_cleanup) (In the case of both VHOST_IOTLB_UPDATE and VHOST_IOTLB_INVALIDATE) ===== ===== vhost_umem_clean(dev->iotlb); if (!dev->iotlb) { ret = -EFAULT; break; } dev->iotlb = NULL; The reason is we don't synchronize between them, fixing by protecting vhost_process_iotlb_msg() with dev mutex. Reported-by: DaeRyong Jeong Fixes: 6b1e6cc7855b0 ("vhost: new device IOTLB API") Signed-off-by: Jason Wang Acked-by: Michael S. Tsirkin Signed-off-by: David S. Miller --- drivers/vhost/vhost.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c index f3bd8e941224..f0be5f35ab28 100644 --- a/drivers/vhost/vhost.c +++ b/drivers/vhost/vhost.c @@ -981,6 +981,7 @@ static int vhost_process_iotlb_msg(struct vhost_dev *dev, { int ret = 0; + mutex_lock(&dev->mutex); vhost_dev_lock_vqs(dev); switch (msg->type) { case VHOST_IOTLB_UPDATE: @@ -1016,6 +1017,8 @@ static int vhost_process_iotlb_msg(struct vhost_dev *dev, } vhost_dev_unlock_vqs(dev); + mutex_unlock(&dev->mutex); + return ret; } ssize_t vhost_chr_write_iter(struct vhost_dev *dev, From f8f4bef322e4600c5856911c7a632c0e3da920d6 Mon Sep 17 00:00:00 2001 From: Or Gerlitz Date: Wed, 23 May 2018 19:24:48 +0300 Subject: [PATCH 168/190] net : sched: cls_api: deal with egdev path only if needed When dealing with ingress rule on a netdev, if we did fine through the conventional path, there's no need to continue into the egdev route, and we can stop right there. Not doing so may cause a 2nd rule to be added by the cls api layer with the ingress being the egdev. For example, under sriov switchdev scheme, a user rule of VFR A --> VFR B will end up with two HW rules (1) VF A --> VF B and (2) uplink --> VF B Fixes: 208c0f4b5237 ('net: sched: use tc_setup_cb_call to call per-block callbacks') Signed-off-by: Or Gerlitz Signed-off-by: David S. Miller --- net/sched/cls_api.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index 963e4bf0aab8..a57e112d9b3e 100644 --- a/net/sched/cls_api.c +++ b/net/sched/cls_api.c @@ -1588,7 +1588,7 @@ int tc_setup_cb_call(struct tcf_block *block, struct tcf_exts *exts, return ret; ok_count = ret; - if (!exts) + if (!exts || ok_count) return ok_count; ret = tc_exts_setup_cb_egdev_call(exts, type, type_data, err_stop); if (ret < 0) From 730c54d59403658a62af6517338fa8d4922c1b28 Mon Sep 17 00:00:00 2001 From: Willem de Bruijn Date: Wed, 23 May 2018 14:29:52 -0400 Subject: [PATCH 169/190] ipv4: remove warning in ip_recv_error A precondition check in ip_recv_error triggered on an otherwise benign race. Remove the warning. The warning triggers when passing an ipv6 socket to this ipv4 error handling function. RaceFuzzer was able to trigger it due to a race in setsockopt IPV6_ADDRFORM. --- CPU0 do_ipv6_setsockopt sk->sk_socket->ops = &inet_dgram_ops; --- CPU1 sk->sk_prot->recvmsg udp_recvmsg ip_recv_error WARN_ON_ONCE(sk->sk_family == AF_INET6); --- CPU0 do_ipv6_setsockopt sk->sk_family = PF_INET; This socket option converts a v6 socket that is connected to a v4 peer to an v4 socket. It updates the socket on the fly, changing fields in sk as well as other structs. This is inherently non-atomic. It races with the lockless udp_recvmsg path. No other code makes an assumption that these fields are updated atomically. It is benign here, too, as ip_recv_error cares only about the protocol of the skbs enqueued on the error queue, for which sk_family is not a precise predictor (thanks to another isue with IPV6_ADDRFORM). Link: http://lkml.kernel.org/r/20180518120826.GA19515@dragonet.kaist.ac.kr Fixes: 7ce875e5ecb8 ("ipv4: warn once on passing AF_INET6 socket to ip_recv_error") Reported-by: DaeRyong Jeong Suggested-by: Eric Dumazet Signed-off-by: Willem de Bruijn Signed-off-by: David S. Miller --- net/ipv4/ip_sockglue.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/net/ipv4/ip_sockglue.c b/net/ipv4/ip_sockglue.c index 5ad2d8ed3a3f..57bbb060faaf 100644 --- a/net/ipv4/ip_sockglue.c +++ b/net/ipv4/ip_sockglue.c @@ -505,8 +505,6 @@ int ip_recv_error(struct sock *sk, struct msghdr *msg, int len, int *addr_len) int err; int copied; - WARN_ON_ONCE(sk->sk_family == AF_INET6); - err = -EAGAIN; skb = sock_dequeue_err_skb(sk); if (!skb) From af8d3c7c001ae7df1ed2b2715f058113efc86187 Mon Sep 17 00:00:00 2001 From: Eric Biggers Date: Wed, 23 May 2018 14:37:38 -0700 Subject: [PATCH 170/190] ppp: remove the PPPIOCDETACH ioctl The PPPIOCDETACH ioctl effectively tries to "close" the given ppp file before f_count has reached 0, which is fundamentally a bad idea. It does check 'f_count < 2', which excludes concurrent operations on the file since they would only be possible with a shared fd table, in which case each fdget() would take a file reference. However, it fails to account for the fact that even with 'f_count == 1' the file can still be linked into epoll instances. As reported by syzbot, this can trivially be used to cause a use-after-free. Yet, the only known user of PPPIOCDETACH is pppd versions older than ppp-2.4.2, which was released almost 15 years ago (November 2003). Also, PPPIOCDETACH apparently stopped working reliably at around the same time, when the f_count check was added to the kernel, e.g. see https://lkml.org/lkml/2002/12/31/83. Also, the current 'f_count < 2' check makes PPPIOCDETACH only work in single-threaded applications; it always fails if called from a multithreaded application. All pppd versions released in the last 15 years just close() the file descriptor instead. Therefore, instead of hacking around this bug by exporting epoll internals to modules, and probably missing other related bugs, just remove the PPPIOCDETACH ioctl and see if anyone actually notices. Leave a stub in place that prints a one-time warning and returns EINVAL. Reported-by: syzbot+16363c99d4134717c05b@syzkaller.appspotmail.com Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Eric Biggers Acked-by: Paul Mackerras Reviewed-by: Guillaume Nault Tested-by: Guillaume Nault Signed-off-by: David S. Miller --- Documentation/networking/ppp_generic.txt | 6 ------ drivers/net/ppp/ppp_generic.c | 27 +++++------------------- include/uapi/linux/ppp-ioctl.h | 2 +- 3 files changed, 6 insertions(+), 29 deletions(-) diff --git a/Documentation/networking/ppp_generic.txt b/Documentation/networking/ppp_generic.txt index 091d20273dcb..61daf4b39600 100644 --- a/Documentation/networking/ppp_generic.txt +++ b/Documentation/networking/ppp_generic.txt @@ -300,12 +300,6 @@ unattached instance are: The ioctl calls available on an instance of /dev/ppp attached to a channel are: -* PPPIOCDETACH detaches the instance from the channel. This ioctl is - deprecated since the same effect can be achieved by closing the - instance. In order to prevent possible races this ioctl will fail - with an EINVAL error if more than one file descriptor refers to this - instance (i.e. as a result of dup(), dup2() or fork()). - * PPPIOCCONNECT connects this channel to a PPP interface. The argument should point to an int containing the interface unit number. It will return an EINVAL error if the channel is already diff --git a/drivers/net/ppp/ppp_generic.c b/drivers/net/ppp/ppp_generic.c index dc7c7ec43202..02ad03a2fab7 100644 --- a/drivers/net/ppp/ppp_generic.c +++ b/drivers/net/ppp/ppp_generic.c @@ -605,30 +605,13 @@ static long ppp_ioctl(struct file *file, unsigned int cmd, unsigned long arg) if (cmd == PPPIOCDETACH) { /* - * We have to be careful here... if the file descriptor - * has been dup'd, we could have another process in the - * middle of a poll using the same file *, so we had - * better not free the interface data structures - - * instead we fail the ioctl. Even in this case, we - * shut down the interface if we are the owner of it. - * Actually, we should get rid of PPPIOCDETACH, userland - * (i.e. pppd) could achieve the same effect by closing - * this fd and reopening /dev/ppp. + * PPPIOCDETACH is no longer supported as it was heavily broken, + * and is only known to have been used by pppd older than + * ppp-2.4.2 (released November 2003). */ + pr_warn_once("%s (%d) used obsolete PPPIOCDETACH ioctl\n", + current->comm, current->pid); err = -EINVAL; - if (pf->kind == INTERFACE) { - ppp = PF_TO_PPP(pf); - rtnl_lock(); - if (file == ppp->owner) - unregister_netdevice(ppp->dev); - rtnl_unlock(); - } - if (atomic_long_read(&file->f_count) < 2) { - ppp_release(NULL, file); - err = 0; - } else - pr_warn("PPPIOCDETACH file->f_count=%ld\n", - atomic_long_read(&file->f_count)); goto out; } diff --git a/include/uapi/linux/ppp-ioctl.h b/include/uapi/linux/ppp-ioctl.h index b19a9c249b15..784c2e3e572e 100644 --- a/include/uapi/linux/ppp-ioctl.h +++ b/include/uapi/linux/ppp-ioctl.h @@ -106,7 +106,7 @@ struct pppol2tp_ioc_stats { #define PPPIOCGIDLE _IOR('t', 63, struct ppp_idle) /* get idle time */ #define PPPIOCNEWUNIT _IOWR('t', 62, int) /* create new ppp unit */ #define PPPIOCATTACH _IOW('t', 61, int) /* attach to ppp unit */ -#define PPPIOCDETACH _IOW('t', 60, int) /* detach from ppp unit/chan */ +#define PPPIOCDETACH _IOW('t', 60, int) /* obsolete, do not use */ #define PPPIOCSMRRU _IOW('t', 59, int) /* set multilink MRU */ #define PPPIOCCONNECT _IOW('t', 58, int) /* connect channel to unit */ #define PPPIOCDISCONN _IO('t', 57) /* disconnect channel */ From 322eaa06d55ebc1402a4a8d140945cff536638b4 Mon Sep 17 00:00:00 2001 From: Govindarajulu Varadarajan Date: Wed, 23 May 2018 11:17:39 -0700 Subject: [PATCH 171/190] enic: set DMA mask to 47 bit In commit 624dbf55a359b ("driver/net: enic: Try DMA 64 first, then failover to DMA") DMA mask was changed from 40 bits to 64 bits. Hardware actually supports only 47 bits. Fixes: 624dbf55a359b ("driver/net: enic: Try DMA 64 first, then failover to DMA") Signed-off-by: Govindarajulu Varadarajan Signed-off-by: David S. Miller --- drivers/net/ethernet/cisco/enic/enic_main.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/cisco/enic/enic_main.c b/drivers/net/ethernet/cisco/enic/enic_main.c index 81684acf52af..8a8b12b720ef 100644 --- a/drivers/net/ethernet/cisco/enic/enic_main.c +++ b/drivers/net/ethernet/cisco/enic/enic_main.c @@ -2747,11 +2747,11 @@ static int enic_probe(struct pci_dev *pdev, const struct pci_device_id *ent) pci_set_master(pdev); /* Query PCI controller on system for DMA addressing - * limitation for the device. Try 64-bit first, and + * limitation for the device. Try 47-bit first, and * fail to 32-bit. */ - err = pci_set_dma_mask(pdev, DMA_BIT_MASK(64)); + err = pci_set_dma_mask(pdev, DMA_BIT_MASK(47)); if (err) { err = pci_set_dma_mask(pdev, DMA_BIT_MASK(32)); if (err) { @@ -2765,10 +2765,10 @@ static int enic_probe(struct pci_dev *pdev, const struct pci_device_id *ent) goto err_out_release_regions; } } else { - err = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(64)); + err = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(47)); if (err) { dev_err(dev, "Unable to obtain %u-bit DMA " - "for consistent allocations, aborting\n", 64); + "for consistent allocations, aborting\n", 47); goto err_out_release_regions; } using_dac = 1; From 1383cb8103bb166e50cbab1543bb3b5118fccf82 Mon Sep 17 00:00:00 2001 From: Qing Huang Date: Wed, 23 May 2018 16:22:46 -0700 Subject: [PATCH 172/190] mlx4_core: allocate ICM memory in page size chunks When a system is under memory presure (high usage with fragments), the original 256KB ICM chunk allocations will likely trigger kernel memory management to enter slow path doing memory compact/migration ops in order to complete high order memory allocations. When that happens, user processes calling uverb APIs may get stuck for more than 120s easily even though there are a lot of free pages in smaller chunks available in the system. Syslog: ... Dec 10 09:04:51 slcc03db02 kernel: [397078.572732] INFO: task oracle_205573_e:205573 blocked for more than 120 seconds. ... With 4KB ICM chunk size on x86_64 arch, the above issue is fixed. However in order to support smaller ICM chunk size, we need to fix another issue in large size kcalloc allocations. E.g. Setting log_num_mtt=30 requires 1G mtt entries. With the 4KB ICM chunk size, each ICM chunk can only hold 512 mtt entries (8 bytes for each mtt entry). So we need a 16MB allocation for a table->icm pointer array to hold 2M pointers which can easily cause kcalloc to fail. The solution is to use kvzalloc to replace kcalloc which will fall back to vmalloc automatically if kmalloc fails. Signed-off-by: Qing Huang Acked-by: Daniel Jurgens Reviewed-by: Zhu Yanjun Reviewed-by: Tariq Toukan Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx4/icm.c | 16 +++++++++------- 1 file changed, 9 insertions(+), 7 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx4/icm.c b/drivers/net/ethernet/mellanox/mlx4/icm.c index a822f7a56bc5..685337d58276 100644 --- a/drivers/net/ethernet/mellanox/mlx4/icm.c +++ b/drivers/net/ethernet/mellanox/mlx4/icm.c @@ -43,12 +43,12 @@ #include "fw.h" /* - * We allocate in as big chunks as we can, up to a maximum of 256 KB - * per chunk. + * We allocate in page size (default 4KB on many archs) chunks to avoid high + * order memory allocations in fragmented/high usage memory situation. */ enum { - MLX4_ICM_ALLOC_SIZE = 1 << 18, - MLX4_TABLE_CHUNK_SIZE = 1 << 18 + MLX4_ICM_ALLOC_SIZE = PAGE_SIZE, + MLX4_TABLE_CHUNK_SIZE = PAGE_SIZE, }; static void mlx4_free_icm_pages(struct mlx4_dev *dev, struct mlx4_icm_chunk *chunk) @@ -398,9 +398,11 @@ int mlx4_init_icm_table(struct mlx4_dev *dev, struct mlx4_icm_table *table, u64 size; obj_per_chunk = MLX4_TABLE_CHUNK_SIZE / obj_size; + if (WARN_ON(!obj_per_chunk)) + return -EINVAL; num_icm = (nobj + obj_per_chunk - 1) / obj_per_chunk; - table->icm = kcalloc(num_icm, sizeof(*table->icm), GFP_KERNEL); + table->icm = kvzalloc(num_icm * sizeof(*table->icm), GFP_KERNEL); if (!table->icm) return -ENOMEM; table->virt = virt; @@ -446,7 +448,7 @@ int mlx4_init_icm_table(struct mlx4_dev *dev, struct mlx4_icm_table *table, mlx4_free_icm(dev, table->icm[i], use_coherent); } - kfree(table->icm); + kvfree(table->icm); return -ENOMEM; } @@ -462,5 +464,5 @@ void mlx4_cleanup_icm_table(struct mlx4_dev *dev, struct mlx4_icm_table *table) mlx4_free_icm(dev, table->icm[i], table->coherent); } - kfree(table->icm); + kvfree(table->icm); } From 24e4b075d899e5376dfa39fecd1dbc12bddc1e98 Mon Sep 17 00:00:00 2001 From: Stefano Brivio Date: Thu, 24 May 2018 16:10:12 +0200 Subject: [PATCH 173/190] selftests/net: Add missing config options for PMTU tests PMTU tests in pmtu.sh need support for VTI, VTI6 and dummy interfaces: add them to config file. Reported-by: Naresh Kamboju Fixes: d1f1b9cbf34c ("selftests: net: Introduce first PMTU test") Signed-off-by: Stefano Brivio Signed-off-by: David S. Miller --- tools/testing/selftests/net/config | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/tools/testing/selftests/net/config b/tools/testing/selftests/net/config index 6a75a3ea44ad..7ba089b33e8b 100644 --- a/tools/testing/selftests/net/config +++ b/tools/testing/selftests/net/config @@ -7,3 +7,8 @@ CONFIG_NET_L3_MASTER_DEV=y CONFIG_IPV6=y CONFIG_IPV6_MULTIPLE_TABLES=y CONFIG_VETH=y +CONFIG_INET_XFRM_MODE_TUNNEL=y +CONFIG_NET_IPVTI=y +CONFIG_INET6_XFRM_MODE_TUNNEL=y +CONFIG_IPV6_VTI=y +CONFIG_DUMMY=y From eb110410b9f6477726026669f3f0c0567e8241e6 Mon Sep 17 00:00:00 2001 From: Thomas Falcon Date: Thu, 24 May 2018 14:37:53 -0500 Subject: [PATCH 174/190] ibmvnic: Fix partial success login retries In its current state, the driver will handle backing device login in a loop for a certain number of retries while the device returns a partial success, indicating that the driver may need to try again using a smaller number of resources. The variable it checks to continue retrying may change over the course of operations, resulting in reallocation of resources but exits without sending the login attempt. Guard against this by introducing a boolean variable that will retain the state indicating that the driver needs to reattempt login with backing device firmware. Signed-off-by: Thomas Falcon Signed-off-by: David S. Miller --- drivers/net/ethernet/ibm/ibmvnic.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/ibm/ibmvnic.c b/drivers/net/ethernet/ibm/ibmvnic.c index 62cd3602c633..5ec1185808e5 100644 --- a/drivers/net/ethernet/ibm/ibmvnic.c +++ b/drivers/net/ethernet/ibm/ibmvnic.c @@ -796,9 +796,11 @@ static int ibmvnic_login(struct net_device *netdev) struct ibmvnic_adapter *adapter = netdev_priv(netdev); unsigned long timeout = msecs_to_jiffies(30000); int retry_count = 0; + bool retry; int rc; do { + retry = false; if (retry_count > IBMVNIC_MAX_QUEUES) { netdev_warn(netdev, "Login attempts exceeded\n"); return -1; @@ -822,6 +824,9 @@ static int ibmvnic_login(struct net_device *netdev) retry_count++; release_sub_crqs(adapter, 1); + retry = true; + netdev_dbg(netdev, + "Received partial success, retrying...\n"); adapter->init_done_rc = 0; reinit_completion(&adapter->init_done); send_cap_queries(adapter); @@ -849,7 +854,7 @@ static int ibmvnic_login(struct net_device *netdev) netdev_warn(netdev, "Adapter login failed\n"); return -1; } - } while (adapter->init_done_rc == PARTIALSUCCESS); + } while (retry); /* handle pending MAC address changes after successful login */ if (adapter->mac_change_pending) { From 7cbf319234adaa4518a28c8c523d3330e06638f0 Mon Sep 17 00:00:00 2001 From: Omar Sandoval Date: Fri, 25 May 2018 14:47:17 -0700 Subject: [PATCH 175/190] mm: fix nr_rotate_swap leak in swapon() error case If swapon() fails after incrementing nr_rotate_swap, we don't decrement it and thus effectively leak it. Make sure we decrement it if we incremented it. Link: http://lkml.kernel.org/r/b6fe6b879f17fa68eee6cbd876f459f6e5e33495.1526491581.git.osandov@fb.com Fixes: 81a0298bdfab ("mm, swap: don't use VMA based swap readahead if HDD is used as swap") Signed-off-by: Omar Sandoval Reviewed-by: Rik van Riel Reviewed-by: "Huang, Ying" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/swapfile.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/mm/swapfile.c b/mm/swapfile.c index cc2cf04d9018..78a015fcec3b 100644 --- a/mm/swapfile.c +++ b/mm/swapfile.c @@ -3112,6 +3112,7 @@ SYSCALL_DEFINE2(swapon, const char __user *, specialfile, int, swap_flags) unsigned long *frontswap_map = NULL; struct page *page = NULL; struct inode *inode = NULL; + bool inced_nr_rotate_swap = false; if (swap_flags & ~SWAP_FLAGS_VALID) return -EINVAL; @@ -3215,8 +3216,10 @@ SYSCALL_DEFINE2(swapon, const char __user *, specialfile, int, swap_flags) cluster = per_cpu_ptr(p->percpu_cluster, cpu); cluster_set_null(&cluster->index); } - } else + } else { atomic_inc(&nr_rotate_swap); + inced_nr_rotate_swap = true; + } error = swap_cgroup_swapon(p->type, maxpages); if (error) @@ -3307,6 +3310,8 @@ SYSCALL_DEFINE2(swapon, const char __user *, specialfile, int, swap_flags) vfree(swap_map); kvfree(cluster_info); kvfree(frontswap_map); + if (inced_nr_rotate_swap) + atomic_dec(&nr_rotate_swap); if (swap_file) { if (inode && S_ISREG(inode->i_mode)) { inode_unlock(inode); From 3373de209cb123462954740f41c324d03ecfb6d0 Mon Sep 17 00:00:00 2001 From: Changwei Ge Date: Fri, 25 May 2018 14:47:20 -0700 Subject: [PATCH 176/190] ocfs2: revert "ocfs2/o2hb: check len for bio_add_page() to avoid getting incorrect bio" This reverts commit ba16ddfbeb9d ("ocfs2/o2hb: check len for bio_add_page() to avoid getting incorrect bio"). In my testing, this patch introduces a problem that mkfs can't have slots more than 16 with 4k block size. And the original logic is safe actually with the situation it mentions so revert this commit. Attach test log: (mkfs.ocfs2,27479,2):o2hb_setup_one_bio:463 page 0, vec_len = 4096, vec_start = 0 (mkfs.ocfs2,27479,2):o2hb_setup_one_bio:463 page 1, vec_len = 4096, vec_start = 0 (mkfs.ocfs2,27479,2):o2hb_setup_one_bio:463 page 2, vec_len = 4096, vec_start = 0 (mkfs.ocfs2,27479,2):o2hb_setup_one_bio:463 page 3, vec_len = 4096, vec_start = 0 (mkfs.ocfs2,27479,2):o2hb_setup_one_bio:463 page 4, vec_len = 4096, vec_start = 0 (mkfs.ocfs2,27479,2):o2hb_setup_one_bio:463 page 5, vec_len = 4096, vec_start = 0 (mkfs.ocfs2,27479,2):o2hb_setup_one_bio:463 page 6, vec_len = 4096, vec_start = 0 (mkfs.ocfs2,27479,2):o2hb_setup_one_bio:463 page 7, vec_len = 4096, vec_start = 0 (mkfs.ocfs2,27479,2):o2hb_setup_one_bio:463 page 8, vec_len = 4096, vec_start = 0 (mkfs.ocfs2,27479,2):o2hb_setup_one_bio:463 page 9, vec_len = 4096, vec_start = 0 (mkfs.ocfs2,27479,2):o2hb_setup_one_bio:463 page 10, vec_len = 4096, vec_start = 0 (mkfs.ocfs2,27479,2):o2hb_setup_one_bio:463 page 11, vec_len = 4096, vec_start = 0 (mkfs.ocfs2,27479,2):o2hb_setup_one_bio:463 page 12, vec_len = 4096, vec_start = 0 (mkfs.ocfs2,27479,2):o2hb_setup_one_bio:463 page 13, vec_len = 4096, vec_start = 0 (mkfs.ocfs2,27479,2):o2hb_setup_one_bio:463 page 14, vec_len = 4096, vec_start = 0 (mkfs.ocfs2,27479,2):o2hb_setup_one_bio:463 page 15, vec_len = 4096, vec_start = 0 (mkfs.ocfs2,27479,2):o2hb_setup_one_bio:463 page 16, vec_len = 4096, vec_start = 0 (mkfs.ocfs2,27479,2):o2hb_setup_one_bio:471 ERROR: Adding page[16] to bio failed, page ffffea0002d7ed40, len 0, vec_len 4096, vec_start 0,bi_sector 8192 (mkfs.ocfs2,27479,2):o2hb_read_slots:500 ERROR: status = -5 (mkfs.ocfs2,27479,2):o2hb_populate_slot_data:1911 ERROR: status = -5 (mkfs.ocfs2,27479,2):o2hb_region_dev_write:2012 ERROR: status = -5 Link: http://lkml.kernel.org/r/SIXPR06MB0461721F398A5A92FC68C39ED5920@SIXPR06MB0461.apcprd06.prod.outlook.com Signed-off-by: Changwei Ge Cc: Jun Piao Cc: Yiwen Jiang Cc: Joseph Qi Cc: Mark Fasheh Cc: Joel Becker Cc: Junxiao Bi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ocfs2/cluster/heartbeat.c | 11 +---------- 1 file changed, 1 insertion(+), 10 deletions(-) diff --git a/fs/ocfs2/cluster/heartbeat.c b/fs/ocfs2/cluster/heartbeat.c index 91a8889abf9b..ea8c551bcd7e 100644 --- a/fs/ocfs2/cluster/heartbeat.c +++ b/fs/ocfs2/cluster/heartbeat.c @@ -570,16 +570,7 @@ static struct bio *o2hb_setup_one_bio(struct o2hb_region *reg, current_page, vec_len, vec_start); len = bio_add_page(bio, page, vec_len, vec_start); - if (len != vec_len) { - mlog(ML_ERROR, "Adding page[%d] to bio failed, " - "page %p, len %d, vec_len %u, vec_start %u, " - "bi_sector %llu\n", current_page, page, len, - vec_len, vec_start, - (unsigned long long)bio->bi_iter.bi_sector); - bio_put(bio); - bio = ERR_PTR(-EIO); - return bio; - } + if (len != vec_len) break; cs += vec_len / (PAGE_SIZE/spp); vec_start = 0; From 7a4deea1aa8bddfed4ef1b35fc2b6732563d8ad5 Mon Sep 17 00:00:00 2001 From: Matthew Wilcox Date: Fri, 25 May 2018 14:47:24 -0700 Subject: [PATCH 177/190] idr: fix invalid ptr dereference on item delete If the radix tree underlying the IDR happens to be full and we attempt to remove an id which is larger than any id in the IDR, we will call __radix_tree_delete() with an uninitialised 'slot' pointer, at which point anything could happen. This was easiest to hit with a single entry at id 0 and attempting to remove a non-0 id, but it could have happened with 64 entries and attempting to remove an id >= 64. Roman said: The syzcaller test boils down to opening /dev/kvm, creating an eventfd, and calling a couple of KVM ioctls. None of this requires superuser. And the result is dereferencing an uninitialized pointer which is likely a crash. The specific path caught by syzbot is via KVM_HYPERV_EVENTD ioctl which is new in 4.17. But I guess there are other user-triggerable paths, so cc:stable is probably justified. Matthew added: We have around 250 calls to idr_remove() in the kernel today. Many of them pass an ID which is embedded in the object they're removing, so they're safe. Picking a few likely candidates: drivers/firewire/core-cdev.c looks unsafe; the ID comes from an ioctl. drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c is similar drivers/atm/nicstar.c could be taken down by a handcrafted packet Link: http://lkml.kernel.org/r/20180518175025.GD6361@bombadil.infradead.org Fixes: 0a835c4f090a ("Reimplement IDR and IDA using the radix tree") Reported-by: Debugged-by: Roman Kagan Signed-off-by: Matthew Wilcox Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- lib/radix-tree.c | 4 +++- tools/testing/radix-tree/idr-test.c | 7 +++++++ 2 files changed, 10 insertions(+), 1 deletion(-) diff --git a/lib/radix-tree.c b/lib/radix-tree.c index 43e0cbedc3a0..a9e41aed6de4 100644 --- a/lib/radix-tree.c +++ b/lib/radix-tree.c @@ -2034,10 +2034,12 @@ void *radix_tree_delete_item(struct radix_tree_root *root, unsigned long index, void *item) { struct radix_tree_node *node = NULL; - void __rcu **slot; + void __rcu **slot = NULL; void *entry; entry = __radix_tree_lookup(root, index, &node, &slot); + if (!slot) + return NULL; if (!entry && (!is_idr(root) || node_tag_get(root, node, IDR_FREE, get_slot_offset(node, slot)))) return NULL; diff --git a/tools/testing/radix-tree/idr-test.c b/tools/testing/radix-tree/idr-test.c index 6c645eb77d42..ee820fcc29b0 100644 --- a/tools/testing/radix-tree/idr-test.c +++ b/tools/testing/radix-tree/idr-test.c @@ -252,6 +252,13 @@ void idr_checks(void) idr_remove(&idr, 3); idr_remove(&idr, 0); + assert(idr_alloc(&idr, DUMMY_PTR, 0, 0, GFP_KERNEL) == 0); + idr_remove(&idr, 1); + for (i = 1; i < RADIX_TREE_MAP_SIZE; i++) + assert(idr_alloc(&idr, DUMMY_PTR, 0, 0, GFP_KERNEL) == i); + idr_remove(&idr, 1 << 30); + idr_destroy(&idr); + for (i = INT_MAX - 3UL; i < INT_MAX + 1UL; i++) { struct item *item = item_create(i, 0); assert(idr_alloc(&idr, item, i, i + 10, GFP_KERNEL) == i); From a73ab244f0dad8fffb3291b905f73e2d3eaa7c00 Mon Sep 17 00:00:00 2001 From: Davidlohr Bueso Date: Fri, 25 May 2018 14:47:27 -0700 Subject: [PATCH 178/190] Revert "ipc/shm: Fix shmat mmap nil-page protection" Patch series "ipc/shm: shmat() fixes around nil-page". These patches fix two issues reported[1] a while back by Joe and Andrea around how shmat(2) behaves with nil-page. The first reverts a commit that it was incorrectly thought that mapping nil-page (address=0) was a no no with MAP_FIXED. This is not the case, with the exception of SHM_REMAP; which is address in the second patch. I chose two patches because it is easier to backport and it explicitly reverts bogus behaviour. Both patches ought to be in -stable and ltp testcases need updated (the added testcase around the cve can be modified to just test for SHM_RND|SHM_REMAP). [1] lkml.kernel.org/r/20180430172152.nfa564pvgpk3ut7p@linux-n805 This patch (of 2): Commit 95e91b831f87 ("ipc/shm: Fix shmat mmap nil-page protection") worked on the idea that we should not be mapping as root addr=0 and MAP_FIXED. However, it was reported that this scenario is in fact valid, thus making the patch both bogus and breaks userspace as well. For example X11's libint10.so relies on shmat(1, SHM_RND) for lowmem initialization[1]. [1] https://cgit.freedesktop.org/xorg/xserver/tree/hw/xfree86/os-support/linux/int10/linux.c#n347 Link: http://lkml.kernel.org/r/20180503203243.15045-2-dave@stgolabs.net Fixes: 95e91b831f87 ("ipc/shm: Fix shmat mmap nil-page protection") Signed-off-by: Davidlohr Bueso Reported-by: Joe Lawrence Reported-by: Andrea Arcangeli Cc: Manfred Spraul Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- ipc/shm.c | 9 ++------- 1 file changed, 2 insertions(+), 7 deletions(-) diff --git a/ipc/shm.c b/ipc/shm.c index 3cf48988d68c..930be3aa80cf 100644 --- a/ipc/shm.c +++ b/ipc/shm.c @@ -1363,13 +1363,8 @@ long do_shmat(int shmid, char __user *shmaddr, int shmflg, if (addr) { if (addr & (shmlba - 1)) { - /* - * Round down to the nearest multiple of shmlba. - * For sane do_mmap_pgoff() parameters, avoid - * round downs that trigger nil-page and MAP_FIXED. - */ - if ((shmflg & SHM_RND) && addr >= shmlba) - addr &= ~(shmlba - 1); + if (shmflg & SHM_RND) + addr &= ~(shmlba - 1); /* round down */ else #ifndef __ARCH_FORCE_SHMLBA if (addr & ~PAGE_MASK) From 8f89c007b6dec16a1793cb88de88fcc02117bbbc Mon Sep 17 00:00:00 2001 From: Davidlohr Bueso Date: Fri, 25 May 2018 14:47:30 -0700 Subject: [PATCH 179/190] ipc/shm: fix shmat() nil address after round-down when remapping shmat()'s SHM_REMAP option forbids passing a nil address for; this is in fact the very first thing we check for. Andrea reported that for SHM_RND|SHM_REMAP cases we can end up bypassing the initial addr check, but we need to check again if the address was rounded down to nil. As of this patch, such cases will return -EINVAL. Link: http://lkml.kernel.org/r/20180503204934.kk63josdu6u53fbd@linux-n805 Signed-off-by: Davidlohr Bueso Reported-by: Andrea Arcangeli Cc: Joe Lawrence Cc: Manfred Spraul Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- ipc/shm.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) diff --git a/ipc/shm.c b/ipc/shm.c index 930be3aa80cf..d73269381ec7 100644 --- a/ipc/shm.c +++ b/ipc/shm.c @@ -1363,9 +1363,17 @@ long do_shmat(int shmid, char __user *shmaddr, int shmflg, if (addr) { if (addr & (shmlba - 1)) { - if (shmflg & SHM_RND) + if (shmflg & SHM_RND) { addr &= ~(shmlba - 1); /* round down */ - else + + /* + * Ensure that the round-down is non-nil + * when remapping. This can happen for + * cases when addr < shmlba. + */ + if (!addr && (shmflg & SHM_REMAP)) + goto out; + } else #ifndef __ARCH_FORCE_SHMLBA if (addr & ~PAGE_MASK) #endif From b9ddff9b85036292f8b6d4ac37e21fc229fedea1 Mon Sep 17 00:00:00 2001 From: Mike Kravetz Date: Fri, 25 May 2018 14:47:35 -0700 Subject: [PATCH 180/190] MAINTAINERS: change hugetlbfs maintainer and update files The current hugetlbfs maintainer has not been active for more than a few years. I have been been active in this area for more than two years and plan to remain active in the foreseeable future. Also, update the hugetlbfs entry to include linux-mm mail list and additional hugetlbfs related files. hugetlb.c and hugetlb.h are not 100% hugetlbfs, but a majority of their content is hugetlbfs related. Link: http://lkml.kernel.org/r/20180518225236.19079-1-mike.kravetz@oracle.com Signed-off-by: Mike Kravetz Acked-by: Naoya Horiguchi Acked-by: Kirill A. Shutemov Acked-by: Vlastimil Babka Acked-by: Michal Hocko Cc: Nadia Yvette Chambers Cc: "Aneesh Kumar K . V" Cc: Jan Kara Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- MAINTAINERS | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index a9ca122957e9..89ead18184a6 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -6503,9 +6503,15 @@ F: Documentation/networking/hinic.txt F: drivers/net/ethernet/huawei/hinic/ HUGETLB FILESYSTEM -M: Nadia Yvette Chambers +M: Mike Kravetz +L: linux-mm@kvack.org S: Maintained F: fs/hugetlbfs/ +F: mm/hugetlb.c +F: include/linux/hugetlb.h +F: Documentation/admin-guide/mm/hugetlbpage.rst +F: Documentation/vm/hugetlbfs_reserv.rst +F: Documentation/ABI/testing/sysfs-kernel-mm-hugepages HVA ST MEDIA DRIVER M: Jean-Christophe Trotin From 0f901dcbc31f88ae41a2aaa365f7802b5d520a28 Mon Sep 17 00:00:00 2001 From: Andrey Ryabinin Date: Fri, 25 May 2018 14:47:38 -0700 Subject: [PATCH 181/190] mm/kasan: don't vfree() nonexistent vm_area KASAN uses different routines to map shadow for hot added memory and memory obtained in boot process. Attempt to offline memory onlined by normal boot process leads to this: Trying to vfree() nonexistent vm area (000000005d3b34b9) WARNING: CPU: 2 PID: 13215 at mm/vmalloc.c:1525 __vunmap+0x147/0x190 Call Trace: kasan_mem_notifier+0xad/0xb9 notifier_call_chain+0x166/0x260 __blocking_notifier_call_chain+0xdb/0x140 __offline_pages+0x96a/0xb10 memory_subsys_offline+0x76/0xc0 device_offline+0xb8/0x120 store_mem_state+0xfa/0x120 kernfs_fop_write+0x1d5/0x320 __vfs_write+0xd4/0x530 vfs_write+0x105/0x340 SyS_write+0xb0/0x140 Obviously we can't call vfree() to free memory that wasn't allocated via vmalloc(). Use find_vm_area() to see if we can call vfree(). Unfortunately it's a bit tricky to properly unmap and free shadow allocated during boot, so we'll have to keep it. If memory will come online again that shadow will be reused. Matthew asked: how can you call vfree() on something that isn't a vmalloc address? vfree() is able to free any address returned by __vmalloc_node_range(). And __vmalloc_node_range() gives you any address you ask. It doesn't have to be an address in [VMALLOC_START, VMALLOC_END] range. That's also how the module_alloc()/module_memfree() works on architectures that have designated area for modules. [aryabinin@virtuozzo.com: improve comments] Link: http://lkml.kernel.org/r/dabee6ab-3a7a-51cd-3b86-5468718e0390@virtuozzo.com [akpm@linux-foundation.org: fix typos, reflow comment] Link: http://lkml.kernel.org/r/20180201163349.8700-1-aryabinin@virtuozzo.com Fixes: fa69b5989bb0 ("mm/kasan: add support for memory hotplug") Signed-off-by: Andrey Ryabinin Reported-by: Paul Menzel Cc: Alexander Potapenko Cc: Dmitry Vyukov Cc: Matthew Wilcox Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/kasan/kasan.c | 63 ++++++++++++++++++++++++++++++++++++++++++++++-- 1 file changed, 61 insertions(+), 2 deletions(-) diff --git a/mm/kasan/kasan.c b/mm/kasan/kasan.c index bc0e68f7dc75..7160028efd12 100644 --- a/mm/kasan/kasan.c +++ b/mm/kasan/kasan.c @@ -792,6 +792,40 @@ DEFINE_ASAN_SET_SHADOW(f5); DEFINE_ASAN_SET_SHADOW(f8); #ifdef CONFIG_MEMORY_HOTPLUG +static bool shadow_mapped(unsigned long addr) +{ + pgd_t *pgd = pgd_offset_k(addr); + p4d_t *p4d; + pud_t *pud; + pmd_t *pmd; + pte_t *pte; + + if (pgd_none(*pgd)) + return false; + p4d = p4d_offset(pgd, addr); + if (p4d_none(*p4d)) + return false; + pud = pud_offset(p4d, addr); + if (pud_none(*pud)) + return false; + + /* + * We can't use pud_large() or pud_huge(), the first one is + * arch-specific, the last one depends on HUGETLB_PAGE. So let's abuse + * pud_bad(), if pud is bad then it's bad because it's huge. + */ + if (pud_bad(*pud)) + return true; + pmd = pmd_offset(pud, addr); + if (pmd_none(*pmd)) + return false; + + if (pmd_bad(*pmd)) + return true; + pte = pte_offset_kernel(pmd, addr); + return !pte_none(*pte); +} + static int __meminit kasan_mem_notifier(struct notifier_block *nb, unsigned long action, void *data) { @@ -813,6 +847,14 @@ static int __meminit kasan_mem_notifier(struct notifier_block *nb, case MEM_GOING_ONLINE: { void *ret; + /* + * If shadow is mapped already than it must have been mapped + * during the boot. This could happen if we onlining previously + * offlined memory. + */ + if (shadow_mapped(shadow_start)) + return NOTIFY_OK; + ret = __vmalloc_node_range(shadow_size, PAGE_SIZE, shadow_start, shadow_end, GFP_KERNEL, PAGE_KERNEL, VM_NO_GUARD, @@ -824,8 +866,25 @@ static int __meminit kasan_mem_notifier(struct notifier_block *nb, kmemleak_ignore(ret); return NOTIFY_OK; } - case MEM_OFFLINE: - vfree((void *)shadow_start); + case MEM_OFFLINE: { + struct vm_struct *vm; + + /* + * shadow_start was either mapped during boot by kasan_init() + * or during memory online by __vmalloc_node_range(). + * In the latter case we can use vfree() to free shadow. + * Non-NULL result of the find_vm_area() will tell us if + * that was the second case. + * + * Currently it's not possible to free shadow mapped + * during boot by kasan_init(). It's because the code + * to do that hasn't been written yet. So we'll just + * leak the memory. + */ + vm = find_vm_area((void *)shadow_start); + if (vm) + vfree((void *)shadow_start); + } } return NOTIFY_OK; From 15c30bc0908514b4a72fadc3995d32313500393a Mon Sep 17 00:00:00 2001 From: Michal Hocko Date: Fri, 25 May 2018 14:47:42 -0700 Subject: [PATCH 182/190] mm, memory_hotplug: make has_unmovable_pages more robust Oscar has reported: : Due to an unfortunate setting with movablecore, memblocks containing bootmem : memory (pages marked by get_page_bootmem()) ended up marked in zone_movable. : So while trying to remove that memory, the system failed in do_migrate_range : and __offline_pages never returned. : : This can be reproduced by running : qemu-system-x86_64 -m 6G,slots=8,maxmem=8G -numa node,mem=4096M -numa node,mem=2048M : and movablecore=4G kernel command line : : linux kernel: BIOS-provided physical RAM map: : linux kernel: BIOS-e820: [mem 0x0000000000000000-0x000000000009fbff] usable : linux kernel: BIOS-e820: [mem 0x000000000009fc00-0x000000000009ffff] reserved : linux kernel: BIOS-e820: [mem 0x00000000000f0000-0x00000000000fffff] reserved : linux kernel: BIOS-e820: [mem 0x0000000000100000-0x00000000bffdffff] usable : linux kernel: BIOS-e820: [mem 0x00000000bffe0000-0x00000000bfffffff] reserved : linux kernel: BIOS-e820: [mem 0x00000000feffc000-0x00000000feffffff] reserved : linux kernel: BIOS-e820: [mem 0x00000000fffc0000-0x00000000ffffffff] reserved : linux kernel: BIOS-e820: [mem 0x0000000100000000-0x00000001bfffffff] usable : linux kernel: NX (Execute Disable) protection: active : linux kernel: SMBIOS 2.8 present. : linux kernel: DMI: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.0.0-prebuilt.qemu-project.org : linux kernel: Hypervisor detected: KVM : linux kernel: e820: update [mem 0x00000000-0x00000fff] usable ==> reserved : linux kernel: e820: remove [mem 0x000a0000-0x000fffff] usable : linux kernel: last_pfn = 0x1c0000 max_arch_pfn = 0x400000000 : : linux kernel: SRAT: PXM 0 -> APIC 0x00 -> Node 0 : linux kernel: SRAT: PXM 1 -> APIC 0x01 -> Node 1 : linux kernel: ACPI: SRAT: Node 0 PXM 0 [mem 0x00000000-0x0009ffff] : linux kernel: ACPI: SRAT: Node 0 PXM 0 [mem 0x00100000-0xbfffffff] : linux kernel: ACPI: SRAT: Node 0 PXM 0 [mem 0x100000000-0x13fffffff] : linux kernel: ACPI: SRAT: Node 1 PXM 1 [mem 0x140000000-0x1bfffffff] : linux kernel: ACPI: SRAT: Node 0 PXM 0 [mem 0x1c0000000-0x43fffffff] hotplug : linux kernel: NUMA: Node 0 [mem 0x00000000-0x0009ffff] + [mem 0x00100000-0xbfffffff] -> [mem 0x0 : linux kernel: NUMA: Node 0 [mem 0x00000000-0xbfffffff] + [mem 0x100000000-0x13fffffff] -> [mem 0 : linux kernel: NODE_DATA(0) allocated [mem 0x13ffd6000-0x13fffffff] : linux kernel: NODE_DATA(1) allocated [mem 0x1bffd3000-0x1bfffcfff] : : zoneinfo shows that the zone movable is placed into both numa nodes: : Node 0, zone Movable : pages free 160140 : min 1823 : low 2278 : high 2733 : spanned 262144 : present 262144 : managed 245670 : Node 1, zone Movable : pages free 448427 : min 3827 : low 4783 : high 5739 : spanned 524288 : present 524288 : managed 515766 Note how only Node 0 has a hutplugable memory region which would rule it out from the early memblock allocations (most likely memmap). Node1 will surely contain memmaps on the same node and those would prevent offlining to succeed. So this is arguably a configuration issue. Although one could argue that we should be more clever and rule early allocations from the zone movable. This would be correct but probably not worth the effort considering what a hack movablecore is. Anyway, We could do better for those cases though. We rely on start_isolate_page_range resp. has_unmovable_pages to do their job. The first one isolates the whole range to be offlined so that we do not allocate from it anymore and the later makes sure we are not stumbling over non-migrateable pages. has_unmovable_pages is overly optimistic, however. It doesn't check all the pages if we are withing zone_movable because we rely that those pages will be always migrateable. As it turns out we are still not perfect there. While bootmem pages in zonemovable sound like a clear bug which should be fixed let's remove the optimization for now and warn if we encounter unmovable pages in zone_movable in the meantime. That should help for now at least. Btw. this wasn't a real problem until commit 72b39cfc4d75 ("mm, memory_hotplug: do not fail offlining too early") because we used to have a small number of retries and then failed. This turned out to be too fragile though. Link: http://lkml.kernel.org/r/20180523125555.30039-2-mhocko@kernel.org Signed-off-by: Michal Hocko Reported-by: Oscar Salvador Tested-by: Oscar Salvador Reviewed-by: Pavel Tatashin Cc: Vlastimil Babka Cc: Reza Arbab Cc: Igor Mammedov Cc: Vitaly Kuznetsov Cc: Anshuman Khandual Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/page_alloc.c | 16 ++++++++++------ 1 file changed, 10 insertions(+), 6 deletions(-) diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 511a7124d7f9..22320ea27489 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -7598,11 +7598,12 @@ bool has_unmovable_pages(struct zone *zone, struct page *page, int count, unsigned long pfn, iter, found; /* - * For avoiding noise data, lru_add_drain_all() should be called - * If ZONE_MOVABLE, the zone never contains unmovable pages + * TODO we could make this much more efficient by not checking every + * page in the range if we know all of them are in MOVABLE_ZONE and + * that the movable zone guarantees that pages are migratable but + * the later is not the case right now unfortunatelly. E.g. movablecore + * can still lead to having bootmem allocations in zone_movable. */ - if (zone_idx(zone) == ZONE_MOVABLE) - return false; /* * CMA allocations (alloc_contig_range) really need to mark isolate @@ -7623,7 +7624,7 @@ bool has_unmovable_pages(struct zone *zone, struct page *page, int count, page = pfn_to_page(check); if (PageReserved(page)) - return true; + goto unmovable; /* * Hugepages are not in LRU lists, but they're movable. @@ -7673,9 +7674,12 @@ bool has_unmovable_pages(struct zone *zone, struct page *page, int count, * page at boot. */ if (found > count) - return true; + goto unmovable; } return false; +unmovable: + WARN_ON_ONCE(zone_idx(zone) == ZONE_MOVABLE); + return true; } bool is_pageblock_removable_nolock(struct page *page) From 8addc2d00fe171c89b5fcbafe583c2b90574d301 Mon Sep 17 00:00:00 2001 From: Michal Hocko Date: Fri, 25 May 2018 14:47:46 -0700 Subject: [PATCH 183/190] mm: do not warn on offline nodes unless the specific node is explicitly requested Oscar has noticed that we splat WARNING: CPU: 0 PID: 64 at ./include/linux/gfp.h:467 vmemmap_alloc_block+0x4e/0xc9 [...] CPU: 0 PID: 64 Comm: kworker/u4:1 Tainted: G W E 4.17.0-rc5-next-20180517-1-default+ #66 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.0.0-prebuilt.qemu-project.org 04/01/2014 Workqueue: kacpi_hotplug acpi_hotplug_work_fn Call Trace: vmemmap_populate+0xf2/0x2ae sparse_mem_map_populate+0x28/0x35 sparse_add_one_section+0x4c/0x187 __add_pages+0xe7/0x1a0 add_pages+0x16/0x70 add_memory_resource+0xa3/0x1d0 add_memory+0xe4/0x110 acpi_memory_device_add+0x134/0x2e0 acpi_bus_attach+0xd9/0x190 acpi_bus_scan+0x37/0x70 acpi_device_hotplug+0x389/0x4e0 acpi_hotplug_work_fn+0x1a/0x30 process_one_work+0x146/0x340 worker_thread+0x47/0x3e0 kthread+0xf5/0x130 ret_from_fork+0x35/0x40 when adding memory to a node that is currently offline. The VM_WARN_ON is just too loud without a good reason. In this particular case we are doing alloc_pages_node(node, GFP_KERNEL|__GFP_RETRY_MAYFAIL|__GFP_NOWARN, order) so we do not insist on allocating from the given node (it is more a hint) so we can fall back to any other populated node and moreover we explicitly ask to not warn for the allocation failure. Soften the warning only to cases when somebody asks for the given node explicitly by __GFP_THISNODE. Link: http://lkml.kernel.org/r/20180523125555.30039-3-mhocko@kernel.org Signed-off-by: Michal Hocko Reported-by: Oscar Salvador Tested-by: Oscar Salvador Reviewed-by: Pavel Tatashin Cc: Vlastimil Babka Cc: Reza Arbab Cc: Igor Mammedov Cc: Vitaly Kuznetsov Cc: Anshuman Khandual Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/gfp.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/gfp.h b/include/linux/gfp.h index 1a4582b44d32..fc5ab85278d5 100644 --- a/include/linux/gfp.h +++ b/include/linux/gfp.h @@ -464,7 +464,7 @@ static inline struct page * __alloc_pages_node(int nid, gfp_t gfp_mask, unsigned int order) { VM_BUG_ON(nid < 0 || nid >= MAX_NUMNODES); - VM_WARN_ON(!node_online(nid)); + VM_WARN_ON((gfp_mask & __GFP_THISNODE) && !node_online(nid)); return __alloc_pages(gfp_mask, order, nid); } From 6c04ab0eddc9a38e6c14ae4b6217db1e34436b8c Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Fri, 25 May 2018 14:47:50 -0700 Subject: [PATCH 184/190] proc: fix smaps and meminfo alignment The 4.17-rc /proc/meminfo and /proc//smaps look ugly: single-digit numbers (commonly 0) are misaligned. Remove seq_put_decimal_ull_width()'s leftover optimization for single digits: it's wrong now that num_to_str() takes care of the width. Link: http://lkml.kernel.org/r/alpine.LSU.2.11.1805241554210.1326@eggly.anvils Fixes: d1be35cb6f96 ("proc: add seq_put_decimal_ull_width to speed up /proc/pid/smaps") Signed-off-by: Hugh Dickins Cc: Andrei Vagin Cc: Alexey Dobriyan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/seq_file.c | 5 ----- 1 file changed, 5 deletions(-) diff --git a/fs/seq_file.c b/fs/seq_file.c index c6c27f1f9c98..4cc090b50cc5 100644 --- a/fs/seq_file.c +++ b/fs/seq_file.c @@ -709,11 +709,6 @@ void seq_put_decimal_ull_width(struct seq_file *m, const char *delimiter, if (m->count + width >= m->size) goto overflow; - if (num < 10) { - m->buf[m->count++] = num + '0'; - return; - } - len = num_to_str(m->buf + m->count, m->size - m->count, num, width); if (!len) goto overflow; From a21558618c5dfc55b6086743a88ce5a9c1588f0a Mon Sep 17 00:00:00 2001 From: Jonathan Cameron Date: Fri, 25 May 2018 14:47:53 -0700 Subject: [PATCH 185/190] mm/memory_hotplug: fix leftover use of struct page during hotplug The case of a new numa node got missed in avoiding using the node info from page_struct during hotplug. In this path we have a call to register_mem_sect_under_node (which allows us to specify it is hotplug so don't change the node), via link_mem_sections which unfortunately does not. Fix is to pass check_nid through link_mem_sections as well and disable it in the new numa node path. Note the bug only 'sometimes' manifests depending on what happens to be in the struct page structures - there are lots of them and it only needs to match one of them. The result of the bug is that (with a new memory only node) we never successfully call register_mem_sect_under_node so don't get the memory associated with the node in sysfs and meminfo for the node doesn't report it. It came up whilst testing some arm64 hotplug patches, but appears to be universal. Whilst I'm triggering it by removing then reinserting memory to a node with no other elements (thus making the node disappear then appear again), it appears it would happen on hotplugging memory where there was none before and it doesn't seem to be related the arm64 patches. These patches call __add_pages (where most of the issue was fixed by Pavel's patch). If there is a node at the time of the __add_pages call then all is well as it calls register_mem_sect_under_node from there with check_nid set to false. Without a node that function returns having not done the sysfs related stuff as there is no node to use. This is expected but it is the resulting path that fails... Exact path to the problem is as follows: mm/memory_hotplug.c: add_memory_resource() The node is not online so we enter the 'if (new_node)' twice, on the second such block there is a call to link_mem_sections which calls into drivers/node.c: link_mem_sections() which calls drivers/node.c: register_mem_sect_under_node() which calls get_nid_for_pfn and keeps trying until the output of that matches the expected node (passed all the way down from add_memory_resource) It is effectively the same fix as the one referred to in the fixes tag just in the code path for a new node where the comments point out we have to rerun the link creation because it will have failed in register_new_memory (as there was no node at the time). (actually that comment is wrong now as we don't have register_new_memory any more it got renamed to hotplug_memory_register in Pavel's patch). Link: http://lkml.kernel.org/r/20180504085311.1240-1-Jonathan.Cameron@huawei.com Fixes: fc44f7f9231a ("mm/memory_hotplug: don't read nid from struct page during hotplug") Signed-off-by: Jonathan Cameron Reviewed-by: Pavel Tatashin Acked-by: Michal Hocko Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/base/node.c | 5 +++-- include/linux/node.h | 8 +++++--- mm/memory_hotplug.c | 2 +- 3 files changed, 9 insertions(+), 6 deletions(-) diff --git a/drivers/base/node.c b/drivers/base/node.c index 7a3a580821e0..a5e821d09656 100644 --- a/drivers/base/node.c +++ b/drivers/base/node.c @@ -490,7 +490,8 @@ int unregister_mem_sect_under_nodes(struct memory_block *mem_blk, return 0; } -int link_mem_sections(int nid, unsigned long start_pfn, unsigned long nr_pages) +int link_mem_sections(int nid, unsigned long start_pfn, unsigned long nr_pages, + bool check_nid) { unsigned long end_pfn = start_pfn + nr_pages; unsigned long pfn; @@ -514,7 +515,7 @@ int link_mem_sections(int nid, unsigned long start_pfn, unsigned long nr_pages) mem_blk = find_memory_block_hinted(mem_sect, mem_blk); - ret = register_mem_sect_under_node(mem_blk, nid, true); + ret = register_mem_sect_under_node(mem_blk, nid, check_nid); if (!err) err = ret; diff --git a/include/linux/node.h b/include/linux/node.h index 41f171861dcc..6d336e38d155 100644 --- a/include/linux/node.h +++ b/include/linux/node.h @@ -32,9 +32,11 @@ extern struct node *node_devices[]; typedef void (*node_registration_func_t)(struct node *); #if defined(CONFIG_MEMORY_HOTPLUG_SPARSE) && defined(CONFIG_NUMA) -extern int link_mem_sections(int nid, unsigned long start_pfn, unsigned long nr_pages); +extern int link_mem_sections(int nid, unsigned long start_pfn, + unsigned long nr_pages, bool check_nid); #else -static inline int link_mem_sections(int nid, unsigned long start_pfn, unsigned long nr_pages) +static inline int link_mem_sections(int nid, unsigned long start_pfn, + unsigned long nr_pages, bool check_nid) { return 0; } @@ -57,7 +59,7 @@ static inline int register_one_node(int nid) if (error) return error; /* link memory sections under this node */ - error = link_mem_sections(nid, pgdat->node_start_pfn, pgdat->node_spanned_pages); + error = link_mem_sections(nid, pgdat->node_start_pfn, pgdat->node_spanned_pages, true); } return error; diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c index f74826cdceea..25982467800b 100644 --- a/mm/memory_hotplug.c +++ b/mm/memory_hotplug.c @@ -1158,7 +1158,7 @@ int __ref add_memory_resource(int nid, struct resource *res, bool online) * nodes have to go through register_node. * TODO clean up this mess. */ - ret = link_mem_sections(nid, start_pfn, nr_pages); + ret = link_mem_sections(nid, start_pfn, nr_pages, false); register_fail: /* * If sysfs file of new node can't create, cpu on the node From 23d6aef74da86a33fa6bb75f79565e0a16ee97c2 Mon Sep 17 00:00:00 2001 From: "Gustavo A. R. Silva" Date: Fri, 25 May 2018 14:47:57 -0700 Subject: [PATCH 186/190] kernel/sys.c: fix potential Spectre v1 issue `resource' can be controlled by user-space, hence leading to a potential exploitation of the Spectre variant 1 vulnerability. This issue was detected with the help of Smatch: kernel/sys.c:1474 __do_compat_sys_old_getrlimit() warn: potential spectre issue 'get_current()->signal->rlim' (local cap) kernel/sys.c:1455 __do_sys_old_getrlimit() warn: potential spectre issue 'get_current()->signal->rlim' (local cap) Fix this by sanitizing *resource* before using it to index current->signal->rlim Notice that given that speculation windows are large, the policy is to kill the speculation on the first load and not worry if it can be completed with a dependent load/store [1]. [1] https://marc.info/?l=linux-kernel&m=152449131114778&w=2 Link: http://lkml.kernel.org/r/20180515030038.GA11822@embeddedor.com Signed-off-by: Gustavo A. R. Silva Reviewed-by: Andrew Morton Cc: Alexei Starovoitov Cc: Dan Williams Cc: Thomas Gleixner Cc: Peter Zijlstra Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/sys.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/kernel/sys.c b/kernel/sys.c index b0eee418ee0d..d1b2b8d934bb 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -71,6 +71,9 @@ #include #include +/* Hardening for Spectre-v1 */ +#include + #include "uid16.h" #ifndef SET_UNALIGN_CTL @@ -1453,6 +1456,7 @@ SYSCALL_DEFINE2(old_getrlimit, unsigned int, resource, if (resource >= RLIM_NLIMITS) return -EINVAL; + resource = array_index_nospec(resource, RLIM_NLIMITS); task_lock(current->group_leader); x = current->signal->rlim[resource]; task_unlock(current->group_leader); @@ -1472,6 +1476,7 @@ COMPAT_SYSCALL_DEFINE2(old_getrlimit, unsigned int, resource, if (resource >= RLIM_NLIMITS) return -EINVAL; + resource = array_index_nospec(resource, RLIM_NLIMITS); task_lock(current->group_leader); r = current->signal->rlim[resource]; task_unlock(current->group_leader); From ae67d58d0546ad3e1ef323d69de6c3708022791f Mon Sep 17 00:00:00 2001 From: Mathieu Malaterre Date: Fri, 25 May 2018 14:48:00 -0700 Subject: [PATCH 187/190] init/main.c: include In commit c7753208a94c ("x86, swiotlb: Add memory encryption support") a call to function `mem_encrypt_init' was added. Include prototype defined in header to prevent a warning reported during compilation with W=1: init/main.c:494:20: warning: no previous prototype for `mem_encrypt_init' [-Wmissing-prototypes] Link: http://lkml.kernel.org/r/20180522195533.31415-1-malat@debian.org Signed-off-by: Mathieu Malaterre Reviewed-by: Andrew Morton Acked-by: Steven Rostedt (VMware) Cc: Tom Lendacky Cc: Ingo Molnar Cc: Thomas Gleixner Cc: Kees Cook Cc: Laura Abbott Cc: Dominik Brodowski Cc: Gargi Sharma Cc: Josh Poimboeuf Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- init/main.c | 1 + 1 file changed, 1 insertion(+) diff --git a/init/main.c b/init/main.c index fd37315835b4..3b4ada11ed52 100644 --- a/init/main.c +++ b/init/main.c @@ -91,6 +91,7 @@ #include #include #include +#include #include #include From d41362ed12b6402f0bba044037653f916757c3b2 Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Fri, 25 May 2018 14:48:04 -0700 Subject: [PATCH 188/190] checkpatch: fix macro argument precedence test checkpatch's macro argument precedence test is broken so fix it. Link: http://lkml.kernel.org/r/5dd900e9197febc1995604bb33c23c136d8b33ce.camel@perches.com Signed-off-by: Joe Perches Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- scripts/checkpatch.pl | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl index e16d6713f236..2d42eb9cd1a5 100755 --- a/scripts/checkpatch.pl +++ b/scripts/checkpatch.pl @@ -5041,7 +5041,7 @@ sub process { $tmp_stmt =~ s/\b(typeof|__typeof__|__builtin\w+|typecheck\s*\(\s*$Type\s*,|\#+)\s*\(*\s*$arg\s*\)*\b//g; $tmp_stmt =~ s/\#+\s*$arg\b//g; $tmp_stmt =~ s/\b$arg\s*\#\#//g; - my $use_cnt = $tmp_stmt =~ s/\b$arg\b//g; + my $use_cnt = () = $tmp_stmt =~ /\b$arg\b/g; if ($use_cnt > 1) { CHK("MACRO_ARG_REUSE", "Macro argument reuse '$arg' - possible side-effects?\n" . "$herectx"); From ed1596f9ab958dd156a66c9ff1029d3761c1786a Mon Sep 17 00:00:00 2001 From: David Hildenbrand Date: Fri, 25 May 2018 14:48:08 -0700 Subject: [PATCH 189/190] kasan: free allocated shadow memory on MEM_CANCEL_ONLINE We have to free memory again when we cancel onlining, otherwise a later onlining attempt will fail. Link: http://lkml.kernel.org/r/20180522100756.18478-2-david@redhat.com Fixes: fa69b5989bb0 ("mm/kasan: add support for memory hotplug") Signed-off-by: David Hildenbrand Acked-by: Andrey Ryabinin Cc: Alexander Potapenko Cc: Dmitry Vyukov Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/kasan/kasan.c | 1 + 1 file changed, 1 insertion(+) diff --git a/mm/kasan/kasan.c b/mm/kasan/kasan.c index 7160028efd12..721e00837f00 100644 --- a/mm/kasan/kasan.c +++ b/mm/kasan/kasan.c @@ -866,6 +866,7 @@ static int __meminit kasan_mem_notifier(struct notifier_block *nb, kmemleak_ignore(ret); return NOTIFY_OK; } + case MEM_CANCEL_ONLINE: case MEM_OFFLINE: { struct vm_struct *vm; From 3f1959721558a976aaf9c2024d5bc884e6411bf7 Mon Sep 17 00:00:00 2001 From: David Hildenbrand Date: Fri, 25 May 2018 14:48:11 -0700 Subject: [PATCH 190/190] kasan: fix memory hotplug during boot Using module_init() is wrong. E.g. ACPI adds and onlines memory before our memory notifier gets registered. This makes sure that ACPI memory detected during boot up will not result in a kernel crash. Easily reproducible with QEMU, just specify a DIMM when starting up. Link: http://lkml.kernel.org/r/20180522100756.18478-3-david@redhat.com Fixes: 786a8959912e ("kasan: disable memory hotplug") Signed-off-by: David Hildenbrand Acked-by: Andrey Ryabinin Cc: Alexander Potapenko Cc: Dmitry Vyukov Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/kasan/kasan.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/mm/kasan/kasan.c b/mm/kasan/kasan.c index 721e00837f00..f185455b3406 100644 --- a/mm/kasan/kasan.c +++ b/mm/kasan/kasan.c @@ -898,5 +898,5 @@ static int __init kasan_memhotplug_init(void) return 0; } -module_init(kasan_memhotplug_init); +core_initcall(kasan_memhotplug_init); #endif