From 0a5191efe06b5103909206e4fbcff81d30283f8e Mon Sep 17 00:00:00 2001 From: Jonas Gorski Date: Sun, 29 Oct 2017 16:27:21 +0100 Subject: [PATCH 01/55] MIPS: AR7: ensure the port type's FCR value is used Since commit aef9a7bd9b67 ("serial/uart/8250: Add tunable RX interrupt trigger I/F of FIFO buffers"), the port's default FCR value isn't used in serial8250_do_set_termios anymore, but copied over once in serial8250_config_port and then modified as needed. Unfortunately, serial8250_config_port will never be called if the port is shared between kernel and userspace, and the port's flag doesn't have UPF_BOOT_AUTOCONF, which would trigger a serial8250_config_port as well. This causes garbled output from userspace: [ 5.220000] random: procd urandom read with 49 bits of entropy available ers [kee Fix this by forcing it to be configured on boot, resulting in the expected output: [ 5.250000] random: procd urandom read with 50 bits of entropy available Press the [f] key and hit [enter] to enter failsafe mode Press the [1], [2], [3] or [4] key and hit [enter] to select the debug level Fixes: aef9a7bd9b67 ("serial/uart/8250: Add tunable RX interrupt trigger I/F of FIFO buffers") Signed-off-by: Jonas Gorski Cc: Greg Kroah-Hartman Cc: Yoshihiro YUNOMAE Cc: Florian Fainelli Cc: Nicolas Schichan Cc: linux-mips@linux-mips.org Cc: linux-serial@vger.kernel.org Patchwork: https://patchwork.linux-mips.org/patch/17544/ Signed-off-by: Ralf Baechle --- arch/mips/ar7/platform.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/mips/ar7/platform.c b/arch/mips/ar7/platform.c index 4674f1efbe7a..e1675c25d5d4 100644 --- a/arch/mips/ar7/platform.c +++ b/arch/mips/ar7/platform.c @@ -575,7 +575,7 @@ static int __init ar7_register_uarts(void) uart_port.type = PORT_AR7; uart_port.uartclk = clk_get_rate(bus_clk) / 2; uart_port.iotype = UPIO_MEM32; - uart_port.flags = UPF_FIXED_TYPE; + uart_port.flags = UPF_FIXED_TYPE | UPF_BOOT_AUTOCONF; uart_port.regshift = 2; uart_port.line = 0; From 9dd79fed1bed6089f9729027e2d8cfa1049266e5 Mon Sep 17 00:00:00 2001 From: Matt Redfearn Date: Tue, 14 Nov 2017 13:29:17 +0000 Subject: [PATCH 02/55] MIPS: ath25: Avoid undefined early_serial_setup() without SERIAL_8250_CONSOLE Currently MIPS allnoconfig with CONFIG_ATH25=y fails to link due to missing support for early_serial_setup(): LD vmlinux arch/mips/ath25/devices.o: In function ath25_serial_setup': devices.c:(.init.text+0x68): undefined reference to 'early_serial_setup' Rather than adding dependencies to the platform to force inclusion of SERIAL_8250_CONSOLE together with it's dependencies like TTY, HAS_IOMEM, etc, just make ath25_serial_setup() a no-op when the dependency is not selected in the kernel config. Signed-off-by: Matt Redfearn Cc: James Hogan Cc: Thomas Gleixner Cc: Philippe Ombredanne Cc: Kate Stewart Cc: Greg Kroah-Hartman Cc: linux-mips@linux-mips.org Cc: linux-kernel@vger.kernel.org Patchwork: https://patchwork.linux-mips.org/patch/17700/ Signed-off-by: Ralf Baechle --- arch/mips/ath25/devices.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/mips/ath25/devices.c b/arch/mips/ath25/devices.c index e1156347da53..301a9028273c 100644 --- a/arch/mips/ath25/devices.c +++ b/arch/mips/ath25/devices.c @@ -73,6 +73,7 @@ const char *get_system_type(void) void __init ath25_serial_setup(u32 mapbase, int irq, unsigned int uartclk) { +#ifdef CONFIG_SERIAL_8250_CONSOLE struct uart_port s; memset(&s, 0, sizeof(s)); @@ -85,6 +86,7 @@ void __init ath25_serial_setup(u32 mapbase, int irq, unsigned int uartclk) s.uartclk = uartclk; early_serial_setup(&s); +#endif /* CONFIG_SERIAL_8250_CONSOLE */ } int __init ath25_add_wmac(int nr, u32 base, int irq) From ff9bed94d0f3b82d0fff3599cf0eb2cadf0fc770 Mon Sep 17 00:00:00 2001 From: Matt Redfearn Date: Tue, 14 Nov 2017 15:44:22 +0000 Subject: [PATCH 03/55] MIPS: RB532: Avoid undefined early_serial_setup() without SERIAL_8250_CONSOLE Currently MIPS allnoconfig with CONFIG_MIKROTIK_RB532=y fails to link due to missing support for early_serial_setup(): LD vmlinux arch/mips/rb532/serial.o: In function `setup_serial_port': serial.c:(.init.text+0x14): undefined reference to `early_serial_setup' Rather than adding dependencies to the platform to force inclusion of SERIAL_8250_CONSOLE together with it's dependencies like TTY, HAS_IOMEM, etc, just exclude arch/mips/rb532/serial.c from the build when it's dependency is not selected in the kernel config. Reported-by: Ralf Baechle Signed-off-by: Matt Redfearn Cc: James Hogan Cc: linux-mips@linux-mips.org Cc: linux-kernel@vger.kernel.org Patchwork: https://patchwork.linux-mips.org/patch/17701/ Signed-off-by: Ralf Baechle --- arch/mips/rb532/Makefile | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/arch/mips/rb532/Makefile b/arch/mips/rb532/Makefile index efdecdb6e3ea..8186afca2234 100644 --- a/arch/mips/rb532/Makefile +++ b/arch/mips/rb532/Makefile @@ -2,4 +2,6 @@ # Makefile for the RB532 board specific parts of the kernel # -obj-y += irq.o time.o setup.o serial.o prom.o gpio.o devices.o +obj-$(CONFIG_SERIAL_8250_CONSOLE) += serial.o + +obj-y += irq.o time.o setup.o prom.o gpio.o devices.o From ecff167cc80dd3c6afff55bdc66b4981d587ba3e Mon Sep 17 00:00:00 2001 From: Matt Redfearn Date: Tue, 14 Nov 2017 15:44:23 +0000 Subject: [PATCH 04/55] MIPS: RB532: Avoid undefined mac_pton without GENERIC_NET_UTILS Currently MIPS allnoconfig with CONFIG_MIKROTIK_RB532=y fails to link due to missing support for mac_pton(): LD vmlinux arch/mips/rb532/devices.o: In function `setup_kmac': devices.c:(.init.text+0xc): undefined reference to `mac_pton' Rather than adding dependencies to the platform to force inclusion of GENERIC_NET_UTILS which is selected by CONFIG_NET, just exclude the setup of the MAC address if CONFIG_NET is not selected in the kernel config. Signed-off-by: Matt Redfearn Cc: James Hogan Cc: Boris Brezillon Cc: Neil Armstrong Cc: Krzysztof Kozlowski Cc: Tony Lindgren Cc: Vladimir Zapolskiy Cc: Shawn Guo Cc: linux-mips@linux-mips.org Cc: linux-kernel@vger.kernel.org Patchwork: https://patchwork.linux-mips.org/patch/17702/ Signed-off-by: Ralf Baechle --- arch/mips/rb532/devices.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/arch/mips/rb532/devices.c b/arch/mips/rb532/devices.c index 32ea3e6731d6..354d258396ff 100644 --- a/arch/mips/rb532/devices.c +++ b/arch/mips/rb532/devices.c @@ -310,6 +310,8 @@ static int __init plat_setup_devices(void) return platform_add_devices(rb532_devs, ARRAY_SIZE(rb532_devs)); } +#ifdef CONFIG_NET + static int __init setup_kmac(char *s) { printk(KERN_INFO "korina mac = %s\n", s); @@ -322,4 +324,6 @@ static int __init setup_kmac(char *s) __setup("kmac=", setup_kmac); +#endif /* CONFIG_NET */ + arch_initcall(plat_setup_devices); From dfe004951b1387f8e91b83c95a5dc964a8545d49 Mon Sep 17 00:00:00 2001 From: Matt Redfearn Date: Tue, 14 Nov 2017 17:16:27 +0000 Subject: [PATCH 05/55] MIPS: BCM47XX Avoid compile error with MIPS allnoconfig Currently MIPS allnoconfig with CONFIG_BCM47XX=y fails to compile due to neither BCM47XX_BCMA nor BCM47XX_SSB being selected. This leads the enumeration in arch/mips/include/asm/mach-bcm47xx/bcm47xx.h to be empty, and compilation fails: In file included from arch/mips/bcm47xx/irq.c:32:0: ./arch/mips/include/asm/mach-bcm47xx/bcm47xx.h:34:1: error: expected identifier before '}' token }; ^ make[2]: *** [scripts/Makefile.build:314: arch/mips/bcm47xx/irq.o] Error 1 Fix this by ensuring that BCM47XX_SSB is selected if BCM47XX_BCMA is not. This allows us to select either system or both, but not neither. Signed-off-by: Matt Redfearn Cc: James Hogan Cc: linux-mips@linux-mips.org Cc: linux-kernel@vger.kernel.org Patchwork: https://patchwork.linux-mips.org/patch/17703/ Signed-off-by: Ralf Baechle --- arch/mips/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig index 350a990fc719..659e0079487f 100644 --- a/arch/mips/Kconfig +++ b/arch/mips/Kconfig @@ -259,6 +259,7 @@ config BCM47XX select LEDS_GPIO_REGISTER select BCM47XX_NVRAM select BCM47XX_SPROM + select BCM47XX_SSB if !BCM47XX_BCMA help Support for BCM47XX based boards From 7e5e371ee5390a8bb7e111c794a334d9bf25ca3d Mon Sep 17 00:00:00 2001 From: James Hogan Date: Tue, 21 Nov 2017 00:02:40 +0000 Subject: [PATCH 06/55] MIPS: Fix CPS SMP NS16550 UART defaults The MIPS_CPS_NS16550_BASE and MIPS_CPS_NS16550_SHIFT options have no defaults for non-Malta platforms which select SYS_SUPPORTS_MIPS_CPS (i.e. the pistachio and generic platforms). This is problematic for automated allyesconfig and allmodconfig builds based on these platforms, since make silentoldconfig tries to ask the user for values, and especially since v4.15 where the default platform was switched to generic. Default these options to 0 and arrange for MIPS_CPS_NS16550 to be no when using that default base address, so that the option only has an effect when the default is provided (i.e. Malta) or when a value is provided by the user. Fixes: 609cf6f2291a ("MIPS: CPS: Early debug using an ns16550-compatible UART") Signed-off-by: James Hogan Reviewed-by: Paul Burton Tested-by: Guenter Roeck Cc: linux-mips@linux-mips.org Cc: linux-kernel@vger.kernel.org Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/17749/ Signed-off-by: Ralf Baechle --- arch/mips/Kconfig.debug | 14 ++++++++++---- 1 file changed, 10 insertions(+), 4 deletions(-) diff --git a/arch/mips/Kconfig.debug b/arch/mips/Kconfig.debug index 464af5e025d6..0749c3724543 100644 --- a/arch/mips/Kconfig.debug +++ b/arch/mips/Kconfig.debug @@ -124,30 +124,36 @@ config SCACHE_DEBUGFS If unsure, say N. -menuconfig MIPS_CPS_NS16550 +menuconfig MIPS_CPS_NS16550_BOOL bool "CPS SMP NS16550 UART output" depends on MIPS_CPS help Output debug information via an ns16550 compatible UART if exceptions occur early in the boot process of a secondary core. -if MIPS_CPS_NS16550 +if MIPS_CPS_NS16550_BOOL + +config MIPS_CPS_NS16550 + def_bool MIPS_CPS_NS16550_BASE != 0 config MIPS_CPS_NS16550_BASE hex "UART Base Address" default 0x1b0003f8 if MIPS_MALTA + default 0 help The base address of the ns16550 compatible UART on which to output debug information from the early stages of core startup. + This is only used if non-zero. + config MIPS_CPS_NS16550_SHIFT int "UART Register Shift" - default 0 if MIPS_MALTA + default 0 help The number of bits to shift ns16550 register indices by in order to form their addresses. That is, log base 2 of the span between adjacent ns16550 registers in the system. -endif # MIPS_CPS_NS16550 +endif # MIPS_CPS_NS16550_BOOL endmenu From b6ab1a138b758c4bdf30d5517e546e1c8aff9e3b Mon Sep 17 00:00:00 2001 From: Arvind Yadav Date: Fri, 24 Nov 2017 07:38:20 +0530 Subject: [PATCH 07/55] MIPS: ralink: Fix platform_get_irq's error checking The platform_get_irq() function returns negative if an error occurs. zero or positive number on success. platform_get_irq() error checking for zero is not correct. Signed-off-by: Arvind Yadav Cc: john@phrozen.org Cc: linux-kernel@vger.kernel.org Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/17783/ Signed-off-by: Ralf Baechle --- arch/mips/ralink/timer.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/mips/ralink/timer.c b/arch/mips/ralink/timer.c index d4469b20d176..4f46a4509f79 100644 --- a/arch/mips/ralink/timer.c +++ b/arch/mips/ralink/timer.c @@ -109,9 +109,9 @@ static int rt_timer_probe(struct platform_device *pdev) } rt->irq = platform_get_irq(pdev, 0); - if (!rt->irq) { + if (rt->irq < 0) { dev_err(&pdev->dev, "failed to load irq\n"); - return -ENOENT; + return rt->irq; } rt->membase = devm_ioremap_resource(&pdev->dev, res); From c04de7b1ad645b61c141df8ca903ba0cc03a57f7 Mon Sep 17 00:00:00 2001 From: James Hogan Date: Tue, 5 Dec 2017 22:28:22 +0000 Subject: [PATCH 08/55] MIPS: CM: Drop WARN_ON(vp != 0) Since commit 68923cdc2eb3 ("MIPS: CM: Add cluster & block args to mips_cm_lock_other()"), mips_smp_send_ipi_mask() has used mips_cm_lock_other_cpu() with each CPU number, rather than mips_cm_lock_other() with the first VPE in each core. Prior to r6, multicore multithreaded systems such as dual-core dual-thread interAptivs with CPU Idle enabled (e.g. MIPS Creator Ci40) results in mips_cm_lock_other() repeatedly hitting WARN_ON(vp != 0). There doesn't appear to be anything fundamentally wrong about passing a non-zero VP/VPE number, even if it is a core's region that is locked into the other region before r6, so remove that particular WARN_ON(). Fixes: 68923cdc2eb3 ("MIPS: CM: Add cluster & block args to mips_cm_lock_other()") Signed-off-by: James Hogan Reviewed-by: Paul Burton Cc: linux-mips@linux-mips.org Cc: stable@vger.kernel.org # 4.14+ Patchwork: https://patchwork.linux-mips.org/patch/17883/ Signed-off-by: Ralf Baechle --- arch/mips/kernel/mips-cm.c | 1 - 1 file changed, 1 deletion(-) diff --git a/arch/mips/kernel/mips-cm.c b/arch/mips/kernel/mips-cm.c index dd5567b1e305..8f5bd04f320a 100644 --- a/arch/mips/kernel/mips-cm.c +++ b/arch/mips/kernel/mips-cm.c @@ -292,7 +292,6 @@ void mips_cm_lock_other(unsigned int cluster, unsigned int core, *this_cpu_ptr(&cm_core_lock_flags)); } else { WARN_ON(cluster != 0); - WARN_ON(vp != 0); WARN_ON(block != CM_GCR_Cx_OTHER_BLOCK_LOCAL); /* From ccf85c744275de0ba40beff0bf9206a094f12e62 Mon Sep 17 00:00:00 2001 From: James Hogan Date: Thu, 7 Dec 2017 07:14:17 +0000 Subject: [PATCH 09/55] MIPS: mm: Fix duplicate "const" on insn_table_MM MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Fix the following gcc 7.x build error on microMIPS builds: arch/mips/mm/uasm-micromips.c:43:26: error: duplicate ‘const’ declaration specifier [-Werror=duplicate-decl-specifier] static const struct insn const insn_table_MM[insn_invalid] = { ^~~~~ The same issue has already been fixed in uasm-mips by commit 00e06297b351 ("MIPS: mm: remove duplicate "const" qualifier on insn_table"). Signed-off-by: James Hogan Fixes: ce807d5f67ed ("MIPS: Optimize uasm insn lookup.") Cc: David Daney Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/17889/ Signed-off-by: Ralf Baechle --- arch/mips/mm/uasm-micromips.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/mips/mm/uasm-micromips.c b/arch/mips/mm/uasm-micromips.c index cdb5a191b9d5..9bb6baa45da3 100644 --- a/arch/mips/mm/uasm-micromips.c +++ b/arch/mips/mm/uasm-micromips.c @@ -40,7 +40,7 @@ #include "uasm.c" -static const struct insn const insn_table_MM[insn_invalid] = { +static const struct insn insn_table_MM[insn_invalid] = { [insn_addu] = {M(mm_pool32a_op, 0, 0, 0, 0, mm_addu32_op), RT | RS | RD}, [insn_addiu] = {M(mm_addiu32_op, 0, 0, 0, 0, 0), RT | RS | SIMM}, [insn_and] = {M(mm_pool32a_op, 0, 0, 0, 0, mm_and_op), RT | RS | RD}, From c9f926000fe3b84135a81602a9f7e63a6a7898e2 Mon Sep 17 00:00:00 2001 From: Hannes Reinecke Date: Wed, 10 Jan 2018 09:34:02 +0100 Subject: [PATCH 10/55] scsi: libsas: Disable asynchronous aborts for SATA devices Handling CD-ROM devices from libsas is decidedly odd, as libata relies on SCSI EH to be started to figure out that no medium is present. So we cannot do asynchronous aborts for SATA devices. Fixes: 909657615d9 ("scsi: libsas: allow async aborts") Cc: # 4.12+ Signed-off-by: Hannes Reinecke Reviewed-by: Christoph Hellwig Tested-by: Yves-Alexis Perez Signed-off-by: Martin K. Petersen --- drivers/scsi/libsas/sas_scsi_host.c | 17 +++++++++++++++-- 1 file changed, 15 insertions(+), 2 deletions(-) diff --git a/drivers/scsi/libsas/sas_scsi_host.c b/drivers/scsi/libsas/sas_scsi_host.c index 91795eb56206..eea94aa4091c 100644 --- a/drivers/scsi/libsas/sas_scsi_host.c +++ b/drivers/scsi/libsas/sas_scsi_host.c @@ -486,15 +486,28 @@ static int sas_queue_reset(struct domain_device *dev, int reset_type, int sas_eh_abort_handler(struct scsi_cmnd *cmd) { - int res; + int res = TMF_RESP_FUNC_FAILED; struct sas_task *task = TO_SAS_TASK(cmd); struct Scsi_Host *host = cmd->device->host; + struct domain_device *dev = cmd_to_domain_dev(cmd); struct sas_internal *i = to_sas_internal(host->transportt); + unsigned long flags; if (!i->dft->lldd_abort_task) return FAILED; - res = i->dft->lldd_abort_task(task); + spin_lock_irqsave(host->host_lock, flags); + /* We cannot do async aborts for SATA devices */ + if (dev_is_sata(dev) && !host->host_eh_scheduled) { + spin_unlock_irqrestore(host->host_lock, flags); + return FAILED; + } + spin_unlock_irqrestore(host->host_lock, flags); + + if (task) + res = i->dft->lldd_abort_task(task); + else + SAS_DPRINTK("no task to abort\n"); if (res == TMF_RESP_FUNC_SUCC || res == TMF_RESP_FUNC_COMPLETE) return SUCCESS; From ebabcf17bcd7ce968b1631ebe08236275698f39b Mon Sep 17 00:00:00 2001 From: James Hogan Date: Thu, 7 Dec 2017 07:20:46 +0000 Subject: [PATCH 11/55] MIPS: Implement __multi3 for GCC7 MIPS64r6 builds GCC7 is a bit too eager to generate suboptimal __multi3 calls (128bit multiply with 128bit result) for MIPS64r6 builds, even in code which doesn't explicitly use 128bit types, such as the following: unsigned long func(unsigned long a, unsigned long b) { return a > (~0UL) / b; } Which GCC rearanges to: return (unsigned __int128)a * (unsigned __int128)b > 0xffffffffffffffff; Therefore implement __multi3, but only for MIPS64r6 with GCC7 as under normal circumstances we wouldn't expect any calls to __multi3 to be generated from kernel code. Reported-by: Thomas Petazzoni Signed-off-by: James Hogan Tested-by: Waldemar Brodkorb Cc: Ralf Baechle Cc: Maciej W. Rozycki Cc: Matthew Fortune Cc: Florian Fainelli Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/17890/ --- arch/mips/lib/Makefile | 3 ++- arch/mips/lib/libgcc.h | 17 +++++++++++++ arch/mips/lib/multi3.c | 54 ++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 73 insertions(+), 1 deletion(-) create mode 100644 arch/mips/lib/multi3.c diff --git a/arch/mips/lib/Makefile b/arch/mips/lib/Makefile index 78c2affeabf8..e84e12655fa8 100644 --- a/arch/mips/lib/Makefile +++ b/arch/mips/lib/Makefile @@ -16,4 +16,5 @@ obj-$(CONFIG_CPU_R3000) += r3k_dump_tlb.o obj-$(CONFIG_CPU_TX39XX) += r3k_dump_tlb.o # libgcc-style stuff needed in the kernel -obj-y += ashldi3.o ashrdi3.o bswapsi.o bswapdi.o cmpdi2.o lshrdi3.o ucmpdi2.o +obj-y += ashldi3.o ashrdi3.o bswapsi.o bswapdi.o cmpdi2.o lshrdi3.o multi3.o \ + ucmpdi2.o diff --git a/arch/mips/lib/libgcc.h b/arch/mips/lib/libgcc.h index 28002ed90c2c..199a7f96282f 100644 --- a/arch/mips/lib/libgcc.h +++ b/arch/mips/lib/libgcc.h @@ -10,10 +10,18 @@ typedef int word_type __attribute__ ((mode (__word__))); struct DWstruct { int high, low; }; + +struct TWstruct { + long long high, low; +}; #elif defined(__LITTLE_ENDIAN) struct DWstruct { int low, high; }; + +struct TWstruct { + long long low, high; +}; #else #error I feel sick. #endif @@ -23,4 +31,13 @@ typedef union { long long ll; } DWunion; +#if defined(CONFIG_64BIT) && defined(CONFIG_CPU_MIPSR6) +typedef int ti_type __attribute__((mode(TI))); + +typedef union { + struct TWstruct s; + ti_type ti; +} TWunion; +#endif + #endif /* __ASM_LIBGCC_H */ diff --git a/arch/mips/lib/multi3.c b/arch/mips/lib/multi3.c new file mode 100644 index 000000000000..111ad475aa0c --- /dev/null +++ b/arch/mips/lib/multi3.c @@ -0,0 +1,54 @@ +// SPDX-License-Identifier: GPL-2.0 +#include + +#include "libgcc.h" + +/* + * GCC 7 suboptimally generates __multi3 calls for mips64r6, so for that + * specific case only we'll implement it here. + * + * See https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82981 + */ +#if defined(CONFIG_64BIT) && defined(CONFIG_CPU_MIPSR6) && (__GNUC__ == 7) + +/* multiply 64-bit values, low 64-bits returned */ +static inline long long notrace dmulu(long long a, long long b) +{ + long long res; + + asm ("dmulu %0,%1,%2" : "=r" (res) : "r" (a), "r" (b)); + return res; +} + +/* multiply 64-bit unsigned values, high 64-bits of 128-bit result returned */ +static inline long long notrace dmuhu(long long a, long long b) +{ + long long res; + + asm ("dmuhu %0,%1,%2" : "=r" (res) : "r" (a), "r" (b)); + return res; +} + +/* multiply 128-bit values, low 128-bits returned */ +ti_type notrace __multi3(ti_type a, ti_type b) +{ + TWunion res, aa, bb; + + aa.ti = a; + bb.ti = b; + + /* + * a * b = (a.lo * b.lo) + * + 2^64 * (a.hi * b.lo + a.lo * b.hi) + * [+ 2^128 * (a.hi * b.hi)] + */ + res.s.low = dmulu(aa.s.low, bb.s.low); + res.s.high = dmuhu(aa.s.low, bb.s.low); + res.s.high += dmulu(aa.s.high, bb.s.low); + res.s.high += dmulu(aa.s.low, bb.s.high); + + return res.ti; +} +EXPORT_SYMBOL(__multi3); + +#endif /* 64BIT && CPU_MIPSR6 && GCC7 */ From c507babf10ead4d5c8cca704539b170752a8ac84 Mon Sep 17 00:00:00 2001 From: Punit Agrawal Date: Thu, 4 Jan 2018 18:24:33 +0000 Subject: [PATCH 12/55] KVM: arm/arm64: Check pagesize when allocating a hugepage at Stage 2 KVM only supports PMD hugepages at stage 2 but doesn't actually check that the provided hugepage memory pagesize is PMD_SIZE before populating stage 2 entries. In cases where the backing hugepage size is smaller than PMD_SIZE (such as when using contiguous hugepages), KVM can end up creating stage 2 mappings that extend beyond the supplied memory. Fix this by checking for the pagesize of userspace vma before creating PMD hugepage at stage 2. Fixes: 66b3923a1a0f77a ("arm64: hugetlb: add support for PTE contiguous bit") Signed-off-by: Punit Agrawal Cc: Marc Zyngier Cc: # v4.5+ Reviewed-by: Christoffer Dall Signed-off-by: Christoffer Dall --- virt/kvm/arm/mmu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/virt/kvm/arm/mmu.c b/virt/kvm/arm/mmu.c index b4b69c2d1012..9dea96380339 100644 --- a/virt/kvm/arm/mmu.c +++ b/virt/kvm/arm/mmu.c @@ -1310,7 +1310,7 @@ static int user_mem_abort(struct kvm_vcpu *vcpu, phys_addr_t fault_ipa, return -EFAULT; } - if (is_vm_hugetlb_page(vma) && !logging_active) { + if (vma_kernel_pagesize(vma) == PMD_SIZE && !logging_active) { hugetlb = true; gfn = (fault_ipa & PMD_MASK) >> PAGE_SHIFT; } else { From f8f85dc00b7427de6222ea3955c52512315d13cd Mon Sep 17 00:00:00 2001 From: Christoffer Dall Date: Fri, 12 Jan 2018 11:40:21 +0100 Subject: [PATCH 13/55] KVM: arm64: Fix GICv4 init when called from vgic_its_create Commit 3d1ad640f8c94 ("KVM: arm/arm64: Fix GICv4 ITS initialization issues") moved the vgic_supports_direct_msis() check in vgic_v4_init(). However when vgic_v4_init is called from vgic_its_create(), the has_its field is not yet set. Hence vgic_supports_direct_msis returns false and vgic_v4_init does nothing. The gic/its init sequence is a bit messy, so let's be specific about the prerequisite checks in the various call paths instead of relying on a common wrapper. Fixes: 3d1ad640f8c94 ("KVM: arm/arm64: Fix GICv4 ITS initialization issues") Reported-by: Eric Auger Signed-off-by: Christoffer Dall --- virt/kvm/arm/vgic/vgic-init.c | 8 +++++--- virt/kvm/arm/vgic/vgic-v4.c | 2 +- 2 files changed, 6 insertions(+), 4 deletions(-) diff --git a/virt/kvm/arm/vgic/vgic-init.c b/virt/kvm/arm/vgic/vgic-init.c index 62310122ee78..743ca5cb05ef 100644 --- a/virt/kvm/arm/vgic/vgic-init.c +++ b/virt/kvm/arm/vgic/vgic-init.c @@ -285,9 +285,11 @@ int vgic_init(struct kvm *kvm) if (ret) goto out; - ret = vgic_v4_init(kvm); - if (ret) - goto out; + if (vgic_has_its(kvm)) { + ret = vgic_v4_init(kvm); + if (ret) + goto out; + } kvm_for_each_vcpu(i, vcpu, kvm) kvm_vgic_vcpu_enable(vcpu); diff --git a/virt/kvm/arm/vgic/vgic-v4.c b/virt/kvm/arm/vgic/vgic-v4.c index 4a37292855bc..bc4265154bac 100644 --- a/virt/kvm/arm/vgic/vgic-v4.c +++ b/virt/kvm/arm/vgic/vgic-v4.c @@ -118,7 +118,7 @@ int vgic_v4_init(struct kvm *kvm) struct kvm_vcpu *vcpu; int i, nr_vcpus, ret; - if (!vgic_supports_direct_msis(kvm)) + if (!kvm_vgic_global_state.has_gicv4) return 0; /* Nothing to see here... move along. */ if (dist->its_vm.vpes) From acfb3b883f6d6a4b5d27ad7fdded11f6a09ae6dd Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Tue, 16 Jan 2018 10:23:47 +0000 Subject: [PATCH 14/55] arm64: KVM: Fix SMCCC handling of unimplemented SMC/HVC calls KVM doesn't follow the SMCCC when it comes to unimplemented calls, and inject an UNDEF instead of returning an error. Since firmware calls are now used for security mitigation, they are becoming more common, and the undef is counter productive. Instead, let's follow the SMCCC which states that -1 must be returned to the caller when getting an unknown function number. Cc: Signed-off-by: Marc Zyngier Signed-off-by: Christoffer Dall --- arch/arm64/kvm/handle_exit.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/arm64/kvm/handle_exit.c b/arch/arm64/kvm/handle_exit.c index 304203fa9e33..e60494f1eef9 100644 --- a/arch/arm64/kvm/handle_exit.c +++ b/arch/arm64/kvm/handle_exit.c @@ -45,7 +45,7 @@ static int handle_hvc(struct kvm_vcpu *vcpu, struct kvm_run *run) ret = kvm_psci_call(vcpu); if (ret < 0) { - kvm_inject_undefined(vcpu); + vcpu_set_reg(vcpu, 0, ~0UL); return 1; } @@ -54,7 +54,7 @@ static int handle_hvc(struct kvm_vcpu *vcpu, struct kvm_run *run) static int handle_smc(struct kvm_vcpu *vcpu, struct kvm_run *run) { - kvm_inject_undefined(vcpu); + vcpu_set_reg(vcpu, 0, ~0UL); return 1; } From 0d83620fd18e5b8d79d390e482583b379a6a986d Mon Sep 17 00:00:00 2001 From: Michael Cree Date: Wed, 3 Jan 2018 21:58:00 +1300 Subject: [PATCH 15/55] alpha: extend memset16 to EV6 optimised routines Commit 92ce4c3ea7c4, "alpha: add support for memset16", renamed the function memsetw() to be memset16() but neglected to do this for the EV6 optimised version, thus when building a kernel optimised for EV6 (or later) link errors result. This extends the memset16 support to EV6. Signed-off-by: Michael Cree Signed-off-by: Matt Turner --- arch/alpha/lib/ev6-memset.S | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/arch/alpha/lib/ev6-memset.S b/arch/alpha/lib/ev6-memset.S index 316a99aa9efe..1cfcfbbea6f0 100644 --- a/arch/alpha/lib/ev6-memset.S +++ b/arch/alpha/lib/ev6-memset.S @@ -18,7 +18,7 @@ * The algorithm for the leading and trailing quadwords remains the same, * however the loop has been unrolled to enable better memory throughput, * and the code has been replicated for each of the entry points: __memset - * and __memsetw to permit better scheduling to eliminate the stalling + * and __memset16 to permit better scheduling to eliminate the stalling * encountered during the mask replication. * A future enhancement might be to put in a byte store loop for really * small (say < 32 bytes) memset()s. Whether or not that change would be @@ -34,7 +34,7 @@ .globl memset .globl __memset .globl ___memset - .globl __memsetw + .globl __memset16 .globl __constant_c_memset .ent ___memset @@ -415,9 +415,9 @@ end: * to mask stalls. Note that entry point names also had to change */ .align 5 - .ent __memsetw + .ent __memset16 -__memsetw: +__memset16: .frame $30,0,$26,0 .prologue 0 @@ -596,8 +596,8 @@ end_w: nop ret $31,($26),1 # L0 : - .end __memsetw - EXPORT_SYMBOL(__memsetw) + .end __memset16 + EXPORT_SYMBOL(__memset16) memset = ___memset __memset = ___memset From 37b95951c58fdf08dc10afa9d02066ed9f176fb5 Mon Sep 17 00:00:00 2001 From: Tianyu Lan Date: Tue, 16 Jan 2018 17:34:07 +0800 Subject: [PATCH 16/55] KVM/x86: Fix wrong macro references of X86_CR0_PG_BIT and X86_CR4_PAE_BIT in kvm_valid_sregs() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit kvm_valid_sregs() should use X86_CR0_PG and X86_CR4_PAE to check bit status rather than X86_CR0_PG_BIT and X86_CR4_PAE_BIT. This patch is to fix it. Fixes: f29810335965a(KVM/x86: Check input paging mode when cs.l is set) Reported-by: Jeremi Piotrowski Cc: Paolo Bonzini Cc: Radim Krčmář Signed-off-by: Tianyu Lan Signed-off-by: Radim Krčmář --- arch/x86/kvm/x86.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 1cec2c62a0b0..c53298dfbf50 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -7496,13 +7496,13 @@ EXPORT_SYMBOL_GPL(kvm_task_switch); int kvm_valid_sregs(struct kvm_vcpu *vcpu, struct kvm_sregs *sregs) { - if ((sregs->efer & EFER_LME) && (sregs->cr0 & X86_CR0_PG_BIT)) { + if ((sregs->efer & EFER_LME) && (sregs->cr0 & X86_CR0_PG)) { /* * When EFER.LME and CR0.PG are set, the processor is in * 64-bit mode (though maybe in a 32-bit code segment). * CR4.PAE and EFER.LMA must be set. */ - if (!(sregs->cr4 & X86_CR4_PAE_BIT) + if (!(sregs->cr4 & X86_CR4_PAE) || !(sregs->efer & EFER_LMA)) return -EINVAL; } else { From 490ae017f54e55bde382d45ea24bddfb6d1a0aaf Mon Sep 17 00:00:00 2001 From: Dennis Yang Date: Tue, 12 Dec 2017 18:21:40 +0800 Subject: [PATCH 17/55] dm thin metadata: THIN_MAX_CONCURRENT_LOCKS should be 6 For btree removal, there is a corner case that a single thread could takes 6 locks which is more than THIN_MAX_CONCURRENT_LOCKS(5) and leads to deadlock. A btree removal might eventually call rebalance_children()->rebalance3() to rebalance entries of three neighbor child nodes when shadow_spine has already acquired two write locks. In rebalance3(), it tries to shadow and acquire the write locks of all three child nodes. However, shadowing a child node requires acquiring a read lock of the original child node and a write lock of the new block. Although the read lock will be released after block shadowing, shadowing the third child node in rebalance3() could still take the sixth lock. (2 write locks for shadow_spine + 2 write locks for the first two child nodes's shadow + 1 write lock for the last child node's shadow + 1 read lock for the last child node) Cc: stable@vger.kernel.org Signed-off-by: Dennis Yang Acked-by: Joe Thornber Signed-off-by: Mike Snitzer --- drivers/md/dm-thin-metadata.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/drivers/md/dm-thin-metadata.c b/drivers/md/dm-thin-metadata.c index d31d18d9727c..36ef284ad086 100644 --- a/drivers/md/dm-thin-metadata.c +++ b/drivers/md/dm-thin-metadata.c @@ -80,10 +80,14 @@ #define SECTOR_TO_BLOCK_SHIFT 3 /* + * For btree insert: * 3 for btree insert + * 2 for btree lookup used within space map + * For btree remove: + * 2 for shadow spine + + * 4 for rebalance 3 child node */ -#define THIN_MAX_CONCURRENT_LOCKS 5 +#define THIN_MAX_CONCURRENT_LOCKS 6 /* This should be plenty */ #define SPACE_MAP_ROOT_SIZE 128 From bc68d0a43560e950850fc69b58f0f8254b28f6d6 Mon Sep 17 00:00:00 2001 From: Joe Thornber Date: Wed, 20 Dec 2017 09:56:06 +0000 Subject: [PATCH 18/55] dm btree: fix serious bug in btree_split_beneath() When inserting a new key/value pair into a btree we walk down the spine of btree nodes performing the following 2 operations: i) space for a new entry ii) adjusting the first key entry if the new key is lower than any in the node. If the _root_ node is full, the function btree_split_beneath() allocates 2 new nodes, and redistibutes the root nodes entries between them. The root node is left with 2 entries corresponding to the 2 new nodes. btree_split_beneath() then adjusts the spine to point to one of the two new children. This means the first key is never adjusted if the new key was lower, ie. operation (ii) gets missed out. This can result in the new key being 'lost' for a period; until another low valued key is inserted that will uncover it. This is a serious bug, and quite hard to make trigger in normal use. A reproducing test case ("thin create devices-in-reverse-order") is available as part of the thin-provision-tools project: https://github.com/jthornber/thin-provisioning-tools/blob/master/functional-tests/device-mapper/dm-tests.scm#L593 Fix the issue by changing btree_split_beneath() so it no longer adjusts the spine. Instead it unlocks both the new nodes, and lets the main loop in btree_insert_raw() relock the appropriate one and make any neccessary adjustments. Cc: stable@vger.kernel.org Reported-by: Monty Pavel Signed-off-by: Joe Thornber Signed-off-by: Mike Snitzer --- drivers/md/persistent-data/dm-btree.c | 19 ++----------------- 1 file changed, 2 insertions(+), 17 deletions(-) diff --git a/drivers/md/persistent-data/dm-btree.c b/drivers/md/persistent-data/dm-btree.c index f21ce6a3d4cf..58b319757b1e 100644 --- a/drivers/md/persistent-data/dm-btree.c +++ b/drivers/md/persistent-data/dm-btree.c @@ -683,23 +683,8 @@ static int btree_split_beneath(struct shadow_spine *s, uint64_t key) pn->keys[1] = rn->keys[0]; memcpy_disk(value_ptr(pn, 1), &val, sizeof(__le64)); - /* - * rejig the spine. This is ugly, since it knows too - * much about the spine - */ - if (s->nodes[0] != new_parent) { - unlock_block(s->info, s->nodes[0]); - s->nodes[0] = new_parent; - } - if (key < le64_to_cpu(rn->keys[0])) { - unlock_block(s->info, right); - s->nodes[1] = left; - } else { - unlock_block(s->info, left); - s->nodes[1] = right; - } - s->count = 2; - + unlock_block(s->info, left); + unlock_block(s->info, right); return 0; } From 27c7003697fc2c78f965984aa224ef26cd6b2949 Mon Sep 17 00:00:00 2001 From: Milan Broz Date: Wed, 3 Jan 2018 22:48:59 +0100 Subject: [PATCH 19/55] dm crypt: fix crash by adding missing check for auth key size If dm-crypt uses authenticated mode with separate MAC, there are two concatenated part of the key structure - key(s) for encryption and authentication key. Add a missing check for authenticated key length. If this key length is smaller than actually provided key, dm-crypt now properly fails instead of crashing. Fixes: ef43aa3806 ("dm crypt: add cryptographic data integrity protection (authenticated encryption)") Cc: stable@vger.kernel.org # 4.12+ Reported-by: Salah Coronya Signed-off-by: Milan Broz Signed-off-by: Mike Snitzer --- drivers/md/dm-crypt.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c index 9fc12f556534..4cc3809b2a3a 100644 --- a/drivers/md/dm-crypt.c +++ b/drivers/md/dm-crypt.c @@ -1954,10 +1954,15 @@ static int crypt_setkey(struct crypt_config *cc) /* Ignore extra keys (which are used for IV etc) */ subkey_size = crypt_subkey_size(cc); - if (crypt_integrity_hmac(cc)) + if (crypt_integrity_hmac(cc)) { + if (subkey_size < cc->key_mac_size) + return -EINVAL; + crypt_copy_authenckey(cc->authenc_key, cc->key, subkey_size - cc->key_mac_size, cc->key_mac_size); + } + for (i = 0; i < cc->tfms_count; i++) { if (crypt_integrity_hmac(cc)) r = crypto_aead_setkey(cc->cipher_tfm.tfms_aead[i], From 717f4b1c52135f279112df82583e0c77e80f90de Mon Sep 17 00:00:00 2001 From: Mikulas Patocka Date: Wed, 10 Jan 2018 09:32:47 -0500 Subject: [PATCH 20/55] dm integrity: don't store cipher request on the stack Some asynchronous cipher implementations may use DMA. The stack may be mapped in the vmalloc area that doesn't support DMA. Therefore, the cipher request and initialization vector shouldn't be on the stack. Fix this by allocating the request and iv with kmalloc. Cc: stable@vger.kernel.org # v4.12+ Signed-off-by: Mikulas Patocka Signed-off-by: Mike Snitzer --- drivers/md/dm-integrity.c | 49 +++++++++++++++++++++++++++++---------- 1 file changed, 37 insertions(+), 12 deletions(-) diff --git a/drivers/md/dm-integrity.c b/drivers/md/dm-integrity.c index 05c7bfd0c9d9..46d7c8749222 100644 --- a/drivers/md/dm-integrity.c +++ b/drivers/md/dm-integrity.c @@ -2559,7 +2559,8 @@ static int create_journal(struct dm_integrity_c *ic, char **error) int r = 0; unsigned i; __u64 journal_pages, journal_desc_size, journal_tree_size; - unsigned char *crypt_data = NULL; + unsigned char *crypt_data = NULL, *crypt_iv = NULL; + struct skcipher_request *req = NULL; ic->commit_ids[0] = cpu_to_le64(0x1111111111111111ULL); ic->commit_ids[1] = cpu_to_le64(0x2222222222222222ULL); @@ -2617,9 +2618,20 @@ static int create_journal(struct dm_integrity_c *ic, char **error) if (blocksize == 1) { struct scatterlist *sg; - SKCIPHER_REQUEST_ON_STACK(req, ic->journal_crypt); - unsigned char iv[ivsize]; - skcipher_request_set_tfm(req, ic->journal_crypt); + + req = skcipher_request_alloc(ic->journal_crypt, GFP_KERNEL); + if (!req) { + *error = "Could not allocate crypt request"; + r = -ENOMEM; + goto bad; + } + + crypt_iv = kmalloc(ivsize, GFP_KERNEL); + if (!crypt_iv) { + *error = "Could not allocate iv"; + r = -ENOMEM; + goto bad; + } ic->journal_xor = dm_integrity_alloc_page_list(ic); if (!ic->journal_xor) { @@ -2641,9 +2653,9 @@ static int create_journal(struct dm_integrity_c *ic, char **error) sg_set_buf(&sg[i], va, PAGE_SIZE); } sg_set_buf(&sg[i], &ic->commit_ids, sizeof ic->commit_ids); - memset(iv, 0x00, ivsize); + memset(crypt_iv, 0x00, ivsize); - skcipher_request_set_crypt(req, sg, sg, PAGE_SIZE * ic->journal_pages + sizeof ic->commit_ids, iv); + skcipher_request_set_crypt(req, sg, sg, PAGE_SIZE * ic->journal_pages + sizeof ic->commit_ids, crypt_iv); init_completion(&comp.comp); comp.in_flight = (atomic_t)ATOMIC_INIT(1); if (do_crypt(true, req, &comp)) @@ -2659,10 +2671,22 @@ static int create_journal(struct dm_integrity_c *ic, char **error) crypto_free_skcipher(ic->journal_crypt); ic->journal_crypt = NULL; } else { - SKCIPHER_REQUEST_ON_STACK(req, ic->journal_crypt); - unsigned char iv[ivsize]; unsigned crypt_len = roundup(ivsize, blocksize); + req = skcipher_request_alloc(ic->journal_crypt, GFP_KERNEL); + if (!req) { + *error = "Could not allocate crypt request"; + r = -ENOMEM; + goto bad; + } + + crypt_iv = kmalloc(ivsize, GFP_KERNEL); + if (!crypt_iv) { + *error = "Could not allocate iv"; + r = -ENOMEM; + goto bad; + } + crypt_data = kmalloc(crypt_len, GFP_KERNEL); if (!crypt_data) { *error = "Unable to allocate crypt data"; @@ -2670,8 +2694,6 @@ static int create_journal(struct dm_integrity_c *ic, char **error) goto bad; } - skcipher_request_set_tfm(req, ic->journal_crypt); - ic->journal_scatterlist = dm_integrity_alloc_journal_scatterlist(ic, ic->journal); if (!ic->journal_scatterlist) { *error = "Unable to allocate sg list"; @@ -2695,12 +2717,12 @@ static int create_journal(struct dm_integrity_c *ic, char **error) struct skcipher_request *section_req; __u32 section_le = cpu_to_le32(i); - memset(iv, 0x00, ivsize); + memset(crypt_iv, 0x00, ivsize); memset(crypt_data, 0x00, crypt_len); memcpy(crypt_data, §ion_le, min((size_t)crypt_len, sizeof(section_le))); sg_init_one(&sg, crypt_data, crypt_len); - skcipher_request_set_crypt(req, &sg, &sg, crypt_len, iv); + skcipher_request_set_crypt(req, &sg, &sg, crypt_len, crypt_iv); init_completion(&comp.comp); comp.in_flight = (atomic_t)ATOMIC_INIT(1); if (do_crypt(true, req, &comp)) @@ -2758,6 +2780,9 @@ static int create_journal(struct dm_integrity_c *ic, char **error) } bad: kfree(crypt_data); + kfree(crypt_iv); + skcipher_request_free(req); + return r; } From dc94902bde1e158cd19c4deab208e5d6eb382a44 Mon Sep 17 00:00:00 2001 From: Ondrej Kozina Date: Fri, 12 Jan 2018 16:30:32 +0100 Subject: [PATCH 21/55] dm crypt: wipe kernel key copy after IV initialization Loading key via kernel keyring service erases the internal key copy immediately after we pass it in crypto layer. This is wrong because IV is initialized later and we use wrong key for the initialization (instead of real key there's just zeroed block). The bug may cause data corruption if key is loaded via kernel keyring service first and later same crypt device is reactivated using exactly same key in hexbyte representation, or vice versa. The bug (and fix) affects only ciphers using following IVs: essiv, lmk and tcw. Fixes: c538f6ec9f56 ("dm crypt: add ability to use keys from the kernel key retention service") Cc: stable@vger.kernel.org # 4.10+ Signed-off-by: Ondrej Kozina Reviewed-by: Milan Broz Signed-off-by: Mike Snitzer --- drivers/md/dm-crypt.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c index 4cc3809b2a3a..971241409c30 100644 --- a/drivers/md/dm-crypt.c +++ b/drivers/md/dm-crypt.c @@ -2058,9 +2058,6 @@ static int crypt_set_keyring_key(struct crypt_config *cc, const char *key_string ret = crypt_setkey(cc); - /* wipe the kernel key payload copy in each case */ - memset(cc->key, 0, cc->key_size * sizeof(u8)); - if (!ret) { set_bit(DM_CRYPT_KEY_VALID, &cc->flags); kzfree(cc->key_string); @@ -2528,6 +2525,10 @@ static int crypt_ctr_cipher(struct dm_target *ti, char *cipher_in, char *key) } } + /* wipe the kernel key payload copy */ + if (cc->key_string) + memset(cc->key, 0, cc->key_size * sizeof(u8)); + return ret; } @@ -2966,6 +2967,9 @@ static int crypt_message(struct dm_target *ti, unsigned argc, char **argv) return ret; if (cc->iv_gen_ops && cc->iv_gen_ops->init) ret = cc->iv_gen_ops->init(cc); + /* wipe the kernel key payload copy */ + if (cc->key_string) + memset(cc->key, 0, cc->key_size * sizeof(u8)); return ret; } if (argc == 2 && !strcasecmp(argv[1], "wipe")) { @@ -3012,7 +3016,7 @@ static void crypt_io_hints(struct dm_target *ti, struct queue_limits *limits) static struct target_type crypt_target = { .name = "crypt", - .version = {1, 18, 0}, + .version = {1, 18, 1}, .module = THIS_MODULE, .ctr = crypt_ctr, .dtr = crypt_dtr, From 3cc2e57c4beabcbbaa46e1ac6d77ca8276a4a42d Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Wed, 17 Jan 2018 11:24:26 +0000 Subject: [PATCH 22/55] dm crypt: fix error return code in crypt_ctr() Fix to return error code -ENOMEM from the mempool_create_kmalloc_pool() error handling case instead of 0, as done elsewhere in this function. Fixes: ef43aa38063a6 ("dm crypt: add cryptographic data integrity protection (authenticated encryption)") Cc: stable@vger.kernel.org # 4.12+ Signed-off-by: Wei Yongjun Signed-off-by: Mike Snitzer --- drivers/md/dm-crypt.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c index 971241409c30..554d60394c06 100644 --- a/drivers/md/dm-crypt.c +++ b/drivers/md/dm-crypt.c @@ -2746,6 +2746,7 @@ static int crypt_ctr(struct dm_target *ti, unsigned int argc, char **argv) cc->tag_pool_max_sectors * cc->on_disk_tag_size); if (!cc->tag_pool) { ti->error = "Cannot allocate integrity tags mempool"; + ret = -ENOMEM; goto bad; } From a0c9259dc4e1923a98356967ce8b732da1979df8 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Wed, 17 Jan 2018 16:01:47 +0100 Subject: [PATCH 23/55] irq/matrix: Spread interrupts on allocation Keith reported an issue with vector space exhaustion on a server machine which is caused by the i40e driver allocating 168 MSI interrupts when the driver is initialized, even when most of these interrupts are not used at all. The x86 vector allocation code tries to avoid the immediate allocation with the reservation mode, but the card uses MSI and does not support MSI entry masking, which prevents reservation mode and requires immediate vector allocation. The matrix allocator is a bit naive and prefers the first CPU in the cpumask which describes the possible target CPUs for an allocation. That results in allocating all 168 vectors on CPU0 which later causes vector space exhaustion when the NVMe driver tries to allocate managed interrupts on each CPU for the per CPU queues. Avoid this by finding the CPU which has the lowest vector allocation count to spread out the non managed interrupt accross the possible target CPUs. Fixes: 2f75d9e1c905 ("genirq: Implement bitmap matrix allocator") Reported-by: Keith Busch Signed-off-by: Thomas Gleixner Tested-by: Keith Busch Link: https://lkml.kernel.org/r/alpine.DEB.2.20.1801171557330.1777@nanos --- kernel/irq/matrix.c | 20 ++++++++++++++------ 1 file changed, 14 insertions(+), 6 deletions(-) diff --git a/kernel/irq/matrix.c b/kernel/irq/matrix.c index 0ba0dd8863a7..5187dfe809ac 100644 --- a/kernel/irq/matrix.c +++ b/kernel/irq/matrix.c @@ -321,15 +321,23 @@ void irq_matrix_remove_reserved(struct irq_matrix *m) int irq_matrix_alloc(struct irq_matrix *m, const struct cpumask *msk, bool reserved, unsigned int *mapped_cpu) { - unsigned int cpu; + unsigned int cpu, best_cpu, maxavl = 0; + struct cpumap *cm; + unsigned int bit; + best_cpu = UINT_MAX; for_each_cpu(cpu, msk) { - struct cpumap *cm = per_cpu_ptr(m->maps, cpu); - unsigned int bit; + cm = per_cpu_ptr(m->maps, cpu); - if (!cm->online) + if (!cm->online || cm->available <= maxavl) continue; + best_cpu = cpu; + maxavl = cm->available; + } + + if (maxavl) { + cm = per_cpu_ptr(m->maps, best_cpu); bit = matrix_alloc_area(m, cm, 1, false); if (bit < m->alloc_end) { cm->allocated++; @@ -338,8 +346,8 @@ int irq_matrix_alloc(struct irq_matrix *m, const struct cpumask *msk, m->global_available--; if (reserved) m->global_reserved--; - *mapped_cpu = cpu; - trace_irq_matrix_alloc(bit, cpu, m, cm); + *mapped_cpu = best_cpu; + trace_irq_matrix_alloc(bit, best_cpu, m, cm); return bit; } } From f23d74f6c66c3697e032550eeef3f640391a3a7d Mon Sep 17 00:00:00 2001 From: Tom Lendacky Date: Wed, 17 Jan 2018 17:41:41 -0600 Subject: [PATCH 24/55] x86/mm: Rework wbinvd, hlt operation in stop_this_cpu() Some issues have been reported with the for loop in stop_this_cpu() that issues the 'wbinvd; hlt' sequence. Reverting this sequence to halt() has been shown to resolve the issue. However, the wbinvd is needed when running with SME. The reason for the wbinvd is to prevent cache flush races between encrypted and non-encrypted entries that have the same physical address. This can occur when kexec'ing from memory encryption active to inactive or vice-versa. The important thing is to not have outside of kernel text memory references (such as stack usage), so the usage of the native_*() functions is needed since these expand as inline asm sequences. So instead of reverting the change, rework the sequence. Move the wbinvd instruction outside of the for loop as native_wbinvd() and make its execution conditional on X86_FEATURE_SME. In the for loop, change the asm 'wbinvd; hlt' sequence back to a halt sequence but use the native_halt() call. Fixes: bba4ed011a52 ("x86/mm, kexec: Allow kexec to be used with SME") Reported-by: Dave Young Signed-off-by: Tom Lendacky Signed-off-by: Thomas Gleixner Tested-by: Dave Young Cc: Juergen Gross Cc: Tony Luck Cc: Yu Chen Cc: Baoquan He Cc: Linus Torvalds Cc: kexec@lists.infradead.org Cc: ebiederm@redhat.com Cc: Borislav Petkov Cc: Rui Zhang Cc: Arjan van de Ven Cc: Boris Ostrovsky Cc: Dan Williams Cc: stable@vger.kernel.org Link: https://lkml.kernel.org/r/20180117234141.21184.44067.stgit@tlendack-t1.amdoffice.net --- arch/x86/kernel/process.c | 25 +++++++++++++++---------- 1 file changed, 15 insertions(+), 10 deletions(-) diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index 832a6acd730f..cb368c2a22ab 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -380,19 +380,24 @@ void stop_this_cpu(void *dummy) disable_local_APIC(); mcheck_cpu_clear(this_cpu_ptr(&cpu_info)); + /* + * Use wbinvd on processors that support SME. This provides support + * for performing a successful kexec when going from SME inactive + * to SME active (or vice-versa). The cache must be cleared so that + * if there are entries with the same physical address, both with and + * without the encryption bit, they don't race each other when flushed + * and potentially end up with the wrong entry being committed to + * memory. + */ + if (boot_cpu_has(X86_FEATURE_SME)) + native_wbinvd(); for (;;) { /* - * Use wbinvd followed by hlt to stop the processor. This - * provides support for kexec on a processor that supports - * SME. With kexec, going from SME inactive to SME active - * requires clearing cache entries so that addresses without - * the encryption bit set don't corrupt the same physical - * address that has the encryption bit set when caches are - * flushed. To achieve this a wbinvd is performed followed by - * a hlt. Even if the processor is not in the kexec/SME - * scenario this only adds a wbinvd to a halting processor. + * Use native_halt() so that memory contents don't change + * (stack usage and variables) after possibly issuing the + * native_wbinvd() above. */ - asm volatile("wbinvd; hlt" : : : "memory"); + native_halt(); } } From f35764e74f0e45e1d89ca9ed9c8299f5e746a4d1 Mon Sep 17 00:00:00 2001 From: James Hogan Date: Mon, 15 Jan 2018 20:54:35 +0000 Subject: [PATCH 25/55] MIPS: Fix undefined reference to physical_memsize Since commit d41e6858ba58 ("MIPS: Kconfig: Set default MIPS system type as generic") switched the default platform to the "generic" platform, allmodconfig has been failing with the following linker error (among other errors): arch/mips/kernel/vpe-mt.o In function `vpe_run': (.text+0x59c): undefined reference to `physical_memsize' The Lantiq platform already worked around the same issue in commit 9050d50e2244 ("MIPS: lantiq: Set physical_memsize") by declaring physical_memsize with the initial value of 0 (on the assumption that the actual memory size will be hard-coded in the loaded VPE firmware), and the Malta platform already provided physical_memsize. Since all other platforms will fail to link with the VPE loader enabled, only allow Lantiq and Malta platforms to enable it, by way of a SYS_SUPPORTS_VPE_LOADER which is selected by those two platforms and which MIPS_VPE_LOADER depends on. SYS_SUPPORTS_MULTITHREADING is now a dependency of SYS_SUPPORTS_VPE_LOADER so that Kconfig emits a warning if SYS_SUPPORTS_VPE_LOADER is selected without SYS_SUPPORTS_MULTITHREADING. Fixes: d41e6858ba58 ("MIPS: Kconfig: Set default MIPS system type as generic") Signed-off-by: James Hogan Cc: Ralf Baechle Cc: John Crispin Cc: Hauke Mehrtens Cc: Paul Burton Cc: Matt Redfearn Cc: Guenter Roeck Cc: linux-mips@linux-mips.org Tested-by: Guenter Roeck Patchwork: https://patchwork.linux-mips.org/patch/18453/ --- arch/mips/Kconfig | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig index 659e0079487f..8e0b3702f1c0 100644 --- a/arch/mips/Kconfig +++ b/arch/mips/Kconfig @@ -390,6 +390,7 @@ config LANTIQ select SYS_SUPPORTS_32BIT_KERNEL select SYS_SUPPORTS_MIPS16 select SYS_SUPPORTS_MULTITHREADING + select SYS_SUPPORTS_VPE_LOADER select SYS_HAS_EARLY_PRINTK select GPIOLIB select SWAP_IO_SPACE @@ -517,6 +518,7 @@ config MIPS_MALTA select SYS_SUPPORTS_MIPS16 select SYS_SUPPORTS_MULTITHREADING select SYS_SUPPORTS_SMARTMIPS + select SYS_SUPPORTS_VPE_LOADER select SYS_SUPPORTS_ZBOOT select SYS_SUPPORTS_RELOCATABLE select USE_OF @@ -2282,9 +2284,16 @@ config MIPSR2_TO_R6_EMULATOR The only reason this is a build-time option is to save ~14K from the final kernel image. +config SYS_SUPPORTS_VPE_LOADER + bool + depends on SYS_SUPPORTS_MULTITHREADING + help + Indicates that the platform supports the VPE loader, and provides + physical_memsize. + config MIPS_VPE_LOADER bool "VPE loader support." - depends on SYS_SUPPORTS_MULTITHREADING && MODULES + depends on SYS_SUPPORTS_VPE_LOADER && MODULES select CPU_MIPSR2_IRQ_VI select CPU_MIPSR2_IRQ_EI select MIPS_MT From 18696edce11e010a1151a779490d6025b497e400 Mon Sep 17 00:00:00 2001 From: James Hogan Date: Fri, 10 Nov 2017 21:04:31 +0000 Subject: [PATCH 26/55] MAINTAINERS: Add James as MIPS co-maintainer I've been taking on some co-maintainer duties already, so lets make it official in the MAINTAINERS file. Link: https://lkml.kernel.org/r/33db77a2-32e4-6b2c-d463-9d116ba55623@imgtec.com Link: https://lkml.kernel.org/r/20171207110549.GM27409@jhogan-linux.mipstec.com Signed-off-by: James Hogan Acked-by: Paul Burton Acked-by: Aaro Koskinen Acked-by: David Daney Acked-by: Florian Fainelli Acked-by: Matt Redfearn Cc: Ralf Baechle Cc: John Crispin Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/18211/ --- MAINTAINERS | 1 + 1 file changed, 1 insertion(+) diff --git a/MAINTAINERS b/MAINTAINERS index d76af75a653a..018d50ed382e 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -9085,6 +9085,7 @@ F: drivers/usb/image/microtek.* MIPS M: Ralf Baechle +M: James Hogan L: linux-mips@linux-mips.org W: http://www.linux-mips.org/ T: git git://git.linux-mips.org/pub/scm/ralf/linux.git From 3214d01f139b7544e870fc0b7fcce8da13c1cb51 Mon Sep 17 00:00:00 2001 From: Paul Mackerras Date: Mon, 15 Jan 2018 16:06:47 +1100 Subject: [PATCH 27/55] KVM: PPC: Book3S: Provide information about hardware/firmware CVE workarounds This adds a new ioctl, KVM_PPC_GET_CPU_CHAR, that gives userspace information about the underlying machine's level of vulnerability to the recently announced vulnerabilities CVE-2017-5715, CVE-2017-5753 and CVE-2017-5754, and whether the machine provides instructions to assist software to work around the vulnerabilities. The ioctl returns two u64 words describing characteristics of the CPU and required software behaviour respectively, plus two mask words which indicate which bits have been filled in by the kernel, for extensibility. The bit definitions are the same as for the new H_GET_CPU_CHARACTERISTICS hypercall. There is also a new capability, KVM_CAP_PPC_GET_CPU_CHAR, which indicates whether the new ioctl is available. Signed-off-by: Paul Mackerras --- Documentation/virtual/kvm/api.txt | 46 ++++++++++ arch/powerpc/include/uapi/asm/kvm.h | 25 ++++++ arch/powerpc/kvm/powerpc.c | 131 ++++++++++++++++++++++++++++ include/uapi/linux/kvm.h | 3 + 4 files changed, 205 insertions(+) diff --git a/Documentation/virtual/kvm/api.txt b/Documentation/virtual/kvm/api.txt index 57d3ee9e4bde..fc3ae951bc07 100644 --- a/Documentation/virtual/kvm/api.txt +++ b/Documentation/virtual/kvm/api.txt @@ -3403,6 +3403,52 @@ invalid, if invalid pages are written to (e.g. after the end of memory) or if no page table is present for the addresses (e.g. when using hugepages). +4.108 KVM_PPC_GET_CPU_CHAR + +Capability: KVM_CAP_PPC_GET_CPU_CHAR +Architectures: powerpc +Type: vm ioctl +Parameters: struct kvm_ppc_cpu_char (out) +Returns: 0 on successful completion + -EFAULT if struct kvm_ppc_cpu_char cannot be written + +This ioctl gives userspace information about certain characteristics +of the CPU relating to speculative execution of instructions and +possible information leakage resulting from speculative execution (see +CVE-2017-5715, CVE-2017-5753 and CVE-2017-5754). The information is +returned in struct kvm_ppc_cpu_char, which looks like this: + +struct kvm_ppc_cpu_char { + __u64 character; /* characteristics of the CPU */ + __u64 behaviour; /* recommended software behaviour */ + __u64 character_mask; /* valid bits in character */ + __u64 behaviour_mask; /* valid bits in behaviour */ +}; + +For extensibility, the character_mask and behaviour_mask fields +indicate which bits of character and behaviour have been filled in by +the kernel. If the set of defined bits is extended in future then +userspace will be able to tell whether it is running on a kernel that +knows about the new bits. + +The character field describes attributes of the CPU which can help +with preventing inadvertent information disclosure - specifically, +whether there is an instruction to flash-invalidate the L1 data cache +(ori 30,30,0 or mtspr SPRN_TRIG2,rN), whether the L1 data cache is set +to a mode where entries can only be used by the thread that created +them, whether the bcctr[l] instruction prevents speculation, and +whether a speculation barrier instruction (ori 31,31,0) is provided. + +The behaviour field describes actions that software should take to +prevent inadvertent information disclosure, and thus describes which +vulnerabilities the hardware is subject to; specifically whether the +L1 data cache should be flushed when returning to user mode from the +kernel, and whether a speculation barrier should be placed between an +array bounds check and the array access. + +These fields use the same bit definitions as the new +H_GET_CPU_CHARACTERISTICS hypercall. + 5. The kvm_run structure ------------------------ diff --git a/arch/powerpc/include/uapi/asm/kvm.h b/arch/powerpc/include/uapi/asm/kvm.h index 61d6049f4c1e..637b7263cb86 100644 --- a/arch/powerpc/include/uapi/asm/kvm.h +++ b/arch/powerpc/include/uapi/asm/kvm.h @@ -443,6 +443,31 @@ struct kvm_ppc_rmmu_info { __u32 ap_encodings[8]; }; +/* For KVM_PPC_GET_CPU_CHAR */ +struct kvm_ppc_cpu_char { + __u64 character; /* characteristics of the CPU */ + __u64 behaviour; /* recommended software behaviour */ + __u64 character_mask; /* valid bits in character */ + __u64 behaviour_mask; /* valid bits in behaviour */ +}; + +/* + * Values for character and character_mask. + * These are identical to the values used by H_GET_CPU_CHARACTERISTICS. + */ +#define KVM_PPC_CPU_CHAR_SPEC_BAR_ORI31 (1ULL << 63) +#define KVM_PPC_CPU_CHAR_BCCTRL_SERIALISED (1ULL << 62) +#define KVM_PPC_CPU_CHAR_L1D_FLUSH_ORI30 (1ULL << 61) +#define KVM_PPC_CPU_CHAR_L1D_FLUSH_TRIG2 (1ULL << 60) +#define KVM_PPC_CPU_CHAR_L1D_THREAD_PRIV (1ULL << 59) +#define KVM_PPC_CPU_CHAR_BR_HINT_HONOURED (1ULL << 58) +#define KVM_PPC_CPU_CHAR_MTTRIG_THR_RECONF (1ULL << 57) +#define KVM_PPC_CPU_CHAR_COUNT_CACHE_DIS (1ULL << 56) + +#define KVM_PPC_CPU_BEHAV_FAVOUR_SECURITY (1ULL << 63) +#define KVM_PPC_CPU_BEHAV_L1D_FLUSH_PR (1ULL << 62) +#define KVM_PPC_CPU_BEHAV_BNDS_CHK_SPEC_BAR (1ULL << 61) + /* Per-vcpu XICS interrupt controller state */ #define KVM_REG_PPC_ICP_STATE (KVM_REG_PPC | KVM_REG_SIZE_U64 | 0x8c) diff --git a/arch/powerpc/kvm/powerpc.c b/arch/powerpc/kvm/powerpc.c index 1915e86cef6f..0a7c88786ec0 100644 --- a/arch/powerpc/kvm/powerpc.c +++ b/arch/powerpc/kvm/powerpc.c @@ -39,6 +39,10 @@ #include #include #include +#ifdef CONFIG_PPC_PSERIES +#include +#include +#endif #include "timing.h" #include "irq.h" @@ -548,6 +552,7 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext) #ifdef CONFIG_KVM_XICS case KVM_CAP_IRQ_XICS: #endif + case KVM_CAP_PPC_GET_CPU_CHAR: r = 1; break; @@ -1759,6 +1764,124 @@ static int kvm_vm_ioctl_enable_cap(struct kvm *kvm, return r; } +#ifdef CONFIG_PPC_BOOK3S_64 +/* + * These functions check whether the underlying hardware is safe + * against attacks based on observing the effects of speculatively + * executed instructions, and whether it supplies instructions for + * use in workarounds. The information comes from firmware, either + * via the device tree on powernv platforms or from an hcall on + * pseries platforms. + */ +#ifdef CONFIG_PPC_PSERIES +static int pseries_get_cpu_char(struct kvm_ppc_cpu_char *cp) +{ + struct h_cpu_char_result c; + unsigned long rc; + + if (!machine_is(pseries)) + return -ENOTTY; + + rc = plpar_get_cpu_characteristics(&c); + if (rc == H_SUCCESS) { + cp->character = c.character; + cp->behaviour = c.behaviour; + cp->character_mask = KVM_PPC_CPU_CHAR_SPEC_BAR_ORI31 | + KVM_PPC_CPU_CHAR_BCCTRL_SERIALISED | + KVM_PPC_CPU_CHAR_L1D_FLUSH_ORI30 | + KVM_PPC_CPU_CHAR_L1D_FLUSH_TRIG2 | + KVM_PPC_CPU_CHAR_L1D_THREAD_PRIV | + KVM_PPC_CPU_CHAR_BR_HINT_HONOURED | + KVM_PPC_CPU_CHAR_MTTRIG_THR_RECONF | + KVM_PPC_CPU_CHAR_COUNT_CACHE_DIS; + cp->behaviour_mask = KVM_PPC_CPU_BEHAV_FAVOUR_SECURITY | + KVM_PPC_CPU_BEHAV_L1D_FLUSH_PR | + KVM_PPC_CPU_BEHAV_BNDS_CHK_SPEC_BAR; + } + return 0; +} +#else +static int pseries_get_cpu_char(struct kvm_ppc_cpu_char *cp) +{ + return -ENOTTY; +} +#endif + +static inline bool have_fw_feat(struct device_node *fw_features, + const char *state, const char *name) +{ + struct device_node *np; + bool r = false; + + np = of_get_child_by_name(fw_features, name); + if (np) { + r = of_property_read_bool(np, state); + of_node_put(np); + } + return r; +} + +static int kvmppc_get_cpu_char(struct kvm_ppc_cpu_char *cp) +{ + struct device_node *np, *fw_features; + int r; + + memset(cp, 0, sizeof(*cp)); + r = pseries_get_cpu_char(cp); + if (r != -ENOTTY) + return r; + + np = of_find_node_by_name(NULL, "ibm,opal"); + if (np) { + fw_features = of_get_child_by_name(np, "fw-features"); + of_node_put(np); + if (!fw_features) + return 0; + if (have_fw_feat(fw_features, "enabled", + "inst-spec-barrier-ori31,31,0")) + cp->character |= KVM_PPC_CPU_CHAR_SPEC_BAR_ORI31; + if (have_fw_feat(fw_features, "enabled", + "fw-bcctrl-serialized")) + cp->character |= KVM_PPC_CPU_CHAR_BCCTRL_SERIALISED; + if (have_fw_feat(fw_features, "enabled", + "inst-l1d-flush-ori30,30,0")) + cp->character |= KVM_PPC_CPU_CHAR_L1D_FLUSH_ORI30; + if (have_fw_feat(fw_features, "enabled", + "inst-l1d-flush-trig2")) + cp->character |= KVM_PPC_CPU_CHAR_L1D_FLUSH_TRIG2; + if (have_fw_feat(fw_features, "enabled", + "fw-l1d-thread-split")) + cp->character |= KVM_PPC_CPU_CHAR_L1D_THREAD_PRIV; + if (have_fw_feat(fw_features, "enabled", + "fw-count-cache-disabled")) + cp->character |= KVM_PPC_CPU_CHAR_COUNT_CACHE_DIS; + cp->character_mask = KVM_PPC_CPU_CHAR_SPEC_BAR_ORI31 | + KVM_PPC_CPU_CHAR_BCCTRL_SERIALISED | + KVM_PPC_CPU_CHAR_L1D_FLUSH_ORI30 | + KVM_PPC_CPU_CHAR_L1D_FLUSH_TRIG2 | + KVM_PPC_CPU_CHAR_L1D_THREAD_PRIV | + KVM_PPC_CPU_CHAR_COUNT_CACHE_DIS; + + if (have_fw_feat(fw_features, "enabled", + "speculation-policy-favor-security")) + cp->behaviour |= KVM_PPC_CPU_BEHAV_FAVOUR_SECURITY; + if (!have_fw_feat(fw_features, "disabled", + "needs-l1d-flush-msr-pr-0-to-1")) + cp->behaviour |= KVM_PPC_CPU_BEHAV_L1D_FLUSH_PR; + if (!have_fw_feat(fw_features, "disabled", + "needs-spec-barrier-for-bound-checks")) + cp->behaviour |= KVM_PPC_CPU_BEHAV_BNDS_CHK_SPEC_BAR; + cp->behaviour_mask = KVM_PPC_CPU_BEHAV_FAVOUR_SECURITY | + KVM_PPC_CPU_BEHAV_L1D_FLUSH_PR | + KVM_PPC_CPU_BEHAV_BNDS_CHK_SPEC_BAR; + + of_node_put(fw_features); + } + + return 0; +} +#endif + long kvm_arch_vm_ioctl(struct file *filp, unsigned int ioctl, unsigned long arg) { @@ -1861,6 +1984,14 @@ long kvm_arch_vm_ioctl(struct file *filp, r = -EFAULT; break; } + case KVM_PPC_GET_CPU_CHAR: { + struct kvm_ppc_cpu_char cpuchar; + + r = kvmppc_get_cpu_char(&cpuchar); + if (r >= 0 && copy_to_user(argp, &cpuchar, sizeof(cpuchar))) + r = -EFAULT; + break; + } default: { struct kvm *kvm = filp->private_data; r = kvm->arch.kvm_ops->arch_vm_ioctl(filp, ioctl, arg); diff --git a/include/uapi/linux/kvm.h b/include/uapi/linux/kvm.h index 496e59a2738b..7a99b98cf88e 100644 --- a/include/uapi/linux/kvm.h +++ b/include/uapi/linux/kvm.h @@ -932,6 +932,7 @@ struct kvm_ppc_resize_hpt { #define KVM_CAP_HYPERV_SYNIC2 148 #define KVM_CAP_HYPERV_VP_INDEX 149 #define KVM_CAP_S390_AIS_MIGRATION 150 +#define KVM_CAP_PPC_GET_CPU_CHAR 151 #ifdef KVM_CAP_IRQ_ROUTING @@ -1261,6 +1262,8 @@ struct kvm_s390_ucas_mapping { #define KVM_PPC_CONFIGURE_V3_MMU _IOW(KVMIO, 0xaf, struct kvm_ppc_mmuv3_cfg) /* Available with KVM_CAP_PPC_RADIX_MMU */ #define KVM_PPC_GET_RMMU_INFO _IOW(KVMIO, 0xb0, struct kvm_ppc_rmmu_info) +/* Available with KVM_CAP_PPC_GET_CPU_CHAR */ +#define KVM_PPC_GET_CPU_CHAR _IOR(KVMIO, 0xb1, struct kvm_ppc_cpu_char) /* ioctl for vm fd */ #define KVM_CREATE_DEVICE _IOWR(KVMIO, 0xe0, struct kvm_create_device) From 6f41c34d69eb005e7848716bbcafc979b35037d5 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Thu, 18 Jan 2018 16:28:26 +0100 Subject: [PATCH 28/55] x86/mce: Make machine check speculation protected The machine check idtentry uses an indirect branch directly from the low level code. This evades the speculation protection. Replace it by a direct call into C code and issue the indirect call there so the compiler can apply the proper speculation protection. Signed-off-by: Thomas Gleixner Reviewed-by:Borislav Petkov Reviewed-by: David Woodhouse Niced-by: Peter Zijlstra Cc: stable@vger.kernel.org Link: https://lkml.kernel.org/r/alpine.DEB.2.20.1801181626290.1847@nanos --- arch/x86/entry/entry_64.S | 2 +- arch/x86/include/asm/traps.h | 1 + arch/x86/kernel/cpu/mcheck/mce.c | 5 +++++ 3 files changed, 7 insertions(+), 1 deletion(-) diff --git a/arch/x86/entry/entry_64.S b/arch/x86/entry/entry_64.S index d54a0ede61d1..63f4320602a3 100644 --- a/arch/x86/entry/entry_64.S +++ b/arch/x86/entry/entry_64.S @@ -1258,7 +1258,7 @@ idtentry async_page_fault do_async_page_fault has_error_code=1 #endif #ifdef CONFIG_X86_MCE -idtentry machine_check has_error_code=0 paranoid=1 do_sym=*machine_check_vector(%rip) +idtentry machine_check do_mce has_error_code=0 paranoid=1 #endif /* diff --git a/arch/x86/include/asm/traps.h b/arch/x86/include/asm/traps.h index 31051f35cbb7..3de69330e6c5 100644 --- a/arch/x86/include/asm/traps.h +++ b/arch/x86/include/asm/traps.h @@ -88,6 +88,7 @@ dotraplinkage void do_simd_coprocessor_error(struct pt_regs *, long); #ifdef CONFIG_X86_32 dotraplinkage void do_iret_error(struct pt_regs *, long); #endif +dotraplinkage void do_mce(struct pt_regs *, long); static inline int get_si_code(unsigned long condition) { diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c index 3b413065c613..a9e898b71208 100644 --- a/arch/x86/kernel/cpu/mcheck/mce.c +++ b/arch/x86/kernel/cpu/mcheck/mce.c @@ -1788,6 +1788,11 @@ static void unexpected_machine_check(struct pt_regs *regs, long error_code) void (*machine_check_vector)(struct pt_regs *, long error_code) = unexpected_machine_check; +dotraplinkage void do_mce(struct pt_regs *regs, long error_code) +{ + machine_check_vector(regs, error_code); +} + /* * Called for each booted CPU to set up machine checks. * Must be called with preempt off: From 736e80a4213e9bbce40a7c050337047128b472ac Mon Sep 17 00:00:00 2001 From: Masami Hiramatsu Date: Fri, 19 Jan 2018 01:14:21 +0900 Subject: [PATCH 29/55] retpoline: Introduce start/end markers of indirect thunk Introduce start/end markers of __x86_indirect_thunk_* functions. To make it easy, consolidate .text.__x86.indirect_thunk.* sections to one .text.__x86.indirect_thunk section and put it in the end of kernel text section and adds __indirect_thunk_start/end so that other subsystem (e.g. kprobes) can identify it. Signed-off-by: Masami Hiramatsu Signed-off-by: Thomas Gleixner Acked-by: David Woodhouse Cc: Andi Kleen Cc: Peter Zijlstra Cc: Ananth N Mavinakayanahalli Cc: Arjan van de Ven Cc: Greg Kroah-Hartman Cc: stable@vger.kernel.org Link: https://lkml.kernel.org/r/151629206178.10241.6828804696410044771.stgit@devbox --- arch/x86/include/asm/nospec-branch.h | 3 +++ arch/x86/kernel/vmlinux.lds.S | 6 ++++++ arch/x86/lib/retpoline.S | 2 +- 3 files changed, 10 insertions(+), 1 deletion(-) diff --git a/arch/x86/include/asm/nospec-branch.h b/arch/x86/include/asm/nospec-branch.h index 7b45d8424150..19ba5ad19c65 100644 --- a/arch/x86/include/asm/nospec-branch.h +++ b/arch/x86/include/asm/nospec-branch.h @@ -194,6 +194,9 @@ enum spectre_v2_mitigation { SPECTRE_V2_IBRS, }; +extern char __indirect_thunk_start[]; +extern char __indirect_thunk_end[]; + /* * On VMEXIT we must ensure that no RSB predictions learned in the guest * can be followed in the host, by overwriting the RSB completely. Both diff --git a/arch/x86/kernel/vmlinux.lds.S b/arch/x86/kernel/vmlinux.lds.S index 1e413a9326aa..9b138a06c1a4 100644 --- a/arch/x86/kernel/vmlinux.lds.S +++ b/arch/x86/kernel/vmlinux.lds.S @@ -124,6 +124,12 @@ SECTIONS ASSERT(. - _entry_trampoline == PAGE_SIZE, "entry trampoline is too big"); #endif +#ifdef CONFIG_RETPOLINE + __indirect_thunk_start = .; + *(.text.__x86.indirect_thunk) + __indirect_thunk_end = .; +#endif + /* End of text section */ _etext = .; } :text = 0x9090 diff --git a/arch/x86/lib/retpoline.S b/arch/x86/lib/retpoline.S index cb45c6cb465f..d3415dc30f82 100644 --- a/arch/x86/lib/retpoline.S +++ b/arch/x86/lib/retpoline.S @@ -9,7 +9,7 @@ #include .macro THUNK reg - .section .text.__x86.indirect_thunk.\reg + .section .text.__x86.indirect_thunk ENTRY(__x86_indirect_thunk_\reg) CFI_STARTPROC From c1804a236894ecc942da7dc6c5abe209e56cba93 Mon Sep 17 00:00:00 2001 From: Masami Hiramatsu Date: Fri, 19 Jan 2018 01:14:51 +0900 Subject: [PATCH 30/55] kprobes/x86: Blacklist indirect thunk functions for kprobes Mark __x86_indirect_thunk_* functions as blacklist for kprobes because those functions can be called from anywhere in the kernel including blacklist functions of kprobes. Signed-off-by: Masami Hiramatsu Signed-off-by: Thomas Gleixner Acked-by: David Woodhouse Cc: Andi Kleen Cc: Peter Zijlstra Cc: Ananth N Mavinakayanahalli Cc: Arjan van de Ven Cc: Greg Kroah-Hartman Cc: stable@vger.kernel.org Link: https://lkml.kernel.org/r/151629209111.10241.5444852823378068683.stgit@devbox --- arch/x86/lib/retpoline.S | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/arch/x86/lib/retpoline.S b/arch/x86/lib/retpoline.S index d3415dc30f82..dfb2ba91b670 100644 --- a/arch/x86/lib/retpoline.S +++ b/arch/x86/lib/retpoline.S @@ -25,7 +25,8 @@ ENDPROC(__x86_indirect_thunk_\reg) * than one per register with the correct names. So we do it * the simple and nasty way... */ -#define EXPORT_THUNK(reg) EXPORT_SYMBOL(__x86_indirect_thunk_ ## reg) +#define __EXPORT_THUNK(sym) _ASM_NOKPROBE(sym); EXPORT_SYMBOL(sym) +#define EXPORT_THUNK(reg) __EXPORT_THUNK(__x86_indirect_thunk_ ## reg) #define GENERATE_THUNK(reg) THUNK reg ; EXPORT_THUNK(reg) GENERATE_THUNK(_ASM_AX) From c86a32c09f8ced67971a2310e3b0dda4d1749007 Mon Sep 17 00:00:00 2001 From: Masami Hiramatsu Date: Fri, 19 Jan 2018 01:15:20 +0900 Subject: [PATCH 31/55] kprobes/x86: Disable optimizing on the function jumps to indirect thunk Since indirect jump instructions will be replaced by jump to __x86_indirect_thunk_*, those jmp instruction must be treated as an indirect jump. Since optprobe prohibits to optimize probes in the function which uses an indirect jump, it also needs to find out the function which jump to __x86_indirect_thunk_* and disable optimization. Add a check that the jump target address is between the __indirect_thunk_start/end when optimizing kprobe. Signed-off-by: Masami Hiramatsu Signed-off-by: Thomas Gleixner Acked-by: David Woodhouse Cc: Andi Kleen Cc: Peter Zijlstra Cc: Ananth N Mavinakayanahalli Cc: Arjan van de Ven Cc: Greg Kroah-Hartman Cc: stable@vger.kernel.org Link: https://lkml.kernel.org/r/151629212062.10241.6991266100233002273.stgit@devbox --- arch/x86/kernel/kprobes/opt.c | 23 ++++++++++++++++++++++- 1 file changed, 22 insertions(+), 1 deletion(-) diff --git a/arch/x86/kernel/kprobes/opt.c b/arch/x86/kernel/kprobes/opt.c index 4f98aad38237..3668f28cf5fc 100644 --- a/arch/x86/kernel/kprobes/opt.c +++ b/arch/x86/kernel/kprobes/opt.c @@ -40,6 +40,7 @@ #include #include #include +#include #include "common.h" @@ -205,7 +206,7 @@ static int copy_optimized_instructions(u8 *dest, u8 *src) } /* Check whether insn is indirect jump */ -static int insn_is_indirect_jump(struct insn *insn) +static int __insn_is_indirect_jump(struct insn *insn) { return ((insn->opcode.bytes[0] == 0xff && (X86_MODRM_REG(insn->modrm.value) & 6) == 4) || /* Jump */ @@ -239,6 +240,26 @@ static int insn_jump_into_range(struct insn *insn, unsigned long start, int len) return (start <= target && target <= start + len); } +static int insn_is_indirect_jump(struct insn *insn) +{ + int ret = __insn_is_indirect_jump(insn); + +#ifdef CONFIG_RETPOLINE + /* + * Jump to x86_indirect_thunk_* is treated as an indirect jump. + * Note that even with CONFIG_RETPOLINE=y, the kernel compiled with + * older gcc may use indirect jump. So we add this check instead of + * replace indirect-jump check. + */ + if (!ret) + ret = insn_jump_into_range(insn, + (unsigned long)__indirect_thunk_start, + (unsigned long)__indirect_thunk_end - + (unsigned long)__indirect_thunk_start); +#endif + return ret; +} + /* Decode whole function to ensure any instructions don't jump into target */ static int can_optimize(unsigned long paddr) { From 98f0fceec7f84d80bc053e49e596088573086421 Mon Sep 17 00:00:00 2001 From: "zhenwei.pi" Date: Thu, 18 Jan 2018 09:04:52 +0800 Subject: [PATCH 32/55] x86/pti: Document fix wrong index In section <2. Runtime Cost>, fix wrong index. Signed-off-by: zhenwei.pi Signed-off-by: Thomas Gleixner Cc: dave.hansen@linux.intel.com Link: https://lkml.kernel.org/r/1516237492-27739-1-git-send-email-zhenwei.pi@youruncloud.com --- Documentation/x86/pti.txt | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Documentation/x86/pti.txt b/Documentation/x86/pti.txt index d11eff61fc9a..5cd58439ad2d 100644 --- a/Documentation/x86/pti.txt +++ b/Documentation/x86/pti.txt @@ -78,7 +78,7 @@ this protection comes at a cost: non-PTI SYSCALL entry code, so requires mapping fewer things into the userspace page tables. The downside is that stacks must be switched at entry time. - d. Global pages are disabled for all kernel structures not + c. Global pages are disabled for all kernel structures not mapped into both kernel and userspace page tables. This feature of the MMU allows different processes to share TLB entries mapping the kernel. Losing the feature means more From 3f7d875566d8e79c5e0b2c9a413e91b2c29e0854 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Wed, 17 Jan 2018 14:53:28 -0800 Subject: [PATCH 33/55] x86/retpoline: Optimize inline assembler for vmexit_fill_RSB The generated assembler for the C fill RSB inline asm operations has several issues: - The C code sets up the loop register, which is then immediately overwritten in __FILL_RETURN_BUFFER with the same value again. - The C code also passes in the iteration count in another register, which is not used at all. Remove these two unnecessary operations. Just rely on the single constant passed to the macro for the iterations. Signed-off-by: Andi Kleen Signed-off-by: Thomas Gleixner Acked-by: David Woodhouse Cc: dave.hansen@intel.com Cc: gregkh@linuxfoundation.org Cc: torvalds@linux-foundation.org Cc: arjan@linux.intel.com Link: https://lkml.kernel.org/r/20180117225328.15414-1-andi@firstfloor.org --- arch/x86/include/asm/nospec-branch.h | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/arch/x86/include/asm/nospec-branch.h b/arch/x86/include/asm/nospec-branch.h index 19ba5ad19c65..4ad41087ce0e 100644 --- a/arch/x86/include/asm/nospec-branch.h +++ b/arch/x86/include/asm/nospec-branch.h @@ -206,16 +206,17 @@ extern char __indirect_thunk_end[]; static inline void vmexit_fill_RSB(void) { #ifdef CONFIG_RETPOLINE - unsigned long loops = RSB_CLEAR_LOOPS / 2; + unsigned long loops; asm volatile (ANNOTATE_NOSPEC_ALTERNATIVE ALTERNATIVE("jmp 910f", __stringify(__FILL_RETURN_BUFFER(%0, RSB_CLEAR_LOOPS, %1)), X86_FEATURE_RETPOLINE) "910:" - : "=&r" (loops), ASM_CALL_CONSTRAINT - : "r" (loops) : "memory" ); + : "=r" (loops), ASM_CALL_CONSTRAINT + : : "memory" ); #endif } + #endif /* __ASSEMBLY__ */ #endif /* __NOSPEC_BRANCH_H__ */ From 35b3fde6203b932b2b1a5b53b3d8808abc9c4f60 Mon Sep 17 00:00:00 2001 From: Christian Borntraeger Date: Wed, 17 Jan 2018 14:44:34 +0100 Subject: [PATCH 34/55] KVM: s390: wire up bpb feature MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The new firmware interfaces for branch prediction behaviour changes are transparently available for the guest. Nevertheless, there is new state attached that should be migrated and properly resetted. Provide a mechanism for handling reset, migration and VSIE. Signed-off-by: Christian Borntraeger Reviewed-by: David Hildenbrand Reviewed-by: Cornelia Huck [Changed capability number to 152. - Radim] Signed-off-by: Radim Krčmář --- arch/s390/include/asm/kvm_host.h | 3 ++- arch/s390/include/uapi/asm/kvm.h | 5 ++++- arch/s390/kvm/kvm-s390.c | 12 ++++++++++++ arch/s390/kvm/vsie.c | 10 ++++++++++ include/uapi/linux/kvm.h | 1 + 5 files changed, 29 insertions(+), 2 deletions(-) diff --git a/arch/s390/include/asm/kvm_host.h b/arch/s390/include/asm/kvm_host.h index e14f381757f6..c1b0a9ac1dc8 100644 --- a/arch/s390/include/asm/kvm_host.h +++ b/arch/s390/include/asm/kvm_host.h @@ -207,7 +207,8 @@ struct kvm_s390_sie_block { __u16 ipa; /* 0x0056 */ __u32 ipb; /* 0x0058 */ __u32 scaoh; /* 0x005c */ - __u8 reserved60; /* 0x0060 */ +#define FPF_BPBC 0x20 + __u8 fpf; /* 0x0060 */ #define ECB_GS 0x40 #define ECB_TE 0x10 #define ECB_SRSI 0x04 diff --git a/arch/s390/include/uapi/asm/kvm.h b/arch/s390/include/uapi/asm/kvm.h index 38535a57fef8..4cdaa55fabfe 100644 --- a/arch/s390/include/uapi/asm/kvm.h +++ b/arch/s390/include/uapi/asm/kvm.h @@ -224,6 +224,7 @@ struct kvm_guest_debug_arch { #define KVM_SYNC_RICCB (1UL << 7) #define KVM_SYNC_FPRS (1UL << 8) #define KVM_SYNC_GSCB (1UL << 9) +#define KVM_SYNC_BPBC (1UL << 10) /* length and alignment of the sdnx as a power of two */ #define SDNXC 8 #define SDNXL (1UL << SDNXC) @@ -247,7 +248,9 @@ struct kvm_sync_regs { }; __u8 reserved[512]; /* for future vector expansion */ __u32 fpc; /* valid on KVM_SYNC_VRS or KVM_SYNC_FPRS */ - __u8 padding1[52]; /* riccb needs to be 64byte aligned */ + __u8 bpbc : 1; /* bp mode */ + __u8 reserved2 : 7; + __u8 padding1[51]; /* riccb needs to be 64byte aligned */ __u8 riccb[64]; /* runtime instrumentation controls block */ __u8 padding2[192]; /* sdnx needs to be 256byte aligned */ union { diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c index 2c93cbbcd15e..2598cf243b86 100644 --- a/arch/s390/kvm/kvm-s390.c +++ b/arch/s390/kvm/kvm-s390.c @@ -421,6 +421,9 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext) case KVM_CAP_S390_GS: r = test_facility(133); break; + case KVM_CAP_S390_BPB: + r = test_facility(82); + break; default: r = 0; } @@ -2198,6 +2201,8 @@ int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu) kvm_s390_set_prefix(vcpu, 0); if (test_kvm_facility(vcpu->kvm, 64)) vcpu->run->kvm_valid_regs |= KVM_SYNC_RICCB; + if (test_kvm_facility(vcpu->kvm, 82)) + vcpu->run->kvm_valid_regs |= KVM_SYNC_BPBC; if (test_kvm_facility(vcpu->kvm, 133)) vcpu->run->kvm_valid_regs |= KVM_SYNC_GSCB; /* fprs can be synchronized via vrs, even if the guest has no vx. With @@ -2339,6 +2344,7 @@ static void kvm_s390_vcpu_initial_reset(struct kvm_vcpu *vcpu) current->thread.fpu.fpc = 0; vcpu->arch.sie_block->gbea = 1; vcpu->arch.sie_block->pp = 0; + vcpu->arch.sie_block->fpf &= ~FPF_BPBC; vcpu->arch.pfault_token = KVM_S390_PFAULT_TOKEN_INVALID; kvm_clear_async_pf_completion_queue(vcpu); if (!kvm_s390_user_cpu_state_ctrl(vcpu->kvm)) @@ -3298,6 +3304,11 @@ static void sync_regs(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) vcpu->arch.sie_block->ecd |= ECD_HOSTREGMGMT; vcpu->arch.gs_enabled = 1; } + if ((kvm_run->kvm_dirty_regs & KVM_SYNC_BPBC) && + test_kvm_facility(vcpu->kvm, 82)) { + vcpu->arch.sie_block->fpf &= ~FPF_BPBC; + vcpu->arch.sie_block->fpf |= kvm_run->s.regs.bpbc ? FPF_BPBC : 0; + } save_access_regs(vcpu->arch.host_acrs); restore_access_regs(vcpu->run->s.regs.acrs); /* save host (userspace) fprs/vrs */ @@ -3344,6 +3355,7 @@ static void store_regs(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) kvm_run->s.regs.pft = vcpu->arch.pfault_token; kvm_run->s.regs.pfs = vcpu->arch.pfault_select; kvm_run->s.regs.pfc = vcpu->arch.pfault_compare; + kvm_run->s.regs.bpbc = (vcpu->arch.sie_block->fpf & FPF_BPBC) == FPF_BPBC; save_access_regs(vcpu->run->s.regs.acrs); restore_access_regs(vcpu->arch.host_acrs); /* Save guest register state */ diff --git a/arch/s390/kvm/vsie.c b/arch/s390/kvm/vsie.c index 5d6ae0326d9e..751348348477 100644 --- a/arch/s390/kvm/vsie.c +++ b/arch/s390/kvm/vsie.c @@ -223,6 +223,12 @@ static void unshadow_scb(struct kvm_vcpu *vcpu, struct vsie_page *vsie_page) memcpy(scb_o->gcr, scb_s->gcr, 128); scb_o->pp = scb_s->pp; + /* branch prediction */ + if (test_kvm_facility(vcpu->kvm, 82)) { + scb_o->fpf &= ~FPF_BPBC; + scb_o->fpf |= scb_s->fpf & FPF_BPBC; + } + /* interrupt intercept */ switch (scb_s->icptcode) { case ICPT_PROGI: @@ -265,6 +271,7 @@ static int shadow_scb(struct kvm_vcpu *vcpu, struct vsie_page *vsie_page) scb_s->ecb3 = 0; scb_s->ecd = 0; scb_s->fac = 0; + scb_s->fpf = 0; rc = prepare_cpuflags(vcpu, vsie_page); if (rc) @@ -324,6 +331,9 @@ static int shadow_scb(struct kvm_vcpu *vcpu, struct vsie_page *vsie_page) prefix_unmapped(vsie_page); scb_s->ecb |= scb_o->ecb & ECB_TE; } + /* branch prediction */ + if (test_kvm_facility(vcpu->kvm, 82)) + scb_s->fpf |= scb_o->fpf & FPF_BPBC; /* SIMD */ if (test_kvm_facility(vcpu->kvm, 129)) { scb_s->eca |= scb_o->eca & ECA_VX; diff --git a/include/uapi/linux/kvm.h b/include/uapi/linux/kvm.h index 7a99b98cf88e..8fb90a0819c3 100644 --- a/include/uapi/linux/kvm.h +++ b/include/uapi/linux/kvm.h @@ -933,6 +933,7 @@ struct kvm_ppc_resize_hpt { #define KVM_CAP_HYPERV_VP_INDEX 149 #define KVM_CAP_S390_AIS_MIGRATION 150 #define KVM_CAP_PPC_GET_CPU_CHAR 151 +#define KVM_CAP_S390_BPB 152 #ifdef KVM_CAP_IRQ_ROUTING From 86be89939d11a84800f66e2a283b915b704bf33d Mon Sep 17 00:00:00 2001 From: Lorenzo Pieralisi Date: Tue, 16 Jan 2018 11:52:59 +0000 Subject: [PATCH 35/55] alpha/PCI: Fix noname IRQ level detection The conversion of the alpha architecture PCI host bridge legacy IRQ mapping/swizzling to the new PCI host bridge map/swizzle hooks carried out through: commit 0e4c2eeb758a ("alpha/PCI: Replace pci_fixup_irqs() call with host bridge IRQ mapping hooks") implies that IRQ for devices are now allocated through pci_assign_irq() function in pci_device_probe() that is called when a driver matching a device is found in order to probe the device through the device driver. Alpha noname platforms required IRQ level programming to be executed in sio_fixup_irq_levels(), that is called in noname_init_pci(), a platform hook called within a subsys_initcall. In noname_init_pci(), present IRQs are detected through sio_collect_irq_levels() that check the struct pci_dev->irq number to detect if an IRQ has been allocated for the device. By the time sio_collect_irq_levels() is called, some devices may still have not a matching driver loaded to match them (eg loadable module) therefore their IRQ allocation is still pending - which means that sio_collect_irq_levels() does not programme the correct IRQ level for those devices, causing their IRQ handling to be broken when the device driver is actually loaded and the device is probed. Fix the issue by adding code in the noname map_irq() function (noname_map_irq()) that, whilst mapping/swizzling the IRQ line, it also ensures that the correct IRQ level programming is executed at platform level, fixing the issue. Fixes: 0e4c2eeb758a ("alpha/PCI: Replace pci_fixup_irqs() call with host bridge IRQ mapping hooks") Reported-by: Mikulas Patocka Signed-off-by: Lorenzo Pieralisi Cc: stable@vger.kernel.org # 4.14 Cc: Bjorn Helgaas Cc: Richard Henderson Cc: Ivan Kokshaysky Cc: Mikulas Patocka Cc: Meelis Roos Signed-off-by: Matt Turner --- arch/alpha/kernel/sys_sio.c | 35 +++++++++++++++++++++++++++++------ 1 file changed, 29 insertions(+), 6 deletions(-) diff --git a/arch/alpha/kernel/sys_sio.c b/arch/alpha/kernel/sys_sio.c index 37bd6d9b8eb9..a6bdc1da47ad 100644 --- a/arch/alpha/kernel/sys_sio.c +++ b/arch/alpha/kernel/sys_sio.c @@ -102,6 +102,15 @@ sio_pci_route(void) alpha_mv.sys.sio.route_tab); } +static bool sio_pci_dev_irq_needs_level(const struct pci_dev *dev) +{ + if ((dev->class >> 16 == PCI_BASE_CLASS_BRIDGE) && + (dev->class >> 8 != PCI_CLASS_BRIDGE_PCMCIA)) + return false; + + return true; +} + static unsigned int __init sio_collect_irq_levels(void) { @@ -110,8 +119,7 @@ sio_collect_irq_levels(void) /* Iterate through the devices, collecting IRQ levels. */ for_each_pci_dev(dev) { - if ((dev->class >> 16 == PCI_BASE_CLASS_BRIDGE) && - (dev->class >> 8 != PCI_CLASS_BRIDGE_PCMCIA)) + if (!sio_pci_dev_irq_needs_level(dev)) continue; if (dev->irq) @@ -120,8 +128,7 @@ sio_collect_irq_levels(void) return level_bits; } -static void __init -sio_fixup_irq_levels(unsigned int level_bits) +static void __sio_fixup_irq_levels(unsigned int level_bits, bool reset) { unsigned int old_level_bits; @@ -139,12 +146,21 @@ sio_fixup_irq_levels(unsigned int level_bits) */ old_level_bits = inb(0x4d0) | (inb(0x4d1) << 8); - level_bits |= (old_level_bits & 0x71ff); + if (reset) + old_level_bits &= 0x71ff; + + level_bits |= old_level_bits; outb((level_bits >> 0) & 0xff, 0x4d0); outb((level_bits >> 8) & 0xff, 0x4d1); } +static inline void +sio_fixup_irq_levels(unsigned int level_bits) +{ + __sio_fixup_irq_levels(level_bits, true); +} + static inline int noname_map_irq(const struct pci_dev *dev, u8 slot, u8 pin) { @@ -181,7 +197,14 @@ noname_map_irq(const struct pci_dev *dev, u8 slot, u8 pin) const long min_idsel = 6, max_idsel = 14, irqs_per_slot = 5; int irq = COMMON_TABLE_LOOKUP, tmp; tmp = __kernel_extbl(alpha_mv.sys.sio.route_tab, irq); - return irq >= 0 ? tmp : -1; + + irq = irq >= 0 ? tmp : -1; + + /* Fixup IRQ level if an actual IRQ mapping is detected */ + if (sio_pci_dev_irq_needs_level(dev) && irq >= 0) + __sio_fixup_irq_levels(1 << irq, false); + + return irq; } static inline int From 91cfc88c66bf8ab95937606569670cf67fa73e09 Mon Sep 17 00:00:00 2001 From: Laura Abbott Date: Sat, 20 Jan 2018 17:14:02 -0800 Subject: [PATCH 36/55] x86: Use __nostackprotect for sme_encrypt_kernel Commit bacf6b499e11 ("x86/mm: Use a struct to reduce parameters for SME PGD mapping") moved some parameters into a structure. The structure was large enough to trigger the stack protection canary in sme_encrypt_kernel which doesn't work this early, causing reboots. Mark sme_encrypt_kernel appropriately to not use the canary. Fixes: bacf6b499e11 ("x86/mm: Use a struct to reduce parameters for SME PGD mapping") Signed-off-by: Laura Abbott Cc: Tom Lendacky Cc: Ingo Molnar Signed-off-by: Linus Torvalds --- arch/x86/mm/mem_encrypt.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/mm/mem_encrypt.c b/arch/x86/mm/mem_encrypt.c index 3ef362f598e3..e1d61e8500f9 100644 --- a/arch/x86/mm/mem_encrypt.c +++ b/arch/x86/mm/mem_encrypt.c @@ -738,7 +738,7 @@ static unsigned long __init sme_pgtable_calc(unsigned long len) return total; } -void __init sme_encrypt_kernel(struct boot_params *bp) +void __init __nostackprotector sme_encrypt_kernel(struct boot_params *bp) { unsigned long workarea_start, workarea_end, workarea_len; unsigned long execute_start, execute_end, execute_len; From 0c5b9b5d9adbad4b60491f9ba0d2af38904bb4b9 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 21 Jan 2018 13:51:26 -0800 Subject: [PATCH 37/55] Linux 4.15-rc9 --- Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Makefile b/Makefile index bf5b8cbb9469..339397b838d3 100644 --- a/Makefile +++ b/Makefile @@ -2,7 +2,7 @@ VERSION = 4 PATCHLEVEL = 15 SUBLEVEL = 0 -EXTRAVERSION = -rc8 +EXTRAVERSION = -rc9 NAME = Fearless Coyote # *DOCUMENTATION* From e58edaa4863583b54409444f11b4f80dff0af1cd Mon Sep 17 00:00:00 2001 From: Talat Batheesh Date: Sun, 21 Jan 2018 05:30:42 +0200 Subject: [PATCH 38/55] net/mlx5e: Fix fixpoint divide exception in mlx5e_am_stats_compare Helmut reported a bug about division by zero while running traffic and doing physical cable pull test. When the cable unplugged the ppms become zero, so when dividing the current ppms by the previous ppms in the next dim iteration there is division by zero. This patch prevent this division for both ppms and epms. Fixes: c3164d2fc48f ("net/mlx5e: Added BW check for DIM decision mechanism") Reported-by: Helmut Grauer Signed-off-by: Talat Batheesh Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx5/core/en_rx_am.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx_am.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx_am.c index e401d9d245f3..b69a705fd787 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx_am.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx_am.c @@ -201,9 +201,15 @@ static int mlx5e_am_stats_compare(struct mlx5e_rx_am_stats *curr, return (curr->bpms > prev->bpms) ? MLX5E_AM_STATS_BETTER : MLX5E_AM_STATS_WORSE; + if (!prev->ppms) + return curr->ppms ? MLX5E_AM_STATS_BETTER : + MLX5E_AM_STATS_SAME; + if (IS_SIGNIFICANT_DIFF(curr->ppms, prev->ppms)) return (curr->ppms > prev->ppms) ? MLX5E_AM_STATS_BETTER : MLX5E_AM_STATS_WORSE; + if (!prev->epms) + return MLX5E_AM_STATS_SAME; if (IS_SIGNIFICANT_DIFF(curr->epms, prev->epms)) return (curr->epms < prev->epms) ? MLX5E_AM_STATS_BETTER : From 0d665e7b109d512b7cae3ccef6e8654714887844 Mon Sep 17 00:00:00 2001 From: "Kirill A. Shutemov" Date: Fri, 19 Jan 2018 15:49:24 +0300 Subject: [PATCH 39/55] mm, page_vma_mapped: Drop faulty pointer arithmetics in check_pte() Tetsuo reported random crashes under memory pressure on 32-bit x86 system and tracked down to change that introduced page_vma_mapped_walk(). The root cause of the issue is the faulty pointer math in check_pte(). As ->pte may point to an arbitrary page we have to check that they are belong to the section before doing math. Otherwise it may lead to weird results. It wasn't noticed until now as mem_map[] is virtually contiguous on flatmem or vmemmap sparsemem. Pointer arithmetic just works against all 'struct page' pointers. But with classic sparsemem, it doesn't because each section memap is allocated separately and so consecutive pfns crossing two sections might have struct pages at completely unrelated addresses. Let's restructure code a bit and replace pointer arithmetic with operations on pfns. Signed-off-by: Kirill A. Shutemov Reported-and-tested-by: Tetsuo Handa Acked-by: Michal Hocko Fixes: ace71a19cec5 ("mm: introduce page_vma_mapped_walk()") Cc: stable@vger.kernel.org Signed-off-by: Linus Torvalds --- include/linux/swapops.h | 21 ++++++++++++++ mm/page_vma_mapped.c | 63 +++++++++++++++++++++++++---------------- 2 files changed, 59 insertions(+), 25 deletions(-) diff --git a/include/linux/swapops.h b/include/linux/swapops.h index 9c5a2628d6ce..1d3877c39a00 100644 --- a/include/linux/swapops.h +++ b/include/linux/swapops.h @@ -124,6 +124,11 @@ static inline bool is_write_device_private_entry(swp_entry_t entry) return unlikely(swp_type(entry) == SWP_DEVICE_WRITE); } +static inline unsigned long device_private_entry_to_pfn(swp_entry_t entry) +{ + return swp_offset(entry); +} + static inline struct page *device_private_entry_to_page(swp_entry_t entry) { return pfn_to_page(swp_offset(entry)); @@ -154,6 +159,11 @@ static inline bool is_write_device_private_entry(swp_entry_t entry) return false; } +static inline unsigned long device_private_entry_to_pfn(swp_entry_t entry) +{ + return 0; +} + static inline struct page *device_private_entry_to_page(swp_entry_t entry) { return NULL; @@ -189,6 +199,11 @@ static inline int is_write_migration_entry(swp_entry_t entry) return unlikely(swp_type(entry) == SWP_MIGRATION_WRITE); } +static inline unsigned long migration_entry_to_pfn(swp_entry_t entry) +{ + return swp_offset(entry); +} + static inline struct page *migration_entry_to_page(swp_entry_t entry) { struct page *p = pfn_to_page(swp_offset(entry)); @@ -218,6 +233,12 @@ static inline int is_migration_entry(swp_entry_t swp) { return 0; } + +static inline unsigned long migration_entry_to_pfn(swp_entry_t entry) +{ + return 0; +} + static inline struct page *migration_entry_to_page(swp_entry_t entry) { return NULL; diff --git a/mm/page_vma_mapped.c b/mm/page_vma_mapped.c index d22b84310f6d..956015614395 100644 --- a/mm/page_vma_mapped.c +++ b/mm/page_vma_mapped.c @@ -30,10 +30,29 @@ static bool map_pte(struct page_vma_mapped_walk *pvmw) return true; } +/** + * check_pte - check if @pvmw->page is mapped at the @pvmw->pte + * + * page_vma_mapped_walk() found a place where @pvmw->page is *potentially* + * mapped. check_pte() has to validate this. + * + * @pvmw->pte may point to empty PTE, swap PTE or PTE pointing to arbitrary + * page. + * + * If PVMW_MIGRATION flag is set, returns true if @pvmw->pte contains migration + * entry that points to @pvmw->page or any subpage in case of THP. + * + * If PVMW_MIGRATION flag is not set, returns true if @pvmw->pte points to + * @pvmw->page or any subpage in case of THP. + * + * Otherwise, return false. + * + */ static bool check_pte(struct page_vma_mapped_walk *pvmw) { + unsigned long pfn; + if (pvmw->flags & PVMW_MIGRATION) { -#ifdef CONFIG_MIGRATION swp_entry_t entry; if (!is_swap_pte(*pvmw->pte)) return false; @@ -41,37 +60,31 @@ static bool check_pte(struct page_vma_mapped_walk *pvmw) if (!is_migration_entry(entry)) return false; - if (migration_entry_to_page(entry) - pvmw->page >= - hpage_nr_pages(pvmw->page)) { + + pfn = migration_entry_to_pfn(entry); + } else if (is_swap_pte(*pvmw->pte)) { + swp_entry_t entry; + + /* Handle un-addressable ZONE_DEVICE memory */ + entry = pte_to_swp_entry(*pvmw->pte); + if (!is_device_private_entry(entry)) return false; - } - if (migration_entry_to_page(entry) < pvmw->page) - return false; -#else - WARN_ON_ONCE(1); -#endif + + pfn = device_private_entry_to_pfn(entry); } else { - if (is_swap_pte(*pvmw->pte)) { - swp_entry_t entry; - - entry = pte_to_swp_entry(*pvmw->pte); - if (is_device_private_entry(entry) && - device_private_entry_to_page(entry) == pvmw->page) - return true; - } - if (!pte_present(*pvmw->pte)) return false; - /* THP can be referenced by any subpage */ - if (pte_page(*pvmw->pte) - pvmw->page >= - hpage_nr_pages(pvmw->page)) { - return false; - } - if (pte_page(*pvmw->pte) < pvmw->page) - return false; + pfn = pte_pfn(*pvmw->pte); } + if (pfn < page_to_pfn(pvmw->page)) + return false; + + /* THP can be referenced by any subpage */ + if (pfn - page_to_pfn(pvmw->page) >= hpage_nr_pages(pvmw->page)) + return false; + return true; } From 7222708e823afc98dfb769d36bf2f26b6420b7ce Mon Sep 17 00:00:00 2001 From: "Kirill A. Shutemov" Date: Mon, 22 Jan 2018 12:22:30 +0300 Subject: [PATCH 40/55] mm, page_vma_mapped: Introduce pfn_in_hpage() The new helper would check if the pfn belongs to the page. For huge pages it checks if the PFN is within range covered by the huge page. The helper is used in check_pte(). The original code the helper replaces had two call to page_to_pfn(). page_to_pfn() is relatively costly. Although current GCC is able to optimize code to have one call, it's better to do this explicitly. Signed-off-by: Kirill A. Shutemov Signed-off-by: Linus Torvalds --- mm/page_vma_mapped.c | 17 +++++++++-------- 1 file changed, 9 insertions(+), 8 deletions(-) diff --git a/mm/page_vma_mapped.c b/mm/page_vma_mapped.c index 956015614395..ae3c2a35d61b 100644 --- a/mm/page_vma_mapped.c +++ b/mm/page_vma_mapped.c @@ -30,6 +30,14 @@ static bool map_pte(struct page_vma_mapped_walk *pvmw) return true; } +static inline bool pfn_in_hpage(struct page *hpage, unsigned long pfn) +{ + unsigned long hpage_pfn = page_to_pfn(hpage); + + /* THP can be referenced by any subpage */ + return pfn >= hpage_pfn && pfn - hpage_pfn < hpage_nr_pages(hpage); +} + /** * check_pte - check if @pvmw->page is mapped at the @pvmw->pte * @@ -78,14 +86,7 @@ static bool check_pte(struct page_vma_mapped_walk *pvmw) pfn = pte_pfn(*pvmw->pte); } - if (pfn < page_to_pfn(pvmw->page)) - return false; - - /* THP can be referenced by any subpage */ - if (pfn - page_to_pfn(pvmw->page) >= hpage_nr_pages(pvmw->page)) - return false; - - return true; + return pfn_in_hpage(pvmw->page, pfn); } /** From d728f13102cd4b2afdbb183babde2b43ac2c3130 Mon Sep 17 00:00:00 2001 From: Kumar Sanghvi Date: Wed, 17 Jan 2018 12:13:33 +0530 Subject: [PATCH 41/55] cxgb4: set filter type to 1 for ETH_P_IPV6 For ethtype_key = ETH_P_IPV6, set filter type as 1 in cxgb4_tc_flower code when processing flow match parameters. Signed-off-by: Kumar Sanghvi Signed-off-by: Rahul Lakkireddy Signed-off-by: Ganesh Goudar Signed-off-by: David S. Miller --- drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c index d4a548a6a55c..276edcbb3259 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c @@ -111,6 +111,9 @@ static void cxgb4_process_flow_match(struct net_device *dev, ethtype_mask = 0; } + if (ethtype_key == ETH_P_IPV6) + fs->type = 1; + fs->val.ethtype = ethtype_key; fs->mask.ethtype = ethtype_mask; fs->val.proto = key->ip_proto; From 100d39af5048aecb00522a128ec264781ba7ecfa Mon Sep 17 00:00:00 2001 From: Kumar Sanghvi Date: Wed, 17 Jan 2018 12:13:34 +0530 Subject: [PATCH 42/55] cxgb4: fix endianness for vlan value in cxgb4_tc_flower Don't change endianness when assigning vlan value in cxgb4_tc_flower code when processing flow match parameters. The value gets converted to network order as part of filtering code in set_filter_wr. Signed-off-by: Kumar Sanghvi Signed-off-by: Rahul Lakkireddy Signed-off-by: Ganesh Goudar Signed-off-by: David S. Miller --- drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c index 276edcbb3259..a452d5a1b0f3 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c @@ -208,8 +208,8 @@ static void cxgb4_process_flow_match(struct net_device *dev, VLAN_PRIO_SHIFT); vlan_tci_mask = mask->vlan_id | (mask->vlan_priority << VLAN_PRIO_SHIFT); - fs->val.ivlan = cpu_to_be16(vlan_tci); - fs->mask.ivlan = cpu_to_be16(vlan_tci_mask); + fs->val.ivlan = vlan_tci; + fs->mask.ivlan = vlan_tci_mask; /* Chelsio adapters use ivlan_vld bit to match vlan packets * as 802.1Q. Also, when vlan tag is present in packets, From ab18a9c9ca27f8208c4960a1b25e8475b9f9dda5 Mon Sep 17 00:00:00 2001 From: Oliver Neukum Date: Wed, 17 Jan 2018 16:33:54 +0100 Subject: [PATCH 43/55] usbnet: silence an unnecessary warning That a kevent could not be scheduled is not an error. Such handlers must be able to deal with multiple events anyway. As the successful scheduling of a work is a debug event, make the failure debug priority, too. V2: coding style Signed-off-by: Oliver Neukum Reported-by: Cristian Caravena Signed-off-by: David S. Miller --- drivers/net/usb/usbnet.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) diff --git a/drivers/net/usb/usbnet.c b/drivers/net/usb/usbnet.c index d56fe32bf48d..8a22ff67b026 100644 --- a/drivers/net/usb/usbnet.c +++ b/drivers/net/usb/usbnet.c @@ -457,12 +457,10 @@ static enum skb_state defer_bh(struct usbnet *dev, struct sk_buff *skb, void usbnet_defer_kevent (struct usbnet *dev, int work) { set_bit (work, &dev->flags); - if (!schedule_work (&dev->kevent)) { - if (net_ratelimit()) - netdev_err(dev->net, "kevent %d may have been dropped\n", work); - } else { + if (!schedule_work (&dev->kevent)) + netdev_dbg(dev->net, "kevent %d may have been dropped\n", work); + else netdev_dbg(dev->net, "kevent %d scheduled\n", work); - } } EXPORT_SYMBOL_GPL(usbnet_defer_kevent); From b589513e6354a5fd6934823b7fd66bffad41137a Mon Sep 17 00:00:00 2001 From: Sowmini Varadhan Date: Thu, 18 Jan 2018 13:11:07 -0800 Subject: [PATCH 44/55] rds: tcp: compute m_ack_seq as offset from ->write_seq rds-tcp uses m_ack_seq to track the tcp ack# that indicates that the peer has received a rds_message. The m_ack_seq is used in rds_tcp_is_acked() to figure out when it is safe to drop the rds_message from the RDS retransmit queue. The m_ack_seq must be calculated as an offset from the right edge of the in-flight tcp buffer, i.e., it should be based on the ->write_seq, not the ->snd_nxt. Signed-off-by: Sowmini Varadhan Signed-off-by: David S. Miller --- net/rds/tcp.c | 5 +++-- net/rds/tcp.h | 2 +- net/rds/tcp_send.c | 4 ++-- 3 files changed, 6 insertions(+), 5 deletions(-) diff --git a/net/rds/tcp.c b/net/rds/tcp.c index 6b7ee71f40c6..ab7356e0ba83 100644 --- a/net/rds/tcp.c +++ b/net/rds/tcp.c @@ -90,9 +90,10 @@ void rds_tcp_nonagle(struct socket *sock) sizeof(val)); } -u32 rds_tcp_snd_nxt(struct rds_tcp_connection *tc) +u32 rds_tcp_write_seq(struct rds_tcp_connection *tc) { - return tcp_sk(tc->t_sock->sk)->snd_nxt; + /* seq# of the last byte of data in tcp send buffer */ + return tcp_sk(tc->t_sock->sk)->write_seq; } u32 rds_tcp_snd_una(struct rds_tcp_connection *tc) diff --git a/net/rds/tcp.h b/net/rds/tcp.h index 1aafbf7c3011..864ca7d8f019 100644 --- a/net/rds/tcp.h +++ b/net/rds/tcp.h @@ -54,7 +54,7 @@ void rds_tcp_set_callbacks(struct socket *sock, struct rds_conn_path *cp); void rds_tcp_reset_callbacks(struct socket *sock, struct rds_conn_path *cp); void rds_tcp_restore_callbacks(struct socket *sock, struct rds_tcp_connection *tc); -u32 rds_tcp_snd_nxt(struct rds_tcp_connection *tc); +u32 rds_tcp_write_seq(struct rds_tcp_connection *tc); u32 rds_tcp_snd_una(struct rds_tcp_connection *tc); u64 rds_tcp_map_seq(struct rds_tcp_connection *tc, u32 seq); extern struct rds_transport rds_tcp_transport; diff --git a/net/rds/tcp_send.c b/net/rds/tcp_send.c index dc860d1bb608..9b76e0fa1722 100644 --- a/net/rds/tcp_send.c +++ b/net/rds/tcp_send.c @@ -86,7 +86,7 @@ int rds_tcp_xmit(struct rds_connection *conn, struct rds_message *rm, * m_ack_seq is set to the sequence number of the last byte of * header and data. see rds_tcp_is_acked(). */ - tc->t_last_sent_nxt = rds_tcp_snd_nxt(tc); + tc->t_last_sent_nxt = rds_tcp_write_seq(tc); rm->m_ack_seq = tc->t_last_sent_nxt + sizeof(struct rds_header) + be32_to_cpu(rm->m_inc.i_hdr.h_len) - 1; @@ -98,7 +98,7 @@ int rds_tcp_xmit(struct rds_connection *conn, struct rds_message *rm, rm->m_inc.i_hdr.h_flags |= RDS_FLAG_RETRANSMITTED; rdsdebug("rm %p tcp nxt %u ack_seq %llu\n", - rm, rds_tcp_snd_nxt(tc), + rm, rds_tcp_write_seq(tc), (unsigned long long)rm->m_ack_seq); } From 896d86959fee58113fc510c70cd8d10e82aa3e6a Mon Sep 17 00:00:00 2001 From: John Allen Date: Thu, 18 Jan 2018 16:26:31 -0600 Subject: [PATCH 45/55] ibmvnic: Modify buffer size and number of queues on failover Using newer backing devices can cause the required padding at the end of buffer as well as the number of queues to change after a failover. Since we currently assume that these values never change, after a failover to a backing device with different capabilities, we can get errors from the vnic server, attempt to free long term buffers that are no longer there, or not free long term buffers that should be freed. This patch resolves the issue by checking whether any of these values change, and if so perform the necessary re-allocations. Signed-off-by: John Allen Reviewed-by: Nathan Fontenot Signed-off-by: David S. Miller --- drivers/net/ethernet/ibm/ibmvnic.c | 45 +++++++++++++++++++++++++----- drivers/net/ethernet/ibm/ibmvnic.h | 2 ++ 2 files changed, 40 insertions(+), 7 deletions(-) diff --git a/drivers/net/ethernet/ibm/ibmvnic.c b/drivers/net/ethernet/ibm/ibmvnic.c index ab2e1917cd04..7f75d01432ef 100644 --- a/drivers/net/ethernet/ibm/ibmvnic.c +++ b/drivers/net/ethernet/ibm/ibmvnic.c @@ -410,6 +410,10 @@ static int reset_rx_pools(struct ibmvnic_adapter *adapter) struct ibmvnic_rx_pool *rx_pool; int rx_scrqs; int i, j, rc; + u64 *size_array; + + size_array = (u64 *)((u8 *)(adapter->login_rsp_buf) + + be32_to_cpu(adapter->login_rsp_buf->off_rxadd_buff_size)); rx_scrqs = be32_to_cpu(adapter->login_rsp_buf->num_rxadd_subcrqs); for (i = 0; i < rx_scrqs; i++) { @@ -417,7 +421,17 @@ static int reset_rx_pools(struct ibmvnic_adapter *adapter) netdev_dbg(adapter->netdev, "Re-setting rx_pool[%d]\n", i); - rc = reset_long_term_buff(adapter, &rx_pool->long_term_buff); + if (rx_pool->buff_size != be64_to_cpu(size_array[i])) { + free_long_term_buff(adapter, &rx_pool->long_term_buff); + rx_pool->buff_size = be64_to_cpu(size_array[i]); + alloc_long_term_buff(adapter, &rx_pool->long_term_buff, + rx_pool->size * + rx_pool->buff_size); + } else { + rc = reset_long_term_buff(adapter, + &rx_pool->long_term_buff); + } + if (rc) return rc; @@ -439,14 +453,12 @@ static int reset_rx_pools(struct ibmvnic_adapter *adapter) static void release_rx_pools(struct ibmvnic_adapter *adapter) { struct ibmvnic_rx_pool *rx_pool; - int rx_scrqs; int i, j; if (!adapter->rx_pool) return; - rx_scrqs = be32_to_cpu(adapter->login_rsp_buf->num_rxadd_subcrqs); - for (i = 0; i < rx_scrqs; i++) { + for (i = 0; i < adapter->num_active_rx_pools; i++) { rx_pool = &adapter->rx_pool[i]; netdev_dbg(adapter->netdev, "Releasing rx_pool[%d]\n", i); @@ -469,6 +481,7 @@ static void release_rx_pools(struct ibmvnic_adapter *adapter) kfree(adapter->rx_pool); adapter->rx_pool = NULL; + adapter->num_active_rx_pools = 0; } static int init_rx_pools(struct net_device *netdev) @@ -493,6 +506,8 @@ static int init_rx_pools(struct net_device *netdev) return -1; } + adapter->num_active_rx_pools = 0; + for (i = 0; i < rxadd_subcrqs; i++) { rx_pool = &adapter->rx_pool[i]; @@ -536,6 +551,8 @@ static int init_rx_pools(struct net_device *netdev) rx_pool->next_free = 0; } + adapter->num_active_rx_pools = rxadd_subcrqs; + return 0; } @@ -586,13 +603,12 @@ static void release_vpd_data(struct ibmvnic_adapter *adapter) static void release_tx_pools(struct ibmvnic_adapter *adapter) { struct ibmvnic_tx_pool *tx_pool; - int i, tx_scrqs; + int i; if (!adapter->tx_pool) return; - tx_scrqs = be32_to_cpu(adapter->login_rsp_buf->num_txsubm_subcrqs); - for (i = 0; i < tx_scrqs; i++) { + for (i = 0; i < adapter->num_active_tx_pools; i++) { netdev_dbg(adapter->netdev, "Releasing tx_pool[%d]\n", i); tx_pool = &adapter->tx_pool[i]; kfree(tx_pool->tx_buff); @@ -603,6 +619,7 @@ static void release_tx_pools(struct ibmvnic_adapter *adapter) kfree(adapter->tx_pool); adapter->tx_pool = NULL; + adapter->num_active_tx_pools = 0; } static int init_tx_pools(struct net_device *netdev) @@ -619,6 +636,8 @@ static int init_tx_pools(struct net_device *netdev) if (!adapter->tx_pool) return -1; + adapter->num_active_tx_pools = 0; + for (i = 0; i < tx_subcrqs; i++) { tx_pool = &adapter->tx_pool[i]; @@ -666,6 +685,8 @@ static int init_tx_pools(struct net_device *netdev) tx_pool->producer_index = 0; } + adapter->num_active_tx_pools = tx_subcrqs; + return 0; } @@ -1548,6 +1569,7 @@ static int ibmvnic_set_mac(struct net_device *netdev, void *p) static int do_reset(struct ibmvnic_adapter *adapter, struct ibmvnic_rwi *rwi, u32 reset_state) { + u64 old_num_rx_queues, old_num_tx_queues; struct net_device *netdev = adapter->netdev; int i, rc; @@ -1557,6 +1579,9 @@ static int do_reset(struct ibmvnic_adapter *adapter, netif_carrier_off(netdev); adapter->reset_reason = rwi->reset_reason; + old_num_rx_queues = adapter->req_rx_queues; + old_num_tx_queues = adapter->req_tx_queues; + if (rwi->reset_reason == VNIC_RESET_MOBILITY) { rc = ibmvnic_reenable_crq_queue(adapter); if (rc) @@ -1601,6 +1626,12 @@ static int do_reset(struct ibmvnic_adapter *adapter, rc = init_resources(adapter); if (rc) return rc; + } else if (adapter->req_rx_queues != old_num_rx_queues || + adapter->req_tx_queues != old_num_tx_queues) { + release_rx_pools(adapter); + release_tx_pools(adapter); + init_rx_pools(netdev); + init_tx_pools(netdev); } else { rc = reset_tx_pools(adapter); if (rc) diff --git a/drivers/net/ethernet/ibm/ibmvnic.h b/drivers/net/ethernet/ibm/ibmvnic.h index 4487f1e2c266..3aec42118db2 100644 --- a/drivers/net/ethernet/ibm/ibmvnic.h +++ b/drivers/net/ethernet/ibm/ibmvnic.h @@ -1091,6 +1091,8 @@ struct ibmvnic_adapter { u64 opt_rxba_entries_per_subcrq; __be64 tx_rx_desc_req; u8 map_id; + u64 num_active_rx_pools; + u64 num_active_tx_pools; struct tasklet_struct tasklet; enum vnic_state state; From e791380340685698dbdd38c7e3f3fcbf70a3c832 Mon Sep 17 00:00:00 2001 From: John Allen Date: Thu, 18 Jan 2018 16:27:12 -0600 Subject: [PATCH 46/55] ibmvnic: Revert to previous mtu when unsupported value requested If we request an unsupported mtu value, the vnic server will suggest a different value. Currently we take the suggested value without question and login with that value. However, the behavior doesn't seem completely sane as attempting to change the mtu to some specific value will change the mtu to some completely different value most of the time. This patch fixes the issue by logging in with the previously used mtu value and printing an error message saying that the given mtu is unsupported. Signed-off-by: John Allen Reviewed-by: Nathan Fontenot Signed-off-by: David S. Miller --- drivers/net/ethernet/ibm/ibmvnic.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/ibm/ibmvnic.c b/drivers/net/ethernet/ibm/ibmvnic.c index 7f75d01432ef..cc36b254917c 100644 --- a/drivers/net/ethernet/ibm/ibmvnic.c +++ b/drivers/net/ethernet/ibm/ibmvnic.c @@ -3623,7 +3623,17 @@ static void handle_request_cap_rsp(union ibmvnic_crq *crq, *req_value, (long int)be64_to_cpu(crq->request_capability_rsp. number), name); - *req_value = be64_to_cpu(crq->request_capability_rsp.number); + + if (be16_to_cpu(crq->request_capability_rsp.capability) == + REQ_MTU) { + pr_err("mtu of %llu is not supported. Reverting.\n", + *req_value); + *req_value = adapter->fallback.mtu; + } else { + *req_value = + be64_to_cpu(crq->request_capability_rsp.number); + } + ibmvnic_send_req_caps(adapter, 1); return; default: From 69d08dcbbe34347cbc044629cf6f25d062593dfe Mon Sep 17 00:00:00 2001 From: John Allen Date: Thu, 18 Jan 2018 16:27:58 -0600 Subject: [PATCH 47/55] ibmvnic: Allocate and request vpd in init_resources In reset events in which our memory allocations need to be reallocated, VPD data is being freed, but never reallocated. This can cause issues if we later attempt to access that memory or reset and attempt to free the memory. This patch moves the allocation of the VPD data to init_resources so that it will be symmetrically freed during release resources. Signed-off-by: John Allen Reviewed-by: Nathan Fontenot Signed-off-by: David S. Miller --- drivers/net/ethernet/ibm/ibmvnic.c | 16 +++++++++------- 1 file changed, 9 insertions(+), 7 deletions(-) diff --git a/drivers/net/ethernet/ibm/ibmvnic.c b/drivers/net/ethernet/ibm/ibmvnic.c index cc36b254917c..b65f5f3ac034 100644 --- a/drivers/net/ethernet/ibm/ibmvnic.c +++ b/drivers/net/ethernet/ibm/ibmvnic.c @@ -881,7 +881,7 @@ static int ibmvnic_get_vpd(struct ibmvnic_adapter *adapter) if (adapter->vpd->buff) len = adapter->vpd->len; - reinit_completion(&adapter->fw_done); + init_completion(&adapter->fw_done); crq.get_vpd_size.first = IBMVNIC_CRQ_CMD; crq.get_vpd_size.cmd = GET_VPD_SIZE; ibmvnic_send_crq(adapter, &crq); @@ -943,6 +943,13 @@ static int init_resources(struct ibmvnic_adapter *adapter) if (!adapter->vpd) return -ENOMEM; + /* Vital Product Data (VPD) */ + rc = ibmvnic_get_vpd(adapter); + if (rc) { + netdev_err(netdev, "failed to initialize Vital Product Data (VPD)\n"); + return rc; + } + adapter->map_id = 1; adapter->napi = kcalloc(adapter->req_rx_queues, sizeof(struct napi_struct), GFP_KERNEL); @@ -1016,7 +1023,7 @@ static int __ibmvnic_open(struct net_device *netdev) static int ibmvnic_open(struct net_device *netdev) { struct ibmvnic_adapter *adapter = netdev_priv(netdev); - int rc, vpd; + int rc; mutex_lock(&adapter->reset_lock); @@ -1039,11 +1046,6 @@ static int ibmvnic_open(struct net_device *netdev) rc = __ibmvnic_open(netdev); netif_carrier_on(netdev); - /* Vital Product Data (VPD) */ - vpd = ibmvnic_get_vpd(adapter); - if (vpd) - netdev_err(netdev, "failed to initialize Vital Product Data (VPD)\n"); - mutex_unlock(&adapter->reset_lock); return rc; From 7c68d1a6b4db9012790af7ac0f0fdc0d2083422a Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 18 Jan 2018 19:59:19 -0800 Subject: [PATCH 48/55] net: qdisc_pkt_len_init() should be more robust Without proper validation of DODGY packets, we might very well feed qdisc_pkt_len_init() with invalid GSO packets. tcp_hdrlen() might access out-of-bound data, so let's use skb_header_pointer() and proper checks. Whole story is described in commit d0c081b49137 ("flow_dissector: properly cap thoff field") We have the goal of validating DODGY packets earlier in the stack, so we might very well revert this fix in the future. Signed-off-by: Eric Dumazet Cc: Willem de Bruijn Cc: Jason Wang Reported-by: syzbot+9da69ebac7dddd804552@syzkaller.appspotmail.com Acked-by: Jason Wang Signed-off-by: David S. Miller --- net/core/dev.c | 19 +++++++++++++++---- 1 file changed, 15 insertions(+), 4 deletions(-) diff --git a/net/core/dev.c b/net/core/dev.c index 0e0ba36eeac9..613fb4066be7 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3151,10 +3151,21 @@ static void qdisc_pkt_len_init(struct sk_buff *skb) hdr_len = skb_transport_header(skb) - skb_mac_header(skb); /* + transport layer */ - if (likely(shinfo->gso_type & (SKB_GSO_TCPV4 | SKB_GSO_TCPV6))) - hdr_len += tcp_hdrlen(skb); - else - hdr_len += sizeof(struct udphdr); + if (likely(shinfo->gso_type & (SKB_GSO_TCPV4 | SKB_GSO_TCPV6))) { + const struct tcphdr *th; + struct tcphdr _tcphdr; + + th = skb_header_pointer(skb, skb_transport_offset(skb), + sizeof(_tcphdr), &_tcphdr); + if (likely(th)) + hdr_len += __tcp_hdrlen(th); + } else { + struct udphdr _udphdr; + + if (skb_header_pointer(skb, skb_transport_offset(skb), + sizeof(_udphdr), &_udphdr)) + hdr_len += sizeof(struct udphdr); + } if (shinfo->gso_type & SKB_GSO_DODGY) gso_segs = DIV_ROUND_UP(skb->len - hdr_len, From 121d57af308d0cf943f08f4738d24d3966c38cd9 Mon Sep 17 00:00:00 2001 From: Willem de Bruijn Date: Fri, 19 Jan 2018 09:29:18 -0500 Subject: [PATCH 49/55] gso: validate gso_type in GSO handlers Validate gso_type during segmentation as SKB_GSO_DODGY sources may pass packets where the gso_type does not match the contents. Syzkaller was able to enter the SCTP gso handler with a packet of gso_type SKB_GSO_TCPV4. On entry of transport layer gso handlers, verify that the gso_type matches the transport protocol. Fixes: 90017accff61 ("sctp: Add GSO support") Link: http://lkml.kernel.org/r/<001a1137452496ffc305617e5fe0@google.com> Reported-by: syzbot+fee64147a25aecd48055@syzkaller.appspotmail.com Signed-off-by: Willem de Bruijn Acked-by: Jason Wang Reviewed-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller --- net/ipv4/esp4_offload.c | 3 +++ net/ipv4/tcp_offload.c | 3 +++ net/ipv4/udp_offload.c | 3 +++ net/ipv6/esp6_offload.c | 3 +++ net/ipv6/tcpv6_offload.c | 3 +++ net/ipv6/udp_offload.c | 3 +++ net/sctp/offload.c | 3 +++ 7 files changed, 21 insertions(+) diff --git a/net/ipv4/esp4_offload.c b/net/ipv4/esp4_offload.c index b1338e576d00..29b333a62ab0 100644 --- a/net/ipv4/esp4_offload.c +++ b/net/ipv4/esp4_offload.c @@ -122,6 +122,9 @@ static struct sk_buff *esp4_gso_segment(struct sk_buff *skb, if (!xo) goto out; + if (!(skb_shinfo(skb)->gso_type & SKB_GSO_ESP)) + goto out; + seq = xo->seq.low; x = skb->sp->xvec[skb->sp->len - 1]; diff --git a/net/ipv4/tcp_offload.c b/net/ipv4/tcp_offload.c index b6a2aa1dcf56..4d58e2ce0b5b 100644 --- a/net/ipv4/tcp_offload.c +++ b/net/ipv4/tcp_offload.c @@ -32,6 +32,9 @@ static void tcp_gso_tstamp(struct sk_buff *skb, unsigned int ts_seq, static struct sk_buff *tcp4_gso_segment(struct sk_buff *skb, netdev_features_t features) { + if (!(skb_shinfo(skb)->gso_type & SKB_GSO_TCPV4)) + return ERR_PTR(-EINVAL); + if (!pskb_may_pull(skb, sizeof(struct tcphdr))) return ERR_PTR(-EINVAL); diff --git a/net/ipv4/udp_offload.c b/net/ipv4/udp_offload.c index 01801b77bd0d..ea6e6e7df0ee 100644 --- a/net/ipv4/udp_offload.c +++ b/net/ipv4/udp_offload.c @@ -203,6 +203,9 @@ static struct sk_buff *udp4_ufo_fragment(struct sk_buff *skb, goto out; } + if (!(skb_shinfo(skb)->gso_type & SKB_GSO_UDP)) + goto out; + if (!pskb_may_pull(skb, sizeof(struct udphdr))) goto out; diff --git a/net/ipv6/esp6_offload.c b/net/ipv6/esp6_offload.c index dd9627490c7c..f52c314d4c97 100644 --- a/net/ipv6/esp6_offload.c +++ b/net/ipv6/esp6_offload.c @@ -149,6 +149,9 @@ static struct sk_buff *esp6_gso_segment(struct sk_buff *skb, if (!xo) goto out; + if (!(skb_shinfo(skb)->gso_type & SKB_GSO_ESP)) + goto out; + seq = xo->seq.low; x = skb->sp->xvec[skb->sp->len - 1]; diff --git a/net/ipv6/tcpv6_offload.c b/net/ipv6/tcpv6_offload.c index d883c9204c01..278e49cd67d4 100644 --- a/net/ipv6/tcpv6_offload.c +++ b/net/ipv6/tcpv6_offload.c @@ -46,6 +46,9 @@ static struct sk_buff *tcp6_gso_segment(struct sk_buff *skb, { struct tcphdr *th; + if (!(skb_shinfo(skb)->gso_type & SKB_GSO_TCPV6)) + return ERR_PTR(-EINVAL); + if (!pskb_may_pull(skb, sizeof(*th))) return ERR_PTR(-EINVAL); diff --git a/net/ipv6/udp_offload.c b/net/ipv6/udp_offload.c index a0f89ad76f9d..2a04dc9c781b 100644 --- a/net/ipv6/udp_offload.c +++ b/net/ipv6/udp_offload.c @@ -42,6 +42,9 @@ static struct sk_buff *udp6_ufo_fragment(struct sk_buff *skb, const struct ipv6hdr *ipv6h; struct udphdr *uh; + if (!(skb_shinfo(skb)->gso_type & SKB_GSO_UDP)) + goto out; + if (!pskb_may_pull(skb, sizeof(struct udphdr))) goto out; diff --git a/net/sctp/offload.c b/net/sctp/offload.c index 275925b93b29..35bc7106d182 100644 --- a/net/sctp/offload.c +++ b/net/sctp/offload.c @@ -45,6 +45,9 @@ static struct sk_buff *sctp_gso_segment(struct sk_buff *skb, struct sk_buff *segs = ERR_PTR(-EINVAL); struct sctphdr *sh; + if (!(skb_shinfo(skb)->gso_type & SKB_GSO_SCTP)) + goto out; + sh = sctp_hdr(skb); if (!pskb_may_pull(skb, sizeof(*sh))) goto out; From ad23b750933ea7bf962678972a286c78a8fa36aa Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Fri, 19 Jan 2018 11:50:46 +0100 Subject: [PATCH 50/55] net: igmp: fix source address check for IGMPv3 reports Commit "net: igmp: Use correct source address on IGMPv3 reports" introduced a check to validate the source address of locally generated IGMPv3 packets. Instead of checking the local interface address directly, it uses inet_ifa_match(fl4->saddr, ifa), which checks if the address is on the local subnet (or equal to the point-to-point address if used). This breaks for point-to-point interfaces, so check against ifa->ifa_local directly. Cc: Kevin Cernekee Fixes: a46182b00290 ("net: igmp: Use correct source address on IGMPv3 reports") Reported-by: Sebastian Gottschall Signed-off-by: Felix Fietkau Signed-off-by: David S. Miller --- net/ipv4/igmp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c index 726f6b608274..2d49717a7421 100644 --- a/net/ipv4/igmp.c +++ b/net/ipv4/igmp.c @@ -332,7 +332,7 @@ static __be32 igmpv3_get_srcaddr(struct net_device *dev, return htonl(INADDR_ANY); for_ifa(in_dev) { - if (inet_ifa_match(fl4->saddr, ifa)) + if (fl4->saddr == ifa->ifa_local) return fl4->saddr; } endfor_ifa(in_dev); From 52acf06451930eb4cefabd5ecea56e2d46c32f76 Mon Sep 17 00:00:00 2001 From: Ivan Vecera Date: Fri, 19 Jan 2018 20:23:50 +0100 Subject: [PATCH 51/55] be2net: restore properly promisc mode after queues reconfiguration The commit 622190669403 ("be2net: Request RSS capability of Rx interface depending on number of Rx rings") modified be_update_queues() so the IFACE (HW representation of the netdevice) is destroyed and then re-created. This causes a regression because potential promiscuous mode is not restored properly during be_open() because the driver thinks that the HW has promiscuous mode already enabled. Note that Lancer is not affected by this bug because RX-filter flags are disabled during be_close() for this chipset. Cc: Sathya Perla Cc: Ajit Khaparde Cc: Sriharsha Basavapatna Cc: Somnath Kotur Fixes: 622190669403 ("be2net: Request RSS capability of Rx interface depending on number of Rx rings") Signed-off-by: Ivan Vecera Signed-off-by: David S. Miller --- drivers/net/ethernet/emulex/benet/be_main.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c index c6e859a27ee6..e180657a02ef 100644 --- a/drivers/net/ethernet/emulex/benet/be_main.c +++ b/drivers/net/ethernet/emulex/benet/be_main.c @@ -4634,6 +4634,15 @@ int be_update_queues(struct be_adapter *adapter) be_schedule_worker(adapter); + /* + * The IF was destroyed and re-created. We need to clear + * all promiscuous flags valid for the destroyed IF. + * Without this promisc mode is not restored during + * be_open() because the driver thinks that it is + * already enabled in HW. + */ + adapter->if_flags &= ~BE_IF_FLAGS_ALL_PROMISCUOUS; + if (netif_running(netdev)) status = be_open(netdev); From 7a8c4dd9be91a7e8f8f0e0419a560663adc694a3 Mon Sep 17 00:00:00 2001 From: Dave Watson Date: Fri, 19 Jan 2018 12:30:13 -0800 Subject: [PATCH 52/55] tls: Correct length of scatterlist in tls_sw_sendpage The scatterlist is reused by both sendmsg and sendfile. If a sendmsg of smaller number of pages is followed by a sendfile of larger number of pages, the scatterlist may be too short, resulting in a crash in gcm_encrypt. Add sg_unmark_end to make the list the correct length. tls_sw_sendmsg already calls sg_unmark_end correctly when it allocates memory in alloc_sg, or in zerocopy_from_iter. Signed-off-by: Dave Watson Signed-off-by: David S. Miller --- net/tls/tls_sw.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/tls/tls_sw.c b/net/tls/tls_sw.c index 61f394d369bf..0a9b72fbd761 100644 --- a/net/tls/tls_sw.c +++ b/net/tls/tls_sw.c @@ -577,6 +577,8 @@ int tls_sw_sendpage(struct sock *sk, struct page *page, get_page(page); sg = ctx->sg_plaintext_data + ctx->sg_plaintext_num_elem; sg_set_page(sg, page, copy, offset); + sg_unmark_end(sg); + ctx->sg_plaintext_num_elem++; sk_mem_charge(sk, copy); From 0afc0decf247f65b7aba666a76a0a68adf4bc435 Mon Sep 17 00:00:00 2001 From: Martin Brandenburg Date: Mon, 22 Jan 2018 15:44:51 -0500 Subject: [PATCH 53/55] orangefs: use list_for_each_entry_safe in purge_waiting_ops set_op_state_purged can delete the op. Signed-off-by: Martin Brandenburg Cc: stable@vger.kernel.org Signed-off-by: Linus Torvalds --- fs/orangefs/waitqueue.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/fs/orangefs/waitqueue.c b/fs/orangefs/waitqueue.c index 835c6e148afc..0577d6dba8c8 100644 --- a/fs/orangefs/waitqueue.c +++ b/fs/orangefs/waitqueue.c @@ -29,10 +29,10 @@ static void orangefs_clean_up_interrupted_operation(struct orangefs_kernel_op_s */ void purge_waiting_ops(void) { - struct orangefs_kernel_op_s *op; + struct orangefs_kernel_op_s *op, *tmp; spin_lock(&orangefs_request_list_lock); - list_for_each_entry(op, &orangefs_request_list, list) { + list_for_each_entry_safe(op, tmp, &orangefs_request_list, list) { gossip_debug(GOSSIP_WAIT_DEBUG, "pvfs2-client-core: purging op tag %llu %s\n", llu(op->tag), From a0ec1ded22e6a6bc41981fae22406835b006a66e Mon Sep 17 00:00:00 2001 From: Martin Brandenburg Date: Mon, 22 Jan 2018 15:44:52 -0500 Subject: [PATCH 54/55] orangefs: initialize op on loop restart in orangefs_devreq_read In orangefs_devreq_read, there is a loop which picks an op off the list of pending ops. If the loop fails to find an op, there is nothing to read, and it returns EAGAIN. If the op has been given up on, the loop is restarted via a goto. The bug is that the variable which the found op is written to is not reinitialized, so if there are no more eligible ops on the list, the code runs again on the already handled op. This is triggered by interrupting a process while the op is being copied to the client-core. It's a fairly small window, but it's there. Signed-off-by: Martin Brandenburg Cc: stable@vger.kernel.org Signed-off-by: Linus Torvalds --- fs/orangefs/devorangefs-req.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/fs/orangefs/devorangefs-req.c b/fs/orangefs/devorangefs-req.c index ded456f17de6..c584ad8d023c 100644 --- a/fs/orangefs/devorangefs-req.c +++ b/fs/orangefs/devorangefs-req.c @@ -162,7 +162,7 @@ static ssize_t orangefs_devreq_read(struct file *file, struct orangefs_kernel_op_s *op, *temp; __s32 proto_ver = ORANGEFS_KERNEL_PROTO_VERSION; static __s32 magic = ORANGEFS_DEVREQ_MAGIC; - struct orangefs_kernel_op_s *cur_op = NULL; + struct orangefs_kernel_op_s *cur_op; unsigned long ret; /* We do not support blocking IO. */ @@ -186,6 +186,7 @@ static ssize_t orangefs_devreq_read(struct file *file, return -EAGAIN; restart: + cur_op = NULL; /* Get next op (if any) from top of list. */ spin_lock(&orangefs_request_list_lock); list_for_each_entry_safe(op, temp, &orangefs_request_list, list) { From 1995266727fa8143897e89b55f5d3c79aa828420 Mon Sep 17 00:00:00 2001 From: Ben Hutchings Date: Mon, 22 Jan 2018 20:11:06 +0000 Subject: [PATCH 55/55] nfsd: auth: Fix gid sorting when rootsquash enabled Commit bdcf0a423ea1 ("kernel: make groups_sort calling a responsibility group_info allocators") appears to break nfsd rootsquash in a pretty major way. It adds a call to groups_sort() inside the loop that copies/squashes gids, which means the valid gids are sorted along with the following garbage. The net result is that the highest numbered valid gids are replaced with any lower-valued garbage gids, possibly including 0. We should sort only once, after filling in all the gids. Fixes: bdcf0a423ea1 ("kernel: make groups_sort calling a responsibility ...") Signed-off-by: Ben Hutchings Acked-by: J. Bruce Fields Signed-off-by: Linus Torvalds --- fs/nfsd/auth.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/fs/nfsd/auth.c b/fs/nfsd/auth.c index f650e475d8f0..fdf2aad73470 100644 --- a/fs/nfsd/auth.c +++ b/fs/nfsd/auth.c @@ -60,10 +60,10 @@ int nfsd_setuser(struct svc_rqst *rqstp, struct svc_export *exp) gi->gid[i] = exp->ex_anon_gid; else gi->gid[i] = rqgi->gid[i]; - - /* Each thread allocates its own gi, no race */ - groups_sort(gi); } + + /* Each thread allocates its own gi, no race */ + groups_sort(gi); } else { gi = get_group_info(rqgi); }