- Selftests fixes and improvements (Chris)
- More work around engine tracking for better handling (Chris, Tvrtko)
- HDCP debug and info improvements (Ram, Ashuman)
- Add DSI properties (Vandita)
- Rework on sdvo support for better debuggability before fixing bugs (Ville)
- Display PLLs fixes and improvements, specially targeting Ice Lake (Imre, Matt, Ville)
- Perf fixes and improvements (Lionel)
- Enumerate scratch buffers (Lionel)
- Add infra to hold off preemption on a request (Lionel)
- Ice Lake color space fixes (Uma)
- Type-C fixes and improvements (Lucas)
- Fix and improvements around workarounds (Chris, John, Tvrtko)
- GuC related fixes and improvements (Chris, Daniele, Michal, Tvrtko)
- Fix on VLV/CHV display power domain (Ville)
- Improvements around Watermark (Ville)
- Favor intel_ types on intel_atomic functions (Ville)
- Don’t pass stack garbage to pcode (Ville)
- Improve display tracepoints (Steven)
- Don’t overestimate 4:2:0 link symbol clock (Ville)
- Add support for 4th pipe and transcoder (Lucas)
- Introduce initial support for Tiger Lake platform (Daniele, Lucas, Mahesh, Jose, Imre, Mika, Vandita, Rodrigo, Michel)
- PPGTT allocation simplification (Chris)
- Standardize function names and suffixes to make clean, symmetric and let checkpatch happy (Janusz)
- Skip SINK_COUNT read on CH7511 (Ville)
- Fix on kernel documentation (Chris, Michal)
- Add modular FIA (Anusha, Lucas)
- Fix EHL display (Matt, Vivek)
- Enable hotplug retry (Imre, Jose)
- Disable preemption under GVT (Chris)
- OA; Reconfigure context on the fly (Chris)
- Fixes and improvements around engine reset. (Chris)
- Small clean up on display pipe fault mask (Ville)
- Make sure cdclk is high enough for DP audio on VLV/CHV (Ville)
- Drop some wmb() and improve pwrite flush (Chris)
- Fix critical PSR regression (DK)
- Remove unused variables (YueHaibing)
- Use dev_get_drvdata for simplification (Chunhong)
- Use upstream version of header tests (Jani)
-----BEGIN PGP SIGNATURE-----
iQEcBAABAgAGBQJdQJHXAAoJEPpiX2QO6xPKIpkH/3lMqbuv6UXyX1zvcYj6Ap4g
c6ocA7O1ooQDfFBfnLJNd6D+Gs3uTt9KROL0WdhmolfgzfLihFnvSx1VP/pvi7gC
kVT1JbwbzuwYbBXQ8WhmtkfqDp/quy3wku/ThNchY9pG1IaqNuRiP35+pXRNLO08
Q+RUHl8j1OkoLTLuzxfYGFtY72F8mIlkki8zMwlthH2Skz9h9d8POh8phOv+3TDx
aQ7CsOfScnLSrEyWlnOeYFexps0LpNC7TAG8fGkVI28Jig16DSwg7QR3MhQ9UtB1
8IC3+Jz8+p83PQHx7mGS7Va/XTERVT4czsoNC/IK7cFMy1yFilzoqpFHH8Is3sk=
=dAkP
-----END PGP SIGNATURE-----
Merge tag 'drm-intel-next-2019-07-30' of git://anongit.freedesktop.org/drm/drm-intel into drm-next
- More changes on simplifying locking mechanisms (Chris)
- Selftests fixes and improvements (Chris)
- More work around engine tracking for better handling (Chris, Tvrtko)
- HDCP debug and info improvements (Ram, Ashuman)
- Add DSI properties (Vandita)
- Rework on sdvo support for better debuggability before fixing bugs (Ville)
- Display PLLs fixes and improvements, specially targeting Ice Lake (Imre, Matt, Ville)
- Perf fixes and improvements (Lionel)
- Enumerate scratch buffers (Lionel)
- Add infra to hold off preemption on a request (Lionel)
- Ice Lake color space fixes (Uma)
- Type-C fixes and improvements (Lucas)
- Fix and improvements around workarounds (Chris, John, Tvrtko)
- GuC related fixes and improvements (Chris, Daniele, Michal, Tvrtko)
- Fix on VLV/CHV display power domain (Ville)
- Improvements around Watermark (Ville)
- Favor intel_ types on intel_atomic functions (Ville)
- Don’t pass stack garbage to pcode (Ville)
- Improve display tracepoints (Steven)
- Don’t overestimate 4:2:0 link symbol clock (Ville)
- Add support for 4th pipe and transcoder (Lucas)
- Introduce initial support for Tiger Lake platform (Daniele, Lucas, Mahesh, Jose, Imre, Mika, Vandita, Rodrigo, Michel)
- PPGTT allocation simplification (Chris)
- Standardize function names and suffixes to make clean, symmetric and let checkpatch happy (Janusz)
- Skip SINK_COUNT read on CH7511 (Ville)
- Fix on kernel documentation (Chris, Michal)
- Add modular FIA (Anusha, Lucas)
- Fix EHL display (Matt, Vivek)
- Enable hotplug retry (Imre, Jose)
- Disable preemption under GVT (Chris)
- OA; Reconfigure context on the fly (Chris)
- Fixes and improvements around engine reset. (Chris)
- Small clean up on display pipe fault mask (Ville)
- Make sure cdclk is high enough for DP audio on VLV/CHV (Ville)
- Drop some wmb() and improve pwrite flush (Chris)
- Fix critical PSR regression (DK)
- Remove unused variables (YueHaibing)
- Use dev_get_drvdata for simplification (Chunhong)
- Use upstream version of header tests (Jani)
drm-intel-next-2019-07-08:
- Signal fence completion from i915_request_wait (Chris)
- Fixes and improvements around rings pin/unpin (Chris)
- Display uncore prep patches (Daniele)
- Execlists preemption improvements (Chris)
- Selftests fixes and improvements (Chris)
- More Elkhartlake enabling work (Vandita, Jose, Matt, Vivek)
- Defer address space cleanup to an RCU worker (Chris)
- Implicit dev_priv removal and GT compartmentalization and other related follow-ups (Tvrtko, Chris)
- Prevent dereference of engine before NULL check in error capture (Chris)
- GuC related fixes (Daniele, Robert)
- Many changes on active tracking, timelines and locking mechanisms (Chris)
- Disable SAMPLER_STATE prefetching on Gen11 (HW W/a) (Kenneth)
- I915_perf fixes (Lionel)
- Add Ice Lake PCI ID (Mika)
- eDP backlight fix (Lee)
- Fix various gen2 tracepoints (Ville)
- Some irq vfunc clean-up and improvements (Ville)
- Move OA files to separated folder (Michal)
- Display self contained headers clean-up (Jani)
- Preparation for 4th pile (Lucas)
- Move atomic commit, watermark and other places to use more intel_crtc_state (Maarten)
- Many Ice Lake Type C and Thunderbolt fixes (Imre)
- Fix some Ice Lake hw w/a whitelist regs (Lionel)
- Fix memleak in runtime wakeref tracking (Mika)
- Remove unused Private PPAT manager (Michal)
- Don't check PPGTT presence on PPGTT-only platforms (Michal)
- Fix ICL DSI suspend/resume (Chris)
- Fix ICL Bandwidth issues (Ville)
- Add N & CTS values for 10/12 bit deep color (Aditya)
- Moving more GT related stuff under gt folder (Chris)
- Forcewake related fixes (Chris)
- Show support for accurate sw PMU busyness tracking (Chris)
- Handle gtt double alloc failures (Chris)
- Upgrade to new GuC version (Michal)
- Improve w/a debug dumps and pull engine w/a initialization into a common (Chris)
- Look for instdone on all engines at hangcheck (Tvrtko)
- Engine lookup simplification (Chris)
- Many plane color formats fixes and improvements (Ville)
- Fix some compilation issues (YueHaibing)
- GTT page directory clean up and improvements (Mika)
Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190801201314.GA23635@intel.com
Wire up the new clone3 syscall.
A fix for the PAPR SCM nvdimm driver, to fix a crash when firmware gives us a
device that's attached to a non-online NUMA node.
A fix for a boot failure on 32-bit with KASAN enabled.
Three fixes for implicit fall through warnings, some of which are errors for us
due to -Werror.
Thanks to:
Aneesh Kumar K.V, Christophe Leroy, Kees Cook, Santosh Sivaraj, Stephen
Rothwell.
-----BEGIN PGP SIGNATURE-----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=WT8n
-----END PGP SIGNATURE-----
Merge tag 'powerpc-5.3-3' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux
Pull powerpc fixes from Michael Ellerman:
"Some more powerpc fixes for 5.3:
- Wire up the new clone3 syscall.
- A fix for the PAPR SCM nvdimm driver, to fix a crash when firmware
gives us a device that's attached to a non-online NUMA node.
- A fix for a boot failure on 32-bit with KASAN enabled.
- Three fixes for implicit fall through warnings, some of which are
errors for us due to -Werror.
Thanks to: Aneesh Kumar K.V, Christophe Leroy, Kees Cook, Santosh
Sivaraj, Stephen Rothwell"
* tag 'powerpc-5.3-3' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:
powerpc/kasan: fix early boot failure on PPC32
drivers/macintosh/smu.c: Mark expected switch fall-through
powerpc/spe: Mark expected switch fall-throughs
powerpc/nvdimm: Pick nearby online node if the device node is not online
powerpc/kvm: Fall through switch case explicitly
powerpc: Wire up clone3 syscall
- fix build for xtensa cores with coprocessors that was broken by
entry/return abstraction patch.
-----BEGIN PGP SIGNATURE-----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=QOUM
-----END PGP SIGNATURE-----
Merge tag 'xtensa-20190803' of git://github.com/jcmvbkbc/linux-xtensa
Pull Xtensa fix from Max Filippov:
"Fix build for xtensa cores with coprocessors that was broken by
entry/return abstraction patch"
* tag 'xtensa-20190803' of git://github.com/jcmvbkbc/linux-xtensa:
xtensa: fix build for cores with coprocessors
Pull vdso timer fixes from Thomas Gleixner:
"A series of commits to deal with the regression caused by the generic
VDSO implementation.
The usage of clock_gettime64() for 32bit compat fallback syscalls
caused seccomp filters to kill innocent processes because they only
allow clock_gettime().
Handle the compat syscalls with clock_gettime() as before, which is
not a functional problem for the VDSO as the legacy compat application
interface is not y2038 safe anyway. It's just extra fallback code
which needs to be implemented on every architecture.
It's opt in for now so that it does not break the compile of already
converted architectures in linux-next. Once these are fixed, the
#ifdeffery goes away.
So much for trying to be smart and reuse code..."
* 'timers-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
arm64: compat: vdso: Use legacy syscalls as fallback
x86/vdso/32: Use 32bit syscall fallback
lib/vdso/32: Provide legacy syscall fallbacks
lib/vdso: Move fallback invocation to the callers
lib/vdso/32: Remove inconsistent NULL pointer checks
Merge misc fixes from Andrew Morton:
"17 fixes"
* emailed patches from Andrew Morton <akpm@linux-foundation.org>:
drivers/acpi/scan.c: document why we don't need the device_hotplug_lock
memremap: move from kernel/ to mm/
lib/test_meminit.c: use GFP_ATOMIC in RCU critical section
asm-generic: fix -Wtype-limits compiler warnings
cgroup: kselftest: relax fs_spec checks
mm/memory_hotplug.c: remove unneeded return for void function
mm/migrate.c: initialize pud_entry in migrate_vma()
coredump: split pipe command whitespace before expanding template
page flags: prioritize kasan bits over last-cpuid
ubsan: build ubsan.c more conservatively
kasan: remove clang version check for KASAN_STACK
mm: compaction: avoid 100% CPU usage during compaction when a task is killed
mm: migrate: fix reference check race between __find_get_block() and migration
mm: vmscan: check if mem cgroup is disabled or not before calling memcg slab shrinker
ocfs2: remove set but not used variable 'last_hash'
Revert "kmemleak: allow to coexist with fault injection"
kernel/signal.c: fix a kernel-doc markup
Three minor RISC-V-related changes for v5.3-rc3:
- Add build ID to VDSO builds to avoid a double-free in perf when
libelf isn't used
- Align the RV64 defconfig to the output of "make savedefconfig" so
subsequent defconfig patches don't get out of hand
- Drop a superfluous DT property from the FU540 SoC DT data (since it
must be already set in board data that includes it)
-----BEGIN PGP SIGNATURE-----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=CIhE
-----END PGP SIGNATURE-----
Merge tag 'riscv/for-v5.3-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux
Pull RISC-V fixes from Paul Walmsley:
"Three minor RISC-V-related changes for v5.3-rc3:
- Add build ID to VDSO builds to avoid a double-free in perf when
libelf isn't used
- Align the RV64 defconfig to the output of "make savedefconfig" so
subsequent defconfig patches don't get out of hand
- Drop a superfluous DT property from the FU540 SoC DT data (since it
must be already set in board data that includes it)"
* tag 'riscv/for-v5.3-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux:
riscv: defconfig: align RV64 defconfig to the output of "make savedefconfig"
riscv: dts: fu540-c000: drop "timebase-frequency"
riscv: Fix perf record without libelf support
ARM64 randdconfig builds regularly run into a build error, especially
when NUMA_BALANCING and SPARSEMEM are enabled but not SPARSEMEM_VMEMMAP:
#error "KASAN: not enough bits in page flags for tag"
The last-cpuid bits are already contitional on the available space, so
the result of the calculation is a bit random on whether they were
already left out or not.
Adding the kasan tag bits before last-cpuid makes it much more likely to
end up with a successful build here, and should be reliable for
randconfig at least, as long as that does not randomize NR_CPUS or
NODES_SHIFT but uses the defaults.
In order for the modified check to not trigger in the x86 vdso32 code
where all constants are wrong (building with -m32), enclose all the
definitions with an #ifdef.
[arnd@arndb.de: build fix]
Link: http://lkml.kernel.org/r/CAK8P3a3Mno1SWTcuAOT0Wa9VS15pdU6EfnkxLbDpyS55yO04+g@mail.gmail.com
Link: http://lkml.kernel.org/r/20190722115520.3743282-1-arnd@arndb.de
Link: https://lore.kernel.org/lkml/20190618095347.3850490-1-arnd@arndb.de/
Fixes: 2813b9c029 ("kasan, mm, arm64: tag non slab memory allocated via pagealloc")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Andrey Konovalov <andreyknvl@google.com>
Reviewed-by: Andrey Ryabinin <aryabinin@virtuozzo.com>
Cc: Andrey Konovalov <andreyknvl@google.com>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
- Update the compat layer to allow single-byte watchpoints on all
addresses (similar to the native support)
- arm_pmu: fix the restoration of the counters on the
CPU_PM_ENTER_FAILED path
- Fix build regression with vDSO and Makefile not stripping
CROSS_COMPILE_COMPAT
- Fix the CTR_EL0 (cache type register) sanitisation on heterogeneous
machines (e.g. big.LITTLE)
- Fix the interrupt controller priority mask value when pseudo-NMIs are
enabled
- arm64 kprobes fixes: recovering of the PSTATE.D flag in the
single-step exception handler, NOKPROBE annotations for unwind_frame()
and walk_stackframe(), remove unneeded rcu_read_lock/unlock from debug
handlers
- Several gcc fall-through warnings
- Unused variable warnings
-----BEGIN PGP SIGNATURE-----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=vA+f
-----END PGP SIGNATURE-----
Merge tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux
Pull arm64 fixes from Catalin Marinas:
- Update the compat layer to allow single-byte watchpoints on all
addresses (similar to the native support)
- arm_pmu: fix the restoration of the counters on the
CPU_PM_ENTER_FAILED path
- Fix build regression with vDSO and Makefile not stripping
CROSS_COMPILE_COMPAT
- Fix the CTR_EL0 (cache type register) sanitisation on heterogeneous
machines (e.g. big.LITTLE)
- Fix the interrupt controller priority mask value when pseudo-NMIs are
enabled
- arm64 kprobes fixes: recovering of the PSTATE.D flag in the
single-step exception handler, NOKPROBE annotations for
unwind_frame() and walk_stackframe(), remove unneeded
rcu_read_lock/unlock from debug handlers
- Several gcc fall-through warnings
- Unused variable warnings
* tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:
arm64: Make debug exception handlers visible from RCU
arm64: kprobes: Recover pstate.D in single-step exception handler
arm64/mm: fix variable 'tag' set but not used
arm64/mm: fix variable 'pud' set but not used
arm64: Remove unneeded rcu_read_lock from debug handlers
arm64: unwind: Prohibit probing on return_address()
arm64: Lower priority mask for GIC_PRIO_IRQON
arm64/efi: fix variable 'si' set but not used
arm64: cpufeature: Fix feature comparison for CTR_EL0.{CWG,ERG}
arm64: vdso: Fix Makefile regression
arm64: module: Mark expected switch fall-through
arm64: smp: Mark expected switch fall-through
arm64: hw_breakpoint: Fix warnings about implicit fallthrough
drivers/perf: arm_pmu: Fix failure path in PM notifier
arm64: compat: Allow single-byte watchpoints on all addresses
Pull parisc fixes from Helge Deller:
"A few small fixes for the parisc architecture:
- Fix fall-through warnings in parisc math emu code
- Fix vmlinuz linking failure with debug-enabled kernels
- Fix a race condition in kernel live-patching code
- Add missing archclean Makefile target & defconfig adjustments"
* 'parisc-5.3-4' of git://git.kernel.org/pub/scm/linux/kernel/git/deller/parisc-linux:
parisc: Add archclean Makefile target
parisc: Strip debug info from kernel before creating compressed vmlinuz
parisc: Fix build of compressed kernel even with debug enabled
parisc: fix race condition in patching code
parisc: rename default_defconfig to defconfig
parisc: Fix fall-through warnings in fpudispatch.c
parisc: Mark expected switch fall-throughs in fault.c
This fixes a cascade of regressions that originally started with
the addition of the ia64 port, but only got fatal once we removed
most uses of block layer bounce buffering in Linux 4.18.
The reason is that while the original i386/PAE code that was the first
architecture that supported > 4GB of memory without an iommu decided to
leave bounce buffering to the subsystems, which in those days just mean
block and networking as no one else consumer arbitrary userspace memory.
Later with ia64, x86_64 and other ports we assumed that either an iommu
or something that fakes it up ("software IOTLB" in beautiful Intel
speak) is present and that subsystems can rely on that for dealing with
addressing limitations in devices. Except that the ARM LPAE scheme
that added larger physical address to 32-bit ARM did not follow that
scheme and thus only worked by chance and only for block and networking
I/O directly to highmem.
Long story, short fix - add swiotlb support to arm when build for LPAE
platforms, which actuallys turns out to be pretty trivial with the
modern dma-direct / swiotlb code to fix the Linux 4.18-ish regression.
-----BEGIN PGP SIGNATURE-----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=D6Uy
-----END PGP SIGNATURE-----
Merge tag 'arm-swiotlb-5.3' of git://git.infradead.org/users/hch/dma-mapping
Pull arm swiotlb support from Christoph Hellwig:
"This fixes a cascade of regressions that originally started with the
addition of the ia64 port, but only got fatal once we removed most
uses of block layer bounce buffering in Linux 4.18.
The reason is that while the original i386/PAE code that was the first
architecture that supported > 4GB of memory without an iommu decided
to leave bounce buffering to the subsystems, which in those days just
mean block and networking as no one else consumed arbitrary userspace
memory.
Later with ia64, x86_64 and other ports we assumed that either an
iommu or something that fakes it up ("software IOTLB" in beautiful
Intel speak) is present and that subsystems can rely on that for
dealing with addressing limitations in devices. Except that the ARM
LPAE scheme that added larger physical address to 32-bit ARM did not
follow that scheme and thus only worked by chance and only for block
and networking I/O directly to highmem.
Long story, short fix - add swiotlb support to arm when build for LPAE
platforms, which actuallys turns out to be pretty trivial with the
modern dma-direct / swiotlb code to fix the Linux 4.18-ish regression"
* tag 'arm-swiotlb-5.3' of git://git.infradead.org/users/hch/dma-mapping:
arm: use swiotlb for bounce buffering on LPAE configs
dma-mapping: check pfn validity in dma_common_{mmap,get_sgtable}
Make debug exceptions visible from RCU so that synchronize_rcu()
correctly track the debug exception handler.
This also introduces sanity checks for user-mode exceptions as same
as x86's ist_enter()/ist_exit().
The debug exception can interrupt in idle task. For example, it warns
if we put a kprobe on a function called from idle task as below.
The warning message showed that the rcu_read_lock() caused this
problem. But actually, this means the RCU is lost the context which
is already in NMI/IRQ.
/sys/kernel/debug/tracing # echo p default_idle_call >> kprobe_events
/sys/kernel/debug/tracing # echo 1 > events/kprobes/enable
/sys/kernel/debug/tracing # [ 135.122237]
[ 135.125035] =============================
[ 135.125310] WARNING: suspicious RCU usage
[ 135.125581] 5.2.0-08445-g9187c508bdc7 #20 Not tainted
[ 135.125904] -----------------------------
[ 135.126205] include/linux/rcupdate.h:594 rcu_read_lock() used illegally while idle!
[ 135.126839]
[ 135.126839] other info that might help us debug this:
[ 135.126839]
[ 135.127410]
[ 135.127410] RCU used illegally from idle CPU!
[ 135.127410] rcu_scheduler_active = 2, debug_locks = 1
[ 135.128114] RCU used illegally from extended quiescent state!
[ 135.128555] 1 lock held by swapper/0/0:
[ 135.128944] #0: (____ptrval____) (rcu_read_lock){....}, at: call_break_hook+0x0/0x178
[ 135.130499]
[ 135.130499] stack backtrace:
[ 135.131192] CPU: 0 PID: 0 Comm: swapper/0 Not tainted 5.2.0-08445-g9187c508bdc7 #20
[ 135.131841] Hardware name: linux,dummy-virt (DT)
[ 135.132224] Call trace:
[ 135.132491] dump_backtrace+0x0/0x140
[ 135.132806] show_stack+0x24/0x30
[ 135.133133] dump_stack+0xc4/0x10c
[ 135.133726] lockdep_rcu_suspicious+0xf8/0x108
[ 135.134171] call_break_hook+0x170/0x178
[ 135.134486] brk_handler+0x28/0x68
[ 135.134792] do_debug_exception+0x90/0x150
[ 135.135051] el1_dbg+0x18/0x8c
[ 135.135260] default_idle_call+0x0/0x44
[ 135.135516] cpu_startup_entry+0x2c/0x30
[ 135.135815] rest_init+0x1b0/0x280
[ 135.136044] arch_call_rest_init+0x14/0x1c
[ 135.136305] start_kernel+0x4d4/0x500
[ 135.136597]
So make debug exception visible to RCU can fix this warning.
Reported-by: Naresh Kamboju <naresh.kamboju@linaro.org>
Acked-by: Paul E. McKenney <paulmck@linux.ibm.com>
Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org>
Signed-off-by: Will Deacon <will@kernel.org>
kprobes manipulates the interrupted PSTATE for single step, and
doesn't restore it. Thus, if we put a kprobe where the pstate.D
(debug) masked, the mask will be cleared after the kprobe hits.
Moreover, in the most complicated case, this can lead a kernel
crash with below message when a nested kprobe hits.
[ 152.118921] Unexpected kernel single-step exception at EL1
When the 1st kprobe hits, do_debug_exception() will be called.
At this point, debug exception (= pstate.D) must be masked (=1).
But if another kprobes hits before single-step of the first kprobe
(e.g. inside user pre_handler), it unmask the debug exception
(pstate.D = 0) and return.
Then, when the 1st kprobe setting up single-step, it saves current
DAIF, mask DAIF, enable single-step, and restore DAIF.
However, since "D" flag in DAIF is cleared by the 2nd kprobe, the
single-step exception happens soon after restoring DAIF.
This has been introduced by commit 7419333fa1 ("arm64: kprobe:
Always clear pstate.D in breakpoint exception handler")
To solve this issue, this stores all DAIF bits and restore it
after single stepping.
Reported-by: Naresh Kamboju <naresh.kamboju@linaro.org>
Fixes: 7419333fa1 ("arm64: kprobe: Always clear pstate.D in breakpoint exception handler")
Reviewed-by: James Morse <james.morse@arm.com>
Tested-by: James Morse <james.morse@arm.com>
Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org>
Signed-off-by: Will Deacon <will@kernel.org>
When CONFIG_KASAN_SW_TAGS=n, set_tag() is compiled away. GCC throws a
warning,
mm/kasan/common.c: In function '__kasan_kmalloc':
mm/kasan/common.c:464:5: warning: variable 'tag' set but not used
[-Wunused-but-set-variable]
u8 tag = 0xff;
^~~
Fix it by making __tag_set() a static inline function the same as
arch_kasan_set_tag() in mm/kasan/kasan.h for consistency because there
is a macro in arch/arm64/include/asm/kasan.h,
#define arch_kasan_set_tag(addr, tag) __tag_set(addr, tag)
However, when CONFIG_DEBUG_VIRTUAL=n and CONFIG_SPARSEMEM_VMEMMAP=y,
page_to_virt() will call __tag_set() with incorrect type of a
parameter, so fix that as well. Also, still let page_to_virt() return
"void *" instead of "const void *", so will not need to add a similar
cast in lowmem_page_address().
Signed-off-by: Qian Cai <cai@lca.pw>
Signed-off-by: Will Deacon <will@kernel.org>
GCC throws a warning,
arch/arm64/mm/mmu.c: In function 'pud_free_pmd_page':
arch/arm64/mm/mmu.c:1033:8: warning: variable 'pud' set but not used
[-Wunused-but-set-variable]
pud_t pud;
^~~
because pud_table() is a macro and compiled away. Fix it by making it a
static inline function and for pud_sect() as well.
Signed-off-by: Qian Cai <cai@lca.pw>
Signed-off-by: Will Deacon <will@kernel.org>
Remove rcu_read_lock()/rcu_read_unlock() from debug exception
handlers since we are sure those are not preemptible and
interrupts are off.
Acked-by: Paul E. McKenney <paulmck@linux.ibm.com>
Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org>
Signed-off-by: Will Deacon <will@kernel.org>
Prohibit probing on return_address() and subroutines which
is called from return_address(), since the it is invoked from
trace_hardirqs_off() which is also kprobe blacklisted.
Reported-by: Naresh Kamboju <naresh.kamboju@linaro.org>
Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org>
Signed-off-by: Will Deacon <will@kernel.org>
On a system with two security states, if SCR_EL3.FIQ is cleared,
non-secure IRQ priorities get shifted to fit the secure view but
priority masks aren't.
On such system, it turns out that GIC_PRIO_IRQON masks the priority of
normal interrupts, which obviously ends up in a hang.
Increase GIC_PRIO_IRQON value (i.e. lower priority) to make sure
interrupts are not blocked by it.
Cc: Oleg Nesterov <oleg@redhat.com>
Fixes: bd82d4bd21 ("arm64: Fix incorrect irqflag restore for priority masking")
Acked-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Julien Thierry <julien.thierry.kdev@gmail.com>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
[will: fixed Fixes: tag]
Signed-off-by: Will Deacon <will@kernel.org>
Apparently we don't have an archclean target in our
arch/parisc/Makefile, so files in there never get cleaned out by make
mrproper. This, in turn means that the sizes.h file in
arch/parisc/boot/compressed never gets removed and worse, when you
transition to an O=build/parisc[64] build model it overrides the
generated file. The upshot being my bzImage was building with a SZ_end
that was too small.
I fixed it by making mrproper clean everything.
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
Cc: stable@vger.kernel.org # v4.20+
Signed-off-by: Helge Deller <deller@gmx.de>
Same as on x86-64, strip the .comment, .note and debug sections from the
Linux kernel before creating the compressed image for the boot loader.
Reported-by: James Bottomley <James.Bottomley@HansenPartnership.com>
Reported-by: Sven Schnelle <svens@stackframe.org>
Cc: stable@vger.kernel.org # v4.20+
Signed-off-by: Helge Deller <deller@gmx.de>
With debug info enabled (CONFIG_DEBUG_INFO=y) the resulting vmlinux may get
that huge that we need to increase the start addresss for the decompression
text section otherwise one will face a linker error.
Reported-by: Sven Schnelle <svens@stackframe.org>
Tested-by: Sven Schnelle <svens@stackframe.org>
Cc: stable@vger.kernel.org # v4.14+
Signed-off-by: Helge Deller <deller@gmx.de>
Align the RV64 defconfig to the output of "make savedefconfig" to
avoid unnecessary deltas for future defconfig patches. This patch
should have no runtime functional impact.
Signed-off-by: Paul Walmsley <paul.walmsley@sifive.com>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
On FU540-based systems, the "timebase-frequency" (RTCCLK) is sourced
from an external crystal located on the PCB. Thus the
timebase-frequency DT property should be defined by the board that
uses the SoC, not the SoC itself. Drop the superfluous
timebase-frequency property from the SoC DT data. (It's already
present in the board DT data.)
Signed-off-by: Paul Walmsley <paul.walmsley@sifive.com>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
This patch fix following perf record error by linking vdso.so with
build id.
perf.data perf.data.old
[ perf record: Woken up 1 times to write data ]
free(): double free detected in tcache 2
Aborted
perf record use filename__read_build_id(util/symbol-minimal.c) to get
build id when libelf is not supported. When vdso.so is linked without
build id, the section size of PT_NOTE will be zero, buf size will
realloc to zero and cause memory corruption.
Signed-off-by: Mao Han <han_mao@c-sky.com>
Cc: Paul Walmsley <paul.walmsley@sifive.com>
Cc: Palmer Dabbelt <palmer@sifive.com>
Cc: Albert Ou <aou@eecs.berkeley.edu>
Signed-off-by: Paul Walmsley <paul.walmsley@sifive.com>
GCC throws out this warning on arm64.
drivers/firmware/efi/libstub/arm-stub.c: In function 'efi_entry':
drivers/firmware/efi/libstub/arm-stub.c:132:22: warning: variable 'si'
set but not used [-Wunused-but-set-variable]
Fix it by making free_screen_info() a static inline function.
Acked-by: Will Deacon <will@kernel.org>
Signed-off-by: Qian Cai <cai@lca.pw>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
If CTR_EL0.{CWG,ERG} are 0b0000 then they must be interpreted to have
their architecturally maximum values, which defeats the use of
FTR_HIGHER_SAFE when sanitising CPU ID registers on heterogeneous
machines.
Introduce FTR_HIGHER_OR_ZERO_SAFE so that these fields effectively
saturate at zero.
Fixes: 3c739b5710 ("arm64: Keep track of CPU feature registers")
Cc: <stable@vger.kernel.org> # 4.4.x-
Reviewed-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Acked-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Will Deacon <will@kernel.org>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Using an old .config in combination with "make oldconfig" can cause
an incorrect detection of the compat compiler:
$ grep CROSS_COMPILE_COMPAT .config
CONFIG_CROSS_COMPILE_COMPAT_VDSO=""
$ make oldconfig && make
arch/arm64/Makefile:58: gcc not found, check CROSS_COMPILE_COMPAT.
Stop.
Accordingly to the section 7.2 of the GNU Make manual "Syntax of
Conditionals", "When the value results from complex expansions of
variables and functions, expansions you would consider empty may
actually contain whitespace characters and thus are not seen as
empty. However, you can use the strip function to avoid interpreting
whitespace as a non-empty value."
Fix the issue adding strip to the CROSS_COMPILE_COMPAT string
evaluation.
Reported-by: Matteo Croce <mcroce@redhat.com>
Tested-by: Matteo Croce <mcroce@redhat.com>
Acked-by: Will Deacon <will@kernel.org>
Signed-off-by: Vincenzo Frascino <vincenzo.frascino@arm.com>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Assume the following ftrace code sequence that was patched in earlier by
ftrace_make_call():
PAGE A:
ffc: addr of ftrace_caller()
PAGE B:
000: 0x6fc10080 /* stw,ma r1,40(sp) */
004: 0x48213fd1 /* ldw -18(r1),r1 */
008: 0xe820c002 /* bv,n r0(r1) */
00c: 0xe83f1fdf /* b,l,n .-c,r1 */
When a Code sequences that is to be patched spans a page break, we might
have already cleared the part on the PAGE A. If an interrupt is coming in
during the remap of the fixed mapping to PAGE B, it might execute the
patched function with only parts of the FTRACE code cleared. To prevent
this, clear the jump to our mini trampoline first, and clear the remaining
parts after this. This might also happen when patch_text() patches a
function that it calls during remap.
Signed-off-by: Sven Schnelle <svens@stackframe.org>
Cc: <stable@vger.kernel.org> # 5.2+
Signed-off-by: Helge Deller <deller@gmx.de>
'default_defconfig' is an awkward name since 'defconfig' is the default.
Let's simply say 'defconfig' like other architectures. You can drop the
KBUILD_DEFCONFIG define by following the standard naming.
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Signed-off-by: Helge Deller <deller@gmx.de>
In fpudispatch.c we see a lot of fall-through warnings, but for this file we
prefer to not mark the switches and instead keep it in it's original state as
it's copied from HP-UX.
Fixes: a035d552a9 ("Makefile: Globally enable fall-through warning")
Signed-off-by: Helge Deller <deller@gmx.de>
The generic VDSO implementation uses the Y2038 safe clock_gettime64() and
clock_getres_time64() syscalls as fallback for 32bit VDSO. This breaks
seccomp setups because these syscalls might be not (yet) allowed.
Implement the 32bit variants which use the legacy syscalls and select the
variant in the core library.
The 64bit time variants are not removed because they are required for the
time64 based vdso accessors.
Fixes: 00b26474c2 ("lib/vdso: Provide generic VDSO implementation")
Reported-by: Sean Christopherson <sean.j.christopherson@intel.com>
Reported-by: Paul Bolle <pebolle@tiscali.nl>
Suggested-by: Andy Lutomirski <luto@kernel.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Tested-by: Vincenzo Frascino <vincenzo.frascino@arm.com>
Reviewed-by: Vincenzo Frascino <vincenzo.frascino@arm.com>
Link: https://lkml.kernel.org/r/20190728131648.971361611@linutronix.de
The generic VDSO implementation uses the Y2038 safe clock_gettime64() and
clock_getres_time64() syscalls as fallback for 32bit VDSO. This breaks
seccomp setups because these syscalls might be not (yet) allowed.
Implement the 32bit variants which use the legacy syscalls and select the
variant in the core library.
The 64bit time variants are not removed because they are required for the
time64 based vdso accessors.
Fixes: 7ac8707479 ("x86/vdso: Switch to generic vDSO implementation")
Reported-by: Sean Christopherson <sean.j.christopherson@intel.com>
Reported-by: Paul Bolle <pebolle@tiscali.nl>
Suggested-by: Andy Lutomirski <luto@kernel.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Vincenzo Frascino <vincenzo.frascino@arm.com>
Reviewed-by: Andy Lutomirski <luto@kernel.org>
Link: https://lkml.kernel.org/r/20190728131648.879156507@linutronix.de
Mark switch cases where we are expecting to fall through.
Fixes errors such as below, seen with mpc85xx_defconfig:
arch/powerpc/kernel/align.c: In function 'emulate_spe':
arch/powerpc/kernel/align.c:178:8: error: this statement may fall through
ret |= __get_user_inatomic(temp.v[3], p++);
^~
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20190730141917.21817-1-mpe@ellerman.id.au
Currently, nvdimm subsystem expects the device numa node for SCM device to be
an online node. It also doesn't try to bring the device numa node online. Hence
if we use a non-online numa node as device node we hit crashes like below. This
is because we try to access uninitialized NODE_DATA in different code paths.
cpu 0x0: Vector: 300 (Data Access) at [c0000000fac53170]
pc: c0000000004bbc50: ___slab_alloc+0x120/0xca0
lr: c0000000004bc834: __slab_alloc+0x64/0xc0
sp: c0000000fac53400
msr: 8000000002009033
dar: 73e8
dsisr: 80000
current = 0xc0000000fabb6d80
paca = 0xc000000003870000 irqmask: 0x03 irq_happened: 0x01
pid = 7, comm = kworker/u16:0
Linux version 5.2.0-06234-g76bd729b2644 (kvaneesh@ltc-boston123) (gcc version 7.4.0 (Ubuntu 7.4.0-1ubuntu1~18.04.1)) #135 SMP Thu Jul 11 05:36:30 CDT 2019
enter ? for help
[link register ] c0000000004bc834 __slab_alloc+0x64/0xc0
[c0000000fac53400] c0000000fac53480 (unreliable)
[c0000000fac53500] c0000000004bc818 __slab_alloc+0x48/0xc0
[c0000000fac53560] c0000000004c30a0 __kmalloc_node_track_caller+0x3c0/0x6b0
[c0000000fac535d0] c000000000cfafe4 devm_kmalloc+0x74/0xc0
[c0000000fac53600] c000000000d69434 nd_region_activate+0x144/0x560
[c0000000fac536d0] c000000000d6b19c nd_region_probe+0x17c/0x370
[c0000000fac537b0] c000000000d6349c nvdimm_bus_probe+0x10c/0x230
[c0000000fac53840] c000000000cf3cc4 really_probe+0x254/0x4e0
[c0000000fac538d0] c000000000cf429c driver_probe_device+0x16c/0x1e0
[c0000000fac53950] c000000000cf0b44 bus_for_each_drv+0x94/0x130
[c0000000fac539b0] c000000000cf392c __device_attach+0xdc/0x200
[c0000000fac53a50] c000000000cf231c bus_probe_device+0x4c/0xf0
[c0000000fac53a90] c000000000ced268 device_add+0x528/0x810
[c0000000fac53b60] c000000000d62a58 nd_async_device_register+0x28/0xa0
[c0000000fac53bd0] c0000000001ccb8c async_run_entry_fn+0xcc/0x1f0
[c0000000fac53c50] c0000000001bcd9c process_one_work+0x46c/0x860
[c0000000fac53d20] c0000000001bd4f4 worker_thread+0x364/0x5f0
[c0000000fac53db0] c0000000001c7260 kthread+0x1b0/0x1c0
[c0000000fac53e20] c00000000000b954 ret_from_kernel_thread+0x5c/0x68
The patch tries to fix this by picking the nearest online node as the SCM node.
This does have a problem of us losing the information that SCM node is
equidistant from two other online nodes. If applications need to understand these
fine-grained details we should express then like x86 does via
/sys/devices/system/node/nodeX/accessY/initiators/
With the patch we get
# numactl -H
available: 2 nodes (0-1)
node 0 cpus:
node 0 size: 0 MB
node 0 free: 0 MB
node 1 cpus: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
node 1 size: 130865 MB
node 1 free: 129130 MB
node distances:
node 0 1
0: 10 20
1: 20 10
# cat /sys/bus/nd/devices/region0/numa_node
0
# dmesg | grep papr_scm
[ 91.332305] papr_scm ibm,persistent-memory:ibm,pmemory@44104001: Region registered with target node 2 and online node 0
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20190729095128.23707-1-aneesh.kumar@linux.ibm.com
Since gmap_test_and_clear_dirty_pmd is not exported and has no reason to
be globally visible make it static to avoid the following sparse warning:
arch/s390/mm/gmap.c:2427:6: warning: symbol 'gmap_test_and_clear_dirty_pmd' was not declared. Should it be static?
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
Include <asm/kexec.h> into machine_kexec_reloc.c to expose
arch_kexec_do_relocs declaration and avoid the following sparse warnings:
arch/s390/kernel/machine_kexec_reloc.c:4:5: warning: symbol 'arch_kexec_do_relocs' was not declared. Should it be static?
arch/s390/boot/../kernel/machine_kexec_reloc.c:4:5: warning: symbol 'arch_kexec_do_relocs' was not declared. Should it be static?
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
Since there is really no reason for cf_diag_csd per cpu variable to be
globally visible make it static to avoid the following sparse warning:
arch/s390/kernel/perf_cpum_cf_diag.c:37:1: warning: symbol 'cf_diag_csd' was not declared. Should it be static?
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
Include <asm/xor.h> into arch/s390/lib/xor.c to expose xor_block_xc
declaration and avoid the following sparse warning:
arch/s390/lib/xor.c:128:27: warning: symbol 'xor_block_xc' was not declared. Should it be static?
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
Add __swsusp_reset_dma declaration to avoid the following sparse warnings:
arch/s390/kernel/setup.c:107:15: warning: symbol '__swsusp_reset_dma' was not declared. Should it be static?
arch/s390/boot/startup.c:52:15: warning: symbol '__swsusp_reset_dma' was not declared. Should it be static?
Add verify_facilities declaration to avoid the following sparse warning:
arch/s390/boot/als.c:105:6: warning: symbol 'verify_facilities' was not declared. Should it be static?
Include "boot.h" into arch/s390/boot/kaslr.c to expose get_random_base
function declaration and avoid the following sparse warning:
arch/s390/boot/kaslr.c:90:15: warning: symbol 'get_random_base' was not declared. Should it be static?
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
Fix two typos, document missing fields in the driver initialization
data and remove the copy&pasted 'pfmt' field from the qdr struct.
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Reviewed-by: Jens Remus <jremus@linux.ibm.com>
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
Implicit fallthrough warning was enabled globally which broke
the build. Make it explicit with a `fall through` comment.
Signed-off-by: Santosh Sivaraj <santosh@fossix.org>
Reviewed-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20190729055536.25591-1-santosh@fossix.org
When fall-through warnings was enabled by default the following warnings
was starting to show up:
../arch/arm64/kernel/module.c: In function ‘apply_relocate_add’:
../arch/arm64/kernel/module.c:316:19: warning: this statement may fall
through [-Wimplicit-fallthrough=]
overflow_check = false;
~~~~~~~~~~~~~~~^~~~~~~
../arch/arm64/kernel/module.c:317:3: note: here
case R_AARCH64_MOVW_UABS_G0:
^~~~
../arch/arm64/kernel/module.c:322:19: warning: this statement may fall
through [-Wimplicit-fallthrough=]
overflow_check = false;
~~~~~~~~~~~~~~~^~~~~~~
../arch/arm64/kernel/module.c:323:3: note: here
case R_AARCH64_MOVW_UABS_G1:
^~~~
Rework so that the compiler doesn't warn about fall-through.
Fixes: d93512ef0f0e ("Makefile: Globally enable fall-through warning")
Signed-off-by: Anders Roxell <anders.roxell@linaro.org>
Signed-off-by: Will Deacon <will@kernel.org>
When fall-through warnings was enabled by default the following warning
was starting to show up:
In file included from ../include/linux/kernel.h:15,
from ../include/linux/list.h:9,
from ../include/linux/kobject.h:19,
from ../include/linux/of.h:17,
from ../include/linux/irqdomain.h:35,
from ../include/linux/acpi.h:13,
from ../arch/arm64/kernel/smp.c:9:
../arch/arm64/kernel/smp.c: In function ‘__cpu_up’:
../include/linux/printk.h:302:2: warning: this statement may fall
through [-Wimplicit-fallthrough=]
printk(KERN_CRIT pr_fmt(fmt), ##__VA_ARGS__)
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
../arch/arm64/kernel/smp.c:156:4: note: in expansion of macro ‘pr_crit’
pr_crit("CPU%u: may not have shut down cleanly\n", cpu);
^~~~~~~
../arch/arm64/kernel/smp.c:157:3: note: here
case CPU_STUCK_IN_KERNEL:
^~~~
Rework so that the compiler doesn't warn about fall-through.
Fixes: d93512ef0f0e ("Makefile: Globally enable fall-through warning")
Signed-off-by: Anders Roxell <anders.roxell@linaro.org>
Signed-off-by: Will Deacon <will@kernel.org>
Now that -Wimplicit-fallthrough is passed to GCC by default, the kernel
build has suddenly got noisy. Annotate the two fall-through cases in our
hw_breakpoint implementation, since they are both intentional.
Reported-by: Anders Roxell <anders.roxell@linaro.org>
Signed-off-by: Will Deacon <will@kernel.org>
Commit d968d2b801 ("ARM: 7497/1: hw_breakpoint: allow single-byte
watchpoints on all addresses") changed the validation requirements for
hardware watchpoints on arch/arm/. Update our compat layer to implement
the same relaxation.
Cc: <stable@vger.kernel.org>
Signed-off-by: Will Deacon <will@kernel.org>