Pull folio updates from Matthew Wilcox:
- Rewrite how munlock works to massively reduce the contention on
i_mmap_rwsem (Hugh Dickins):
https://lore.kernel.org/linux-mm/8e4356d-9622-a7f0-b2c-f116b5f2efea@google.com/
- Sort out the page refcount mess for ZONE_DEVICE pages (Christoph
Hellwig):
https://lore.kernel.org/linux-mm/20220210072828.2930359-1-hch@lst.de/
- Convert GUP to use folios and make pincount available for order-1
pages. (Matthew Wilcox)
- Convert a few more truncation functions to use folios (Matthew
Wilcox)
- Convert page_vma_mapped_walk to use PFNs instead of pages (Matthew
Wilcox)
- Convert rmap_walk to use folios (Matthew Wilcox)
- Convert most of shrink_page_list() to use a folio (Matthew Wilcox)
- Add support for creating large folios in readahead (Matthew Wilcox)
* tag 'folio-5.18c' of git://git.infradead.org/users/willy/pagecache: (114 commits)
mm/damon: minor cleanup for damon_pa_young
selftests/vm/transhuge-stress: Support file-backed PMD folios
mm/filemap: Support VM_HUGEPAGE for file mappings
mm/readahead: Switch to page_cache_ra_order
mm/readahead: Align file mappings for non-DAX
mm/readahead: Add large folio readahead
mm: Support arbitrary THP sizes
mm: Make large folios depend on THP
mm: Fix READ_ONLY_THP warning
mm/filemap: Allow large folios to be added to the page cache
mm: Turn can_split_huge_page() into can_split_folio()
mm/vmscan: Convert pageout() to take a folio
mm/vmscan: Turn page_check_references() into folio_check_references()
mm/vmscan: Account large folios correctly
mm/vmscan: Optimise shrink_page_list for non-PMD-sized folios
mm/vmscan: Free non-shmem folios without splitting them
mm/rmap: Constify the rmap_walk_control argument
mm/rmap: Convert rmap_walk() to take a folio
mm: Turn page_anon_vma() into folio_anon_vma()
mm/rmap: Turn page_lock_anon_vma_read() into folio_lock_anon_vma_read()
...
With the advent of various new memory types, some machines will have
multiple types of memory, e.g. DRAM and PMEM (persistent memory). The
memory subsystem of these machines can be called memory tiering system,
because the performance of the different types of memory are usually
different.
In such system, because of the memory accessing pattern changing etc,
some pages in the slow memory may become hot globally. So in this
patch, the NUMA balancing mechanism is enhanced to optimize the page
placement among the different memory types according to hot/cold
dynamically.
In a typical memory tiering system, there are CPUs, fast memory and slow
memory in each physical NUMA node. The CPUs and the fast memory will be
put in one logical node (called fast memory node), while the slow memory
will be put in another (faked) logical node (called slow memory node).
That is, the fast memory is regarded as local while the slow memory is
regarded as remote. So it's possible for the recently accessed pages in
the slow memory node to be promoted to the fast memory node via the
existing NUMA balancing mechanism.
The original NUMA balancing mechanism will stop to migrate pages if the
free memory of the target node becomes below the high watermark. This
is a reasonable policy if there's only one memory type. But this makes
the original NUMA balancing mechanism almost do not work to optimize
page placement among different memory types. Details are as follows.
It's the common cases that the working-set size of the workload is
larger than the size of the fast memory nodes. Otherwise, it's
unnecessary to use the slow memory at all. So, there are almost always
no enough free pages in the fast memory nodes, so that the globally hot
pages in the slow memory node cannot be promoted to the fast memory
node. To solve the issue, we have 2 choices as follows,
a. Ignore the free pages watermark checking when promoting hot pages
from the slow memory node to the fast memory node. This will
create some memory pressure in the fast memory node, thus trigger
the memory reclaiming. So that, the cold pages in the fast memory
node will be demoted to the slow memory node.
b. Define a new watermark called wmark_promo which is higher than
wmark_high, and have kswapd reclaiming pages until free pages reach
such watermark. The scenario is as follows: when we want to promote
hot-pages from a slow memory to a fast memory, but fast memory's free
pages would go lower than high watermark with such promotion, we wake
up kswapd with wmark_promo watermark in order to demote cold pages and
free us up some space. So, next time we want to promote hot-pages we
might have a chance of doing so.
The choice "a" may create high memory pressure in the fast memory node.
If the memory pressure of the workload is high, the memory pressure
may become so high that the memory allocation latency of the workload
is influenced, e.g. the direct reclaiming may be triggered.
The choice "b" works much better at this aspect. If the memory
pressure of the workload is high, the hot pages promotion will stop
earlier because its allocation watermark is higher than that of the
normal memory allocation. So in this patch, choice "b" is implemented.
A new zone watermark (WMARK_PROMO) is added. Which is larger than the
high watermark and can be controlled via watermark_scale_factor.
In addition to the original page placement optimization among sockets,
the NUMA balancing mechanism is extended to be used to optimize page
placement according to hot/cold among different memory types. So the
sysctl user space interface (numa_balancing) is extended in a backward
compatible way as follow, so that the users can enable/disable these
functionality individually.
The sysctl is converted from a Boolean value to a bits field. The
definition of the flags is,
- 0: NUMA_BALANCING_DISABLED
- 1: NUMA_BALANCING_NORMAL
- 2: NUMA_BALANCING_MEMORY_TIERING
We have tested the patch with the pmbench memory accessing benchmark
with the 80:20 read/write ratio and the Gauss access address
distribution on a 2 socket Intel server with Optane DC Persistent
Memory Model. The test results shows that the pmbench score can
improve up to 95.9%.
Thanks Andrew Morton to help fix the document format error.
Link: https://lkml.kernel.org/r/20220221084529.1052339-3-ying.huang@intel.com
Signed-off-by: "Huang, Ying" <ying.huang@intel.com>
Tested-by: Baolin Wang <baolin.wang@linux.alibaba.com>
Reviewed-by: Baolin Wang <baolin.wang@linux.alibaba.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Reviewed-by: Oscar Salvador <osalvador@suse.de>
Reviewed-by: Yang Shi <shy828301@gmail.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Rik van Riel <riel@surriel.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Zi Yan <ziy@nvidia.com>
Cc: Wei Xu <weixugc@google.com>
Cc: Shakeel Butt <shakeelb@google.com>
Cc: zhongjiang-ali <zhongjiang-ali@linux.alibaba.com>
Cc: Randy Dunlap <rdunlap@infradead.org>
Cc: Feng Tang <feng.tang@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Patch series "mm: enforce pageblock_order < MAX_ORDER".
Having pageblock_order >= MAX_ORDER seems to be able to happen in corner
cases and some parts of the kernel are not prepared for it.
For example, Aneesh has shown [1] that such kernels can be compiled on
ppc64 with 64k base pages by setting FORCE_MAX_ZONEORDER=8, which will
run into a WARN_ON_ONCE(order >= MAX_ORDER) in comapction code right
during boot.
We can get pageblock_order >= MAX_ORDER when the default hugetlb size is
bigger than the maximum allocation granularity of the buddy, in which
case we are no longer talking about huge pages but instead gigantic
pages.
Having pageblock_order >= MAX_ORDER can only make alloc_contig_range()
of such gigantic pages more likely to succeed.
Reliable use of gigantic pages either requires boot time allcoation or
CMA, no need to overcomplicate some places in the kernel to optimize for
corner cases that are broken in other areas of the kernel.
This patch (of 2):
Let's enforce pageblock_order < MAX_ORDER and simplify.
Especially patch #1 can be regarded a cleanup before:
[PATCH v5 0/6] Use pageblock_order for cma and alloc_contig_range
alignment. [2]
[1] https://lkml.kernel.org/r/87r189a2ks.fsf@linux.ibm.com
[2] https://lkml.kernel.org/r/20220211164135.1803616-1-zi.yan@sent.com
Link: https://lkml.kernel.org/r/20220214174132.219303-2-david@redhat.com
Signed-off-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Zi Yan <ziy@nvidia.com>
Acked-by: Rob Herring <robh@kernel.org>
Cc: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Frank Rowand <frowand.list@gmail.com>
Cc: Michael S. Tsirkin <mst@redhat.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Marek Szyprowski <m.szyprowski@samsung.com>
Cc: Robin Murphy <robin.murphy@arm.com>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: John Garry via iommu <iommu@lists.linux-foundation.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull scheduler updates from Ingo Molnar:
- Cleanups for SCHED_DEADLINE
- Tracing updates/fixes
- CPU Accounting fixes
- First wave of changes to optimize the overhead of the scheduler
build, from the fast-headers tree - including placeholder *_api.h
headers for later header split-ups.
- Preempt-dynamic using static_branch() for ARM64
- Isolation housekeeping mask rework; preperatory for further changes
- NUMA-balancing: deal with CPU-less nodes
- NUMA-balancing: tune systems that have multiple LLC cache domains per
node (eg. AMD)
- Updates to RSEQ UAPI in preparation for glibc usage
- Lots of RSEQ/selftests, for same
- Add Suren as PSI co-maintainer
* tag 'sched-core-2022-03-22' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (81 commits)
sched/headers: ARM needs asm/paravirt_api_clock.h too
sched/numa: Fix boot crash on arm64 systems
headers/prep: Fix header to build standalone: <linux/psi.h>
sched/headers: Only include <linux/entry-common.h> when CONFIG_GENERIC_ENTRY=y
cgroup: Fix suspicious rcu_dereference_check() usage warning
sched/preempt: Tell about PREEMPT_DYNAMIC on kernel headers
sched/topology: Remove redundant variable and fix incorrect type in build_sched_domains
sched/deadline,rt: Remove unused parameter from pick_next_[rt|dl]_entity()
sched/deadline,rt: Remove unused functions for !CONFIG_SMP
sched/deadline: Use __node_2_[pdl|dle]() and rb_first_cached() consistently
sched/deadline: Merge dl_task_can_attach() and dl_cpu_busy()
sched/deadline: Move bandwidth mgmt and reclaim functions into sched class source file
sched/deadline: Remove unused def_dl_bandwidth
sched/tracing: Report TASK_RTLOCK_WAIT tasks as TASK_UNINTERRUPTIBLE
sched/tracing: Don't re-read p->state when emitting sched_switch event
sched/rt: Plug rt_mutex_setprio() vs push_rt_task() race
sched/cpuacct: Remove redundant RCU read lock
sched/cpuacct: Optimize away RCU read lock
sched/cpuacct: Fix charge percpu cpuusage
sched/headers: Reorganize, clean up and optimize kernel/sched/sched.h dependencies
...
Pull locking updates from Ingo Molnar:
"Changes in this cycle were:
Bitops & cpumask:
- Always inline various generic helpers, to improve code generation,
but also for instrumentation, found by noinstr validation.
- Add a x86-specific cpumask_clear_cpu() helper to improve code
generation
Atomics:
- Fix atomic64_{read_acquire,set_release} fallbacks
Lockdep:
- Fix /proc/lockdep output loop iteration for classes
- Fix /proc/lockdep potential access to invalid memory
- Add Mark Rutland as reviewer for atomic primitives
- Minor cleanups
Jump labels:
- Clean up the code a bit
Misc:
- Add __sched annotations to percpu rwsem primitives
- Enable RT_MUTEXES on PREEMPT_RT by default
- Stray v8086_mode() inlining fix, result of noinstr objtool
validation"
* tag 'locking-core-2022-03-21' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
jump_label: Refactor #ifdef of struct static_key
jump_label: Avoid unneeded casts in STATIC_KEY_INIT_{TRUE,FALSE}
locking/lockdep: Iterate lock_classes directly when reading lockdep files
x86/ptrace: Always inline v8086_mode() for instrumentation
cpumask: Add a x86-specific cpumask_clear_cpu() helper
locking: Enable RT_MUTEXES by default on PREEMPT_RT.
locking/local_lock: Make the empty local_lock_*() function a macro.
atomics: Fix atomic64_{read_acquire,set_release} fallbacks
locking: Add missing __sched attributes
cpumask: Always inline helpers which use bit manipulation functions
asm-generic/bitops: Always inline all bit manipulation helpers
locking/lockdep: Avoid potential access of invalid memory in lock_class
lockdep: Use memset_startat() helper in reinit_class()
MAINTAINERS: add myself as reviewer for atomics
Pull x86 perf event updates from Ingo Molnar:
- Fix address filtering for Intel/PT,ARM/CoreSight
- Enable Intel/PEBS format 5
- Allow more fixed-function counters for x86
- Intel/PT: Enable not recording Taken-Not-Taken packets
- Add a few branch-types
* tag 'perf-core-2022-03-21' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
perf/x86/intel/uncore: Fix the build on !CONFIG_PHYS_ADDR_T_64BIT
perf: Add irq and exception return branch types
perf/x86/intel/uncore: Make uncore_discovery clean for 64 bit addresses
perf/x86/intel/pt: Add a capability and config bit for disabling TNTs
perf/x86/intel/pt: Add a capability and config bit for event tracing
perf/x86/intel: Increase max number of the fixed counters
KVM: x86: use the KVM side max supported fixed counter
perf/x86/intel: Enable PEBS format 5
perf/core: Allow kernel address filter when not filtering the kernel
perf/x86/intel/pt: Fix address filter config for 32-bit kernel
perf/core: Fix address filter parser for multiple filters
x86: Share definition of __is_canonical_address()
perf/x86/intel/pt: Relax address filter validation
Pull nfsd updates from Chuck Lever:
"New features:
- NFSv3 support in NFSD is now always built
- Added NFSD support for the NFSv4 birth-time file attribute
- Added support for storing and displaying sockaddrs in trace points
- NFSD now recognizes RPC_AUTH_TLS probes
Performance improvements:
- Optimized the svc transport enqueuing mechanism
- Added micro-optimizations for the duplicate reply cache
Notable bug fixes:
- Allocation of the NFSD file cache hash table is more reliable"
* tag 'nfsd-5.18' of git://git.kernel.org/pub/scm/linux/kernel/git/cel/linux: (30 commits)
nfsd: fix using the correct variable for sizeof()
nfsd: use correct format characters
NFSD: prevent integer overflow on 32 bit systems
NFSD: prevent underflow in nfssvc_decode_writeargs()
fs/lock: documentation cleanup. Replace inode->i_lock with flc_lock.
NFSD: Fix nfsd_breaker_owns_lease() return values
NFSD: Clean up _lm_ operation names
arch: Remove references to CONFIG_NFSD_V3 in the default configs
NFSD: Remove CONFIG_NFSD_V3
nfsd: more robust allocation failure handling in nfsd_file_cache_init
SUNRPC: Teach server to recognize RPC_AUTH_TLS
NFSD: Move svc_serv_ops::svo_function into struct svc_serv
NFSD: Remove svc_serv_ops::svo_module
SUNRPC: Remove svc_shutdown_net()
SUNRPC: Rename svc_close_xprt()
SUNRPC: Rename svc_create_xprt()
SUNRPC: Remove svo_shutdown method
SUNRPC: Merge svc_do_enqueue_xprt() into svc_enqueue_xprt()
SUNRPC: Remove the .svo_enqueue_xprt method
SUNRPC: Record endpoint information in trace log
...
Qian Cai reported a boot crash on arm64 systems, caused by:
0fb3978b0a ("sched/numa: Fix NUMA topology for systems with CPU-less nodes")
The bug is that node_state() must be supplied a valid node_states[] array index,
but in task_numa_placement() the max_nid search can fail with NUMA_NO_NODE,
which is not a valid index.
Fix it by checking that max_nid is a valid index.
[ mingo: Added changelog. ]
Fixes: 0fb3978b0a ("sched/numa: Fix NUMA topology for systems with CPU-less nodes")
Reported-by: Qian Cai <quic_qiancai@quicinc.com>
Tested-by: Qian Cai <quic_qiancai@quicinc.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: "Huang, Ying" <ying.huang@intel.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Pull audit update from Paul Moore:
"Just one audit patch queued for v5.18:
- Change the AUDIT_TIME_* record generation so that they are
generated at syscall exit time and subject to all of the normal
syscall exit filtering.
This should help reduce noise and ensure those records which are
most relevant to the admin's audit configuration are recorded in
the audit log"
* tag 'audit-pr-20220321' of git://git.kernel.org/pub/scm/linux/kernel/git/pcmoore/audit:
audit: log AUDIT_TIME_* records only from rules
Pull watch_queue fixes from David Howells:
"Here are fixes for a couple more watch_queue bugs, both found by syzbot:
- Fix error cleanup in watch_queue_set_size() where it tries to clean
up all the pointers in the page list, even if they've not been
allocated yet[1]. Unfortunately, __free_page() doesn't treat a NULL
pointer as being "do nothing".
A second report[2] looks like it's probably the same bug, but on
arm64 rather than x86_64, but there's no reproducer.
- Fix a missing kfree in free_watch() to actually free the watch[3]"
Link: https://lore.kernel.org/r/000000000000b1807c05daad8f98@google.com/ [1]
Link: https://lore.kernel.org/r/000000000000035b9c05daae8a5e@google.com/ [2]
Link: https://lore.kernel.org/r/000000000000bc8eaf05dab91c63@google.com/ [3]
* 'keys-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-fs:
watch_queue: Actually free the watch
watch_queue: Fix NULL dereference in error cleanup
Pull random number generator updates from Jason Donenfeld:
"There have been a few important changes to the RNG's crypto, but the
intent for 5.18 has been to shore up the existing design as much as
possible with modern cryptographic functions and proven constructions,
rather than actually changing up anything fundamental to the RNG's
design.
So it's still the same old RNG at its core as before: it still counts
entropy bits, and collects from the various sources with the same
heuristics as before, and so forth. However, the cryptographic
algorithms that transform that entropic data into safe random numbers
have been modernized.
Just as important, if not more, is that the code has been cleaned up
and re-documented. As one of the first drivers in Linux, going back to
1.3.30, its general style and organization was showing its age and
becoming both a maintenance burden and an auditability impediment.
Hopefully this provides a more solid foundation to build on for the
future. I encourage you to open up the file in full, and maybe you'll
remark, "oh, that's what it's doing," and enjoy reading it. That, at
least, is the eventual goal, which this pull begins working toward.
Here's a summary of the various patches in this pull:
- /dev/urandom and /dev/random now do the same thing, per the patch
we discussed on the list. I think this is worth trying out. If it
does appear problematic, I've made sure to keep it standalone and
revertible without any conflicts.
- Fixes and cleanups for numerous integer type problems, locking
issues, and general code quality concerns.
- The input pool's LFSR has been replaced with a cryptographically
secure hash function, which has security and performance benefits
alike, and consequently allows us to count entropy bits linearly.
- The pre-init injection now uses a real hash function too, instead
of an LFSR or vanilla xor.
- The interrupt handler's fast_mix() function now uses one round of
SipHash, rather than the fake crypto that was there before.
- All additions of RDRAND and RDSEED now go through the input pool's
hash function, in part to mitigate ridiculous hypothetical CPU
backdoors, but more so to have a consistent interface for ingesting
entropy that's easy to analyze, making everything happen one way,
instead of a potpourri of different ways.
- The crng now works on per-cpu data, while also being in accordance
with the actual "fast key erasure RNG" design. This allows us to
fix several boot-time race complications associated with the prior
dynamically allocated model, eliminates much locking, and makes our
backtrack protection more robust.
- Batched entropy now erases doled out values so that it's backtrack
resistant.
- Working closely with Sebastian, the interrupt handler no longer
needs to take any locks at all, as we punt the
synchronized/expensive operations to a workqueue. This is
especially nice for PREEMPT_RT, where taking spinlocks in irq
context is problematic. It also makes the handler faster for the
rest of us.
- Also working with Sebastian, we now do the right thing on CPU
hotplug, so that we don't use stale entropy or fail to accumulate
new entropy when CPUs come back online.
- We handle virtual machines that fork / clone / snapshot, using the
"vmgenid" ACPI specification for retrieving a unique new RNG seed,
which we can use to also make WireGuard (and in the future, other
things) safe across VM forks.
- Around boot time, we now try to reseed more often if enough entropy
is available, before settling on the usual 5 minute schedule.
- Last, but certainly not least, the documentation in the file has
been updated considerably"
* tag 'random-5.18-rc1-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/crng/random: (60 commits)
random: check for signal and try earlier when generating entropy
random: reseed more often immediately after booting
random: make consistent usage of crng_ready()
random: use SipHash as interrupt entropy accumulator
wireguard: device: clear keys on VM fork
random: provide notifier for VM fork
random: replace custom notifier chain with standard one
random: do not export add_vmfork_randomness() unless needed
virt: vmgenid: notify RNG of VM fork and supply generation ID
ACPI: allow longer device IDs
random: add mechanism for VM forks to reinitialize crng
random: don't let 644 read-only sysctls be written to
random: give sysctl_random_min_urandom_seed a more sensible value
random: block in /dev/urandom
random: do crng pre-init loading in worker rather than irq
random: unify cycles_t and jiffies usage and types
random: cleanup UUID handling
random: only wake up writers after zap if threshold was passed
random: round-robin registers as ulong, not u32
random: clear fast pool, crng, and batches in cpuhp bring up
...
Pull power management updates from Rafael Wysocki:
"These are mostly fixes and cleanups all over the code and a new piece
of documentation for Intel uncore frequency scaling.
Functionality-wise, the intel_idle driver will support Sapphire Rapids
Xeons natively now (with some extra facilities for controlling
C-states more precisely on those systems), virtual guests will take
the ACPI S4 hardware signature into account by default, the
intel_pstate driver will take the defualt EPP value from the firmware,
cpupower utility will support the AMD P-state driver added in the
previous cycle, and there is a new tracer utility for that driver.
Specifics:
- Allow device_pm_check_callbacks() to be called from interrupt
context without issues (Dmitry Baryshkov).
- Modify devm_pm_runtime_enable() to automatically handle
pm_runtime_dont_use_autosuspend() at driver exit time (Douglas
Anderson).
- Make the schedutil cpufreq governor use to_gov_attr_set() instead
of open coding it (Kevin Hao).
- Replace acpi_bus_get_device() with acpi_fetch_acpi_dev() in the
cpufreq longhaul driver (Rafael Wysocki).
- Unify show() and store() naming in cpufreq and make it use
__ATTR_XX (Lianjie Zhang).
- Make the intel_pstate driver use the EPP value set by the firmware
by default (Srinivas Pandruvada).
- Re-order the init checks in the powernow-k8 cpufreq driver (Mario
Limonciello).
- Make the ACPI processor idle driver check for architectural support
for LPI to avoid using it on x86 by mistake (Mario Limonciello).
- Add Sapphire Rapids Xeon support to the intel_idle driver (Artem
Bityutskiy).
- Add 'preferred_cstates' module argument to the intel_idle driver to
work around C1 and C1E handling issue on Sapphire Rapids (Artem
Bityutskiy).
- Add core C6 optimization on Sapphire Rapids to the intel_idle
driver (Artem Bityutskiy).
- Optimize the haltpoll cpuidle driver a bit (Li RongQing).
- Remove leftover text from intel_idle() kerneldoc comment and fix up
white space in intel_idle (Rafael Wysocki).
- Fix load_image_and_restore() error path (Ye Bin).
- Fix typos in comments in the system wakeup hadling code (Tom Rix).
- Clean up non-kernel-doc comments in hibernation code (Jiapeng
Chong).
- Fix __setup handler error handling in system-wide suspend and
hibernation core code (Randy Dunlap).
- Add device name to suspend_report_result() (Youngjin Jang).
- Make virtual guests honour ACPI S4 hardware signature by default
(David Woodhouse).
- Block power off of a parent PM domain unless child is in deepest
state (Ulf Hansson).
- Use dev_err_probe() to simplify error handling for generic PM
domains (Ahmad Fatoum).
- Fix sleep-in-atomic bug caused by genpd_debug_remove() (Shawn Guo).
- Document Intel uncore frequency scaling (Srinivas Pandruvada).
- Add DTPM hierarchy description (Daniel Lezcano).
- Change the locking scheme in DTPM (Daniel Lezcano).
- Fix dtpm_cpu cleanup at exit time and missing virtual DTPM pointer
release (Daniel Lezcano).
- Make dtpm_node_callback[] static (kernel test robot).
- Fix spelling mistake "initialze" -> "initialize" in
dtpm_create_hierarchy() (Colin Ian King).
- Add tracer tool for the amd-pstate driver (Jinzhou Su).
- Fix PC6 displaying in turbostat on some systems (Artem Bityutskiy).
- Add AMD P-State support to the cpupower utility (Huang Rui)"
* tag 'pm-5.18-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm: (58 commits)
cpufreq: powernow-k8: Re-order the init checks
cpuidle: intel_idle: Drop redundant backslash at line end
cpuidle: intel_idle: Update intel_idle() kerneldoc comment
PM: hibernate: Honour ACPI hardware signature by default for virtual guests
cpufreq: intel_pstate: Use firmware default EPP
cpufreq: unify show() and store() naming and use __ATTR_XX
PM: core: keep irq flags in device_pm_check_callbacks()
cpuidle: haltpoll: Call cpuidle_poll_state_init() later
Documentation: amd-pstate: add tracer tool introduction
tools/power/x86/amd_pstate_tracer: Add tracer tool for AMD P-state
tools/power/x86/intel_pstate_tracer: make tracer as a module
cpufreq: amd-pstate: Add more tracepoint for AMD P-State module
PM: sleep: Add device name to suspend_report_result()
turbostat: fix PC6 displaying on some systems
intel_idle: add core C6 optimization for SPR
intel_idle: add 'preferred_cstates' module argument
intel_idle: add SPR support
PM: runtime: Have devm_pm_runtime_enable() handle pm_runtime_dont_use_autosuspend()
ACPI: processor idle: Check for architectural support for LPI
cpuidle: PSCI: Move the `has_lpi` check to the beginning of the function
...
Pull RCU updates from Paul McKenney:
- Fix idle detection (Neeraj Upadhyay) and missing access marking
detected by KCSAN.
- Reduce coupling between rcu_barrier() and CPU-hotplug operations, so
that rcu_barrier() no longer needs to do cpus_read_lock(). This may
also someday allow system boot to bring CPUs online concurrently.
- Enable more aggressive movement to per-CPU queueing when reacting to
excessive lock contention due to workloads placing heavy update-side
stress on RCU tasks.
- Improvements to RCU priority boosting, including changes from Neeraj
Upadhyay, Zqiang, and Alison Chaiken.
- Various fixes improving test robustness and debug information.
- Add tests for SRCU size transitions, further compress torture.sh
build products, and improve debug output.
- Miscellaneous fixes.
* tag 'rcu.2022.03.13a' of git://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-rcu: (49 commits)
rcu: Replace cpumask_weight with cpumask_empty where appropriate
rcu: Remove __read_mostly annotations from rcu_scheduler_active externs
rcu: Uninline multi-use function: finish_rcuwait()
rcu: Mark writes to the rcu_segcblist structure's ->flags field
kasan: Record work creation stack trace with interrupts enabled
rcu: Inline __call_rcu() into call_rcu()
rcu: Add mutex for rcu boost kthread spawning and affinity setting
rcu: Fix description of kvfree_rcu()
MAINTAINERS: Add Frederic and Neeraj to their RCU files
rcutorture: Provide non-power-of-two Tasks RCU scenarios
rcutorture: Test SRCU size transitions
torture: Make torture.sh help message match reality
rcu-tasks: Set ->percpu_enqueue_shift to zero upon contention
rcu-tasks: Use order_base_2() instead of ilog2()
rcu: Create and use an rcu_rdp_cpu_online()
rcu: Make rcu_barrier() no longer block CPU-hotplug operations
rcu: Rework rcu_barrier() and callback-migration logic
rcu: Refactor rcu_barrier() empty-list handling
rcu: Kill rnp->ofl_seq and use only rcu_state.ofl_lock for exclusion
torture: Change KVM environment variable to RCUTORTURE
...
Pull interrupt updates from Thomas Gleixner:
"Core code:
- Provide generic_handle_irq_safe() which can be invoked from any
context (hard interrupt or threaded). This allows to remove ugly
workarounds in drivers all over the place.
- Use generic_handle_irq_safe() in the affected drivers.
- The usual cleanups and improvements.
Interrupt chip drivers:
- Support for new interrupt chips or not yet supported variants:
STM32MP14, Meson GPIO, Apple M1 PMU, Apple M1 AICv2, Qualcomm MPM
- Convert the Xilinx driver to generic interrupt domains
- Cleanup the irq_chip::name handling
- The usual cleanups and improvements all over the place"
* tag 'irq-core-2022-03-21' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (52 commits)
irqchip: Add Qualcomm MPM controller driver
dt-bindings: interrupt-controller: Add Qualcomm MPM support
irqchip/apple-aic: Add support for AICv2
irqchip/apple-aic: Support multiple dies
irqchip/apple-aic: Dynamically compute register offsets
irqchip/apple-aic: Switch to irq_domain_create_tree and sparse hwirqs
irqchip/apple-aic: Add Fast IPI support
dt-bindings: interrupt-controller: apple,aic2: New binding for AICv2
PCI: apple: Change MSI handling to handle 4-cell AIC fwspec form
irqchip/apple-aic: Fix cpumask allocation for FIQs
irqchip/meson-gpio: Add support for meson s4 SoCs
irqchip/meson-gpio: add select trigger type callback
irqchip/meson-gpio: support more than 8 channels gpio irq
dt-bindings: interrupt-controller: New binding for Meson-S4 SoCs
irqchip/xilinx: Switch to GENERIC_IRQ_MULTI_HANDLER
staging: greybus: gpio: Use generic_handle_irq_safe().
net: usb: lan78xx: Use generic_handle_irq_safe().
mfd: ezx-pcap: Use generic_handle_irq_safe().
misc: hi6421-spmi-pmic: Use generic_handle_irq_safe().
irqchip/sifive-plic: Disable S-mode IRQs if running in M-mode
...
Pull timer and timekeeping updates from Thomas Gleixner:
"Core code:
- Make the NOHZ handling of the timekeeping/tick core more robust to
prevent a rare jiffies update stall.
- Handle softirqs in the NOHZ/idle case correctly
Drivers:
- Add support for event stream scaling of the 1GHz counter on ARM(64)
- Correct an error code check in the timer-of layer
- The usual cleanups and improvements all over the place"
* tag 'timers-core-2022-03-21' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (23 commits)
lib/irq_poll: Declare IRQ_POLL softirq vector as ksoftirqd-parking safe
tick/rcu: Stop allowing RCU_SOFTIRQ in idle
tick/rcu: Remove obsolete rcu_needs_cpu() parameters
tick: Detect and fix jiffies update stall
clocksource/drivers/timer-of: Check return value of of_iomap in timer_of_base_init()
clocksource/drivers/timer-microchip-pit64b: Use 5MHz for clockevent
clocksource/drivers/timer-microchip-pit64b: Use notrace
clocksource/drivers/timer-microchip-pit64b: Remove mmio selection
dt-bindings: timer: Tegra: Convert text bindings to yaml
clocksource/drivers/imx-tpm: Move tpm_read_sched_clock() under CONFIG_ARM
clocksource/drivers/arm_arch_timer: Use event stream scaling when available
clocksource/drivers/exynos_mct: Increase the size of name array
clocksource/drivers/exynos_mct: Bump up mct max irq number
clocksource/drivers/exynos_mct: Remove mct interrupt index enum
clocksource/drivers/exynos_mct: Handle DTS with higher number of interrupts
clocksource/drivers/timer-ti-dm: Fix regression from errata i940 fix
clocksource/drivers/imx-tpm: Exclude sched clock for ARM64
clocksource: Add a Kconfig option for WATCHDOG_MAX_SKEW
clocksource/drivers/imx-tpm: Update name of clkevt
clocksource/drivers/imx-tpm: Add CLOCK_EVT_FEAT_DYNIRQ
...
Pull core process handling RT latency updates from Thomas Gleixner:
- Reduce the amount of work to release a task stack in context switch.
There is no real reason to do cgroup accounting and memory freeing in
this performance sensitive context.
Aside of this the invoked functions cannot be called from this
preemption disabled context on PREEMPT_RT enabled kernels. Solve this
by moving the accounting into do_exit() and delaying the freeing of
the stack unless the vmap stack can be cached.
- Provide a mechanism to delay raising signals from atomic context on
PREEMPT_RT enabled kernels as sighand::lock cannot be acquired. Store
the information in the task struct and raise it in the exit path.
* tag 'core-core-2022-03-21' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
signal, x86: Delay calling signals in atomic on RT enabled kernels
fork: Use IS_ENABLED() in account_kernel_stack()
fork: Only cache the VMAP stack in finish_task_switch()
fork: Move task stack accounting to do_exit()
fork: Move memcg_charge_kernel_stack() into CONFIG_VMAP_STACK
fork: Don't assign the stack pointer in dup_task_struct()
fork, IA64: Provide alloc_thread_stack_node() for IA64
fork: Duplicate task_struct before stack allocation
fork: Redo ifdefs around task stack handling
Pull x86 PASID support from Thomas Gleixner:
"Reenable ENQCMD/PASID support:
- Simplify the PASID handling to allocate the PASID once, associate
it to the mm of a process and free it on mm_exit().
The previous attempt of refcounted PASIDs and dynamic
alloc()/free() turned out to be error prone and too complex. The
PASID space is 20bits, so the case of resource exhaustion is a pure
academic concern.
- Populate the PASID MSR on demand via #GP to avoid racy updates via
IPIs.
- Reenable ENQCMD and let objtool check for the forbidden usage of
ENQCMD in the kernel.
- Update the documentation for Shared Virtual Addressing accordingly"
* tag 'x86-pasid-2022-03-21' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
Documentation/x86: Update documentation for SVA (Shared Virtual Addressing)
tools/objtool: Check for use of the ENQCMD instruction in the kernel
x86/cpufeatures: Re-enable ENQCMD
x86/traps: Demand-populate PASID MSR via #GP
sched: Define and initialize a flag to identify valid PASID in the task
x86/fpu: Clear PASID when copying fpstate
iommu/sva: Assign a PASID to mm on PASID allocation and free it on mm exit
kernel/fork: Initialize mm's PASID
iommu/ioasid: Introduce a helper to check for valid PASIDs
mm: Change CONFIG option for mm->pasid field
iommu/sva: Rename CONFIG_IOMMU_SVA_LIB to CONFIG_IOMMU_SVA
Instead of declaring a struct page_vma_mapped_walk directly,
use these helpers to allow us to transition to a PFN approach in the
following patches.
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
invalidate_inode_page() is the only caller of invalidate_complete_page()
and inlining it reveals that the first check is unnecessary (because we
hold the page locked, and we just retrieved the mapping from the page).
Actually, it does make a difference, in that tail pages no longer fail
at this check, so it's now possible to remove a tail page from a mapping.
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Reviewed-by: John Hubbard <jhubbard@nvidia.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
In watch_queue_set_size(), the error cleanup code doesn't take account of
the fact that __free_page() can't handle a NULL pointer when trying to free
up buffer pages that did get allocated.
Fix this by only calling __free_page() on the pages actually allocated.
Without the fix, this can lead to something like the following:
BUG: KASAN: null-ptr-deref in __free_pages+0x1f/0x1b0 mm/page_alloc.c:5473
Read of size 4 at addr 0000000000000034 by task syz-executor168/3599
...
Call Trace:
<TASK>
__dump_stack lib/dump_stack.c:88 [inline]
dump_stack_lvl+0xcd/0x134 lib/dump_stack.c:106
__kasan_report mm/kasan/report.c:446 [inline]
kasan_report.cold+0x66/0xdf mm/kasan/report.c:459
check_region_inline mm/kasan/generic.c:183 [inline]
kasan_check_range+0x13d/0x180 mm/kasan/generic.c:189
instrument_atomic_read include/linux/instrumented.h:71 [inline]
atomic_read include/linux/atomic/atomic-instrumented.h:27 [inline]
page_ref_count include/linux/page_ref.h:67 [inline]
put_page_testzero include/linux/mm.h:717 [inline]
__free_pages+0x1f/0x1b0 mm/page_alloc.c:5473
watch_queue_set_size+0x499/0x630 kernel/watch_queue.c:275
pipe_ioctl+0xac/0x2b0 fs/pipe.c:632
vfs_ioctl fs/ioctl.c:51 [inline]
__do_sys_ioctl fs/ioctl.c:874 [inline]
__se_sys_ioctl fs/ioctl.c:860 [inline]
__x64_sys_ioctl+0x193/0x200 fs/ioctl.c:860
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x44/0xae
Fixes: c73be61ced ("pipe: Add general notification queue support")
Reported-and-tested-by: syzbot+d55757faa9b80590767b@syzkaller.appspotmail.com
Signed-off-by: David Howells <dhowells@redhat.com>
Reviewed-by: Fabio M. De Francesco <fmdefrancesco@gmail.com>
Merge changes related to system sleep, PM domains changes and power
management documentation changes for 5.18-rc1:
- Fix load_image_and_restore() error path (Ye Bin).
- Fix typos in comments in the system wakeup hadling code (Tom Rix).
- Clean up non-kernel-doc comments in hibernation code (Jiapeng
Chong).
- Fix __setup handler error handling in system-wide suspend and
hibernation core code (Randy Dunlap).
- Add device name to suspend_report_result() (Youngjin Jang).
- Make virtual guests honour ACPI S4 hardware signature by
default (David Woodhouse).
- Block power off of a parent PM domain unless child is in deepest
state (Ulf Hansson).
- Use dev_err_probe() to simplify error handling for generic PM
domains (Ahmad Fatoum).
- Fix sleep-in-atomic bug caused by genpd_debug_remove() (Shawn Guo).
- Document Intel uncore frequency scaling (Srinivas Pandruvada).
* pm-sleep:
PM: hibernate: Honour ACPI hardware signature by default for virtual guests
PM: sleep: Add device name to suspend_report_result()
PM: suspend: fix return value of __setup handler
PM: hibernate: fix __setup handler error handling
PM: hibernate: Clean up non-kernel-doc comments
PM: sleep: wakeup: Fix typos in comments
PM: hibernate: fix load_image_and_restore() error path
* pm-domains:
PM: domains: Fix sleep-in-atomic bug caused by genpd_debug_remove()
PM: domains: use dev_err_probe() to simplify error handling
PM: domains: Prevent power off for parent unless child is in deepest state
* pm-docs:
Documentation: admin-guide: pm: Document uncore frequency scaling
Merge cpufreq and cpuidle changes for 5.18-rc1:
- Make the schedutil cpufreq governor use to_gov_attr_set() instead
of open coding it (Kevin Hao).
- Replace acpi_bus_get_device() with acpi_fetch_acpi_dev() in the
cpufreq longhaul driver (Rafael Wysocki).
- Unify show() and store() naming in cpufreq and make it use
__ATTR_XX (Lianjie Zhang).
- Make the intel_pstate driver use the EPP value set by the firmware
by default (Srinivas Pandruvada).
- Re-order the init checks in the powernow-k8 cpufreq driver (Mario
Limonciello).
- Make the ACPI processor idle driver check for architectural
support for LPI to avoid using it on x86 by mistake (Mario
Limonciello).
- Add Sapphire Rapids Xeon support to the intel_idle driver (Artem
Bityutskiy).
- Add 'preferred_cstates' module argument to the intel_idle driver
to work around C1 and C1E handling issue on Sapphire Rapids (Artem
Bityutskiy).
- Add core C6 optimization on Sapphire Rapids to the intel_idle
driver (Artem Bityutskiy).
- Optimize the haltpoll cpuidle driver a bit (Li RongQing).
- Remove leftover text from intel_idle() kerneldoc comment and fix
up white space in intel_idle (Rafael Wysocki).
* pm-cpufreq:
cpufreq: powernow-k8: Re-order the init checks
cpufreq: intel_pstate: Use firmware default EPP
cpufreq: unify show() and store() naming and use __ATTR_XX
cpufreq: longhaul: Replace acpi_bus_get_device()
cpufreq: schedutil: Use to_gov_attr_set() to get the gov_attr_set
cpufreq: Move to_gov_attr_set() to cpufreq.h
* pm-cpuidle:
cpuidle: intel_idle: Drop redundant backslash at line end
cpuidle: intel_idle: Update intel_idle() kerneldoc comment
cpuidle: haltpoll: Call cpuidle_poll_state_init() later
intel_idle: add core C6 optimization for SPR
intel_idle: add 'preferred_cstates' module argument
intel_idle: add SPR support
ACPI: processor idle: Check for architectural support for LPI
cpuidle: PSCI: Move the `has_lpi` check to the beginning of the function
Pull irqchip updates from Marc Zyngier:
- Add support for the STM32MP13 variant
- Move parent device away from struct irq_chip
- Remove all instances of non-const strings assigned to
struct irq_chip::name, enabling a nice cleanup for VIC and GIC)
- Simplify the Qualcomm PDC driver
- A bunch of SiFive PLIC cleanups
- Add support for a new variant of the Meson GPIO block
- Add support for the irqchip side of the Apple M1 PMU
- Add support for the Apple M1 Pro/Max AICv2 irqchip
- Add support for the Qualcomm MPM wakeup gadget
- Move the Xilinx driver over to the generic irqdomain handling
- Tiny speedup for IPIs on GICv3 systems
- The usual odd cleanups
Link: https://lore.kernel.org/all/20220313105142.704579-1-maz@kernel.org
Merge misc fixes from David Howells:
"A set of patches for watch_queue filter issues noted by Jann. I've
added in a cleanup patch from Christophe Jaillet to convert to using
formal bitmap specifiers for the note allocation bitmap.
Also two filesystem fixes (afs and cachefiles)"
* emailed patches from David Howells <dhowells@redhat.com>:
cachefiles: Fix volume coherency attribute
afs: Fix potential thrashing in afs writeback
watch_queue: Make comment about setting ->defunct more accurate
watch_queue: Fix lack of barrier/sync/lock between post and read
watch_queue: Free the alloc bitmap when the watch_queue is torn down
watch_queue: Fix the alloc bitmap size to reflect notes allocated
watch_queue: Use the bitmap API when applicable
watch_queue: Fix to always request a pow-of-2 pipe ring size
watch_queue: Fix to release page in ->release()
watch_queue, pipe: Free watchqueue state after clearing pipe ring
watch_queue: Fix filter limit check
watch_queue_clear() has a comment stating that setting ->defunct to true
preventing new additions as well as preventing notifications. Whilst
the latter is true, the first bit is superfluous since at the time this
function is called, the pipe cannot be accessed to add new event
sources.
Remove the "new additions" bit from the comment.
Fixes: c73be61ced ("pipe: Add general notification queue support")
Reported-by: Jann Horn <jannh@google.com>
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
There's nothing to synchronise post_one_notification() versus
pipe_read(). Whilst posting is done under pipe->rd_wait.lock, the
reader only takes pipe->mutex which cannot bar notification posting as
that may need to be made from contexts that cannot sleep.
Fix this by setting pipe->head with a barrier in post_one_notification()
and reading pipe->head with a barrier in pipe_read().
If that's not sufficient, the rd_wait.lock will need to be taken,
possibly in a ->confirm() op so that it only applies to notifications.
The lock would, however, have to be dropped before copy_page_to_iter()
is invoked.
Fixes: c73be61ced ("pipe: Add general notification queue support")
Reported-by: Jann Horn <jannh@google.com>
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Currently, watch_queue_set_size() sets the number of notes available in
wqueue->nr_notes according to the number of notes allocated, but sets
the size of the bitmap to the unrounded number of notes originally asked
for.
Fix this by setting the bitmap size to the number of notes we're
actually going to make available (ie. the number allocated).
Fixes: c73be61ced ("pipe: Add general notification queue support")
Reported-by: Jann Horn <jannh@google.com>
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Use bitmap_alloc() to simplify code, improve the semantic and reduce
some open-coded arithmetic in allocator arguments.
Also change a memset(0xff) into an equivalent bitmap_fill() to keep
consistency.
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The pipe ring size must always be a power of 2 as the head and tail
pointers are masked off by AND'ing with the size of the ring - 1.
watch_queue_set_size(), however, lets you specify any number of notes
between 1 and 511. This number is passed through to pipe_resize_ring()
without checking/forcing its alignment.
Fix this by rounding the number of slots required up to the nearest
power of two. The request is meant to guarantee that at least that many
notifications can be generated before the queue is full, so rounding
down isn't an option, but, alternatively, it may be better to give an
error if we aren't allowed to allocate that much ring space.
Fixes: c73be61ced ("pipe: Add general notification queue support")
Reported-by: Jann Horn <jannh@google.com>
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
When a pipe ring descriptor points to a notification message, the
refcount on the backing page is incremented by the generic get function,
but the release function, which marks the bitmap, doesn't drop the page
ref.
Fix this by calling generic_pipe_buf_release() at the end of
watch_queue_pipe_buf_release().
Fixes: c73be61ced ("pipe: Add general notification queue support")
Reported-by: Jann Horn <jannh@google.com>
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
In watch_queue_set_filter(), there are a couple of places where we check
that the filter type value does not exceed what the type_filter bitmap
can hold. One place calculates the number of bits by:
if (tf[i].type >= sizeof(wfilter->type_filter) * 8)
which is fine, but the second does:
if (tf[i].type >= sizeof(wfilter->type_filter) * BITS_PER_LONG)
which is not. This can lead to a couple of out-of-bounds writes due to
a too-large type:
(1) __set_bit() on wfilter->type_filter
(2) Writing more elements in wfilter->filters[] than we allocated.
Fix this by just using the proper WATCH_TYPE__NR instead, which is the
number of types we actually know about.
The bug may cause an oops looking something like:
BUG: KASAN: slab-out-of-bounds in watch_queue_set_filter+0x659/0x740
Write of size 4 at addr ffff88800d2c66bc by task watch_queue_oob/611
...
Call Trace:
<TASK>
dump_stack_lvl+0x45/0x59
print_address_description.constprop.0+0x1f/0x150
...
kasan_report.cold+0x7f/0x11b
...
watch_queue_set_filter+0x659/0x740
...
__x64_sys_ioctl+0x127/0x190
do_syscall_64+0x43/0x90
entry_SYSCALL_64_after_hwframe+0x44/0xae
Allocated by task 611:
kasan_save_stack+0x1e/0x40
__kasan_kmalloc+0x81/0xa0
watch_queue_set_filter+0x23a/0x740
__x64_sys_ioctl+0x127/0x190
do_syscall_64+0x43/0x90
entry_SYSCALL_64_after_hwframe+0x44/0xae
The buggy address belongs to the object at ffff88800d2c66a0
which belongs to the cache kmalloc-32 of size 32
The buggy address is located 28 bytes inside of
32-byte region [ffff88800d2c66a0, ffff88800d2c66c0)
Fixes: c73be61ced ("pipe: Add general notification queue support")
Reported-by: Jann Horn <jannh@google.com>
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull tracing fixes from Steven Rostedt:
"Minor tracing fixes:
- Fix unregistering the same event twice. A user could disable the
same event that osnoise will disable on unregistering.
- Inform RCU of a quiescent state in the osnoise testing thread.
- Fix some kerneldoc comments"
* tag 'trace-v5.17-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
ftrace: Fix some W=1 warnings in kernel doc comments
tracing/osnoise: Force quiescent states while tracing
tracing/osnoise: Do not unregister events twice
At the moment running osnoise on a nohz_full CPU or uncontested FIFO
priority and a PREEMPT_RCU kernel might have the side effect of
extending grace periods too much. This will entice RCU to force a
context switch on the wayward CPU to end the grace period, all while
introducing unwarranted noise into the tracer. This behaviour is
unavoidable as overly extending grace periods might exhaust the system's
memory.
This same exact problem is what extended quiescent states (EQS) were
created for, conversely, rcu_momentary_dyntick_idle() emulates them by
performing a zero duration EQS. So let's make use of it.
In the common case rcu_momentary_dyntick_idle() is fairly inexpensive:
atomically incrementing a local per-CPU counter and doing a store. So it
shouldn't affect osnoise's measurements (which has a 1us granularity),
so we'll call it unanimously.
The uncommon case involve calling rcu_momentary_dyntick_idle() after
having the osnoise process:
- Receive an expedited quiescent state IPI with preemption disabled or
during an RCU critical section. (activates rdp->cpu_no_qs.b.exp
code-path).
- Being preempted within in an RCU critical section and having the
subsequent outermost rcu_read_unlock() called with interrupts
disabled. (t->rcu_read_unlock_special.b.blocked code-path).
Neither of those are possible at the moment, and are unlikely to be in
the future given the osnoise's loop design. On top of this, the noise
generated by the situations described above is unavoidable, and if not
exposed by rcu_momentary_dyntick_idle() will be eventually seen in
subsequent rcu_read_unlock() calls or schedule operations.
Link: https://lkml.kernel.org/r/20220307180740.577607-1-nsaenzju@redhat.com
Cc: stable@vger.kernel.org
Fixes: bce29ac9ce ("trace: Add osnoise tracer")
Signed-off-by: Nicolas Saenz Julienne <nsaenzju@redhat.com>
Acked-by: Paul E. McKenney <paulmck@kernel.org>
Acked-by: Daniel Bristot de Oliveira <bristot@kernel.org>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Nicolas reported that using:
# trace-cmd record -e all -M 10 -p osnoise --poll
Resulted in the following kernel warning:
------------[ cut here ]------------
WARNING: CPU: 0 PID: 1217 at kernel/tracepoint.c:404 tracepoint_probe_unregister+0x280/0x370
[...]
CPU: 0 PID: 1217 Comm: trace-cmd Not tainted 5.17.0-rc6-next-20220307-nico+ #19
RIP: 0010:tracepoint_probe_unregister+0x280/0x370
[...]
CR2: 00007ff919b29497 CR3: 0000000109da4005 CR4: 0000000000170ef0
Call Trace:
<TASK>
osnoise_workload_stop+0x36/0x90
tracing_set_tracer+0x108/0x260
tracing_set_trace_write+0x94/0xd0
? __check_object_size.part.0+0x10a/0x150
? selinux_file_permission+0x104/0x150
vfs_write+0xb5/0x290
ksys_write+0x5f/0xe0
do_syscall_64+0x3b/0x90
entry_SYSCALL_64_after_hwframe+0x44/0xae
RIP: 0033:0x7ff919a18127
[...]
---[ end trace 0000000000000000 ]---
The warning complains about an attempt to unregister an
unregistered tracepoint.
This happens on trace-cmd because it first stops tracing, and
then switches the tracer to nop. Which is equivalent to:
# cd /sys/kernel/tracing/
# echo osnoise > current_tracer
# echo 0 > tracing_on
# echo nop > current_tracer
The osnoise tracer stops the workload when no trace instance
is actually collecting data. This can be caused both by
disabling tracing or disabling the tracer itself.
To avoid unregistering events twice, use the existing
trace_osnoise_callback_enabled variable to check if the events
(and the workload) are actually active before trying to
deactivate them.
Link: https://lore.kernel.org/all/c898d1911f7f9303b7e14726e7cc9678fbfb4a0e.camel@redhat.com/
Link: https://lkml.kernel.org/r/938765e17d5a781c2df429a98f0b2e7cc317b022.1646823913.git.bristot@kernel.org
Cc: stable@vger.kernel.org
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Fixes: 2fac8d6486 ("tracing/osnoise: Allow multiple instances of the same tracer")
Reported-by: Nicolas Saenz Julienne <nsaenzju@redhat.com>
Signed-off-by: Daniel Bristot de Oliveira <bristot@kernel.org>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
While investigating the sparse warning reported by the LKP bot [1],
observed that we have a redundant variable "top" in the function
build_sched_domains that was introduced in the recent commit
e496132ebe ("sched/fair: Adjust the allowed NUMA imbalance when
SD_NUMA spans multiple LLCs")
The existing variable "sd" suffices which allows us to remove the
redundant variable "top" while annotating the other variable "top_p"
with the "__rcu" annotation to silence the sparse warning.
[1] https://lore.kernel.org/lkml/202202170853.9vofgC3O-lkp@intel.com/
Fixes: e496132ebe ("sched/fair: Adjust the allowed NUMA imbalance when SD_NUMA spans multiple LLCs")
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: K Prateek Nayak <kprateek.nayak@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Valentin Schneider <valentin.schneider@arm.com>
Link: https://lore.kernel.org/r/20220218162743.1134-1-kprateek.nayak@amd.com
Both functions are doing almost the same, that is checking if admission
control is still respected.
With exclusive cpusets, dl_task_can_attach() checks if the destination
cpuset (i.e. its root domain) has enough CPU capacity to accommodate the
task.
dl_cpu_busy() checks if there is enough CPU capacity in the cpuset in
case the CPU is hot-plugged out.
dl_task_can_attach() is used to check if a task can be admitted while
dl_cpu_busy() is used to check if a CPU can be hotplugged out.
Make dl_cpu_busy() able to deal with a task and use it instead of
dl_task_can_attach() in task_can_attach().
Signed-off-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Juri Lelli <juri.lelli@redhat.com>
Link: https://lore.kernel.org/r/20220302183433.333029-4-dietmar.eggemann@arm.com
