Support for the CPU PMUs on the Apple M1.
* for-next/perf-m1:
drivers/perf: Add Apple icestorm/firestorm CPU PMU driver
drivers/perf: arm_pmu: Handle 47 bit counters
irqchip/apple-aic: Move PMU-specific registers to their own include file
arm64: dts: apple: Add t8303 PMU nodes
arm64: dts: apple: Add t8103 PMU interrupt affinities
irqchip/apple-aic: Wire PMU interrupts
irqchip/apple-aic: Parse FIQ affinities from device-tree
dt-bindings: apple,aic: Add affinity description for per-cpu pseudo-interrupts
dt-bindings: apple,aic: Add CPU PMU per-cpu pseudo-interrupts
dt-bindings: arm-pmu: Document Apple PMU compatible strings
Add a new, weird and wonderful driver for the equally weird Apple
PMU HW. Although the PMU itself is functional, we don't know much
about the events yet, so this can be considered as yet another
random number generator...
Nonetheless, it can reliably count at least cycles and instructions
in the usually wonky big-little way. For anything else, it of course
supports raw event numbers.
Signed-off-by: Marc Zyngier <maz@kernel.org>
Signed-off-by: Will Deacon <will@kernel.org>
The current ARM PMU framework can only deal with 32 or 64bit counters.
Teach it about a 47bit flavour.
Yes, this is odd.
Reviewed-by: Hector Martin <marcan@marcan.st>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Signed-off-by: Will Deacon <will@kernel.org>
Pull in Apple AIC rework from Marc Zyngier to support PMU interrupts on
the M1 platform.
* 'irq/aic-pmu' of git://git.kernel.org/pub/scm/linux/kernel/git/maz/arm-platforms:
irqchip/apple-aic: Move PMU-specific registers to their own include file
arm64: dts: apple: Add t8303 PMU nodes
arm64: dts: apple: Add t8103 PMU interrupt affinities
irqchip/apple-aic: Wire PMU interrupts
irqchip/apple-aic: Parse FIQ affinities from device-tree
dt-bindings: apple,aic: Add affinity description for per-cpu pseudo-interrupts
dt-bindings: apple,aic: Add CPU PMU per-cpu pseudo-interrupts
dt-bindings: arm-pmu: Document Apple PMU compatible strings
CN10k DSS h/w perfmon does not support event overflow interrupt, so
periodic timer is being used. Each event counter is 48bit, which in worst
case scenario can increment at maximum 5.6 GT/s. At this rate it may take
many hours to overflow these counters. Therefore polling period for
overflow is set to 100 sec, which can be changed using sysfs parameter.
Two fixed event counters starts counting from zero on overflow, so
overflow condition is when new count less than previous count. While
eight programmable event counters freezes at maximum value. Also individual
counter cannot be restarted, so need to restart all eight counters.
Signed-off-by: Bharat Bhushan <bbhushan2@marvell.com>
Reviewed-by: Bhaskara Budiredla <bbudiredla@marvell.com>
Link: https://lore.kernel.org/r/20220211045346.17894-4-bbhushan2@marvell.com
Signed-off-by: Will Deacon <will@kernel.org>
Marvell CN10k DRAM Subsystem (DSS) supports eight event counters for
monitoring performance and software can program each counter to monitor
any of the defined performance event. Performance events are for
interface between the DDR controller and the PHY, interface between the
DDR Controller and the CHI interconnect, or within the DDR Controller.
Additionally DSS also supports two fixed performance event counters, one
for number of ddr reads and other for ddr writes.
This patch add basic support for these performance monitoring events
on CN10k.
Signed-off-by: Bharat Bhushan <bbhushan2@marvell.com>
Reviewed-by: Bhaskara Budiredla <bbudiredla@marvell.com>
Link: https://lore.kernel.org/r/20220211045346.17894-3-bbhushan2@marvell.com
Signed-off-by: Will Deacon <will@kernel.org>
In some places, drivers/perf code calls bitmap_weight() to check if any
bit of a given bitmap is set. It's better to use bitmap_empty() in that
case because bitmap_empty() stops traversing the bitmap as soon as it
finds first set bit, while bitmap_weight() counts all bits unconditionally.
Signed-off-by: Yury Norov <yury.norov@gmail.com>
Acked-by: Mark Rutland <mark.rutland@arm.com>
Link: https://lore.kernel.org/r/20220210224933.379149-13-yury.norov@gmail.com
Signed-off-by: Will Deacon <will@kernel.org>
The kbuild helpfully reports that the Marvell CN10K TAD PMU driver emits
a warning when building with W=1 and CONFIG_OF=n:
| >> drivers/perf/marvell_cn10k_tad_pmu.c:371:34: warning: unused variable 'tad_pmu_of_match' [-Wunused-const-variable]
static const struct of_device_id tad_pmu_of_match[] = {
Guard the match table with CONFIG_OF to squash the warning.
Link: https://lore.kernel.org/r/202201292349.zRQLcDDD-lkp@intel.com
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Will Deacon <will@kernel.org>
platform_get_resource(pdev, IORESOURCE_IRQ, ..) relies on static
allocation of IRQ resources in DT core code, this causes an issue
when using hierarchical interrupt domains using "interrupts" property
in the node as this bypasses the hierarchical setup and messes up the
irq chaining.
In preparation for removal of static setup of IRQ resource from DT core
code use platform_get_irq().
Link: https://lore.kernel.org/r/20211224161334.31123-1-prabhakar.mahadev-lad.rj@bp.renesas.com
Signed-off-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
Signed-off-by: Will Deacon <will@kernel.org>
As we are about to have a PMU driver, move the PMU bits from the AIC
driver into a common include file.
Reviewed-by: Hector Martin <marcan@marcan.st>
Signed-off-by: Marc Zyngier <maz@kernel.org>
The two PMU pseudo interrupts have specific affinities. One set
is affine to the small cores, and the other set affine to the
big ones.
Signed-off-by: Marc Zyngier <maz@kernel.org>
Add the necessary code to configure and P and E-core PMU interrupts
with their respective affinities. When such an interrupt fires, map
it onto the right pseudo-interrupt.
Reviewed-by: Hector Martin <marcan@marcan.st>
Signed-off-by: Marc Zyngier <maz@kernel.org>
In order to be able to tell the core IRQ code about the affinity
of the PMU interrupt in later patches, parse the affinities kindly
provided in the device-tree.
Signed-off-by: Marc Zyngier <maz@kernel.org>
Some of the FIQ per-cpu pseudo-interrupts are better described with
a specific affinity, the most obvious candidate being the CPU PMUs.
Augment the AIC binding to be able to specify that affinity in the
interrupt controller node.
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Advertise the two pseudo-interrupts that tied to the two PMU
flavours present in the Apple M1 SoC.
We choose the expose two different pseudo-interrupts to the OS
as the e-core PMU is obviously different from the p-core one,
effectively presenting two different devices.
Acked-by: Rob Herring <robh@kernel.org>
Reviewed-by: Hector Martin <marcan@marcan.st>
Signed-off-by: Marc Zyngier <maz@kernel.org>
As we are about to add support fur the Apple PMUs, document the compatible
strings associated with the two micro-architectures present in the Apple M1.
Acked-by: Rob Herring <robh@kernel.org>
Reviewed-by: Hector Martin <marcan@marcan.st>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Pull ext4 fixes from Ted Ts'o:
"Various bug fixes for ext4 fast commit and inline data handling.
Also fix regression introduced as part of moving to the new mount API"
* tag 'ext4_for_linus_stable' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4:
fs/ext4: fix comments mentioning i_mutex
ext4: fix incorrect type issue during replay_del_range
jbd2: fix kernel-doc descriptions for jbd2_journal_shrink_{scan,count}()
ext4: fix potential NULL pointer dereference in ext4_fill_super()
jbd2: refactor wait logic for transaction updates into a common function
jbd2: cleanup unused functions declarations from jbd2.h
ext4: fix error handling in ext4_fc_record_modified_inode()
ext4: remove redundant max inline_size check in ext4_da_write_inline_data_begin()
ext4: fix error handling in ext4_restore_inline_data()
ext4: fast commit may miss file actions
ext4: fast commit may not fallback for ineligible commit
ext4: modify the logic of ext4_mb_new_blocks_simple
ext4: prevent used blocks from being allocated during fast commit replay
Pull perf tools fixes from Arnaldo Carvalho de Melo:
- Fix display of grouped aliased events in 'perf stat'.
- Add missing branch_sample_type to perf_event_attr__fprintf().
- Apply correct label to user/kernel symbols in branch mode.
- Fix 'perf ftrace' system_wide tracing, it has to be set before
creating the maps.
- Return error if procfs isn't mounted for PID namespaces when
synthesizing records for pre-existing processes.
- Set error stream of objdump process for 'perf annotate' TUI, to avoid
garbling the screen.
- Add missing arm64 support to perf_mmap__read_self(), the kernel part
got into 5.17.
- Check for NULL pointer before dereference writing debug info about a
sample.
- Update UAPI copies for asound, perf_event, prctl and kvm headers.
- Fix a typo in bpf_counter_cgroup.c.
* tag 'perf-tools-fixes-for-v5.17-2022-02-06' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux:
perf ftrace: system_wide collection is not effective by default
libperf: Add arm64 support to perf_mmap__read_self()
tools include UAPI: Sync sound/asound.h copy with the kernel sources
perf stat: Fix display of grouped aliased events
perf tools: Apply correct label to user/kernel symbols in branch mode
perf bpf: Fix a typo in bpf_counter_cgroup.c
perf synthetic-events: Return error if procfs isn't mounted for PID namespaces
perf session: Check for NULL pointer before dereference
perf annotate: Set error stream of objdump process for TUI
perf tools: Add missing branch_sample_type to perf_event_attr__fprintf()
tools headers UAPI: Sync linux/kvm.h with the kernel sources
tools headers UAPI: Sync linux/prctl.h with the kernel sources
perf beauty: Make the prctl arg regexp more strict to cope with PR_SET_VMA
tools headers cpufeatures: Sync with the kernel sources
tools headers UAPI: Sync linux/perf_event.h with the kernel sources
tools include UAPI: Sync sound/asound.h copy with the kernel sources
Pull perf fixes from Borislav Petkov:
- Intel/PT: filters could crash the kernel
- Intel: default disable the PMU for SMM, some new-ish EFI firmware has
started using CPL3 and the PMU CPL filters don't discriminate against
SMM, meaning that CPL3 (userspace only) events now also count EFI/SMM
cycles.
- Fixup for perf_event_attr::sig_data
* tag 'perf_urgent_for_v5.17_rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
perf/x86/intel/pt: Fix crash with stop filters in single-range mode
perf: uapi: Document perf_event_attr::sig_data truncation on 32 bit architectures
selftests/perf_events: Test modification of perf_event_attr::sig_data
perf: Copy perf_event_attr::sig_data on modification
x86/perf: Default set FREEZE_ON_SMI for all
Pull objtool fix from Borislav Petkov:
"Fix a potential truncated string warning triggered by gcc12"
* tag 'objtool_urgent_for_v5.17_rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
objtool: Fix truncated string warning
Pull irq fix from Borislav Petkov:
"Remove a bogus warning introduced by the recent PCI MSI irq affinity
overhaul"
* tag 'irq_urgent_for_v5.17_rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
PCI/MSI: Remove bogus warning in pci_irq_get_affinity()
Pull EDAC fixes from Borislav Petkov:
"Fix altera and xgene EDAC drivers to propagate the correct error code
from platform_get_irq() so that deferred probing still works"
* tag 'edac_urgent_for_v5.17_rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/ras/ras:
EDAC/xgene: Fix deferred probing
EDAC/altera: Fix deferred probing
Picking the changes from:
06feec6005 ("ASoC: hdmi-codec: Fix OOB memory accesses")
Which entails no changes in the tooling side as it doesn't introduce new
SNDRV_PCM_IOCTL_ ioctls.
To silence this perf tools build warning:
Warning: Kernel ABI header at 'tools/include/uapi/sound/asound.h' differs from latest version at 'include/uapi/sound/asound.h'
diff -u tools/include/uapi/sound/asound.h include/uapi/sound/asound.h
Cc: Dmitry Osipenko <digetx@gmail.com>
Cc: Mark Brown <broonie@kernel.org>
Cc: Takashi Iwai <tiwai@suse.de>
Link: https://lore.kernel.org/lkml/Yf+6OT+2eMrYDEeX@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
For perf recording, it retrieves process info by iterating nodes in proc
fs. If we run perf in a non-root PID namespace with command:
# unshare --fork --pid perf record -e cycles -a -- test_program
... in this case, unshare command creates a child PID namespace and
launches perf tool in it, but the issue is the proc fs is not mounted
for the non-root PID namespace, this leads to the perf tool gathering
process info from its parent PID namespace.
We can use below command to observe the process nodes under proc fs:
# unshare --pid --fork ls /proc
1 137 1968 2128 3 342 48 62 78 crypto kcore net uptime
10 138 2 2142 30 35 49 63 8 devices keys pagetypeinfo version
11 139 20 2143 304 36 50 64 82 device-tree key-users partitions vmallocinfo
12 14 2011 22 305 37 51 65 83 diskstats kmsg self vmstat
128 140 2038 23 307 39 52 656 84 driver kpagecgroup slabinfo zoneinfo
129 15 2074 24 309 4 53 67 9 execdomains kpagecount softirqs
13 16 2094 241 31 40 54 68 asound fb kpageflags stat
130 164 2096 242 310 41 55 69 buddyinfo filesystems loadavg swaps
131 17 2098 25 317 42 56 70 bus fs locks sys
132 175 21 26 32 43 57 71 cgroups interrupts meminfo sysrq-trigger
133 179 2102 263 329 44 58 75 cmdline iomem misc sysvipc
134 1875 2103 27 330 45 59 76 config.gz ioports modules thread-self
135 19 2117 29 333 46 6 77 consoles irq mounts timer_list
136 1941 2121 298 34 47 60 773 cpuinfo kallsyms mtd tty
So it shows many existed tasks, since unshared command has not mounted
the proc fs for the new created PID namespace, it still accesses the
proc fs of the root PID namespace. This leads to two prominent issues:
- Firstly, PID values are mismatched between thread info and samples.
The gathered thread info are coming from the proc fs of the root PID
namespace, but samples record its PID from the child PID namespace.
- The second issue is profiled program 'test_program' returns its forked
PID number from the child PID namespace, perf tool wrongly uses this
PID number to retrieve the process info via the proc fs of the root
PID namespace.
To avoid issues, we need to mount proc fs for the child PID namespace
with the option '--mount-proc' when use unshare command:
# unshare --fork --pid --mount-proc perf record -e cycles -a -- test_program
Conversely, when the proc fs of the root PID namespace is used by child
namespace, perf tool can detect the multiple PID levels and
nsinfo__is_in_root_namespace() returns false, this patch reports error
for this case:
# unshare --fork --pid perf record -e cycles -a -- test_program
Couldn't synthesize bpf events.
Perf runs in non-root PID namespace but it tries to gather process info from its parent PID namespace.
Please mount the proc file system properly, e.g. add the option '--mount-proc' for unshare command.
Reviewed-by: James Clark <james.clark@arm.com>
Signed-off-by: Leo Yan <leo.yan@linaro.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Andrii Nakryiko <andrii@kernel.org>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Cc: Ian Rogers <irogers@google.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: John Fastabend <john.fastabend@gmail.com>
Cc: KP Singh <kpsingh@kernel.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Martin KaFai Lau <kafai@fb.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Song Liu <songliubraving@fb.com>
Cc: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/r/20211224124014.2492751-1-leo.yan@linaro.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
To pick the changes in:
f6c6804c43 ("kvm: Move KVM_GET_XSAVE2 IOCTL definition at the end of kvm.h")
That just rebuilds perf, as these patches don't add any new KVM ioctl to
be harvested for the the 'perf trace' ioctl syscall argument
beautifiers.
This is also by now used by tools/testing/selftests/kvm/, a simple test
build succeeded.
This silences this perf build warning:
Warning: Kernel ABI header at 'tools/include/uapi/linux/kvm.h' differs from latest version at 'include/uapi/linux/kvm.h'
diff -u tools/include/uapi/linux/kvm.h include/uapi/linux/kvm.h
Cc: Janosch Frank <frankja@linux.ibm.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Link: http://lore.kernel.org/lkml/Yf+4k5Fs5Q3HdSG9@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
To check if more kernel API sync is needed and also to see if the perf
build tests continue to pass.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Pull xen fixes from Juergen Gross:
- documentation fixes related to Xen
- enable x2apic mode when available when running as hardware
virtualized guest under Xen
- cleanup and fix a corner case of vcpu enumeration when running a
paravirtualized Xen guest
* tag 'for-linus-5.17a-rc3-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip:
x86/Xen: streamline (and fix) PV CPU enumeration
xen: update missing ioctl magic numers documentation
Improve docs for IOCTL_GNTDEV_MAP_GRANT_REF
xen: xenbus_dev.h: delete incorrect file name
xen/x2apic: enable x2apic mode when supported for HVM
Pull kvm fixes from Paolo Bonzini:
"ARM:
- A couple of fixes when handling an exception while a SError has
been delivered
- Workaround for Cortex-A510's single-step erratum
RISC-V:
- Make CY, TM, and IR counters accessible in VU mode
- Fix SBI implementation version
x86:
- Report deprecation of x87 features in supported CPUID
- Preparation for fixing an interrupt delivery race on AMD hardware
- Sparse fix
All except POWER and s390:
- Rework guest entry code to correctly mark noinstr areas and fix
vtime' accounting (for x86, this was already mostly correct but not
entirely; for ARM, MIPS and RISC-V it wasn't)"
* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
KVM: x86: Use ERR_PTR_USR() to return -EFAULT as a __user pointer
KVM: x86: Report deprecated x87 features in supported CPUID
KVM: arm64: Workaround Cortex-A510's single-step and PAC trap errata
KVM: arm64: Stop handle_exit() from handling HVC twice when an SError occurs
KVM: arm64: Avoid consuming a stale esr value when SError occur
RISC-V: KVM: Fix SBI implementation version
RISC-V: KVM: make CY, TM, and IR counters accessible in VU mode
kvm/riscv: rework guest entry logic
kvm/arm64: rework guest entry logic
kvm/x86: rework guest entry logic
kvm/mips: rework guest entry logic
kvm: add guest_state_{enter,exit}_irqoff()
KVM: x86: Move delivery of non-APICv interrupt into vendor code
kvm: Move KVM_GET_XSAVE2 IOCTL definition at the end of kvm.h
Pull xfs fixes from Darrick Wong:
"I was auditing operations in XFS that clear file privileges, and
realized that XFS' fallocate implementation drops suid/sgid but
doesn't clear file capabilities the same way that file writes and
reflink do.
There are VFS helpers that do it correctly, so refactor XFS to use
them. I also noticed that we weren't flushing the log at the correct
point in the fallocate operation, so that's fixed too.
Summary:
- Fix fallocate so that it drops all file privileges when files are
modified instead of open-coding that incompletely.
- Fix fallocate to flush the log if the caller wanted synchronous
file updates"
* tag 'xfs-5.17-fixes-1' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux:
xfs: ensure log flush at the end of a synchronous fallocate call
xfs: move xfs_update_prealloc_flags() to xfs_pnfs.c
xfs: set prealloc flag in xfs_alloc_file_space()
xfs: fallocate() should call file_modified()
xfs: remove XFS_PREALLOC_SYNC
xfs: reject crazy array sizes being fed to XFS_IOC_GETBMAP*
Pull vfs fixes from Darrick Wong:
"I was auditing the sync_fs code paths recently and noticed that most
callers of ->sync_fs ignore its return value (and many implementations
never return nonzero even if the fs is broken!), which means that
internal fs errors and corruption are not passed up to userspace
callers of syncfs(2) or FIFREEZE. Hence fixing the common code and
XFS, and I'll start working on the ext4/btrfs folks if this is merged.
Summary:
- Fix a bug where callers of ->sync_fs (e.g. sync_filesystem and
syncfs(2)) ignore the return value.
- Fix a bug where callers of sync_filesystem (e.g. fs freeze) ignore
the return value.
- Fix a bug in XFS where xfs_fs_sync_fs never passed back error
returns"
* tag 'vfs-5.17-fixes-2' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux:
xfs: return errors in xfs_fs_sync_fs
quota: make dquot_quota_sync return errors from ->sync_fs
vfs: make sync_filesystem return errors from ->sync_fs
vfs: make freeze_super abort when sync_filesystem returns error
Pull iomap fix from Darrick Wong:
"A single bugfix for iomap.
The fix should eliminate occasional complaints about stall warnings
when a lot of writeback IO completes all at once and we have to then
go clearing status on a large number of folios.
Summary:
- Limit the length of ioend chains in writeback so that we don't trip
the softlockup watchdog and to limit long tail latency on clearing
PageWriteback"
* tag 'iomap-5.17-fixes-1' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux:
xfs, iomap: limit individual ioend chain lengths in writeback
