From: Namhyung Kim <[email protected]>
It was unnecessarily disabling and enabling PMUs for each event. It
should be done at PMU level, otherwise it'd increase the interrupt
latency when there are slow (uncore or virtual) PMUs and lots of events
on the PMUs.
Add pmu_ctx->nr_freq counter to check it at each PMU. As pmu context
has separate active lists for pinned group and flexible group, factor
out a new function to do the job. Also it can skip pmu contexts with
no active events (due to multiplexing or hybrid PMUs).
Another minor optimization is that it can skip PMUs w/ CAP_NO_INTERRUPT
even if it needs to unthrottle sampling events.
Reviewed-by: Ian Rogers <[email protected]>
Reviewed-by: Kan Liang <[email protected]>
Tested-by: Mingwei Zhang <[email protected]>
Signed-off-by: Namhyung Kim <[email protected]>
---
v2: Check if pmu context is active (Mark)
include/linux/perf_event.h | 6 ++++
kernel/events/core.c | 70 ++++++++++++++++++++++++--------------
2 files changed, 50 insertions(+), 26 deletions(-)
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index d2a15c0c6f8a..3e33b366347a 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -883,6 +883,7 @@ struct perf_event_pmu_context {
unsigned int nr_events;
unsigned int nr_cgroups;
+ unsigned int nr_freq;
atomic_t refcount; /* event <-> epc */
struct rcu_head rcu_head;
@@ -897,6 +898,11 @@ struct perf_event_pmu_context {
int rotate_necessary;
};
+static inline bool perf_pmu_ctx_is_active(struct perf_event_pmu_context *epc)
+{
+ return !list_empty(&epc->flexible_active) || !list_empty(&epc->pinned_active);
+}
+
struct perf_event_groups {
struct rb_root tree;
u64 index;
diff --git a/kernel/events/core.c b/kernel/events/core.c
index 59b332cce9e7..e9ce79c8c145 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -2277,8 +2277,10 @@ event_sched_out(struct perf_event *event, struct perf_event_context *ctx)
if (!is_software_event(event))
cpc->active_oncpu--;
- if (event->attr.freq && event->attr.sample_freq)
+ if (event->attr.freq && event->attr.sample_freq) {
ctx->nr_freq--;
+ epc->nr_freq--;
+ }
if (event->attr.exclusive || !cpc->active_oncpu)
cpc->exclusive = 0;
@@ -2533,9 +2535,10 @@ event_sched_in(struct perf_event *event, struct perf_event_context *ctx)
if (!is_software_event(event))
cpc->active_oncpu++;
- if (event->attr.freq && event->attr.sample_freq)
+ if (event->attr.freq && event->attr.sample_freq) {
ctx->nr_freq++;
-
+ epc->nr_freq++;
+ }
if (event->attr.exclusive)
cpc->exclusive = 1;
@@ -4098,30 +4101,14 @@ static void perf_adjust_period(struct perf_event *event, u64 nsec, u64 count, bo
}
}
-/*
- * combine freq adjustment with unthrottling to avoid two passes over the
- * events. At the same time, make sure, having freq events does not change
- * the rate of unthrottling as that would introduce bias.
- */
-static void
-perf_adjust_freq_unthr_context(struct perf_event_context *ctx, bool unthrottle)
+static void perf_adjust_freq_unthr_events(struct list_head *event_list)
{
struct perf_event *event;
struct hw_perf_event *hwc;
u64 now, period = TICK_NSEC;
s64 delta;
- /*
- * only need to iterate over all events iff:
- * - context have events in frequency mode (needs freq adjust)
- * - there are events to unthrottle on this cpu
- */
- if (!(ctx->nr_freq || unthrottle))
- return;
-
- raw_spin_lock(&ctx->lock);
-
- list_for_each_entry_rcu(event, &ctx->event_list, event_entry) {
+ list_for_each_entry(event, event_list, active_list) {
if (event->state != PERF_EVENT_STATE_ACTIVE)
continue;
@@ -4129,8 +4116,6 @@ perf_adjust_freq_unthr_context(struct perf_event_context *ctx, bool unthrottle)
if (!event_filter_match(event))
continue;
- perf_pmu_disable(event->pmu);
-
hwc = &event->hw;
if (hwc->interrupts == MAX_INTERRUPTS) {
@@ -4140,7 +4125,7 @@ perf_adjust_freq_unthr_context(struct perf_event_context *ctx, bool unthrottle)
}
if (!event->attr.freq || !event->attr.sample_freq)
- goto next;
+ continue;
/*
* stop the event and update event->count
@@ -4162,8 +4147,41 @@ perf_adjust_freq_unthr_context(struct perf_event_context *ctx, bool unthrottle)
perf_adjust_period(event, period, delta, false);
event->pmu->start(event, delta > 0 ? PERF_EF_RELOAD : 0);
- next:
- perf_pmu_enable(event->pmu);
+ }
+}
+
+/*
+ * combine freq adjustment with unthrottling to avoid two passes over the
+ * events. At the same time, make sure, having freq events does not change
+ * the rate of unthrottling as that would introduce bias.
+ */
+static void
+perf_adjust_freq_unthr_context(struct perf_event_context *ctx, bool unthrottle)
+{
+ struct perf_event_pmu_context *pmu_ctx;
+
+ /*
+ * only need to iterate over all events iff:
+ * - context have events in frequency mode (needs freq adjust)
+ * - there are events to unthrottle on this cpu
+ */
+ if (!(ctx->nr_freq || unthrottle))
+ return;
+
+ raw_spin_lock(&ctx->lock);
+
+ list_for_each_entry(pmu_ctx, &ctx->pmu_ctx_list, pmu_ctx_entry) {
+ if (!(pmu_ctx->nr_freq || unthrottle))
+ continue;
+ if (!perf_pmu_ctx_is_active(pmu_ctx))
+ continue;
+ if (pmu_ctx->pmu->capabilities & PERF_PMU_CAP_NO_INTERRUPT)
+ continue;
+
+ perf_pmu_disable(pmu_ctx->pmu);
+ perf_adjust_freq_unthr_events(&pmu_ctx->pinned_active);
+ perf_adjust_freq_unthr_events(&pmu_ctx->flexible_active);
+ perf_pmu_enable(pmu_ctx->pmu);
}
raw_spin_unlock(&ctx->lock);
--
2.43.0.275.g3460e3d667-goog
From: Namhyung Kim <[email protected]>
For throttled events, it first starts the event and then stop
unnecessarily. As it's already stopped, it can directly adjust
the frequency and then move on.
Reviewed-by: Ian Rogers <[email protected]>
Reviewed-by: Kan Liang <[email protected]>
Signed-off-by: Namhyung Kim <[email protected]>
---
kernel/events/core.c | 13 ++++++++++---
1 file changed, 10 insertions(+), 3 deletions(-)
diff --git a/kernel/events/core.c b/kernel/events/core.c
index e9ce79c8c145..243cd09ba7fe 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -4121,10 +4121,15 @@ static void perf_adjust_freq_unthr_events(struct list_head *event_list)
if (hwc->interrupts == MAX_INTERRUPTS) {
hwc->interrupts = 0;
perf_log_throttle(event, 1);
- event->pmu->start(event, 0);
- }
- if (!event->attr.freq || !event->attr.sample_freq)
+ if (!event->attr.freq || !event->attr.sample_freq) {
+ delta = 0;
+ goto next;
+ }
+
+ if (event->hw.state & PERF_HES_STOPPED)
+ goto adjust;
+ } else if (!event->attr.freq || !event->attr.sample_freq)
continue;
/*
@@ -4132,6 +4137,7 @@ static void perf_adjust_freq_unthr_events(struct list_head *event_list)
*/
event->pmu->stop(event, PERF_EF_UPDATE);
+adjust:
now = local64_read(&event->count);
delta = now - hwc->freq_count_stamp;
hwc->freq_count_stamp = now;
@@ -4146,6 +4152,7 @@ static void perf_adjust_freq_unthr_events(struct list_head *event_list)
if (delta > 0)
perf_adjust_period(event, period, delta, false);
+next:
event->pmu->start(event, delta > 0 ? PERF_EF_RELOAD : 0);
}
}
--
2.43.0.275.g3460e3d667-goog
Hello,
kernel test robot noticed "WARNING:at_arch/x86/events/core.c:#x86_pmu_start" on:
commit: d6da92786f901cc4ce3588f101182758da295dbb ("[PATCH v2 2/2] perf/core: Reduce PMU access to adjust sample freq")
url: https://github.com/intel-lab-lkp/linux/commits/namhyung-kernel-org/perf-core-Reduce-PMU-access-to-adjust-sample-freq/20240112-044505
base: https://git.kernel.org/cgit/linux/kernel/git/perf/perf-tools-next.git perf-tools-next
patch link: https://lore.kernel.org/all/[email protected]/
patch subject: [PATCH v2 2/2] perf/core: Reduce PMU access to adjust sample freq
in testcase: will-it-scale
version: will-it-scale-x86_64-75f66e4-1_20240111
with following parameters:
nr_task: 16
mode: thread
test: pipe1
cpufreq_governor: performance
compiler: gcc-12
test machine: 104 threads 2 sockets (Skylake) with 192G memory
(please refer to attached dmesg/kmsg for entire log/backtrace)
If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <[email protected]>
| Closes: https://lore.kernel.org/oe-lkp/[email protected]
[ 102.087071][ C94] ------------[ cut here ]------------
[ 102.092623][ C94] WARNING: CPU: 94 PID: 0 at arch/x86/events/core.c:1507 x86_pmu_start (arch/x86/events/core.c:1507 (discriminator 1))
[ 102.101826][ C94] Modules linked in: intel_rapl_msr intel_rapl_common skx_edac nfit libnvdimm x86_pkg_temp_thermal intel_powerclamp btrfs coretemp blake2b_generic xor kvm_intel kvm raid6_pq libcrc32c irqbypass crct10dif_pclmul crc32_pclmul crc32c_intel ghash_clmulni_intel sha512_ssse3 sd_mod sg rapl ipmi_ssif nvme nvme_core ahci intel_cstate acpi_ipmi t10_pi libahci ast crc64_rocksoft_generic drm_shmem_helper mei_me ipmi_si crc64_rocksoft i2c_i801 ioatdma libata intel_uncore drm_kms_helper joydev crc64 mei ipmi_devintf lpc_ich i2c_smbus intel_pch_thermal dca wmi ipmi_msghandler acpi_pad acpi_power_meter drm fuse ip_tables
[ 102.158393][ C94] CPU: 94 PID: 0 Comm: swapper/94 Not tainted 6.7.0-rc6-00192-gd6da92786f90 #1
[ 102.167472][ C94] RIP: 0010:x86_pmu_start (arch/x86/events/core.c:1507 (discriminator 1))
[ 102.172832][ C94] Code: 00 00 4c 0f ab 65 00 48 89 df e8 16 08 01 00 48 89 df 5b 5d 41 5c e9 4a c6 33 00 0f 0b 5b 5d 41 5c c3 cc cc cc cc 0f 0b eb f3 <0f> 0b eb b6 66 66 2e 0f 1f 84 00 00 00 00 00 66 66 2e 0f 1f 84 00
All code
========
0: 00 00 add %al,(%rax)
2: 4c 0f ab 65 00 bts %r12,0x0(%rbp)
7: 48 89 df mov %rbx,%rdi
a: e8 16 08 01 00 callq 0x10825
f: 48 89 df mov %rbx,%rdi
12: 5b pop %rbx
13: 5d pop %rbp
14: 41 5c pop %r12
16: e9 4a c6 33 00 jmpq 0x33c665
1b: 0f 0b ud2
1d: 5b pop %rbx
1e: 5d pop %rbp
1f: 41 5c pop %r12
21: c3 retq
22: cc int3
23: cc int3
24: cc int3
25: cc int3
26: 0f 0b ud2
28: eb f3 jmp 0x1d
2a:* 0f 0b ud2 <-- trapping instruction
2c: eb b6 jmp 0xffffffffffffffe4
2e: 66 66 2e 0f 1f 84 00 data16 nopw %cs:0x0(%rax,%rax,1)
35: 00 00 00 00
39: 66 data16
3a: 66 data16
3b: 2e cs
3c: 0f .byte 0xf
3d: 1f (bad)
3e: 84 00 test %al,(%rax)
Code starting with the faulting instruction
===========================================
0: 0f 0b ud2
2: eb b6 jmp 0xffffffffffffffba
4: 66 66 2e 0f 1f 84 00 data16 nopw %cs:0x0(%rax,%rax,1)
b: 00 00 00 00
f: 66 data16
10: 66 data16
11: 2e cs
12: 0f .byte 0xf
13: 1f (bad)
14: 84 00 test %al,(%rax)
[ 102.192917][ C94] RSP: 0018:ffffc9000ddb0e00 EFLAGS: 00010046
[ 102.199175][ C94] RAX: 0000000000000001 RBX: ffff88b01d17a290 RCX: 0000000000000349
[ 102.207339][ C94] RDX: 0000000000002ff0 RSI: 0000000000000002 RDI: ffff88b01d17a290
[ 102.215509][ C94] RBP: ffff88afa149a220 R08: 0000000000000000 R09: 0000000000000014
[ 102.223684][ C94] R10: 000000000000000f R11: 00000000000f4240 R12: 0000000000000003
[ 102.231855][ C94] R13: 0000000000000001 R14: ffff88afa14b9680 R15: 000000000000005e
[ 102.240038][ C94] FS: 0000000000000000(0000) GS:ffff88afa1480000(0000) knlGS:0000000000000000
[ 102.249178][ C94] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 102.255986][ C94] CR2: 00007f9cdf69ec98 CR3: 000000303e01c002 CR4: 00000000007706f0
[ 102.264179][ C94] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 102.272365][ C94] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 102.280552][ C94] PKRU: 55555554
[ 102.284322][ C94] Call Trace:
[ 102.287830][ C94] <IRQ>
[ 102.290895][ C94] ? x86_pmu_start (arch/x86/events/core.c:1507 (discriminator 1))
[ 102.295695][ C94] ? __warn (kernel/panic.c:677)
[ 102.299980][ C94] ? x86_pmu_start (arch/x86/events/core.c:1507 (discriminator 1))
[ 102.304768][ C94] ? report_bug (lib/bug.c:180 lib/bug.c:219)
[ 102.309473][ C94] ? handle_bug (arch/x86/kernel/traps.c:237)
[ 102.314006][ C94] ? exc_invalid_op (arch/x86/kernel/traps.c:258 (discriminator 1))
[ 102.318879][ C94] ? asm_exc_invalid_op (arch/x86/include/asm/idtentry.h:568)
[ 102.324101][ C94] ? x86_pmu_start (arch/x86/events/core.c:1507 (discriminator 1))
[ 102.328888][ C94] perf_adjust_freq_unthr_events (kernel/events/core.c:4181 (discriminator 4))
[ 102.335069][ C94] perf_adjust_freq_unthr_context (kernel/events/core.c:4216)
[ 102.341244][ C94] perf_event_task_tick (arch/x86/include/asm/current.h:41 kernel/events/core.c:4363)
[ 102.346458][ C94] scheduler_tick (kernel/sched/core.c:5665)
[ 102.351240][ C94] update_process_times (kernel/time/timer.c:2079)
[ 102.356442][ C94] tick_sched_handle (kernel/time/tick-sched.c:256)
[ 102.361381][ C94] tick_nohz_highres_handler (kernel/time/tick-sched.c:1525)
[ 102.367021][ C94] ? __pfx_tick_nohz_highres_handler (kernel/time/tick-sched.c:1503)
[ 102.373345][ C94] __hrtimer_run_queues (kernel/time/hrtimer.c:1688 kernel/time/hrtimer.c:1752)
[ 102.378720][ C94] hrtimer_interrupt (kernel/time/hrtimer.c:1817)
[ 102.383748][ C94] __sysvec_apic_timer_interrupt (arch/x86/kernel/apic/apic.c:1065 arch/x86/kernel/apic/apic.c:1082)
[ 102.389818][ C94] sysvec_apic_timer_interrupt (arch/x86/kernel/apic/apic.c:1076 (discriminator 14))
[ 102.395636][ C94] </IRQ>
[ 102.398759][ C94] <TASK>
[ 102.401872][ C94] asm_sysvec_apic_timer_interrupt (arch/x86/include/asm/idtentry.h:649)
[ 102.408032][ C94] RIP: 0010:cpuidle_enter_state (drivers/cpuidle/cpuidle.c:291)
[ 102.414008][ C94] Code: 00 e8 9e 46 19 ff e8 d9 f1 ff ff 8b 53 04 49 89 c5 0f 1f 44 00 00 31 ff e8 07 2e 18 ff 45 84 ff 0f 85 d2 00 00 00 fb 45 85 f6 <0f> 88 83 01 00 00 49 63 d6 48 8d 04 52 48 8d 04 82 49 8d 0c c4 48
All code
========
0: 00 e8 add %ch,%al
2: 9e sahf
3: 46 19 ff rex.RX sbb %r15d,%edi
6: e8 d9 f1 ff ff callq 0xfffffffffffff1e4
b: 8b 53 04 mov 0x4(%rbx),%edx
e: 49 89 c5 mov %rax,%r13
11: 0f 1f 44 00 00 nopl 0x0(%rax,%rax,1)
16: 31 ff xor %edi,%edi
18: e8 07 2e 18 ff callq 0xffffffffff182e24
1d: 45 84 ff test %r15b,%r15b
20: 0f 85 d2 00 00 00 jne 0xf8
26: fb sti
27: 45 85 f6 test %r14d,%r14d
2a:* 0f 88 83 01 00 00 js 0x1b3 <-- trapping instruction
30: 49 63 d6 movslq %r14d,%rdx
33: 48 8d 04 52 lea (%rdx,%rdx,2),%rax
37: 48 8d 04 82 lea (%rdx,%rax,4),%rax
3b: 49 8d 0c c4 lea (%r12,%rax,8),%rcx
3f: 48 rex.W
Code starting with the faulting instruction
===========================================
0: 0f 88 83 01 00 00 js 0x189
6: 49 63 d6 movslq %r14d,%rdx
9: 48 8d 04 52 lea (%rdx,%rdx,2),%rax
d: 48 8d 04 82 lea (%rdx,%rax,4),%rax
11: 49 8d 0c c4 lea (%r12,%rax,8),%rcx
15: 48 rex.W
The kernel config and materials to reproduce are available at:
https://download.01.org/0day-ci/archive/20240119/[email protected]
--
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki
Hello,
On Thu, Jan 18, 2024 at 6:41 PM kernel test robot <[email protected]> wrote:
>
>
>
> Hello,
>
> kernel test robot noticed "WARNING:at_arch/x86/events/core.c:#x86_pmu_start" on:
>
> commit: d6da92786f901cc4ce3588f101182758da295dbb ("[PATCH v2 2/2] perf/core: Reduce PMU access to adjust sample freq")
> url: https://github.com/intel-lab-lkp/linux/commits/namhyung-kernel-org/perf-core-Reduce-PMU-access-to-adjust-sample-freq/20240112-044505
> base: https://git.kernel.org/cgit/linux/kernel/git/perf/perf-tools-next.git perf-tools-next
> patch link: https://lore.kernel.org/all/[email protected]/
> patch subject: [PATCH v2 2/2] perf/core: Reduce PMU access to adjust sample freq
>
> in testcase: will-it-scale
> version: will-it-scale-x86_64-75f66e4-1_20240111
> with following parameters:
>
> nr_task: 16
> mode: thread
> test: pipe1
> cpufreq_governor: performance
>
>
>
> compiler: gcc-12
> test machine: 104 threads 2 sockets (Skylake) with 192G memory
>
> (please refer to attached dmesg/kmsg for entire log/backtrace)
Thanks for the report. It seems the code calls x86_pmu_stop() without
PERF_EF_UPDATE so we cannot simply skip the stop callback for
frequency events. I'll update it in v3.
Thanks,
Namhyung
>
>
>
> If you fix the issue in a separate patch/commit (i.e. not just a new version of
> the same patch/commit), kindly add following tags
> | Reported-by: kernel test robot <[email protected]>
> | Closes: https://lore.kernel.org/oe-lkp/[email protected]
>
>
> [ 102.087071][ C94] ------------[ cut here ]------------
> [ 102.092623][ C94] WARNING: CPU: 94 PID: 0 at arch/x86/events/core.c:1507 x86_pmu_start (arch/x86/events/core.c:1507 (discriminator 1))
> [ 102.101826][ C94] Modules linked in: intel_rapl_msr intel_rapl_common skx_edac nfit libnvdimm x86_pkg_temp_thermal intel_powerclamp btrfs coretemp blake2b_generic xor kvm_intel kvm raid6_pq libcrc32c irqbypass crct10dif_pclmul crc32_pclmul crc32c_intel ghash_clmulni_intel sha512_ssse3 sd_mod sg rapl ipmi_ssif nvme nvme_core ahci intel_cstate acpi_ipmi t10_pi libahci ast crc64_rocksoft_generic drm_shmem_helper mei_me ipmi_si crc64_rocksoft i2c_i801 ioatdma libata intel_uncore drm_kms_helper joydev crc64 mei ipmi_devintf lpc_ich i2c_smbus intel_pch_thermal dca wmi ipmi_msghandler acpi_pad acpi_power_meter drm fuse ip_tables
> [ 102.158393][ C94] CPU: 94 PID: 0 Comm: swapper/94 Not tainted 6.7.0-rc6-00192-gd6da92786f90 #1
> [ 102.167472][ C94] RIP: 0010:x86_pmu_start (arch/x86/events/core.c:1507 (discriminator 1))
> [ 102.172832][ C94] Code: 00 00 4c 0f ab 65 00 48 89 df e8 16 08 01 00 48 89 df 5b 5d 41 5c e9 4a c6 33 00 0f 0b 5b 5d 41 5c c3 cc cc cc cc 0f 0b eb f3 <0f> 0b eb b6 66 66 2e 0f 1f 84 00 00 00 00 00 66 66 2e 0f 1f 84 00
> All code
> ========
> 0: 00 00 add %al,(%rax)
> 2: 4c 0f ab 65 00 bts %r12,0x0(%rbp)
> 7: 48 89 df mov %rbx,%rdi
> a: e8 16 08 01 00 callq 0x10825
> f: 48 89 df mov %rbx,%rdi
> 12: 5b pop %rbx
> 13: 5d pop %rbp
> 14: 41 5c pop %r12
> 16: e9 4a c6 33 00 jmpq 0x33c665
> 1b: 0f 0b ud2
> 1d: 5b pop %rbx
> 1e: 5d pop %rbp
> 1f: 41 5c pop %r12
> 21: c3 retq
> 22: cc int3
> 23: cc int3
> 24: cc int3
> 25: cc int3
> 26: 0f 0b ud2
> 28: eb f3 jmp 0x1d
> 2a:* 0f 0b ud2 <-- trapping instruction
> 2c: eb b6 jmp 0xffffffffffffffe4
> 2e: 66 66 2e 0f 1f 84 00 data16 nopw %cs:0x0(%rax,%rax,1)
> 35: 00 00 00 00
> 39: 66 data16
> 3a: 66 data16
> 3b: 2e cs
> 3c: 0f .byte 0xf
> 3d: 1f (bad)
> 3e: 84 00 test %al,(%rax)
>
> Code starting with the faulting instruction
> ===========================================
> 0: 0f 0b ud2
> 2: eb b6 jmp 0xffffffffffffffba
> 4: 66 66 2e 0f 1f 84 00 data16 nopw %cs:0x0(%rax,%rax,1)
> b: 00 00 00 00
> f: 66 data16
> 10: 66 data16
> 11: 2e cs
> 12: 0f .byte 0xf
> 13: 1f (bad)
> 14: 84 00 test %al,(%rax)
> [ 102.192917][ C94] RSP: 0018:ffffc9000ddb0e00 EFLAGS: 00010046
> [ 102.199175][ C94] RAX: 0000000000000001 RBX: ffff88b01d17a290 RCX: 0000000000000349
> [ 102.207339][ C94] RDX: 0000000000002ff0 RSI: 0000000000000002 RDI: ffff88b01d17a290
> [ 102.215509][ C94] RBP: ffff88afa149a220 R08: 0000000000000000 R09: 0000000000000014
> [ 102.223684][ C94] R10: 000000000000000f R11: 00000000000f4240 R12: 0000000000000003
> [ 102.231855][ C94] R13: 0000000000000001 R14: ffff88afa14b9680 R15: 000000000000005e
> [ 102.240038][ C94] FS: 0000000000000000(0000) GS:ffff88afa1480000(0000) knlGS:0000000000000000
> [ 102.249178][ C94] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> [ 102.255986][ C94] CR2: 00007f9cdf69ec98 CR3: 000000303e01c002 CR4: 00000000007706f0
> [ 102.264179][ C94] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> [ 102.272365][ C94] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
> [ 102.280552][ C94] PKRU: 55555554
> [ 102.284322][ C94] Call Trace:
> [ 102.287830][ C94] <IRQ>
> [ 102.290895][ C94] ? x86_pmu_start (arch/x86/events/core.c:1507 (discriminator 1))
> [ 102.295695][ C94] ? __warn (kernel/panic.c:677)
> [ 102.299980][ C94] ? x86_pmu_start (arch/x86/events/core.c:1507 (discriminator 1))
> [ 102.304768][ C94] ? report_bug (lib/bug.c:180 lib/bug.c:219)
> [ 102.309473][ C94] ? handle_bug (arch/x86/kernel/traps.c:237)
> [ 102.314006][ C94] ? exc_invalid_op (arch/x86/kernel/traps.c:258 (discriminator 1))
> [ 102.318879][ C94] ? asm_exc_invalid_op (arch/x86/include/asm/idtentry.h:568)
> [ 102.324101][ C94] ? x86_pmu_start (arch/x86/events/core.c:1507 (discriminator 1))
> [ 102.328888][ C94] perf_adjust_freq_unthr_events (kernel/events/core.c:4181 (discriminator 4))
> [ 102.335069][ C94] perf_adjust_freq_unthr_context (kernel/events/core.c:4216)
> [ 102.341244][ C94] perf_event_task_tick (arch/x86/include/asm/current.h:41 kernel/events/core.c:4363)
> [ 102.346458][ C94] scheduler_tick (kernel/sched/core.c:5665)
> [ 102.351240][ C94] update_process_times (kernel/time/timer.c:2079)
> [ 102.356442][ C94] tick_sched_handle (kernel/time/tick-sched.c:256)
> [ 102.361381][ C94] tick_nohz_highres_handler (kernel/time/tick-sched.c:1525)
> [ 102.367021][ C94] ? __pfx_tick_nohz_highres_handler (kernel/time/tick-sched.c:1503)
> [ 102.373345][ C94] __hrtimer_run_queues (kernel/time/hrtimer.c:1688 kernel/time/hrtimer.c:1752)
> [ 102.378720][ C94] hrtimer_interrupt (kernel/time/hrtimer.c:1817)
> [ 102.383748][ C94] __sysvec_apic_timer_interrupt (arch/x86/kernel/apic/apic.c:1065 arch/x86/kernel/apic/apic.c:1082)
> [ 102.389818][ C94] sysvec_apic_timer_interrupt (arch/x86/kernel/apic/apic.c:1076 (discriminator 14))
> [ 102.395636][ C94] </IRQ>
> [ 102.398759][ C94] <TASK>
> [ 102.401872][ C94] asm_sysvec_apic_timer_interrupt (arch/x86/include/asm/idtentry.h:649)
> [ 102.408032][ C94] RIP: 0010:cpuidle_enter_state (drivers/cpuidle/cpuidle.c:291)
> [ 102.414008][ C94] Code: 00 e8 9e 46 19 ff e8 d9 f1 ff ff 8b 53 04 49 89 c5 0f 1f 44 00 00 31 ff e8 07 2e 18 ff 45 84 ff 0f 85 d2 00 00 00 fb 45 85 f6 <0f> 88 83 01 00 00 49 63 d6 48 8d 04 52 48 8d 04 82 49 8d 0c c4 48
> All code
> ========
> 0: 00 e8 add %ch,%al
> 2: 9e sahf
> 3: 46 19 ff rex.RX sbb %r15d,%edi
> 6: e8 d9 f1 ff ff callq 0xfffffffffffff1e4
> b: 8b 53 04 mov 0x4(%rbx),%edx
> e: 49 89 c5 mov %rax,%r13
> 11: 0f 1f 44 00 00 nopl 0x0(%rax,%rax,1)
> 16: 31 ff xor %edi,%edi
> 18: e8 07 2e 18 ff callq 0xffffffffff182e24
> 1d: 45 84 ff test %r15b,%r15b
> 20: 0f 85 d2 00 00 00 jne 0xf8
> 26: fb sti
> 27: 45 85 f6 test %r14d,%r14d
> 2a:* 0f 88 83 01 00 00 js 0x1b3 <-- trapping instruction
> 30: 49 63 d6 movslq %r14d,%rdx
> 33: 48 8d 04 52 lea (%rdx,%rdx,2),%rax
> 37: 48 8d 04 82 lea (%rdx,%rax,4),%rax
> 3b: 49 8d 0c c4 lea (%r12,%rax,8),%rcx
> 3f: 48 rex.W
>
> Code starting with the faulting instruction
> ===========================================
> 0: 0f 88 83 01 00 00 js 0x189
> 6: 49 63 d6 movslq %r14d,%rdx
> 9: 48 8d 04 52 lea (%rdx,%rdx,2),%rax
> d: 48 8d 04 82 lea (%rdx,%rax,4),%rax
> 11: 49 8d 0c c4 lea (%r12,%rax,8),%rcx
> 15: 48 rex.W
>
>
> The kernel config and materials to reproduce are available at:
> https://download.01.org/0day-ci/archive/20240119/[email protected]
>
>
>
> --
> 0-DAY CI Kernel Test Service
> https://github.com/intel/lkp-tests/wiki
>