Linux Perf Users
 help / color / mirror / Atom feed
From: "Mi, Dapeng" <dapeng1.mi@linux.intel.com>
To: Peter Zijlstra <peterz@infradead.org>
Cc: Ingo Molnar <mingo@redhat.com>,
	Arnaldo Carvalho de Melo <acme@kernel.org>,
	Namhyung Kim <namhyung@kernel.org>,
	Ian Rogers <irogers@google.com>,
	Adrian Hunter <adrian.hunter@intel.com>,
	Alexander Shishkin <alexander.shishkin@linux.intel.com>,
	Andi Kleen <ak@linux.intel.com>,
	Eranian Stephane <eranian@google.com>,
	linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org,
	Dapeng Mi <dapeng1.mi@intel.com>, Zide Chen <zide.chen@intel.com>,
	Falcon Thomas <thomas.falcon@intel.com>,
	Xudong Hao <xudong.hao@intel.com>, Yi Lai <yi1.lai@intel.com>
Subject: Re: [Patch v2 7/9] perf/x86/intel: Drop fixed-counter PEBS constraints for baseline PEBS
Date: Wed, 10 Jun 2026 16:50:45 +0800	[thread overview]
Message-ID: <b3485f13-039d-44d9-af02-5cea084ca304@linux.intel.com> (raw)
In-Reply-To: <20260610082051.GF49951@noisy.programming.kicks-ass.net>


On 6/10/2026 4:20 PM, Peter Zijlstra wrote:
> On Tue, Jun 09, 2026 at 01:02:20PM +0800, Dapeng Mi wrote:
>> On SPR guests where pebs_baseline is not advertised, running:
>>
>> $ ./perf record -e cpu/event=0x00,umask=0x01,i\
>> 	 name=INST_RETIRED.PREC_DIST/p -c 10000 sleep 1
>>
>> can trigger:
>>
>> unchecked MSR access error: WRMSR to 0x3f1 ... in\
>> 	 intel_pmu_pebs_enable_all()
>>
>> Root cause:
>> SPR-specific PEBS constraints allow fixed-counter scheduling,
>> for example INST_RETIRED.PREC_DIST on fixed counter 0. In guests without
>> pebs_baseline, KVM does not support PEBS sampling on fixed counters,
>> so enabling such events reaches an invalid MSR programming path.
>>
>> Fix:
>> Drop fixed-counter entries from the PEBS constraint table. Without
>> pebs_baseline, those fixed-counter PEBS events now resolve to empty
>> constraints and are not scheduled/enabled, avoiding the warning and the
>> broken guest PEBS path.
>>
>> This is safe because, in pebs_baseline-capable cases, PEBS constraint
>> lookup already falls back to non-PEBS constraints when needed, and
>> fixed-counter constraints are effectively shared there.
> I am confused, this works outside of KVM? (It appears to work fine on my
> spr).. so removing this to fix some guest only issue seems wrong.

The reason that it works on bare metal is currently the constraint lookup
would fallback into non-PEBS constraints if there is no matched entry in
the PEBS constraints as long as the PEBS supports sampling on all counters
including fixed counters (what the flag "PMU_FL_PEBS_ALL" indicates), like
the below code shows.

```

struct event_constraint *intel_pebs_constraints(struct perf_event *event)
{
    struct event_constraint *pebs_constraints = hybrid(event->pmu,
pebs_constraints);
    struct event_constraint *c;

    if (!event->attr.precise_ip)
        return NULL;

    if (pebs_constraints) {
        for_each_event_constraint(c, pebs_constraints) {
            if (constraint_match(c, event->hw.config)) {
                event->hw.flags |= c->flags;
                return c;
            }
        }
    }

    /*
     * Extended PEBS support
     * Makes the PEBS code search the normal constraints.
     */
    if (x86_pmu.flags & PMU_FL_PEBS_ALL)
        return NULL;

    return &emptyconstraint;
}

static struct event_constraint *
__intel_get_event_constraints(struct cpu_hw_events *cpuc, int idx,
                struct perf_event *event)
{
    struct event_constraint *c;

    c = intel_vlbr_constraints(event);
    if (c)
        return c;

    c = intel_bts_constraints(event);
    if (c)
        return c;

    c = intel_shared_regs_constraints(cpuc, event);
    if (c)
        return c;

    c = intel_pebs_constraints(event);
    if (c)
        return c;

    return x86_get_event_constraints(cpuc, idx, event);
}

```

if intel_pebs_constraints() can't find the matched entry from PEBS
constraints and PMU_FL_PEBS_ALL is set, NULL would be returned and it would
lead to fallback the non-PEBS constraints lookup.

From Icelake starts, regardless of the extended PEBS or the arch-PEBS, all
counters including the fixed counters support PEBS  sampling and the
flag PMU_FL_PEBS_ALL would set by default.

Since the non-PEBS constraints contain the similar fixed counter
constraints, the fixed counter events would still be correctly scheduled on
the fixed counters.

for example, here is the fixed counter constraints in
intel_icl_event_constraints[],

```

static struct event_constraint intel_icl_event_constraints[] = {
    FIXED_EVENT_CONSTRAINT(0x00c0, 0),    /* INST_RETIRED.ANY */
    FIXED_EVENT_CONSTRAINT(0x01c0, 0),    /* old INST_RETIRED.PREC_DIST */
    FIXED_EVENT_CONSTRAINT(0x0100, 0),    /* pseudo INST_RETIRED.ANY */
    FIXED_EVENT_CONSTRAINT(0x003c, 1),    /* CPU_CLK_UNHALTED.CORE */
    FIXED_EVENT_CONSTRAINT(0x0200, 1),    /* pseudo CPU_CLK_UNHALTED.THREAD */
    FIXED_EVENT_CONSTRAINT(0x0300, 2),    /* pseudo CPU_CLK_UNHALTED.REF_TSC */
    FIXED_EVENT_CONSTRAINT(0x0400, 3),    /* pseudo TOPDOWN.SLOTS */

    ......

```

>
>
>> Reported-by: Yi Lai <yi1.lai@intel.com>
>> Signed-off-by: Dapeng Mi <dapeng1.mi@linux.intel.com>
>> ---
>>  arch/x86/events/intel/ds.c | 13 -------------
>>  1 file changed, 13 deletions(-)
>>
>> diff --git a/arch/x86/events/intel/ds.c b/arch/x86/events/intel/ds.c
>> index cb72af9b61ce..5db15a92017a 100644
>> --- a/arch/x86/events/intel/ds.c
>> +++ b/arch/x86/events/intel/ds.c
>> @@ -1447,10 +1447,6 @@ struct event_constraint intel_skl_pebs_event_constraints[] = {
>>  };
>>  
>>  struct event_constraint intel_icl_pebs_event_constraints[] = {
>> -	INTEL_FLAGS_UEVENT_CONSTRAINT(0x01c0, 0x100000000ULL),	/* old INST_RETIRED.PREC_DIST */
>> -	INTEL_FLAGS_UEVENT_CONSTRAINT(0x0100, 0x100000000ULL),	/* INST_RETIRED.PREC_DIST */
>> -	INTEL_FLAGS_UEVENT_CONSTRAINT(0x0400, 0x800000000ULL),	/* SLOTS */
>> -
>>  	INTEL_PLD_CONSTRAINT(0x1cd, 0xff),			/* MEM_TRANS_RETIRED.LOAD_LATENCY */
>>  	INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_LD(0x11d0, 0xf),	/* MEM_INST_RETIRED.STLB_MISS_LOADS */
>>  	INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_ST(0x12d0, 0xf),	/* MEM_INST_RETIRED.STLB_MISS_STORES */
>> @@ -1473,9 +1469,6 @@ struct event_constraint intel_icl_pebs_event_constraints[] = {
>>  };
>>  
>>  struct event_constraint intel_glc_pebs_event_constraints[] = {
>> -	INTEL_FLAGS_UEVENT_CONSTRAINT(0x100, 0x100000000ULL),	/* INST_RETIRED.PREC_DIST */
>> -	INTEL_FLAGS_UEVENT_CONSTRAINT(0x0400, 0x800000000ULL),
>> -
>>  	INTEL_FLAGS_EVENT_CONSTRAINT(0xc0, 0xfe),
>>  	INTEL_PLD_CONSTRAINT(0x1cd, 0xfe),
>>  	INTEL_PSD_CONSTRAINT(0x2cd, 0x1),
>> @@ -1500,9 +1493,6 @@ struct event_constraint intel_glc_pebs_event_constraints[] = {
>>  };
>>  
>>  struct event_constraint intel_lnc_pebs_event_constraints[] = {
>> -	INTEL_FLAGS_UEVENT_CONSTRAINT(0x100, 0x100000000ULL),	/* INST_RETIRED.PREC_DIST */
>> -	INTEL_FLAGS_UEVENT_CONSTRAINT(0x0400, 0x800000000ULL),
>> -
>>  	INTEL_FLAGS_UEVENT_CONSTRAINT(0x012a, 0x1),		/* OCR.* events */
>>  	INTEL_FLAGS_UEVENT_CONSTRAINT(0x012b, 0x1),		/* OCR.* events */
>>  
>> @@ -1534,9 +1524,6 @@ struct event_constraint intel_lnc_pebs_event_constraints[] = {
>>  };
>>  
>>  struct event_constraint intel_pnc_pebs_event_constraints[] = {
>> -	INTEL_FLAGS_UEVENT_CONSTRAINT(0x100, 0x100000000ULL),	/* INST_RETIRED.PREC_DIST */
>> -	INTEL_FLAGS_UEVENT_CONSTRAINT(0x0400, 0x800000000ULL),
>> -
>>  	INTEL_HYBRID_LDLAT_CONSTRAINT(0x1cd, 0xfc),
>>  	INTEL_HYBRID_STLAT_CONSTRAINT(0x2cd, 0x3),
>>  	INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_LD(0x11d0, 0xf),	/* MEM_INST_RETIRED.STLB_MISS_LOADS */
>> -- 
>> 2.34.1
>>

  parent reply	other threads:[~2026-06-10  8:50 UTC|newest]

Thread overview: 32+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-06-09  5:02 [Patch v2 0/9] perf/x86: Miscellaneous PMU bug fixes Dapeng Mi
2026-06-09  5:02 ` [Patch v2 1/9] perf/x86/intel: Remove anythread_deprecated bit from perf_capabilities Dapeng Mi
2026-06-09  5:02 ` [Patch v2 2/9] perf/x86: Introduce is_x86_pmu() helper Dapeng Mi
2026-06-09  5:02 ` [Patch v2 3/9] perf/x86: Update cap_user_rdpmc base on rdpmc user disable state Dapeng Mi
2026-06-09 14:48   ` Peter Zijlstra
2026-06-10  1:47     ` Mi, Dapeng
2026-06-09  5:02 ` [Patch v2 4/9] perf/x86/intel: Fallback to sw branch type decoding if no hw decoding Dapeng Mi
2026-06-09  5:24   ` sashiko-bot
2026-06-09 10:04     ` Mi, Dapeng
2026-06-09 14:49   ` Peter Zijlstra
2026-06-10  1:53     ` Mi, Dapeng
2026-06-09  5:02 ` [Patch v2 5/9] perf/x86/intel: Drop LBR entries whose privilege level mismatches br_sel Dapeng Mi
2026-06-09  5:21   ` sashiko-bot
2026-06-09  9:40     ` Mi, Dapeng
2026-06-09 14:52   ` Peter Zijlstra
2026-06-10  1:57     ` Mi, Dapeng
2026-06-09  5:02 ` [Patch v2 6/9] perf/x86/intel: Validate return value of intel_pmu_init_hybrid() Dapeng Mi
2026-06-09  5:25   ` sashiko-bot
2026-06-09  9:44     ` Mi, Dapeng
2026-06-10  8:16   ` Peter Zijlstra
2026-06-10  8:34     ` Mi, Dapeng
2026-06-09  5:02 ` [Patch v2 7/9] perf/x86/intel: Drop fixed-counter PEBS constraints for baseline PEBS Dapeng Mi
2026-06-10  8:20   ` Peter Zijlstra
2026-06-10  8:23     ` Peter Zijlstra
2026-06-10  8:50     ` Mi, Dapeng [this message]
2026-06-10 11:21       ` Peter Zijlstra
2026-06-10 11:42         ` Mi, Dapeng
2026-06-09  5:02 ` [Patch v2 8/9] perf/core: Fix kernel register info leak via hardware skid Dapeng Mi
2026-06-10  9:16   ` Peter Zijlstra
2026-06-09  5:02 ` [Patch v2 9/9] perf/core: Check kernel access when kernel callchains are requested Dapeng Mi
2026-06-09  5:24   ` sashiko-bot
2026-06-09  9:49     ` Mi, Dapeng

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=b3485f13-039d-44d9-af02-5cea084ca304@linux.intel.com \
    --to=dapeng1.mi@linux.intel.com \
    --cc=acme@kernel.org \
    --cc=adrian.hunter@intel.com \
    --cc=ak@linux.intel.com \
    --cc=alexander.shishkin@linux.intel.com \
    --cc=dapeng1.mi@intel.com \
    --cc=eranian@google.com \
    --cc=irogers@google.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-perf-users@vger.kernel.org \
    --cc=mingo@redhat.com \
    --cc=namhyung@kernel.org \
    --cc=peterz@infradead.org \
    --cc=thomas.falcon@intel.com \
    --cc=xudong.hao@intel.com \
    --cc=yi1.lai@intel.com \
    --cc=zide.chen@intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox