From: tip-bot for Kan Liang <tipbot@zytor.com>
To: linux-tip-commits@vger.kernel.org
Cc: eranian@google.com, mingo@kernel.org,
torvalds@linux-foundation.org, acme@redhat.com, jolsa@redhat.com,
kan.liang@linux.intel.com, tglx@linutronix.de, hpa@zytor.com,
vincent.weaver@maine.edu, alexander.shishkin@linux.intel.com,
linux-kernel@vger.kernel.org, peterz@infradead.org
Subject: [tip:perf/urgent] perf/x86: Disable extended registers for non-supported PMUs
Date: Tue, 25 Jun 2019 01:19:55 -0700 [thread overview]
Message-ID: <tip-e321d02db87af7840da29ef833a2a71fc0eab198@git.kernel.org> (raw)
In-Reply-To: <1559081314-9714-1-git-send-email-kan.liang@linux.intel.com>
Commit-ID: e321d02db87af7840da29ef833a2a71fc0eab198
Gitweb: https://git.kernel.org/tip/e321d02db87af7840da29ef833a2a71fc0eab198
Author: Kan Liang <kan.liang@linux.intel.com>
AuthorDate: Tue, 28 May 2019 15:08:30 -0700
Committer: Ingo Molnar <mingo@kernel.org>
CommitDate: Mon, 24 Jun 2019 19:19:23 +0200
perf/x86: Disable extended registers for non-supported PMUs
The perf fuzzer caused Skylake machine to crash:
[ 9680.085831] Call Trace:
[ 9680.088301] <IRQ>
[ 9680.090363] perf_output_sample_regs+0x43/0xa0
[ 9680.094928] perf_output_sample+0x3aa/0x7a0
[ 9680.099181] perf_event_output_forward+0x53/0x80
[ 9680.103917] __perf_event_overflow+0x52/0xf0
[ 9680.108266] ? perf_trace_run_bpf_submit+0xc0/0xc0
[ 9680.113108] perf_swevent_hrtimer+0xe2/0x150
[ 9680.117475] ? check_preempt_wakeup+0x181/0x230
[ 9680.122091] ? check_preempt_curr+0x62/0x90
[ 9680.126361] ? ttwu_do_wakeup+0x19/0x140
[ 9680.130355] ? try_to_wake_up+0x54/0x460
[ 9680.134366] ? reweight_entity+0x15b/0x1a0
[ 9680.138559] ? __queue_work+0x103/0x3f0
[ 9680.142472] ? update_dl_rq_load_avg+0x1cd/0x270
[ 9680.147194] ? timerqueue_del+0x1e/0x40
[ 9680.151092] ? __remove_hrtimer+0x35/0x70
[ 9680.155191] __hrtimer_run_queues+0x100/0x280
[ 9680.159658] hrtimer_interrupt+0x100/0x220
[ 9680.163835] smp_apic_timer_interrupt+0x6a/0x140
[ 9680.168555] apic_timer_interrupt+0xf/0x20
[ 9680.172756] </IRQ>
The XMM registers can only be collected by PEBS hardware events on the
platforms with PEBS baseline support, e.g. Icelake, not software/probe
events.
Add capabilities flag PERF_PMU_CAP_EXTENDED_REGS to indicate the PMU
which support extended registers. For X86, the extended registers are
XMM registers.
Add has_extended_regs() to check if extended registers are applied.
The generic code define the mask of extended registers as 0 if arch
headers haven't overridden it.
Originally-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reported-by: Vince Weaver <vincent.weaver@maine.edu>
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: 878068ea270e ("perf/x86: Support outputting XMM registers")
Link: https://lkml.kernel.org/r/1559081314-9714-1-git-send-email-kan.liang@linux.intel.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
---
arch/x86/events/intel/ds.c | 1 +
arch/x86/include/uapi/asm/perf_regs.h | 3 +++
include/linux/perf_event.h | 1 +
include/linux/perf_regs.h | 8 ++++++++
kernel/events/core.c | 18 ++++++++++++++----
5 files changed, 27 insertions(+), 4 deletions(-)
diff --git a/arch/x86/events/intel/ds.c b/arch/x86/events/intel/ds.c
index 7acc526b4ad2..6cb38ab02c8a 100644
--- a/arch/x86/events/intel/ds.c
+++ b/arch/x86/events/intel/ds.c
@@ -2020,6 +2020,7 @@ void __init intel_ds_init(void)
PERF_SAMPLE_TIME;
x86_pmu.flags |= PMU_FL_PEBS_ALL;
pebs_qual = "-baseline";
+ x86_get_pmu()->capabilities |= PERF_PMU_CAP_EXTENDED_REGS;
} else {
/* Only basic record supported */
x86_pmu.pebs_no_xmm_regs = 1;
diff --git a/arch/x86/include/uapi/asm/perf_regs.h b/arch/x86/include/uapi/asm/perf_regs.h
index ac67bbea10ca..7c9d2bb3833b 100644
--- a/arch/x86/include/uapi/asm/perf_regs.h
+++ b/arch/x86/include/uapi/asm/perf_regs.h
@@ -52,4 +52,7 @@ enum perf_event_x86_regs {
/* These include both GPRs and XMMX registers */
PERF_REG_X86_XMM_MAX = PERF_REG_X86_XMM15 + 2,
};
+
+#define PERF_REG_EXTENDED_MASK (~((1ULL << PERF_REG_X86_XMM0) - 1))
+
#endif /* _ASM_X86_PERF_REGS_H */
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 0ab99c7b652d..2bca72f3028b 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -241,6 +241,7 @@ struct perf_event;
#define PERF_PMU_CAP_NO_INTERRUPT 0x01
#define PERF_PMU_CAP_NO_NMI 0x02
#define PERF_PMU_CAP_AUX_NO_SG 0x04
+#define PERF_PMU_CAP_EXTENDED_REGS 0x08
#define PERF_PMU_CAP_EXCLUSIVE 0x10
#define PERF_PMU_CAP_ITRACE 0x20
#define PERF_PMU_CAP_HETEROGENEOUS_CPUS 0x40
diff --git a/include/linux/perf_regs.h b/include/linux/perf_regs.h
index 476747456bca..2d12e97d5e7b 100644
--- a/include/linux/perf_regs.h
+++ b/include/linux/perf_regs.h
@@ -11,6 +11,11 @@ struct perf_regs {
#ifdef CONFIG_HAVE_PERF_REGS
#include <asm/perf_regs.h>
+
+#ifndef PERF_REG_EXTENDED_MASK
+#define PERF_REG_EXTENDED_MASK 0
+#endif
+
u64 perf_reg_value(struct pt_regs *regs, int idx);
int perf_reg_validate(u64 mask);
u64 perf_reg_abi(struct task_struct *task);
@@ -18,6 +23,9 @@ void perf_get_regs_user(struct perf_regs *regs_user,
struct pt_regs *regs,
struct pt_regs *regs_user_copy);
#else
+
+#define PERF_REG_EXTENDED_MASK 0
+
static inline u64 perf_reg_value(struct pt_regs *regs, int idx)
{
return 0;
diff --git a/kernel/events/core.c b/kernel/events/core.c
index 8d1c62df20a7..f85929ce13be 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -10036,6 +10036,12 @@ void perf_pmu_unregister(struct pmu *pmu)
}
EXPORT_SYMBOL_GPL(perf_pmu_unregister);
+static inline bool has_extended_regs(struct perf_event *event)
+{
+ return (event->attr.sample_regs_user & PERF_REG_EXTENDED_MASK) ||
+ (event->attr.sample_regs_intr & PERF_REG_EXTENDED_MASK);
+}
+
static int perf_try_init_event(struct pmu *pmu, struct perf_event *event)
{
struct perf_event_context *ctx = NULL;
@@ -10067,12 +10073,16 @@ static int perf_try_init_event(struct pmu *pmu, struct perf_event *event)
perf_event_ctx_unlock(event->group_leader, ctx);
if (!ret) {
+ if (!(pmu->capabilities & PERF_PMU_CAP_EXTENDED_REGS) &&
+ has_extended_regs(event))
+ ret = -EOPNOTSUPP;
+
if (pmu->capabilities & PERF_PMU_CAP_NO_EXCLUDE &&
- event_has_any_exclude_flag(event)) {
- if (event->destroy)
- event->destroy(event);
+ event_has_any_exclude_flag(event))
ret = -EINVAL;
- }
+
+ if (ret && event->destroy)
+ event->destroy(event);
}
if (ret)
prev parent reply other threads:[~2019-06-25 8:21 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-05-28 22:08 [PATCH V3 1/5] perf: Disable extended registers for non-support PMUs kan.liang
2019-05-28 22:08 ` [PATCH V3 2/5] perf/x86/regs: Check reserved bits kan.liang
2019-06-25 8:20 ` [tip:perf/urgent] " tip-bot for Kan Liang
2019-05-28 22:08 ` [PATCH V3 3/5] perf/x86: Clean up PEBS_XMM_REGS kan.liang
2019-06-25 8:21 ` [tip:perf/urgent] " tip-bot for Kan Liang
2019-05-28 22:08 ` [PATCH V3 4/5] perf/x86: Clean up pebs_no_xmm_regs kan.liang
2019-06-25 8:22 ` [tip:perf/urgent] perf/x86: Remove pmu->pebs_no_xmm_regs tip-bot for Kan Liang
2019-05-28 22:08 ` [PATCH V3 5/5] perf regs x86: Use PERF_REG_EXTENDED_MASK kan.liang
2019-06-25 8:22 ` [tip:perf/urgent] perf/x86/regs: " tip-bot for Kan Liang
2019-06-20 12:24 ` [PATCH V3 1/5] perf: Disable extended registers for non-support PMUs Peter Zijlstra
2019-06-25 8:19 ` tip-bot for Kan Liang [this message]
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=tip-e321d02db87af7840da29ef833a2a71fc0eab198@git.kernel.org \
--to=tipbot@zytor.com \
--cc=acme@redhat.com \
--cc=alexander.shishkin@linux.intel.com \
--cc=eranian@google.com \
--cc=hpa@zytor.com \
--cc=jolsa@redhat.com \
--cc=kan.liang@linux.intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-tip-commits@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=peterz@infradead.org \
--cc=tglx@linutronix.de \
--cc=torvalds@linux-foundation.org \
--cc=vincent.weaver@maine.edu \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.