Patch "perf/x86: Enable free running PEBS for REGS_USER/INTR" has been added to the 4.14-stable tree
Subsystems (3)
This is a note to let you know that I've just added the patch titled
perf/x86: Enable free running PEBS for REGS_USER/INTR
to the 4.14-stable tree which can be found at:
http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary
The filename of the patch is:
perf-x86-enable-free-running-pebs-for-regs_user-intr.patch
and it can be found in the queue-4.14 subdirectory.
If you, or anyone else, feels it should not be added to the stable tree, please let <[email protected]> know about it.
Quoted 1 line
From
2fe1bc1f501d55e5925b4035bcd85781adc76c63Mon Sep 17 00:00:00 2001
From: Andi Kleen <[email protected]>
Date: Thu, 31 Aug 2017 14:46:30 -0700
Subject: perf/x86: Enable free running PEBS for REGS_USER/INTR
From: Andi Kleen <[email protected]>
commit 2fe1bc1f501d55e5925b4035bcd85781adc76c63 upstream.
[ Note, this is a Git cherry-pick of the following commit:
a47ba4d77e12 ("perf/x86: Enable free running PEBS for REGS_USER/INTR")
... for easier x86 PTI code testing and back-porting. ]
Currently free running PEBS is disabled when user or interrupt registers are requested. Most of the registers are actually available in the PEBS record and can be supported.
So we just need to check for the supported registers and then allow it: it is all except for the segment register.
For user registers this only works when the counter is limited to ring 3 only, so this also needs to be checked.
--- arch/x86/events/intel/core.c | 4 ++++ arch/x86/events/perf_event.h | 24 +++++++++++++++++++++++- 2 files changed, 27 insertions(+), 1 deletion(-)
arch/x86/events/intel/core.c +4 -0
@@ -2958,6 +2958,10 @@ static unsigned long intel_pmu_free_runnif (event->attr.use_clockid) flags &= ~PERF_SAMPLE_TIME ; +if (!event->attr.exclude_kernel) + flags &= ~PERF_SAMPLE_REGS_USER ; +if (event->attr.sample_regs_user & ~PEBS_REGS ) + flags &= ~(PERF_SAMPLE_REGS_USER |PERF_SAMPLE_REGS_INTR );return flags; }
arch/x86/events/perf_event.h +23 -1
@@ -85,13 +85,15 @@ struct amd_nb {* Flags PEBS can handle without an PMI. * * TID can only be handled by flushing at context switch. +* REGS_USER can be handled for events limited to ring 3. * */ #define PEBS_FREERUNNING_FLAGS \ (PERF_SAMPLE_IP | PERF_SAMPLE_TID | PERF_SAMPLE_ADDR | \ PERF_SAMPLE_ID | PERF_SAMPLE_CPU | PERF_SAMPLE_STREAM_ID | \ PERF_SAMPLE_DATA_SRC | PERF_SAMPLE_IDENTIFIER | \ - PERF_SAMPLE_TRANSACTION | PERF_SAMPLE_PHYS_ADDR) + PERF_SAMPLE_TRANSACTION | PERF_SAMPLE_PHYS_ADDR | \ + PERF_SAMPLE_REGS_INTR | PERF_SAMPLE_REGS_USER)/* * A debug store configuration.
@@ -110,6 +112,26 @@ struct debug_store {u64 pebs_event_reset[MAX_PEBS_EVENTS ]; }; +#define PEBS_REGS \ + (PERF_REG_X86_AX | \ + PERF_REG_X86_BX | \ + PERF_REG_X86_CX | \ + PERF_REG_X86_DX | \ + PERF_REG_X86_DI | \ + PERF_REG_X86_SI | \ + PERF_REG_X86_SP | \ + PERF_REG_X86_BP | \ + PERF_REG_X86_IP | \ + PERF_REG_X86_FLAGS | \ + PERF_REG_X86_R8 | \ + PERF_REG_X86_R9 | \ + PERF_REG_X86_R10 | \ + PERF_REG_X86_R11 | \ + PERF_REG_X86_R12 | \ + PERF_REG_X86_R13 | \ + PERF_REG_X86_R14 | \ + PERF_REG_X86_R15) +/* * Per register state. */
Patches currently in stable-queue which might be from [email protected] are
queue-4.14/bitops-add-clear-set_bit32-to-linux-bitops.h.patch queue-4.14/bitops-revert-cbe96375025e-bitops-add-clear-set_bit32-to-linux-bitops.h.patch queue-4.14/x86-fpu-remove-the-explicit-clearing-of-xsave-dependent-features.patch queue-4.14/x86-cpuid-add-generic-table-for-cpuid-dependencies.patch queue-4.14/x86-cpuid-prevent-out-of-bound-access-in-do_clear_cpu_cap.patch queue-4.14/x86-fpu-parse-clearcpuid-as-early-xsave-argument.patch queue-4.14/x86-fpu-make-xsave-check-the-base-cpuid-features-before-enabling.patch queue-4.14/x86-cpuid-replace-set-clear_bit32.patch queue-4.14/perf-x86-enable-free-running-pebs-for-regs_user-intr.patch