lkml.org 
[lkml]   [2022]   [Sep]   [27]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
From
Date
SubjectRe: [PATCH V2 3/6] perf: Use sample_flags for branch stack
On Thu, Sep 01, 2022 at 06:09:56AM -0700, kan.liang@linux.intel.com wrote:
> From: Kan Liang <kan.liang@linux.intel.com>
>
> Use the new sample_flags to indicate whether the branch stack is filled
> by the PMU driver.
>
> Remove the br_stack from the perf_sample_data_init() to minimize the number
> of cache lines touched.
>
> Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
> ---
> arch/powerpc/perf/core-book3s.c | 1 +
> arch/x86/events/amd/core.c | 4 +++-
> arch/x86/events/core.c | 4 +++-
> arch/x86/events/intel/core.c | 4 +++-
> arch/x86/events/intel/ds.c | 5 ++++-
> include/linux/perf_event.h | 4 ++--
> kernel/events/core.c | 4 ++--
> 7 files changed, 18 insertions(+), 8 deletions(-)
>
> diff --git a/arch/powerpc/perf/core-book3s.c b/arch/powerpc/perf/core-book3s.c
> index 13919eb96931..1ad1efdb33f9 100644
> --- a/arch/powerpc/perf/core-book3s.c
> +++ b/arch/powerpc/perf/core-book3s.c
> @@ -2297,6 +2297,7 @@ static void record_and_restart(struct perf_event *event, unsigned long val,
> cpuhw = this_cpu_ptr(&cpu_hw_events);
> power_pmu_bhrb_read(event, cpuhw);
> data.br_stack = &cpuhw->bhrb_stack;
> + data.sample_flags |= PERF_SAMPLE_BRANCH_STACK;
> }
>
> if (event->attr.sample_type & PERF_SAMPLE_DATA_SRC &&
> diff --git a/arch/x86/events/amd/core.c b/arch/x86/events/amd/core.c
> index 9fbdfbcaf25a..8b70237c33f7 100644
> --- a/arch/x86/events/amd/core.c
> +++ b/arch/x86/events/amd/core.c
> @@ -929,8 +929,10 @@ static int amd_pmu_v2_handle_irq(struct pt_regs *regs)
> if (!x86_perf_event_set_period(event))
> continue;
>
> - if (has_branch_stack(event))
> + if (has_branch_stack(event)) {
> data.br_stack = &cpuc->lbr_stack;
> + data.sample_flags |= PERF_SAMPLE_BRANCH_STACK;
> + }
>
> if (perf_event_overflow(event, &data, regs))
> x86_pmu_stop(event, 0);
> diff --git a/arch/x86/events/core.c b/arch/x86/events/core.c
> index c3d6d139a3c1..b30b8bbcd1e2 100644
> --- a/arch/x86/events/core.c
> +++ b/arch/x86/events/core.c
> @@ -1701,8 +1701,10 @@ int x86_pmu_handle_irq(struct pt_regs *regs)
>
> perf_sample_data_init(&data, 0, event->hw.last_period);
>
> - if (has_branch_stack(event))
> + if (has_branch_stack(event)) {
> data.br_stack = &cpuc->lbr_stack;
> + data.sample_flags |= PERF_SAMPLE_BRANCH_STACK;
> + }
>
> if (perf_event_overflow(event, &data, regs))
> x86_pmu_stop(event, 0);
> diff --git a/arch/x86/events/intel/core.c b/arch/x86/events/intel/core.c
> index 7c1e3d36bc65..b5c02627a155 100644
> --- a/arch/x86/events/intel/core.c
> +++ b/arch/x86/events/intel/core.c
> @@ -3017,8 +3017,10 @@ static int handle_pmi_common(struct pt_regs *regs, u64 status)
>
> perf_sample_data_init(&data, 0, event->hw.last_period);
>
> - if (has_branch_stack(event))
> + if (has_branch_stack(event)) {
> data.br_stack = &cpuc->lbr_stack;
> + data.sample_flags |= PERF_SAMPLE_BRANCH_STACK;
> + }
>
> if (perf_event_overflow(event, &data, regs))
> x86_pmu_stop(event, 0);
> diff --git a/arch/x86/events/intel/ds.c b/arch/x86/events/intel/ds.c
> index 01cbe26225c2..5dcfd2de6ef8 100644
> --- a/arch/x86/events/intel/ds.c
> +++ b/arch/x86/events/intel/ds.c
> @@ -1648,8 +1648,10 @@ static void setup_pebs_fixed_sample_data(struct perf_event *event,
> data->sample_flags |= PERF_SAMPLE_TIME;
> }
>
> - if (has_branch_stack(event))
> + if (has_branch_stack(event)) {
> data->br_stack = &cpuc->lbr_stack;
> + data->sample_flags |= PERF_SAMPLE_BRANCH_STACK;
> + }
> }
>
> static void adaptive_pebs_save_regs(struct pt_regs *regs,
> @@ -1799,6 +1801,7 @@ static void setup_pebs_adaptive_sample_data(struct perf_event *event,
> if (has_branch_stack(event)) {
> intel_pmu_store_pebs_lbrs(lbr);
> data->br_stack = &cpuc->lbr_stack;
> + data->sample_flags |= PERF_SAMPLE_BRANCH_STACK;
> }
> }
>
> diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
> index 0978165a2d87..1e12e79454e0 100644
> --- a/include/linux/perf_event.h
> +++ b/include/linux/perf_event.h
> @@ -1011,7 +1011,6 @@ struct perf_sample_data {
> u64 sample_flags;
> u64 addr;
> struct perf_raw_record *raw;
> - struct perf_branch_stack *br_stack;
> u64 period;
> union perf_sample_weight weight;
> u64 txn;
> @@ -1021,6 +1020,8 @@ struct perf_sample_data {
> * The other fields, optionally {set,used} by
> * perf_{prepare,output}_sample().
> */
> + struct perf_branch_stack *br_stack;
> +
> u64 type;
> u64 ip;
> struct {
> @@ -1061,7 +1062,6 @@ static inline void perf_sample_data_init(struct perf_sample_data *data,
> data->sample_flags = 0;
> data->addr = addr;
> data->raw = NULL;
> - data->br_stack = NULL;

hi,
there's one more place, I'll send full patch for that

jirka


---
diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c
index b05f0310dbd3..98abc6ebb8ea 100644
--- a/kernel/trace/bpf_trace.c
+++ b/kernel/trace/bpf_trace.c
@@ -1687,6 +1687,9 @@ BPF_CALL_4(bpf_read_branch_records, struct bpf_perf_event_data_kern *, ctx,
if (unlikely(flags & ~BPF_F_GET_BRANCH_RECORDS_SIZE))
return -EINVAL;

+ if (unlikely(!(ctx->data->sample_flags & PERF_SAMPLE_BRANCH_STACK)))
+ return -ENOENT;
+
if (unlikely(!br_stack))
return -ENOENT;

\
 
 \ /
  Last update: 2022-09-27 13:07    [W:3.982 / U:0.712 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site