Skip to contentSkip to conversation

perf/core: fix restoring of Intel LBR call stack on a context switch

PATCH Superseded Alexey Budankov 2 participants 3 messages Oct 14, 2019 – Oct 15, 2019 perf events

v1 of 5 Superseded Latest is v5 →

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 perf/core: fix restoring of Intel LBR call stack on a context switch

Restore Intel LBR call stack from cloned inactive task perf context on a context switch. This change inherently addresses inconsistency in LBR call stack data provided on a sample in record profiling mode for example like this:

  $ perf record -N -B -T -R --call-graph lbr \
         -e cpu/period=0xcdfe60,event=0x3c,name=\'CPU_CLK_UNHALTED.THREAD\'/Duk \
         --clockid=monotonic_raw -- ./miniFE.x nx 25 ny 25 nz 25

Let's assume threads A, B, C belonging to the same process.
B and C are siblings of A and their perf contexts are treated as equivalent. At some point B blocks on a futex (non preempt context switch). B's LBRs are preserved at B's perf context task_ctx_data and B's events are removed from PMU and disabled. B's perf context becomes inactive.

Later C gets on a cpu, runs, gets profiled and eventually switches to the awaken but not yet running B. The optimized context switch path is executed coping B's task_ctx_data to C's one and updating B's perf context pointer to refer to C's task_ctx_data that contains preserved B's LBRs after coping.

However, as far B's perf context is inactive there is no enabled events in there and B's task_ctx_data->lbr_callstack_users is equal to 0. When B gets on the cpu B's events reviving is skipped following the optimized context switch path and B's task_ctx_data->lbr_callstack_users remains 0. Thus B's LBR's are not restored by pmu sched_task() code called in the end of perf context switch sched_in callback for B.

In the report that manifests as having short fragments of B's call stack, still tracked by LBR's HW between adjacent samples, but the whole thread call tree doesn't aggregate.

The fix has been evaluated when profiling miniFE [1] (C++, OpenMP) workload running 64 threads on Intel Skylake EP(64 core, 2 sockets):

  $ perf report --call-graph callee,flat

5.3.0-rc6+ (tip perf/core) - fixed

- 92.66% 82.64% miniFE.x libiomp5.so [.] _INTERNAL_25_______src_kmp_barrier_cpp_1d20fae8::__kmp_hyper_barrier_release
- 69.14% _INTERNAL_25_______src_kmp_barrier_cpp_1d20fae8::__kmp_hyper_barrier_release

        __kmp_fork_barrier
        __kmp_launch_thread
        _INTERNAL_24_______src_z_Linux_util_c_3e0095e6::__kmp_launch_worker
        start_thread
        __clone
   - 21.89% _INTERNAL_25_______src_kmp_barrier_cpp_1d20fae8::__kmp_hyper_barrier_release
        __kmp_barrier
        __kmpc_reduce_nowait
        miniFE::cg_solve<miniFE::CSRMatrix<double, int, int>, miniFE::Vector<double, int, int>, miniFE::matvec_std<miniFE::CSRMatrix<double, int, int>, miniFE::Vector<double, int, in
        __kmp_invoke_microtask
        __kmp_invoke_task_func
        __kmp_launch_thread
        _INTERNAL_24_______src_z_Linux_util_c_3e0095e6::__kmp_launch_worker
        start_thread
        __clone
   - 1.63% _INTERNAL_25_______src_kmp_barrier_cpp_1d20fae8::__kmp_hyper_barrier_release
        __kmp_barrier
        __kmpc_reduce_nowait
        main
        __kmp_invoke_microtask
        __kmp_invoke_task_func
        __kmp_launch_thread
        _INTERNAL_24_______src_z_Linux_util_c_3e0095e6::__kmp_launch_worker
        start_thread
        __clone

5.0.13-300.fc30.x86_64 - no fix

- 90.29% 81.01% miniFE.x libiomp5.so [.] _INTERNAL_25_______src_kmp_barrier_cpp_1d20fae8::__kmp_hyper_barrier_release
- 33.45% _INTERNAL_25_______src_kmp_barrier_cpp_1d20fae8::__kmp_hyper_barrier_release

        __kmp_fork_barrier
        __kmp_launch_thread
        _INTERNAL_24_______src_z_Linux_util_c_3e0095e6::__kmp_launch_worker
        start_thread
        __clone
     87.63% _INTERNAL_25_______src_kmp_barrier_cpp_1d20fae8::__kmp_hyper_barrier_release
   - 54.79% _INTERNAL_25_______src_kmp_barrier_cpp_1d20fae8::__kmp_hyper_barrier_release
        __kmp_fork_barrier
        __kmp_launch_thread
   - 9.18% _INTERNAL_25_______src_kmp_barrier_cpp_1d20fae8::__kmp_hyper_barrier_release
        __kmp_barrier
        __kmpc_reduce_nowait
        miniFE::cg_solve<miniFE::CSRMatrix<double, int, int>, miniFE::Vector<double, int, int>, miniFE::matvec_std<miniFE::CSRMatrix<double, int, int>, miniFE::Vector<double, int, in
        __kmp_invoke_microtask
        __kmp_invoke_task_func
        __kmp_launch_thread
        _INTERNAL_24_______src_z_Linux_util_c_3e0095e6::__kmp_launch_worker
        start_thread
        __clone
   - 41.28% _INTERNAL_25_______src_kmp_barrier_cpp_1d20fae8::__kmp_hyper_barrier_release
        __kmp_fork_barrier
        __kmp_launch_thread
        _INTERNAL_24_______src_z_Linux_util_c_3e0095e6::__kmp_launch_worker
   - 15.77% _INTERNAL_25_______src_kmp_barrier_cpp_1d20fae8::__kmp_hyper_barrier_release
        __kmp_barrier
        __kmpc_reduce_nowait
        miniFE::cg_solve<miniFE::CSRMatrix<double, int, int>, miniFE::Vector<double, int, int>, miniFE::matvec_std<miniFE::CSRMatrix<double, int, int>, miniFE::Vector<double, int, in
        __kmp_invoke_microtask
        __kmp_invoke_task_func
        __kmp_launch_thread
   - 11.56% _INTERNAL_25_______src_kmp_barrier_cpp_1d20fae8::__kmp_hyper_barrier_release
        __kmp_barrier
        __kmpc_reduce_nowait
        miniFE::cg_solve<miniFE::CSRMatrix<double, int, int>, miniFE::Vector<double, int, int>, miniFE::matvec_std<miniFE::CSRMatrix<double, int, int>, miniFE::Vector<double, int, in
        __kmp_invoke_microtask
        __kmp_invoke_task_func
        __kmp_launch_thread
        _INTERNAL_24_______src_z_Linux_util_c_3e0095e6::__kmp_launch_worker
   - 2.33% _INTERNAL_25_______src_kmp_barrier_cpp_1d20fae8::__kmp_hyper_barrier_release
        __kmp_barrier
        __kmpc_reduce_nowait
        main
        __kmp_invoke_microtask
        __kmp_invoke_task_func
        __kmp_launch_thread
        _INTERNAL_24_______src_z_Linux_util_c_3e0095e6::__kmp_launch_worker
        start_thread
        __clone
     0.67% _INTERNAL_25_______src_kmp_barrier_cpp_1d20fae8::__kmp_hyper_barrier_gather
     0.57% __kmp_hardware_timestamp

[1] https://www.hpcadvisorycouncil.com/pdf/miniFE_Analysis_and_Profiling.pdf

---
 kernel/events/core.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
kernel/events/core.c +1 -1
index 2aad959e6def..74c2ff38e079 100644
@@ -3181,7 +3181,7 @@ static void perf_event_context_sched_out(struct task_struct *task, int ctxn,
 
 	rcu_read_lock();
 	next_ctx = next->perf_event_ctxp[ctxn];
-	if (!next_ctx)
+	if (!next_ctx || !next_ctx->is_active)
 		goto unlock;
 
 	parent = rcu_dereference(ctx->parent_ctx);
Signature · 1 line
2.20.1
maintainer
#1 lore raw
Quoted 44 lines · Alexey Budankov

Restore Intel LBR call stack from cloned inactive task perf context on a context switch. This change inherently addresses inconsistency in LBR call stack data provided on a sample in record profiling mode for example like this:

  $ perf record -N -B -T -R --call-graph lbr \
         -e cpu/period=0xcdfe60,event=0x3c,name=\'CPU_CLK_UNHALTED.THREAD\'/Duk \
         --clockid=monotonic_raw -- ./miniFE.x nx 25 ny 25 nz 25

Let's assume threads A, B, C belonging to the same process.
B and C are siblings of A and their perf contexts are treated as equivalent. At some point B blocks on a futex (non preempt context switch). B's LBRs are preserved at B's perf context task_ctx_data and B's events are removed from PMU and disabled. B's perf context becomes inactive.

Later C gets on a cpu, runs, gets profiled and eventually switches to the awaken but not yet running B. The optimized context switch path is executed coping B's task_ctx_data to C's one and updating B's perf context pointer to refer to C's task_ctx_data that contains preserved B's LBRs after coping.

However, as far B's perf context is inactive there is no enabled events in there and B's task_ctx_data->lbr_callstack_users is equal to 0. When B gets on the cpu B's events reviving is skipped following the optimized context switch path and B's task_ctx_data->lbr_callstack_users remains 0. Thus B's LBR's are not restored by pmu sched_task() code called in the end of perf context switch sched_in callback for B.

In the report that manifests as having short fragments of B's call stack, still tracked by LBR's HW between adjacent samples, but the whole thread call tree doesn't aggregate.

---
 kernel/events/core.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
kernel/events/core.c +0 -0
index 2aad959e6def..74c2ff38e079 100644

kernel/events/core.c static void perf_event_context_sched_out(struct task_struct *task, int ctxn, · line 3185

Quoted 6 lines
kernel/events/core.c +1 -1
@@ -3181,7 +3181,7 @@ static void perf_event_context_sched_out(struct task_struct *task, int ctxn,
 
 	rcu_read_lock();
 	next_ctx = next->perf_event_ctxp[ctxn];
-	if (!next_ctx)
+	if (!next_ctx || !next_ctx->is_active)
 		goto unlock;

AFAICT this completely kills off the optimization. next_ctx->is_active cannot be set at this point.

#2 lore raw
Quoted 46 lines · Peter Zijlstra
Quoted 45 lines · Alexey Budankov

Restore Intel LBR call stack from cloned inactive task perf context on a context switch. This change inherently addresses inconsistency in LBR call stack data provided on a sample in record profiling mode for example like this:

  $ perf record -N -B -T -R --call-graph lbr \
         -e cpu/period=0xcdfe60,event=0x3c,name=\'CPU_CLK_UNHALTED.THREAD\'/Duk \
         --clockid=monotonic_raw -- ./miniFE.x nx 25 ny 25 nz 25

Let's assume threads A, B, C belonging to the same process.
B and C are siblings of A and their perf contexts are treated as equivalent. At some point B blocks on a futex (non preempt context switch). B's LBRs are preserved at B's perf context task_ctx_data and B's events are removed from PMU and disabled. B's perf context becomes inactive.

Later C gets on a cpu, runs, gets profiled and eventually switches to the awaken but not yet running B. The optimized context switch path is executed coping B's task_ctx_data to C's one and updating B's perf context pointer to refer to C's task_ctx_data that contains preserved B's LBRs after coping.

However, as far B's perf context is inactive there is no enabled events in there and B's task_ctx_data->lbr_callstack_users is equal to 0. When B gets on the cpu B's events reviving is skipped following the optimized context switch path and B's task_ctx_data->lbr_callstack_users remains 0. Thus B's LBR's are not restored by pmu sched_task() code called in the end of perf context switch sched_in callback for B.

In the report that manifests as having short fragments of B's call stack, still tracked by LBR's HW between adjacent samples, but the whole thread call tree doesn't aggregate.

---
 kernel/events/core.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
kernel/events/core.c +0 -0
index 2aad959e6def..74c2ff38e079 100644
@@ -3181,7 +3181,7 @@ static void perf_event_context_sched_out(struct task_struct *task, int ctxn,
 
Quoted 8 lines

kernel/events/core.c static void perf_event_context_sched_out(struct task_struct *task, int ctxn, · line 3185

Quoted 5 lines
kernel/events/core.c +1 -1
 	rcu_read_lock();
 	next_ctx = next->perf_event_ctxp[ctxn];
-	if (!next_ctx)
+	if (!next_ctx || !next_ctx->is_active)
 		goto unlock;

AFAICT this completely kills off the optimization. next_ctx->is_active cannot be set at this point.

Hmm, the intention was to skip optimization path only once when switching to just resumed thread. Thanks for observation.

~Alexey