perf, x86: Prefer fixed-purpose counters when scheduling
Subsystems (2)
Commit-ID: 4defea8559bc0f97a899d94c8d19d3b8bb802bc4 Gitweb: http://git.kernel.org/tip/4defea8559bc0f97a899d94c8d19d3b8bb802bc4 Author: Peter Zijlstra <[email protected]> AuthorDate: Thu, 10 Nov 2011 15:15:42 +0100 Committer: Ingo Molnar <[email protected]> CommitDate: Tue, 6 Dec 2011 08:33:58 +0100
perf, x86: Prefer fixed-purpose counters when scheduling
This avoids a scheduling failure for cases like:
cycles, cycles, instructions, instructions (on Core2)
Which would end up being programmed like:
PMC0, PMC1, FP-instructions, fail
Because all events will have the same weight.
Signed-off-by Peter Zijlstra <[email protected]>
Signed-off-by Ingo Molnar <[email protected]>
--- arch/x86/kernel/cpu/perf_event.c | 19 ++++++++++++++----- 1 files changed, 14 insertions(+), 5 deletions(-)
arch/x86/kernel/cpu/perf_event.c +14 -5
@@ -574,16 +574,25 @@ static bool __perf_sched_find_counter(struct perf_sched *sched) c = sched->constraints[sched->state.event]; +/* Prefer fixed purpose counters */ +if (x86_pmu.num_counters_fixed) { + idx =X86_PMC_IDX_FIXED ; +for_each_set_bit_cont (idx, c->idxmsk,X86_PMC_IDX_MAX ) { +if (!__test_and_set_bit (idx, sched->state.used)) + goto done; + } + }/* Grab the first unused counter starting with idx */ idx = sched->state.counter; -for_each_set_bit_cont (idx, c->idxmsk,X86_PMC_IDX_MAX ) { +for_each_set_bit_cont (idx, c->idxmsk,X86_PMC_IDX_FIXED ) {if (!__test_and_set_bit (idx, sched->state.used)) -break ; + goto done; } - sched->state.counter = idx; -if (idx >=X86_PMC_IDX_MAX ) -return false; +return false; + +done: + sched->state.counter = idx;if (c->overlap)perf_sched_save_state (sched);