perf sample: Make user_regs and intr_regs optional

The struct dump_regs contains 512 bytes of cache_regs, meaning the two
values in perf_sample contribute 1088 bytes of its total 1384 bytes
size. Initializing this much memory has a cost reported by Tavian
Barnes <tavianator@tavianator.com> as about 2.5% when running `perf
script --itrace=i0`:
https://lore.kernel.org/lkml/d841b97b3ad2ca8bcab07e4293375fb7c32dfce7.1736618095.git.tavianator@tavianator.com/

Adrian Hunter <adrian.hunter@intel.com> replied that the zero
initialization was necessary and couldn't simply be removed.

This patch aims to strike a middle ground of still zeroing the
perf_sample, but removing 79% of its size by make user_regs and
intr_regs optional pointers to zalloc-ed memory. To support the
allocation accessors are created for user_regs and intr_regs. To
support correct cleanup perf_sample__init and perf_sample__exit
functions are created and added throughout the code base.

Signed-off-by: Ian Rogers <irogers@google.com>
Link: https://lore.kernel.org/r/20250113194345.1537821-1-irogers@google.com
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
This commit is contained in:
Ian Rogers
2025-01-13 11:43:45 -08:00
committed by Namhyung Kim
parent 08d9e88348
commit dc6d2bc2d8
34 changed files with 448 additions and 191 deletions

View File

@@ -479,19 +479,25 @@ static int process_sample_event(struct machine *machine,
struct thread *thread;
int ret;
if (evlist__parse_sample(evlist, event, &sample)) {
perf_sample__init(&sample, /*all=*/false);
ret = evlist__parse_sample(evlist, event, &sample);
if (ret) {
pr_debug("evlist__parse_sample failed\n");
return -1;
ret = -1;
goto out;
}
thread = machine__findnew_thread(machine, sample.pid, sample.tid);
if (!thread) {
pr_debug("machine__findnew_thread failed\n");
return -1;
ret = -1;
goto out;
}
ret = read_object_code(sample.ip, READLEN, sample.cpumode, thread, state);
thread__put(thread);
out:
perf_sample__exit(&sample);
return ret;
}