Re: [PATCH v3 2/2]: perf record: enable asynchronous trace writing

From: Jiri Olsa
Date: Tue Aug 28 2018 - 05:01:50 EST


On Mon, Aug 27, 2018 at 09:16:55PM +0300, Alexey Budankov wrote:

SNIP

> +static int record__mmap_read_sync(int trace_fd, struct aiocb **cblocks,
> + int cblocks_size, struct record *rec)
> +{
> + size_t rem;
> + ssize_t size;
> + off_t rem_off;
> + int i, aio_ret, aio_errno, do_suspend;
> + struct perf_mmap *md;
> + struct timespec timeout0 = { 0, 0 };
> + struct timespec timeoutS = { 0, 1000 * 500 * 1 }; // 0.5ms
> +
> + if (!cblocks_size)
> + return 0;
> +
> + do {
> + do_suspend = 0;
> + /* aio_suspend() implementation inside glibc (as of v2.27) is
> + * intrusive and not just blocks waiting io requests completion
> + * but polls requests queue inducing context switches in perf
> + * tool process. When profiling in system wide mode with tracing
> + * context switches the trace may be polluted by context switches
> + * from the perf process and the trace size becomes about 3-5
> + * times bigger than that of when writing the trace serially.
> + * To limit the volume of context switches from perf tool
> + * process nanosleep() call below is added prior aio_suspend()
> + * calling till every half of the kernel timer tick which is
> + * usually 1ms (depends on CONFIG_HZ value).

nice, any idea this is intentional for some reason?

was there some impact on the overall performance or
this is purely for the sanity of the trace size?

thanks,
jirka