From: Brian Geffon <bgeffon@google.com>
To: Peter Zijlstra <peterz@infradead.org>, Ingo Molnar <mingo@redhat.com>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>,
linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org,
Adrian Hunter <adrian.hunter@intel.com>,
Jiri Olsa <jolsa@kernel.org>, Namhyung Kim <namhyung@kernel.org>,
Steven Rostedt <rostedt@goodmis.org>,
Brian Geffon <bgeffon@google.com>
Subject: [PATCH] perf tools: sched-pipe bench: add (-n) nonblocking benchmark
Date: Wed, 16 Oct 2024 15:00:09 -0400 [thread overview]
Message-ID: <20241016190009.866615-1-bgeffon@google.com> (raw)
The -n mode will benchmark pipes in a non-blocking mode using
epoll_wait.
This specific mode was added to demonstrate the broken sync nature
of epoll: https://lore.kernel.org/lkml/20240426-zupfen-jahrzehnt-5be786bcdf04@brauner
Signed-off-by: Brian Geffon <bgeffon@google.com>
---
tools/perf/bench/sched-pipe.c | 43 +++++++++++++++++++++++++++++------
1 file changed, 36 insertions(+), 7 deletions(-)
diff --git a/tools/perf/bench/sched-pipe.c b/tools/perf/bench/sched-pipe.c
index 3af6d3c55aba..e2562677df96 100644
--- a/tools/perf/bench/sched-pipe.c
+++ b/tools/perf/bench/sched-pipe.c
@@ -23,6 +23,7 @@
#include <errno.h>
#include <fcntl.h>
#include <assert.h>
+#include <sys/epoll.h>
#include <sys/time.h>
#include <sys/types.h>
#include <sys/syscall.h>
@@ -34,6 +35,8 @@ struct thread_data {
int nr;
int pipe_read;
int pipe_write;
+ struct epoll_event epoll_ev;
+ int epoll_fd;
bool cgroup_failed;
pthread_t pthread;
};
@@ -44,6 +47,7 @@ static int loops = LOOPS_DEFAULT;
/* Use processes by default: */
static bool threaded;
+static bool nonblocking;
static char *cgrp_names[2];
static struct cgroup *cgrps[2];
@@ -81,6 +85,7 @@ static int parse_two_cgroups(const struct option *opt __maybe_unused,
}
static const struct option options[] = {
+ OPT_BOOLEAN('n', "nonblocking", &nonblocking, "Use non-blocking operations"),
OPT_INTEGER('l', "loop", &loops, "Specify number of loops"),
OPT_BOOLEAN('T', "threaded", &threaded, "Specify threads/process based task setup"),
OPT_CALLBACK('G', "cgroups", NULL, "SEND,RECV",
@@ -165,11 +170,25 @@ static void exit_cgroup(int nr)
free(cgrp_names[nr]);
}
+static inline int read_pipe(struct thread_data *td)
+{
+ int ret, m;
+retry:
+ if (nonblocking) {
+ ret = epoll_wait(td->epoll_fd, &td->epoll_ev, 1, -1);
+ if (ret < 0)
+ return ret;
+ }
+ ret = read(td->pipe_read, &m, sizeof(int));
+ if (nonblocking && ret < 0 && errno == EWOULDBLOCK)
+ goto retry;
+ return ret;
+}
+
static void *worker_thread(void *__tdata)
{
struct thread_data *td = __tdata;
- int m = 0, i;
- int ret;
+ int i, ret, m = 0;
ret = enter_cgroup(td->nr);
if (ret < 0) {
@@ -177,16 +196,23 @@ static void *worker_thread(void *__tdata)
return NULL;
}
+ if (nonblocking) {
+ td->epoll_ev.events = EPOLLIN;
+ td->epoll_fd = epoll_create(1);
+ BUG_ON(td->epoll_fd < 0);
+ BUG_ON(epoll_ctl(td->epoll_fd, EPOLL_CTL_ADD, td->pipe_read, &td->epoll_ev) < 0);
+ }
+
for (i = 0; i < loops; i++) {
if (!td->nr) {
- ret = read(td->pipe_read, &m, sizeof(int));
+ ret = read_pipe(td);
BUG_ON(ret != sizeof(int));
ret = write(td->pipe_write, &m, sizeof(int));
BUG_ON(ret != sizeof(int));
} else {
ret = write(td->pipe_write, &m, sizeof(int));
BUG_ON(ret != sizeof(int));
- ret = read(td->pipe_read, &m, sizeof(int));
+ ret = read_pipe(td);
BUG_ON(ret != sizeof(int));
}
}
@@ -209,13 +235,16 @@ int bench_sched_pipe(int argc, const char **argv)
* discarding returned value of read(), write()
* causes error in building environment for perf
*/
- int __maybe_unused ret, wait_stat;
+ int __maybe_unused ret, wait_stat, flags = 0;
pid_t pid, retpid __maybe_unused;
argc = parse_options(argc, argv, options, bench_sched_pipe_usage, 0);
- BUG_ON(pipe(pipe_1));
- BUG_ON(pipe(pipe_2));
+ if (nonblocking)
+ flags |= O_NONBLOCK;
+
+ BUG_ON(pipe2(pipe_1, flags));
+ BUG_ON(pipe2(pipe_2, flags));
gettimeofday(&start, NULL);
--
2.47.0.rc1.288.g06298d1525-goog
next reply other threads:[~2024-10-16 19:00 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-10-16 19:00 Brian Geffon [this message]
2024-10-16 19:27 ` [PATCH] perf tools: sched-pipe bench: add (-n) nonblocking benchmark Ian Rogers
2024-10-23 18:00 ` Namhyung Kim
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20241016190009.866615-1-bgeffon@google.com \
--to=bgeffon@google.com \
--cc=acme@kernel.org \
--cc=adrian.hunter@intel.com \
--cc=jolsa@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-perf-users@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=namhyung@kernel.org \
--cc=peterz@infradead.org \
--cc=rostedt@goodmis.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox