yes: use tee syscall as pipe only fast-path by oech3 · Pull Request #11458 · uutils/coreutils

oech3 · 2026-03-23T00:00:37Z

Add fast-path only valid for pipe output:
tee is unstable compared with vmsplice (about 28~35GiB) but better than sendfile.

~~sendfile code-path is still needed for the case GNU added -z for fast zero generator.~~

Closes #11454 .
I don't know how to align pipe by 1 MiB.

codspeed-hq · 2026-03-23T00:18:05Z

Merging this PR will not alter performance

✅ 10 untouched benchmarks
⏩ 338 skipped benchmarks¹

_{Comparing oech3:yes-tee (a3a747c) with main (15f57d0)}

338 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩

github-actions · 2026-03-23T00:36:16Z

GNU testsuite comparison:

Skip an intermittent issue tests/cut/bounded-memory (fails in this run but passes in the 'main' branch)
Skip an intermittent issue tests/pr/bounded-memory (fails in this run but passes in the 'main' branch)
Skipping an intermittent issue tests/date/date-locale-hour (passes in this run but fails in the 'main' branch)
Congrats! The gnu test tests/seq/seq-epipe is no longer failing!

github-actions · 2026-03-23T01:04:48Z

GNU testsuite comparison:

Skipping an intermittent issue tests/date/date-locale-hour (passes in this run but fails in the 'main' branch)
Congrats! The gnu test tests/seq/seq-epipe is no longer failing!
Note: The gnu test tests/expand/bounded-memory is now being skipped but was previously passing.

github-actions · 2026-03-23T01:47:48Z

GNU testsuite comparison:

Note: The gnu test tests/expand/bounded-memory is now being skipped but was previously passing.
Note: The gnu test tests/printf/printf-surprise is now being skipped but was previously passing.

github-actions · 2026-03-23T06:36:53Z

GNU testsuite comparison:

GNU test failed: tests/misc/io-errors. tests/misc/io-errors is passing on 'main'. Maybe you have to rebase?
Skip an intermittent issue tests/tail/symlink (fails in this run but passes in the 'main' branch)
Skipping an intermittent issue tests/date/resolution (passes in this run but fails in the 'main' branch)
Skipping an intermittent issue tests/pr/bounded-memory (passes in this run but fails in the 'main' branch)
Congrats! The gnu test tests/basenc/bounded-memory is now passing!
Congrats! The gnu test tests/dd/no-allocate is now passing!
Congrats! The gnu test tests/expand/bounded-memory is now passing!

github-actions · 2026-03-23T08:01:54Z

GNU testsuite comparison:

Skipping an intermittent issue tests/pr/bounded-memory (passes in this run but fails in the 'main' branch)
Note: The gnu test tests/unexpand/bounded-memory is now being skipped but was previously passing.
Congrats! The gnu test tests/printf/printf-surprise is now passing!

github-actions · 2026-03-23T09:43:07Z

GNU testsuite comparison:

Skip an intermittent issue tests/cut/bounded-memory (fails in this run but passes in the 'main' branch)
Note: The gnu test tests/cp/link-heap is now being skipped but was previously passing.
Note: The gnu test tests/seq/seq-epipe is now being skipped but was previously passing.
Note: The gnu test tests/unexpand/bounded-memory is now being skipped but was previously passing.
Congrats! The gnu test tests/printf/printf-surprise is now passing!

github-actions · 2026-03-23T10:36:46Z

GNU testsuite comparison:

GNU test failed: tests/timeout/timeout-group. tests/timeout/timeout-group is passing on 'main'. Maybe you have to rebase?
Skipping an intermittent issue tests/tty/tty-eof (passes in this run but fails in the 'main' branch)
Note: The gnu test tests/tail/tail-n0f is now being skipped but was previously passing.
Congrats! The gnu test tests/basenc/bounded-memory is now passing!

oech3 · 2026-03-23T11:07:07Z

CI fails by unrelated reason too many times. But this is ready.

github-actions · 2026-03-23T11:33:58Z

GNU testsuite comparison:

Skip an intermittent issue tests/pr/bounded-memory (fails in this run but passes in the 'main' branch)
Skipping an intermittent issue tests/date/date-locale-hour (passes in this run but fails in the 'main' branch)
Congrats! The gnu test tests/basenc/bounded-memory is now passing!

oech3 · 2026-03-23T12:52:26Z

It seems tee is faster than vmsplice with enough RAM (bandwidth?)

> taskset -c 1 yes-tee|taskset -c 2 pv>/dev/null
[50.9GiB/s]
> taskset -c 1 yes-vmsplice|taskset -c 2 pv>/dev/null
[47.5GiB/s]

> taskset -c 1 gnu5ec45a1aa/yes|taskset -c 2 pv>/dev/null
[49.6GiB/s]

and LD_PRELOAD=/usr/lib/libmimalloc.so does not help with enough RAM.
@ArniDagur

github-actions · 2026-03-23T12:52:28Z

GNU testsuite comparison:

Skip an intermittent issue tests/date/date-locale-hour (fails in this run but passes in the 'main' branch)
Congrats! The gnu test tests/printf/printf-surprise is now passing!
Note: The gnu test tests/env/env-signal-handler was skipped on 'main' but is now failing.

github-actions · 2026-03-23T13:06:57Z

GNU testsuite comparison:

GNU test failed: tests/rm/isatty. tests/rm/isatty is passing on 'main'. Maybe you have to rebase?
Congrats! The gnu test tests/misc/io-errors is no longer failing!
Note: The gnu test tests/seq/seq-epipe is now being skipped but was previously passing.
Note: The gnu test tests/tail/tail-n0f is now being skipped but was previously passing.
Congrats! The gnu test tests/csplit/csplit-heap is now passing!

github-actions · 2026-03-23T15:18:03Z

GNU testsuite comparison:

GNU test failed: tests/misc/io-errors. tests/misc/io-errors is passing on 'main'. Maybe you have to rebase?
Skip an intermittent issue tests/date/date-locale-hour (fails in this run but passes in the 'main' branch)
Skipping an intermittent issue tests/tail/follow-name (passes in this run but fails in the 'main' branch)
Congrats! The gnu test tests/tail/pipe-f2 is no longer failing!
Note: The gnu test tests/tail/pipe-f is now being skipped but was previously passing.
Congrats! The gnu test tests/tail/tail-n0f is now passing!

github-actions · 2026-03-23T18:31:37Z

GNU testsuite comparison:

Skip an intermittent issue tests/date/date-locale-hour (fails in this run but passes in the 'main' branch)
Congrats! The gnu test tests/tail/pipe-f2 is no longer failing!
Congrats! The gnu test tests/tail/tail-n0f is now passing!

github-actions · 2026-03-25T08:14:05Z

GNU testsuite comparison:

Skipping an intermittent issue tests/date/resolution (passes in this run but fails in the 'main' branch)
Note: The gnu test tests/printf/printf-surprise is now being skipped but was previously passing.
Note: The gnu test tests/tail/tail-n0f is now being skipped but was previously passing.

github-actions · 2026-03-27T08:46:36Z

GNU testsuite comparison:

Skip an intermittent issue tests/tty/tty-eof (fails in this run but passes in the 'main' branch)
Skipping an intermittent issue tests/rm/isatty (passes in this run but fails in the 'main' branch)
Note: The gnu test tests/tail/pipe-f is now being skipped but was previously passing.
Note: The gnu test tests/tail/tail-n0f is now being skipped but was previously passing.
Congrats! The gnu test tests/csplit/csplit-heap is now passing!
Congrats! The gnu test tests/printf/printf-surprise is now passing!
Congrats! The gnu test tests/unexpand/bounded-memory is now passing!

oech3 · 2026-03-28T13:38:24Z

Hmm... I considered that tee()ing references of small y\n pages generated by tee() is faster that splice()ing a sparse file. But it is few GiB slower than later...

io_uring might needed.

github-actions · 2026-03-28T17:17:51Z

GNU testsuite comparison:

Skip an intermittent issue tests/cut/bounded-memory (fails in this run but passes in the 'main' branch)
Congrats! The gnu test tests/misc/io-errors is no longer failing!
Congrats! The gnu test tests/unexpand/bounded-memory is now passing!

github-actions · 2026-03-28T17:45:32Z

GNU testsuite comparison:

Skipping an intermittent issue tests/pr/bounded-memory (passes in this run but fails in the 'main' branch)
Congrats! The gnu test tests/misc/io-errors is no longer failing!
Note: The gnu test tests/tail/pipe-f is now being skipped but was previously passing.
Congrats! The gnu test tests/unexpand/bounded-memory is now passing!

github-actions · 2026-03-28T20:44:07Z

GNU testsuite comparison:

Skipping an intermittent issue tests/date/date-locale-hour (passes in this run but fails in the 'main' branch)
Congrats! The gnu test tests/misc/io-errors is no longer failing!
Note: The gnu test tests/expand/bounded-memory is now being skipped but was previously passing.
Note: The gnu test tests/printf/printf-surprise is now being skipped but was previously passing.
Note: The gnu test tests/rm/many-dir-entries-vs-OOM is now being skipped but was previously passing.
Note: The gnu test tests/tail/tail-n0f is now being skipped but was previously passing.
Congrats! The gnu test tests/unexpand/bounded-memory is now passing!

github-actions · 2026-03-28T22:19:20Z

GNU testsuite comparison:

Skipping an intermittent issue tests/date/date-locale-hour (passes in this run but fails in the 'main' branch)
Skipping an intermittent issue tests/tty/tty-eof (passes in this run but fails in the 'main' branch)
Congrats! The gnu test tests/misc/io-errors is no longer failing!
Congrats! The gnu test tests/unexpand/bounded-memory is now passing!

github-actions · 2026-03-28T23:37:50Z

GNU testsuite comparison:

Skipping an intermittent issue tests/date/date-locale-hour (passes in this run but fails in the 'main' branch)
Skipping an intermittent issue tests/pr/bounded-memory (passes in this run but fails in the 'main' branch)
Congrats! The gnu test tests/misc/io-errors is no longer failing!
Note: The gnu test tests/tail/tail-n0f is now being skipped but was previously passing.
Congrats! The gnu test tests/unexpand/bounded-memory is now passing!

oech3 force-pushed the yes-tee branch 3 times, most recently from 57ce5eb to 1506d10 Compare March 23, 2026 00:15

oech3 force-pushed the yes-tee branch 2 times, most recently from 0c0949b to 32606d4 Compare March 23, 2026 00:28

oech3 force-pushed the yes-tee branch from 32606d4 to 9dd7151 Compare March 23, 2026 00:55

oech3 marked this pull request as ready for review March 23, 2026 01:31

oech3 force-pushed the yes-tee branch from 9dd7151 to 6afeb8c Compare March 23, 2026 01:39

oech3 force-pushed the yes-tee branch from 6afeb8c to 48e65e9 Compare March 23, 2026 06:27

oech3 force-pushed the yes-tee branch from 48e65e9 to 10449b6 Compare March 23, 2026 07:51

This comment was marked as resolved.

Sign in to view

oech3 force-pushed the yes-tee branch from 10449b6 to d90f019 Compare March 23, 2026 09:30

oech3 force-pushed the yes-tee branch from d90f019 to 58285e2 Compare March 23, 2026 10:23

oech3 force-pushed the yes-tee branch from 58285e2 to fea6807 Compare March 23, 2026 11:22

oech3 force-pushed the yes-tee branch from fea6807 to 6a9505a Compare March 23, 2026 12:40

oech3 force-pushed the yes-tee branch from 6a9505a to 6be7b39 Compare March 23, 2026 12:52

oech3 mentioned this pull request Mar 23, 2026

tests, yes: Support more syscall failure coreutils/coreutils#227

Closed

oech3 force-pushed the yes-tee branch from 6be7b39 to 0d98290 Compare March 23, 2026 15:08

oech3 force-pushed the yes-tee branch from 0d98290 to c5f3e96 Compare March 23, 2026 18:22

This comment was marked as outdated.

Sign in to view

oech3 force-pushed the yes-tee branch from c5f3e96 to 98c6a12 Compare March 25, 2026 08:03

oech3 force-pushed the yes-tee branch 2 times, most recently from cef5200 to 60e1cdf Compare March 27, 2026 08:36

oech3 marked this pull request as draft March 28, 2026 16:56

oech3 force-pushed the yes-tee branch from 60e1cdf to 4b723f4 Compare March 28, 2026 17:09

oech3 marked this pull request as ready for review March 28, 2026 17:30

oech3 force-pushed the yes-tee branch from 4b723f4 to 8761095 Compare March 28, 2026 17:36

oech3 force-pushed the yes-tee branch 2 times, most recently from 09891c1 to 690cfbf Compare March 28, 2026 20:34

oech3 force-pushed the yes-tee branch from 690cfbf to 5ada346 Compare March 28, 2026 22:09

oech3 force-pushed the yes-tee branch 2 times, most recently from c706ec2 to 5f79f2a Compare March 28, 2026 23:21

yes: use tee syscall as pipe only fast-path

a3a747c

oech3 force-pushed the yes-tee branch from 5f79f2a to a3a747c Compare March 28, 2026 23:28

Uh oh!

Conversation

oech3 commented Mar 23, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

codspeed-hq bot commented Mar 23, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Merging this PR will not alter performance

Footnotes

Uh oh!

github-actions bot commented Mar 23, 2026

Uh oh!

github-actions bot commented Mar 23, 2026

Uh oh!

github-actions bot commented Mar 23, 2026

Uh oh!

github-actions bot commented Mar 23, 2026

Uh oh!

github-actions bot commented Mar 23, 2026

Uh oh!

This comment was marked as resolved.

github-actions bot commented Mar 23, 2026

Uh oh!

github-actions bot commented Mar 23, 2026

Uh oh!

oech3 commented Mar 23, 2026

Uh oh!

github-actions bot commented Mar 23, 2026

Uh oh!

oech3 commented Mar 23, 2026

Uh oh!

github-actions bot commented Mar 23, 2026

Uh oh!

github-actions bot commented Mar 23, 2026

Uh oh!

github-actions bot commented Mar 23, 2026

Uh oh!

github-actions bot commented Mar 23, 2026

Uh oh!

This comment was marked as outdated.

github-actions bot commented Mar 25, 2026

Uh oh!

github-actions bot commented Mar 27, 2026

Uh oh!

oech3 commented Mar 28, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

github-actions bot commented Mar 28, 2026

Uh oh!

github-actions bot commented Mar 28, 2026

Uh oh!

github-actions bot commented Mar 28, 2026

Uh oh!

github-actions bot commented Mar 28, 2026

Uh oh!

github-actions bot commented Mar 28, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

oech3 commented Mar 23, 2026 •

edited

Loading

codspeed-hq bot commented Mar 23, 2026 •

edited

Loading

oech3 commented Mar 28, 2026 •

edited

Loading