tee: improve throughput by raw syscall

[oech3]

Remove flush overhead and use raw syscall for POSIX tee requirement.

$ truncate -s 1PB huge
$ gnu9.11/tee<huge|pv>/dev/null
7.13GiB 0:00:02 [3.61GiB/s]
$ target/release/tee<huge|pv>/dev/null
2.56GiB 0:00:01 [2.56GiB/s]
$ target/release/tee-raw<huge|pv>/dev/null
3.35GiB 0:00:01 [3.35GiB/s]

$ cat huge|gnu9.11/tee|pv>/dev/null
2.15GiB 0:00:01 [2.15GiB/s]
$ cat huge|target/release/tee|pv>/dev/null
2.31GiB 0:00:01 [2.31GiB/s]
$ cat huge|target/release/tee-raw|pv>/dev/null
3.00GiB 0:00:01 [3.00GiB/s]

I was really surprised that std::io has such overhead.

[sylvestre]

is it possible to have a codspeed benchmark to evaluate the win?

[oech3]

I don't know how to connect stdin to CodSpeed.

[sylvestre]

i asked on discord

[oech3]

We should also support syscall failure on I/O benchmarks (inparticular for splice(2)).
Some benches with splice(2) code path are invalid for other platforms.

[oech3]

It might syscall-time bounded bench. If so, we cannot use CodSpeed.
But I made #12134 anyway.

[github-actions]

GNU testsuite comparison:

GNU test failed: tests/tail/pipe-f2. tests/tail/pipe-f2 is passing on 'main'. Maybe you have to rebase?
Skipping an intermittent issue tests/date/date-locale-hour (passes in this run but fails in the 'main' branch)

[oech3]

https://github.com/oech3/coreutils/actions/runs/25298780392/job/74161900855#step:8:686

[oech3]

I think unsafe is needed if we want do same things for for WIndows.

[xtqqczze]

The Write implementation is somewhat unintuitive; it could be removed in favor of a simpler approach with a single method:

fn write(&mut self, buf: &[u8]) -> Result<()>

Additionally, avoiding Write may be preferable if we plan to support a MaybeUninit buffer in the future.

[oech3]

I don't want to manually write loop writing all of buf.

rustix::io::read can generate uninit safely by splitting init part, but rustix::io::write cannot. So uninit is unrelated for Writer.

[oech3]

This impl is because rustix::io::write does not support writing to stdout on Windows. If we can override write by similar way, it is preffered.

[xtqqczze]

I still think its's confusing to implement Write on Unix, but not Windows...

[oech3]

I can add comment instead. We need 2 duplicated write_all definitiols for Stdout anf File if we manually loop.

[xtqqczze]

I was thinking something like this: b0edaae

[oech3]

No. Because write syscall does not guarantee writing everything. Needs loop and take sum of returned values.
(also rustix::io::write_all is not provided...)

[xtqqczze]

It would solve many problems if we could use std::io::stdio::StdoutRaw. Perhaps this functionality could be implemented as a crate.

[oech3]

Private? Is it need to copy and paste unsafe code to uucore?

[xtqqczze]

The issue we’re working around is that std::io::Stdout uses buffering, which doesn’t match our requirements. Ideally, we should provide our own Stdout implementation that performs unbuffered I/O instead. But that would be a much larger piece of work.

[oech3]

https://docs.rs/rustix/latest/rustix/stdio/fn.stdout.html but unix only...

[xtqqczze]

We already have a solution in uucore::io::OwnedFileDescriptorOrHandle, see #10235.

[oech3]

Hmm... it has try_clone. So introducing new overhead?

[oech3]

We cannot use it #12157

[codspeed-hq]

Merging this PR will improve performance by ×17

⚡ 1 improved benchmark
✅ 312 untouched benchmarks
⏩ 46 skipped benchmarks¹

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
⚡	Simulation	tee_stdin_file[10000000]	3,970.9 µs	233 µs	×17

---

_{Comparing oech3:tee-throughput (bcca96f) with main (2e04477)}

Footnotes

1. 46 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. ↩

[oech3]

+syscall time is still better