Sam James 83a27ac516 app-arch/xz-utils: improve USE=pgo for 9999
Takes a lot of inspiration from https://packages.altlinux.org/en/sisyphus/srpms/xz/specfiles/#line-80.

IIRC at least one other distro does this for xz or gzip (was it Clear Linux?)

The testsuite could do with better coverage so supplement it by creating
a tar archive of a bunch of files made during the build, then repeatedly
compress/decompress it to gather profiling data.

That said, I don't get any binary differences at all for liblzma and xz
for my usual flags (-O3 -march=znver2 ...). I do get some small differences
for just -O2. Both tests were done with GCC 14.0.1 20240415.

It would be nice if we could drop -fprofile-partial-training and that's
perhaps the thing to try next. I see this commit as mostly a step towards
being able to do that which should let PGO give smaller binaries with -O2
at least.

Signed-off-by: Sam James <sam@gentoo.org>
2024-04-16 02:24:25 +01:00
2024-04-15 19:47:27 -04:00
2024-04-15 21:56:15 +02:00
2024-04-15 07:51:30 +01:00
2024-04-15 16:12:49 -04:00
2024-04-13 13:26:06 +03:00
2024-04-14 16:47:18 -07:00
2024-04-13 12:53:40 +03:00
2024-04-15 17:58:46 +00:00
2024-04-15 23:50:10 +00:00
2024-04-13 19:48:00 +01:00
2024-04-09 13:11:04 -04:00
2024-04-14 08:04:59 +02:00
2024-04-13 16:15:31 +02:00
2024-04-15 13:20:24 +01:00
2024-04-11 16:51:07 +02:00
2024-04-15 23:50:17 +00:00
2024-04-11 22:14:41 +02:00
2024-04-12 20:23:05 +03:00
2024-04-15 10:52:53 +02:00
2024-04-15 21:32:31 +02:00
2024-04-15 17:23:55 +02:00
2024-04-11 22:14:17 +02:00
2024-04-15 07:55:22 +03:00
2024-04-13 10:51:32 -04:00
2024-04-10 09:02:06 -07:00
2024-04-15 20:43:33 +03:00
2024-04-14 13:50:13 -07:00
2024-04-14 22:36:25 -04:00
2024-04-09 20:19:24 +03:00
…
S
Description
No description provided
9.8 GiB
Languages
Shell 98%
Roff 1.4%
Emacs Lisp 0.2%
C 0.1%