Bug 296588 - sound becomes choppy with snd_uaudio(4) between base d516e52373e1...407c7c339adb
Summary: sound becomes choppy with snd_uaudio(4) between base d516e52373e1...407c7c339adb
Status: Closed FIXED
Alias: None
Product: Base System
Classification: Unclassified
Component: kern (show other bugs)
Version: 16.0-CURRENT
Hardware: Any Any
: --- Affects Only Me
Assignee: freebsd-multimedia (Nobody)
URL:
Keywords: regression
Depends on:
Blocks:
 
Reported: 2026-07-08 02:56 UTC by Evgenii Khramtsov
Modified: 2026-07-09 16:17 UTC (History)
2 users (show)

See Also:


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Evgenii Khramtsov 2026-07-08 02:56:58 UTC
Very weird regression. I run multimedia/audacious with OSS4 backend set in settings, and after updating from base d516e52373e1 to base 407c7c339adb I get choppy playback with every audio file played. I didn't check other ports using audio yet, as well as other things. I don't have time to bisect this yet.

I have confirmed that this issue is related to base alone by rebooting into the d516e52373e1 boot environment and updating base and rebuilding drm-kmod alone, rebooting into 407c7c339adb (and drm-kmod rebuilt against it), observing the choppy sound issue.
Comment 1 Christos Margiolis freebsd_committer freebsd_triage 2026-07-08 12:41:56 UTC
There is no commit to sound(4) or snd_uaudio(4) between d516e52373e1 and 407c7c339adb. Could it be that some other commit increases CPU usage and sound becomes choppy as a result of that?
Comment 2 Evgenii Khramtsov 2026-07-09 16:15:49 UTC
I have updated to base 05639575483e and can no longer reproduce this.

(In reply to Christos Margiolis from comment #1)

I suspected changes to sendfile(2) or something interrupt related between base d516e52373e1...407c7c339adb but couldn't confirm. Now that base 05639575483e included several fixes for recent sendfile(2) regressions I assume that it was that.

Maybe multimedia/audacious uses unix(4) somehow? E.g. couldn't use OSS for some reason and fallbacked to pulseaudio.

Anyway, this is now fixed.
Comment 3 Evgenii Khramtsov 2026-07-09 16:17:38 UTC
(In reply to Evgenii Khramtsov from comment #2)

> base 05639575483e

err, pasted local branch hash by mistake, mine is rebased on top of base 0b95429fde2a