Paths

Table of Contentst

Differential D48394

sound: Simplify pcm/feeder_mixer.c
Needs ReviewPublic
Actions

Authored by christos on Thu, Jan 9, 2:37 PM.

Tags

None

Referenced Files

	Unknown Object (File)
	Mon, Jan 27, 9:05 PM

	Unknown Object (File)
	Mon, Jan 27, 7:06 PM

	Unknown Object (File)
	Fri, Jan 24, 11:59 PM

	Unknown Object (File)
	Fri, Jan 17, 4:17 AM

	Unknown Object (File)
	Wed, Jan 15, 6:02 PM

	Unknown Object (File)
	Wed, Jan 15, 2:08 AM

Subscribers

Details

Reviewers

markj
emaste
dev_submerge.ch

Summary

Get rid of macro magic.
Make feed_mixer_info handling similar to most feeders.

Sponsored by: The FreeBSD Foundation
MFC after: 1 week

Diff Detail

Repository

rG FreeBSD src repository

Lint

Lint Skipped

Unit

Tests Skipped

Build Status

Buildable 62146
Build 59030: arc lint + arc unit

Event Timeline

christos created this revision.Thu, Jan 9, 2:37 PM

Herald added a subscriber: imp. · View Herald TranscriptThu, Jan 9, 2:37 PM

christos requested review of this revision.Thu, Jan 9, 2:37 PM

Harbormaster completed remote builds in B61607: Diff 148992.Thu, Jan 9, 2:37 PM

christos added a parent revision: D47638: sound: Implement AFMT_FLOAT support.Thu, Jan 9, 2:37 PM

christos added inline comments.

sys/dev/sound/pcm/feeder_mixer.c
52	We don't need to care about this, since all these types are defined as `int32_t`.

christos added inline comments.Thu, Jan 9, 2:40 PM

sys/dev/sound/pcm/feeder_mixer.c
52	Actually if `SND_PCM_64` is set, `intpcm32_t` is defined as `int64_t`. @dev_submerge.ch I think we might need to clean up these types altogether.

christos added inline comments.Thu, Jan 9, 2:44 PM

sys/dev/sound/pcm/feeder_mixer.c
52	I think we can just use `int32_t` and `int64_t` in place of `intpcm*_t` directly, and in this particular case, a simple `#ifdef SND_PCM_64` to choose the right type. I'm not sure about the benefit of these types, since they are mostly just aliases for `int32_t`.

dev_submerge.ch added inline comments.Thu, Jan 9, 4:34 PM

sys/dev/sound/pcm/feeder_mixer.c
52	Hm, I have to investigate and look at all the use cases first. Also it looks like we lose precision here when the volume is changed somewhere else in the chain. But that's probably a general issue with the current feeder design.

christos added inline comments.Thu, Jan 9, 4:42 PM

sys/dev/sound/pcm/feeder_mixer.c
52	Yeah, this patch, as well as the others, are just a step towards making this codebase easier to work with. I think there is quite a lot of room for improvements.

christos added a parent revision: D48421: sound: Turn PCM_CLAMP_* macros into a function.Fri, Jan 10, 6:39 PM

Rebase on top of D48421.

Harbormaster completed remote builds in B61656: Diff 149086.Fri, Jan 10, 6:43 PM

Update to reflect changes in D47932 (use of pcm_sample_read_24bit()).

Harbormaster completed remote builds in B61684: Diff 149159.Sun, Jan 12, 9:54 PM

christos mentioned this in D47932: sound: Refactor the format conversion framework.Thu, Jan 16, 1:16 PM

Reflect D47932 changes.

Harbormaster completed remote builds in B61757: Diff 149331.Thu, Jan 16, 1:19 PM

Reflect changes in D47932.

Harbormaster completed remote builds in B61862: Diff 149655.Tue, Jan 21, 11:38 AM

christos added inline comments.Sat, Jan 25, 4:58 PM

sys/dev/sound/pcm/feeder_mixer.c
54	Maybe this should be `intpcm32_t` for now.

Make z intpcm32_t.

Harbormaster completed remote builds in B62145: Diff 150193.Thu, Jan 30, 3:40 PM

Use the right pcm_sample_* variants.

Harbormaster completed remote builds in B62146: Diff 150194.Thu, Jan 30, 3:43 PM

Revision Contents
Changeset List

Path

Size

sys/

dev/

sound/

pcm/

177 lines

Diff 150194

sys/dev/sound/pcm/feeder_mixer.c

Loading...