diff options
| author | Arne Coucheron <arco68@gmail.com> | 2015-12-04 22:26:44 +0100 |
|---|---|---|
| committer | Steve Kondik <shade@chemlab.org> | 2015-12-07 20:35:35 -0800 |
| commit | 02d88d66c02982deec3aedee78c1f64352c539bc (patch) | |
| tree | 2d0ef2ce131f048e77ffe5b766c2fd4fe8634726 /services/audioflinger/Threads.h | |
| parent | 41ccd68b80a426236ee8ff511c22119e955bc548 (diff) | |
| download | frameworks_av-02d88d66c02982deec3aedee78c1f64352c539bc.zip frameworks_av-02d88d66c02982deec3aedee78c1f64352c539bc.tar.gz frameworks_av-02d88d66c02982deec3aedee78c1f64352c539bc.tar.bz2 | |
audioflinger: Don't do float conversion in upmix/downmix for legacy ALSA
Legacy ALSA really hates floating point, and it's breaking
mic input when doing things like audio recording.
Use the old conversion routine for legacy ALSA.
Change-Id: I616f4cd42fa0e4d7595dd61ed2d36c4fa7052c53
Diffstat (limited to 'services/audioflinger/Threads.h')
| -rw-r--r-- | services/audioflinger/Threads.h | 5 |
1 files changed, 5 insertions, 0 deletions
diff --git a/services/audioflinger/Threads.h b/services/audioflinger/Threads.h index 9e32ea1..6182364 100644 --- a/services/audioflinger/Threads.h +++ b/services/audioflinger/Threads.h @@ -1160,11 +1160,16 @@ public: } private: +#ifdef LEGACY_ALSA_AUDIO + // internal convert function for format and channel mask. + void convert(void *dst, /*const*/ void *src, size_t frames); +#else // format conversion when not using resampler void convertNoResampler(void *dst, const void *src, size_t frames); // format conversion when using resampler; modifies src in-place void convertResampler(void *dst, /*not-a-const*/ void *src, size_t frames); +#endif // user provided information audio_channel_mask_t mSrcChannelMask; |