Skip to content

avauddec: fix unnecessary reconfiguration if the audio layout isn't specified

Tim-Philipp Müller requested to merge tpm/gstreamer:avaudec-reconfiguration-fix-no-layout into main

It would constantly want to renegotiate (and spam the debug log) even though the channel layout hasn't actually changed. We use the same fallback in gst_ffmpegauddec_negotiate() already.

This happens with WMA files for example:

Debug log 
gstavauddec.c:671:gst_ffmpegauddec_handle_frame:<avdec_wmav2-0> Received new data of size 5945, offset:18446744073709551615, ts:0:03:39.149000000, dur:0:00:00.368156500
 gstavauddec.c:394:gst_ffmpegauddec_negotiate:<avdec_wmav2-0> Renegotiating audio from 44100Hz@2channels (28, interleaved=0) to 44100Hz@2channels (28, interleaved=0)
 gstavauddec.c:496:gst_ffmpegauddec_audio_frame:<avdec_wmav2-0> Creating output buffer
 gstavauddec.c:520:gst_ffmpegauddec_audio_frame:<avdec_wmav2-0> Buffer created. Size: 131072
 gstavauddec.c:546:gst_ffmpegauddec_audio_frame:<avdec_wmav2-0> return flow 0, out 0x7fb8d8003b40, got_frame 1
 gstavauddec.c:573:gst_ffmpegauddec_frame:<avdec_wmav2-0> Decoded data, buffer buffer: 0x7fb8d8003b40, pts 99:99:99.999999999, dts 99:99:99.999999999, dur 99:99:99.999999999, size 131072, offset none, offset_end none, flags 0x0
 gstavauddec.c:546:gst_ffmpegauddec_audio_frame:<avdec_wmav2-0> return flow 0, out (nil), got_frame 0
 gstavauddec.c:578:gst_ffmpegauddec_frame:<avdec_wmav2-0> We didn't get a decoded buffer
 gstavauddec.c:671:gst_ffmpegauddec_handle_frame:<avdec_wmav2-0> Received new data of size 5945, offset:18446744073709551615, ts:0:03:39.521000000, dur:0:00:00.368156500
 gstavauddec.c:394:gst_ffmpegauddec_negotiate:<avdec_wmav2-0> Renegotiating audio from 44100Hz@2channels (28, interleaved=0) to 44100Hz@2channels (28, interleaved=0)
 gstavauddec.c:496:gst_ffmpegauddec_audio_frame:<avdec_wmav2-0> Creating output buffer
 gstavauddec.c:520:gst_ffmpegauddec_audio_frame:<avdec_wmav2-0> Buffer created. Size: 131072
 gstavauddec.c:546:gst_ffmpegauddec_audio_frame:<avdec_wmav2-0> return flow 0, out 0x7fb8d8003120, got_frame 1
 gstavauddec.c:573:gst_ffmpegauddec_frame:<avdec_wmav2-0> Decoded data, buffer buffer: 0x7fb8d8003120, pts 99:99:99.999999999, dts 99:99:99.999999999, dur 99:99:99.999999999, size 131072, offset none, offset_end none, flags 0x0
 gstavauddec.c:546:gst_ffmpegauddec_audio_frame:<avdec_wmav2-0> return flow 0, out (nil), got_frame 0
 gstavauddec.c:578:gst_ffmpegauddec_frame:<avdec_wmav2-0> We didn't get a decoded buffer
 gstavauddec.c:671:gst_ffmpegauddec_handle_frame:<avdec_wmav2-0> Received new data of size 5945, offset:18446744073709551615, ts:0:03:39.892000000, dur:0:00:00.368156500
 gstavauddec.c:394:gst_ffmpegauddec_negotiate:<avdec_wmav2-0> Renegotiating audio from 44100Hz@2channels (28, interleaved=0) to 44100Hz@2channels (28, interleaved=0)
 gstavauddec.c:496:gst_ffmpegauddec_audio_frame:<avdec_wmav2-0> Creating output buffer

Merge request reports