Changeset 88159 in vbox for trunk/src/VBox
- Timestamp:
- Mar 17, 2021 2:42:40 PM (4 years ago)
- svn:sync-xref-src-repo-rev:
- 143308
- File:
-
- 1 edited
Legend:
- Unmodified
- Added
- Removed
-
trunk/src/VBox/Devices/Audio/HDAStream.cpp
r88158 r88159 1573 1573 static void hdaR3StreamPullFromMixer(PHDASTREAM pStreamShared, PHDASTREAMR3 pStreamR3, PAUDMIXSINK pSink) 1574 1574 { 1575 RT_NOREF(pStreamShared); 1575 1576 int rc = AudioMixerSinkUpdate(pSink); 1576 1577 AssertRC(rc); … … 2286 2287 if (pStreamShared->State.fInputPreBuffered && cbStreamUsed >= cbPeriod) 2287 2288 { /*likely*/ } 2288 /* Because it may take a while for the input stream to get going (at 2289 least with pulseaudio), we feed the guest silence till we've 2290 pre-buffer a reasonable amount of audio. */ 2289 /* 2290 * Because it may take a while for the input stream to get going (at 2291 * least with pulseaudio), we feed the guest silence till we've 2292 * pre-buffer a reasonable amount of audio. 2293 */ 2291 2294 else if (!pStreamShared->State.fInputPreBuffered) 2292 2295 { … … 2312 2315 else 2313 2316 { 2317 /** @todo We're ending up here to frequently with pulse audio at least (just 2318 * watch the stream stats in the statistcs viewer, and way to often we 2319 * have to inject silence bytes. I suspect part of the problem is 2320 * that the HDA device require a much better latency than what the 2321 * pulse audio is configured for by default (10 ms vs 150ms). */ 2314 2322 STAM_REL_COUNTER_INC(&pStreamR3->State.StatDmaFlowProblems); 2315 2323 Log(("hdaR3StreamUpdate: Warning! Stream #%u has insufficient data available: %u bytes, need %u. Will try move pull more data into the buffer...\n",
Note:
See TracChangeset
for help on using the changeset viewer.