A NOTE has been added to this issue. ====================================================================== <https://bugtrack.alsa-project.org/alsa-bug/view.php?id=2158> ====================================================================== Reported By: martin Assigned To: ====================================================================== Project: ALSA - driver Issue ID: 2158 Category: ISA - miro Reproducibility: always Severity: major Priority: normal Status: new Distribution: Kernel Version: ====================================================================== Date Submitted: 05-24-2006 15:13 CEST Last Modified: 05-29-2006 23:21 CEST ====================================================================== Summary: play produces i/o error on ad1845 (miro) Description: At first I will mention that 1.0.10final works without problems. And 1.0.11rc3 or any newer version fails always during the fifth hw_writei with this Input/Output error: aplay -Dplughw:1 /data/audio/lisboa_reloaded_1_20060421.wav Playing WAVE '/data/audio/lisboa_reloaded_1_20060421.wav' : Signed 16 bit Little Endian, Rate 44100 Hz, Stereo hw_writei: frames = 4096, xferi.result = 4096, err = 0 hw_writei: frames = 4096, xferi.result = 4096, err = 0 hw_writei: frames = 4096, xferi.result = 4096, err = 0 hw_writei: frames = 4096, xferi.result = 4096, err = 0 hw_writei: frames = 4096, xferi.result = -5, err = -5 aplay: pcm_write:1264: write error: Input/output error My snd-miro uses an AD1845 which will be driven by the cs4231-lib. It's really possible that other cards with an ad1845 will have the same problem. BTW, recording seems to work without problems. Unfortunately I haven't found the bug, yet. It was introduced between 1.0.10 and 1.0.11rc3. Well, if rc1/rc2 would compile, I could be more precise here. A strace log is attached. Any ideas? (Just before I start the hard way: fixing the rc1/2 compilation bug and hunting for the bad patch...) ====================================================================== ---------------------------------------------------------------------- martin - 05-29-06 23:21 ---------------------------------------------------------------------- I have to correct myself. After trying out a bunch of things it's obvious that it fails on x86. OTOH it works on an alpha cpu without problems. Sorry, I've mixed x86/alpha results in my former report. Trying out an old and new version I can confirm for same miro pcm20 hw: 1.0.9b and 1.0.11: snd-miro works fine on alpha 1.0.9b and 1.0.11: snd-miro fails on x86 with this i/o error Both systems are using a general ISA bus (I guess), but the AlphaBIOS doesn't support PnP. Ok, let's play a little bit with pnp stuff. Alsa compile option --with-isapnp=no doesn't make a difference. Disable PNP support in the BIOS doesn't make a difference again. Building a new kernel with PNP support disabled doesn't help, but at least I saw another message now: "miro.c:1265: unable to detect OPTi9xx chip". Weird. It goes through all loops of snd_card_miro_detect() and can't find my 924. I guess it's time to look for an opti924 datasheet now. Hmm, but why does this card work on alphas then? Issue History Date Modified Username Field Change ====================================================================== 05-24-06 15:13 martin New Issue 05-24-06 15:13 martin File Added: miro.log.gz 05-29-06 23:21 martin Note Added: 0010014 ====================================================================== _______________________________________________ Alsa-devel mailing list Alsa-devel@xxxxxxxxxxxxxxxxxxxxx https://lists.sourceforge.net/lists/listinfo/alsa-devel
