On 16-07-08 15:38, Landis McGauhey wrote:
"pop" in speaker as alsaconf loads snd-ens1371
running alsamixer-- no mute toggle underneath mic, that's a first
and just so you know, there's no mute toggle under "master", "master-m", and "PCM", either, but that's nominal. There was one time, however, that "master" and/or "PCM" had a mute toggle. Perhaps "master" and/or "PCM" are muted and we can't tell because there are supposed to be mute toggles there, but they are missing? Something to consider...
I'm afraid that codec communication is just really messed up and the last two patches seem to have made things worse, so in preparation for any possible follow ups, you might as well revert those:
from /usr/src/linux-2.6.25.9:
# moodprobe -r ens1371-ac97_reset_hack.diff # moodprobe -r ens1371-ac97.diff
do please again run "make" and "make modules_install" after this but that's just so that I'm satisfied you're back at the start. No need to test anything, you'd just find the old behaviour again.
One thing I am noticing now that we've been looking at the codec ID in ALSA in the OSS driver output there:
=== es1371: found chip, vendor id 0x1102 device id 0x8938 revision 0x00 es1371: found es1371 rev 0 at io 0xde80 irq 23 es1371: Enabling internal amplifier. ac97_codec: AC97 Audio codec, id: (Unknown) gameport: ESS1371 Gameport is isa0218/gameport0, io 0x218, speed 1125kHz ===
is the "id: (Unknown)" bit. The OSS driver also definitely knows about the Sigmatel STAC9704 codec so it seems _this_ problam might not be unique to the ALSA driver. The OSS driver does produce sound, so perhaps we're chasing something not maximally productive right now...
It would be sort of interesting to have you look at the physical card and verify that it's indeed a Sigmatel STAC9704. It should be a tiny chip with a greek sigma sign and "STAC9704" in small enough letters that you might even need a magnifying glass. If it's hard or clumsy to take it out, don't bother though, it's not that important. You had 76058384 as an ID and a STAC9704 is 83847605 (upper/lower 16-bit switched) but that will not have gotten to be there by accident.
Yes, before anyone wonders, I verified that the ALSA driver does that "Enabling internal amplifier" thing (that might be that pop that you hear; is the pop also present when loading the OSS driver?)
Rene & Takashi: I fear this is taking up too much of our time and this effort may have long passed the point of diminishing returns. Personally, my goal is to have audio with Skype, whether with OSS or ALSA, I don't care. If you want to continue to work on this for the sake of the community at large and others who have this same card and this same problem, then I'm willing and would be honored to be the guinea pig and be an active participant in the community. I'm sorry this problem is so stubborn and if you have better things to do, I understand. If you don't care to work on this anymore, I would understand and if you could perhaps just help me get sound in Skype (that is, to have /dev/dsp appear as I/O options in Skype configuration) with OSS, then I would be a happy camper indeed.
I'm afraid I don't know Skype. Older versions of it couldn't use ALSA but needed the OSS emulation. If newer version now can't use OSS but need ALSA, lovely...
I'll keep looking for a bit but ofcourse you don't have to wait around for anything. If I find anything with a good chance of success I'll let you know. Please conform that you've reverted the last two patches as mentioned above.
Rene.