[linux-audio-dev] Re: [Alsa-devel] Re: more bad low latency results
At Thu, 15 Nov 2001 18:16:44 +0100, Maarten de Boer wrote: I've used this setting: echo 6 /proc/sys/vm/vm_scan_ratio echo 2 /proc/sys/vm/vm_mapped_ratio echo 4 /proc/sys/vm/vm_balance_ratio Uuhh... I don't have these files... $ ls /proc/sys/vm/ bdflush kswapd overcommit_memory page-cluster pagetable_cache Did I miss something in my kernel configuration? Oops, sorry, they are still in AA kernel only.. Or, will you try AA patches? It's not a bad idea, since the current (vanilla) VM is based on Andrea's code. Linus still doesn't include all his patches. Maybe my 2.4.13 LL patches at ftp.suse.com/pub/people/tiwai/suse-patches can be applied without rejection. That's really weird.. How about to measure the time of read/write responce in latency.c so that we can know at least whether it's related with the kernel latency. It's a different problem. The alsa capture adds random values at random places (not periodically as for a I can see) in a way that it sounds like a dusty vinyl record. This doesn't happen when I use arecord, it only happens with the alsa-lib/test/latency. I write each captured buffer to disk, and the random values are clearly visible (and hearable). It happens both in non-block and in poll mode. Hmm sounds like h/w problem, then.. It's interesting to know where the data is contaminated, whether on the driver level or transfer between capture and playback on user-space, or what else.. Takashi
[linux-audio-dev] Re: [Alsa-devel] Re: more bad low latency results
Or, will you try AA patches? It's not a bad idea, since the current (vanilla) VM is based on Andrea's code. Linus still doesn't include all his patches. Yes, I will try the AA patches, though I am not very sure which to use, and on which kernel. The most recent one on a standard kernel (and not on a pre-release) seems to be http://kernel.org/pub/linux/kernel/people/andrea/patches/v2.4/2.4.12aa1.bz2 Or do you think it is better to jump into the deep, and try to apply 2.4.15pre1aa1.bz2 on 2.4.14 ? Maybe my 2.4.13 LL patches at ftp.suse.com/pub/people/tiwai/suse-patches can be applied without rejection. It's a different problem. The alsa capture adds random values at random places (not periodically as for a I can see) in a way that it sounds like a dusty vinyl record. This doesn't happen when I use arecord, it only happens with the alsa-lib/test/latency. I write each captured buffer to disk, and the random values are clearly visible (and hearable). It happens both in non-block and in poll mode. Hmm sounds like h/w problem, then.. It's interesting to know where the data is contaminated, whether on the driver level or transfer between capture and playback on user-space, or what else.. I would say on driver level: I write the data to disk right after the snd_pcm_readi call. Maarten
Re: [linux-audio-dev] Re: [Alsa-devel] Re: more bad low latency results
Hmm sounds like h/w problem, then.. It's interesting to know where the data is contaminated, whether on the driver level or transfer between capture and playback on user-space, or what else.. I would say on driver level: I write the data to disk right after the snd_pcm_readi call. i hope i'm forgetting something. you're not writing to disk from the same thread as called snd_pcm_readi are you? that's guaranteed to eliminate low latency performance ... --p
[linux-audio-dev] Re: [Alsa-devel] Re: more bad low latency results
At Thu, 15 Nov 2001 19:17:19 +0100, Maarten de Boer wrote: Or, will you try AA patches? It's not a bad idea, since the current (vanilla) VM is based on Andrea's code. Linus still doesn't include all his patches. Yes, I will try the AA patches, though I am not very sure which to use, and on which kernel. The most recent one on a standard kernel (and not on a pre-release) seems to be http://kernel.org/pub/linux/kernel/people/andrea/patches/v2.4/2.4.12aa1.bz2 Or do you think it is better to jump into the deep, and try to apply 2.4.15pre1aa1.bz2 on 2.4.14 ? I think 2.4.14 or 15pre are ok, too. The same LL patch should work. Takashi