Gentoo Archives: gentoo-amd64

From: Mark Knecht <markknecht@×××××.com>
To: gentoo-amd64@l.g.o
Subject: Re: [gentoo-amd64] Re: kernel.org vs. Gentoo-64 bit kernels (xruns)
Date: Tue, 20 Sep 2005 21:53:55
Message-Id: 5bdc1c8b050920145234b3e425@mail.gmail.com
In Reply to: Re: [gentoo-amd64] Re: kernel.org vs. Gentoo-64 bit kernels (xruns) by Paul de Vrieze
1 On 9/20/05, Paul de Vrieze <pauldv@g.o> wrote:
2 > On Tuesday 20 September 2005 14:25, Mark Knecht wrote:
3 > > On 9/20/05, Duncan <1i5t5.duncan@×××.net> wrote:
4 > > > I'm not familiar with the term "xrun", so this may be entirely off the
5 > > > wall, but have you confirmed the hard drive is running DMA? If your
6 > > > chipset or SATA drivers are wrong, and your hard drive is having to run
7 > > > in legacy interrupt mode instead of DMA mode, it *WILL* destroy latency
8 > > > and generally make the system unusable for any sort of real-time work at
9 > > > all, regardless of the other kernel patches applied. So... in addition
10 > > > to checking the network drivers, investigate the hard drive and chipset
11 > > > I/O drivers as well, and confirm you ARE running DMA mode.
12 > >
13 > > Thanks, yes, DMA is running, as far as I can tell. hdparm -tT returns
14 > > numbers that are >50MB/S.
15 > >
16 > > xruns are a term specific to the Jack server
17 > > (jack-audio-connection-kit) that tell us whether we've had and overrun
18 > > or an underrun. It's would be off topic to go deeply into how Jack
19 > > operates when talking to sound cards, but take it to mean something
20 > > bad has happened with real-time audio data.
21 >
22 > Nah, it's more alsa specific. What soundcard do you use? Some soundcards are
23 > more crappy than others (especially onboard ones). I guess it should support
24 > DMA as even the soundblaster pro did so. Soundcards do however provide
25 > various levels of hardware accelleration.
26
27 That's not the problem here. The card is an RME HDSP 9652. It ran in
28 my older Asus/Via/Athlon-XP/Gentoo-32-bit machine just fine. I've
29 built the new Asus/NForce4/Athlon64/Gentoo-64-bit machine and
30 transfered the same card of here so I know the card and drivers are
31 fine, at least in a 32-bit environment.
32 >
33 > > Interestingly Jack runs from memory so hard drive performance should
34 > > not cause major problems unless it's not interruptable in a more or
35 > > less real-time way. On my Gentoo 32-bit machines (using Via and ATI
36 > > chipsets) I've not had to install any real-time patches and can still
37 > > run reliable at sub-2mS latencies. On those machines I can do pretty
38 > > much anything, browse the web with firefox, do and emerge sync, etc.,
39 > > and I get no xruns. On this AMD64/NForce4 machine and emerge sync
40 > > causes xruns immediately, indicating the sound card is getting starved
41 > > for data.
42 >
43 > Good chance the soundcard buffer is smaller or the driver is crappy. You could
44 > try to take the soundcard from the old machine and put it in the new one.
45
46 As you can see, that's what I did! :-)
47
48 - Mark
49
50 --
51 gentoo-amd64@g.o mailing list