Date: Tue, 7 Sep 2004 09:56:26 -0700 (PDT) From: Don Lewis <truckman@FreeBSD.org> To: conrads@cox.net Cc: freebsd-current@FreeBSD.org Subject: Re: pcm0:play:0: play interrupt timeout, channel dead Message-ID: <200409071656.i87GuQbN039098@gw.catspoiler.org> In-Reply-To: <20040907093421.0bbce221@dolphin.local.net>
next in thread | previous in thread | raw e-mail | index | archive | help
On 7 Sep, Conrad J. Sabatier wrote: > On Tue, 7 Sep 2004 11:46:35 +0200 > Guido van Rooij <guido@gvr.org> wrote: > >> On Tue, Sep 07, 2004 at 01:56:42AM -0700, Don Lewis wrote: >> > >> > Looks like this problem is not limited to amd64 ... >> >> Indeed. In fact, I have seen numerous mentioning of it and no >> solution. > > Are you experiencing the problem both with and without ACPI enabled? > > For me, disabling ACPI cures it on this amd64 box with nVidia nForce3 > 250 chipset. > > Cameron Grant recently sent me the URL for a patch he had come up with > for the snd_ich driver: > > http://people.freebsd.org/~cg/ich-uncrackify.diff.gz That looks more like a fix for a distortion/skipping problem. > I tried it, but the problem still persisted as long as ACPI is enabled. > Nonetheless, you may want to try it and see how it works for you. > > This is really one mysterious problem! The interrupt timeout problem appears to be caused by ich_intr() no longer seeing what it thinks are valid interrupts from the device. Guido sent me some useful debug info and we are homing in on the problem. The ACPI vs. non-ACPI difference that you are seeing may indicated that the problem is triggered by interrupt routing and/or what other devices share the same IRQ as the sound hardware.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200409071656.i87GuQbN039098>