Comment 45 for bug 87652

I'm running on a G4 Quicksilver (Tumbler audio), and this bug is still present in vanilla kernels (and Ubuntu, Debian, Fedora, and every other distro I've tried lately). I used git-bisect on a vanilla kernel and it spat out this commit:

http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=68a64357d15ae4f596e92715719071952006e83c
    [POWERPC] Merge 32 and 64 bits asm-powerpc/io.h
    Benjamin Herrenschmidt [Sun, 12 Nov 2006 22:27:39 +0000 (09:27 +1100)]
    powerpc: Merge 32 and 64 bits asm-powerpc/io.h

    The rework on io.h done for the new hookable accessors made it easier,
    so I just finished the work and merged 32 and 64 bits io.h for arch/powerpc.

If I add some logging in sound/ppc/pmac.c, I see that snd_pmac_pcm_update (in sound/ppc/pmac.c) eventually gets a struct dbdma command with xfer_status == 0x8088, which as far as I can tell is nowhere near a valid value. Normally it should continue to get xfrer_status == RUN | ACTIVE, ie: 0x84.

I unfortunately am quite new to the kernel, so I have no idea how to continue looking for a fix. I guess I'll just point benh over here and see if he's able to help.