Re: Linux 2.6.15-rc5 and Alsa 1.0.10

[Hugh Dickins <hugh@veritas.com>]

On Fri, 9 Dec 2005, Jacek Luczak wrote:
> 
> I'm using now 2.6.15-rc5 kernel with latest Alsa 1.0.10 and I received a lot
> of 'bad page state at free_hot_cold_page' (see example below) messages. Is
> this kernel or alsa error?
> 
> System:
> Slackware Linux, GCC 3.4.4, Binutils 2.16.1.
> CPU: Pentium 4 3Ghz HT
> Sound card: CMI9880 (HDA)
> 
> Dec  9 16:53:20 slawek kernel: Bad page state at free_hot_cold_page (in
> process 'xmms', page c12da9c0)
> Dec  9 16:53:20 slawek kernel: flags:0x80000414 mapping:00000000 mapcount:0
> count:0

I think that means you have a mismatch: that you're using core/memalloc.c
from alsa-driver-1.0.10/alsa-kernel rather than from 2.6.15-rc5/sound.
Applying the update below should eliminate your "Bad page state"s:
I've no idea whether there are other mismatches, quite possibly not.

Hugh

--- alsa-driver-1.0.10/alsa-kernel/core/memalloc.c	2005-10-31 13:11:53.000000000 +0000
+++ 2.6.15-rc5/sound/core/memalloc.c	2005-12-04 06:48:20.000000000 +0000
@@ -197,6 +197,7 @@ void *snd_malloc_pages(size_t size, gfp_
 
 	snd_assert(size > 0, return NULL);
 	snd_assert(gfp_flags != 0, return NULL);
+	gfp_flags |= __GFP_COMP;	/* compound page lets parts be mapped */
 	pg = get_order(size);
 	if ((res = (void *) __get_free_pages(gfp_flags, pg)) != NULL) {
 		mark_pages(virt_to_page(res), pg);
@@ -241,6 +242,7 @@ static void *snd_malloc_dev_pages(struct
 	snd_assert(dma != NULL, return NULL);
 	pg = get_order(size);
 	gfp_flags = GFP_KERNEL
+		| __GFP_COMP	/* compound page lets parts be mapped */
 		| __GFP_NORETRY /* don't trigger OOM-killer */
 		| __GFP_NOWARN; /* no stack trace print - this call is non-critical */
 	res = dma_alloc_coherent(dev, PAGE_SIZE << pg, dma, gfp_flags);