> This makes sound compile on my system. I just copied the pattern in the
>other __get_free_pages() diffs, a real kernel hacker might want to check this
>before it's accepted...
I did the same thing but used __get_dma_pages instead of __get_free_pages.
It works for me and I believe it should work for everyone. The original call
appeared to be asking for DMAable memory which the __get_dma_pages function
should return. I've got 64meg of RAM and a 16bit sound card, I'm not sure
whether the version you posted would work for me, but I expect it would work
well for everyone who's got 16megs of RAM.
BTW Does anyone know of a good PCI sound card that has Linux support? I'd
like to go all-PCI and avoid the problems of 24bit DMA etc altogether.
>--- soundcard.c.orig Tue Jan 20 11:21:52 1998
>+++ soundcard.c Tue Jan 20 11:23:48 1998
>@@ -1100,7 +1100,7 @@
>
> dmap->buffsize = PAGE_SIZE * (1 << sz);
>
>- if ((start_addr = (char *) __get_free_pages(GFP_ATOMIC, sz,
>MAX_DMA_ADDRESS)) == NULL) + if ((start_addr = (char *)
>__get_free_pages(GFP_ATOMIC, sz)) == NULL)
> dmap->buffsize /= 2;
> }
>
-- ----------------------------------------------------------- In return for "mailbag contention" errors from buggy Exchange servers I'll set my mail server to refuse mail from your domain. The same response applies when a message to a postmaster account bounces. "Russell Coker - mailing lists account" <bofh@snoopy.virtual.net.au> -----------------------------------------------------------