Messages in this thread | | | Date | Tue, 15 Dec 2009 20:10:48 +0900 | Subject | Re: bisected crash due to "x86: Move swiotlb initialization before dma32_free_bootmem" | From | FUJITA Tomonori <> |
| |
On Tue, 15 Dec 2009 11:56:50 +0100 Ingo Molnar <mingo@elte.hu> wrote:
> > * FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> wrote: > > > There are a few people who hit this. How many people use a box with over > > 256GB memory? > > > > And you can work around this with "dma32_size" kernel boot option. > > Well, since the kernel has not crashed before this change there's really just > two options as per upstream kernel regression policy: either we fix it or we > revert it.
As I wrote, here is a patch that can be applied to cleanly to the git head:
http://www.kernel.org/pub/linux/kernel/people/tomo/misc/0001-x86-two-stage-swiotlb-initialization.patch
It fixes the problem. Yinghai, can you test it? It should work but it's good to confirm it.
I simply wanted to say that it's not a bug that breaks lots of boxes or leads to something serious like data corruption (no need to say something like "revert it now!"). It's also worth investigating why it breaks, I think.
| |