On Mon, 2007-10-29 at 12:47 -0700, Jesse Barnes wrote: > In this case, we're performing basically a dma_sync*(...DMA_TO_DEVICE) > right? But this is just for the GPU; every other DMA device in the system is cache-coherent. > Can we be sure that a single flush is sufficient? Is there any > window between when we flush and when we start accessing memory with > the device that we could get into more caching trouble? An uncached write to this page will not complete until the buffers are completely flushed. > Looks reasonable, I'm not sure we can do much better. The only concern > I have is that allocating some more PCI space like that may end up > clobbering some *other* hidden BIOS mapping, but there's not a whole > lot we can do about that. This isn't a hidden mapping; the i965 doesn't allocate space for it in the BIOS. -- [email protected]
Attachment:
signature.asc
Description: This is a digitally signed message part
- Follow-Ups:
- Re: [RFC] AGP initial support for chipset flushing..
- From: Jesse Barnes <[email protected]>
- Re: [RFC] AGP initial support for chipset flushing..
- References:
- [RFC] AGP initial support for chipset flushing..
- From: Dave Airlie <[email protected]>
- Re: [RFC] AGP initial support for chipset flushing..
- From: Jesse Barnes <[email protected]>
- [RFC] AGP initial support for chipset flushing..
- Prev by Date: Re: [stable] 2.6.23 boot failures on x86-64.
- Next by Date: Re: [PATCH 1/2] [CRYPTO] tcrypt: Move sg_init_table out of timing loops
- Previous by thread: Re: [RFC] AGP initial support for chipset flushing..
- Next by thread: Re: [RFC] AGP initial support for chipset flushing..
- Index(es):
 |