Messages in this thread | | | Date | Fri, 29 Jun 2007 23:39:22 +0100 | From | Alan Cox <> | Subject | Re: [PATCH] Containment measures for slab objects on scatter gather lists |
| |
On Fri, 29 Jun 2007 13:45:29 -0700 Andrew Morton <akpm@linux-foundation.org> wrote:
> On Fri, 29 Jun 2007 13:16:57 +0100 > Alan Cox <alan@lxorguk.ukuu.org.uk> wrote: > > > > If those operations involve modifying that slab page's pageframe then what > > > stops concurrent dma'ers from stomping on each other's changes? As in: > > > why aren't we already buggy? > > > > Or DMA operations falling out with CPU operations in the same memory > > area. Not all platforms have hardware consistency and some will blat the > > entire page out of cache. > > Is that just a performance problem, or can data be lost here? It depends > on the meaning of "blat": writeback? invalidate? More details, please.
Invalidate. Sorry didn't realise it they hadn't discovered that word down under.
If you've got something packing objects in tight we are going to have fun with cache handling simply because the CPU cache granularity may mean that the invalidate also invalidates a few bytes on (ie a 12 byte object will invalidate 16 bytes of memory) and you've just removed any CPU held changes in the start of the next object.
Alan - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |