On Fri, Nov 21, 2008 at 12:25:39AM +1100, Nick Piggin wrote: > > > - The page is sent to the block layer, which stores into the page. Some > > > block devices like 'brd' will potentially store via the kernel linear > > > map here, and they probably don't do enough cache flushing. > > > > btw., if someone is curious, here is another case of what may happen > > on VIPT systems when someone uses a "virtual" block device (like > > 'brd') as, heh, a swap :-) > > > > http://www.linux-mips.org/archives/linux-mips/2008-11/msg00038.html > > Right... Now I'm lacking knowledge when it comes to devices, but I > think it is probably reasonable for the block device layer to ensure > the physical memory is uptodate after it signals request completion. > > That is, there shouldn't be any potentially aliasing dirty lines. > Block devices which do any writeout via the kernel linear address > (eg. brd) should do a flush_dcache_page. It's better to avoid aliases than dealing with them by flushing. A way to avoid aliases whenever a page is mapped to userspace, one creates a mapping at a carefully choosen address that doesn't alias. On architectures with software reload TLBs such as MIPS that's very cheap and the entire cacheflush with all it's associated pains can go away. Right now MIPS uses such a mechanism: void *kmap_coherent(struct page *page, unsigned long addr); void kunmap_coherent(void); within the architecture private implementation but it could be use beyond that, probably on all architectures though I know that there would be some solvable issues on PARISC. Lightweight, no ordering constraints between kernel and userspace accesses, so also no locking needed. Does this look like a possible avenue? Ralf -- To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
