On Mon, 2007-07-16 at 17:03 -0500, James Bottomley wrote: > On Tue, 2007-07-17 at 07:49 +1000, Benjamin Herrenschmidt wrote: > > > No ... that was the point of flush_kernel_dcache_page(). The page in > > > question is page cache backed and contains user mappings. However, the > > > block layer has already done a flush_dcache_page() in get_user_pages() > > > and the user shouldn't be touching memory under I/O (unless they want > > > self induced aliasing problems) so we're free to assume all the user > > > cachelines are purged, hence all we have to do is flush the kernel alias > > > to bring the page up to date and make the users see it correctly. > > > > The block layer will have done that even in the swap-out path ? (Just > > asking... I'm not very familiar with the block layer) > > Er ... not really, this is the I/O path for user initiated I/O. The > page out path, by definition, can't have any extant user mappings. For > page out, the relevant page is flushed before its mapping is detached, > and then it can be paged to the backing store (or for anonymous pages to > the swap device) when no mappings remain. Ok, thanks. Ben. - To unsubscribe from this list: send the line "unsubscribe linux-scsi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
