The patch titled readahead: trigger mmap sequential readahead on PG_readahead has been removed from the -mm tree. Its filename was readahead-trigger-mmap-sequential-readahead-on-pg_readahead.patch This patch was dropped because it was merged into mainline or a subsystem tree The current -mm tree may be found at http://userweb.kernel.org/~akpm/mmotm/ ------------------------------------------------------ Subject: readahead: trigger mmap sequential readahead on PG_readahead From: Wu Fengguang <fengguang.wu@xxxxxxxxx> Previously the mmap sequential readahead is triggered by updating ra->prev_pos on each page fault and compare it with current page offset. It costs dirtying the cache line on each _minor_ page fault. So remove the ra->prev_pos recording, and instead tag PG_readahead to trigger the possible sequential readahead. It's not only more simple, but also will work more reliably and reduce cache line bouncing on concurrent page faults on shared struct file. In the mosbench exim benchmark which does multi-threaded page faults on shared struct file, the ra->mmap_miss and ra->prev_pos updates are found to cause excessive cache line bouncing on tmpfs, which actually disabled readahead totally (shmem_backing_dev_info.ra_pages == 0). So remove the ra->prev_pos recording, and instead tag PG_readahead to trigger the possible sequential readahead. It's not only more simple, but also will work more reliably on concurrent reads on shared struct file. Signed-off-by: Wu Fengguang <fengguang.wu@xxxxxxxxx> Tested-by: Tim Chen <tim.c.chen@xxxxxxxxx> Reported-by: Andi Kleen <ak@xxxxxxxxxxxxxxx> Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> --- mm/filemap.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff -puN mm/filemap.c~readahead-trigger-mmap-sequential-readahead-on-pg_readahead mm/filemap.c --- a/mm/filemap.c~readahead-trigger-mmap-sequential-readahead-on-pg_readahead +++ a/mm/filemap.c @@ -1559,8 +1559,7 @@ static void do_sync_mmap_readahead(struc if (!ra->ra_pages) return; - if (VM_SequentialReadHint(vma) || - offset - 1 == (ra->prev_pos >> PAGE_CACHE_SHIFT)) { + if (VM_SequentialReadHint(vma)) { page_cache_sync_readahead(mapping, ra, file, offset, ra->ra_pages); return; @@ -1583,7 +1582,7 @@ static void do_sync_mmap_readahead(struc ra_pages = max_sane_readahead(ra->ra_pages); ra->start = max_t(long, 0, offset - ra_pages / 2); ra->size = ra_pages; - ra->async_size = 0; + ra->async_size = ra_pages / 4; ra_submit(ra, mapping, file); } @@ -1689,7 +1688,6 @@ retry_find: return VM_FAULT_SIGBUS; } - ra->prev_pos = (loff_t)offset << PAGE_CACHE_SHIFT; vmf->page = page; return ret | VM_FAULT_LOCKED; _ Patches currently in -mm which might be from fengguang.wu@xxxxxxxxx are origin.patch writeback-pass-writeback_control-down-to-move_expired_inodes.patch writeback-introduce-writeback_controlinodes_cleaned.patch writeback-try-more-writeback-as-long-as-something-was-written.patch writeback-the-kupdate-expire-timestamp-should-be-a-moving-target.patch writeback-sync-expired-inodes-first-in-background-writeback.patch writeback-sync-expired-inodes-first-in-background-writeback-fix.patch writeback-refill-b_io-iff-empty.patch writeback-split-inode_wb_list_lock-into-bdi_writebacklist_lock.patch writeback-split-inode_wb_list_lock-into-bdi_writebacklist_lock-fix.patch writeback-split-inode_wb_list_lock-into-bdi_writebacklist_lock-fix-fix.patch writeback-split-inode_wb_list_lock-into-bdi_writebacklist_lock-fix-fix-fix.patch writeback-elevate-queue_io-into-wb_writeback.patch writeback-introduce-wbctagged_sync-for-the-wb_sync_none-sync-stage.patch writeback-update-dirtied_when-for-synced-inode-to-prevent-livelock.patch writeback-avoid-extra-sync-work-at-enqueue-time.patch getdelays-show-average-cpu-io-swap-reclaim-delays.patch -- To unsubscribe from this list: send the line "unsubscribe mm-commits" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html