Revision ad576e63e0c8b274a8558b8e05a6d0526e804dc0 authored by Nick Piggin on 05 May 2005, 23:15:46 UTC, committed by Linus Torvalds on 05 May 2005, 23:36:40 UTC
When running fsstress -v -d $DIR/tmp -n 1000 -p 1000 -l 2 on an ext2 filesystem with 1024 byte block size, on SMP i386 with 4096 byte page size over loopback to an image file on a tmpfs filesystem, I would very quickly hit BUG_ON(!buffer_async_write(bh)); in fs/buffer.c:end_buffer_async_write It seems that more than one request would be submitted for a given bh at a time. What would happen is the following: 2 threads doing __mpage_writepages on the same page. Thread 1 - lock the page first, and enter __block_write_full_page. Thread 1 - (eg.) mark_buffer_async_write on the first 2 buffers. Thread 1 - set page writeback, unlock page. Thread 2 - lock page, wait on page writeback Thread 1 - submit_bh on the first 2 buffers. => both requests complete, none of the page buffers are async_write, end_page_writeback is called. Thread 2 - wakes up. enters __block_write_full_page. Thread 2 - mark_buffer_async_write on (eg.) the last buffer Thread 1 - finds the last buffer has async_write set, submit_bh on that. Thread 2 - submit_bh on the last buffer. => oops. So change __block_write_full_page to explicitly keep track of the last bh we need to issue, so we don't touch anything after issuing the last request. Signed-off-by: Nick Piggin <nickpiggin@yahoo.com.au> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
1 parent f3ddbdc
File | Mode | Size |
---|---|---|
Makefile | -rw-r--r-- | 615 bytes |
bootmem.c | -rw-r--r-- | 10.5 KB |
fadvise.c | -rw-r--r-- | 2.4 KB |
filemap.c | -rw-r--r-- | 57.9 KB |
fremap.c | -rw-r--r-- | 6.3 KB |
highmem.c | -rw-r--r-- | 13.7 KB |
hugetlb.c | -rw-r--r-- | 6.1 KB |
internal.h | -rw-r--r-- | 486 bytes |
madvise.c | -rw-r--r-- | 6.4 KB |
memory.c | -rw-r--r-- | 59.1 KB |
mempolicy.c | -rw-r--r-- | 28.8 KB |
mempool.c | -rw-r--r-- | 7.7 KB |
mincore.c | -rw-r--r-- | 4.4 KB |
mlock.c | -rw-r--r-- | 5.3 KB |
mmap.c | -rw-r--r-- | 52.7 KB |
mprotect.c | -rw-r--r-- | 6.5 KB |
mremap.c | -rw-r--r-- | 10.4 KB |
msync.c | -rw-r--r-- | 5.5 KB |
nommu.c | -rw-r--r-- | 27.6 KB |
oom_kill.c | -rw-r--r-- | 7.4 KB |
page-writeback.c | -rw-r--r-- | 22.7 KB |
page_alloc.c | -rw-r--r-- | 54.9 KB |
page_io.c | -rw-r--r-- | 3.6 KB |
pdflush.c | -rw-r--r-- | 5.8 KB |
prio_tree.c | -rw-r--r-- | 6.3 KB |
readahead.c | -rw-r--r-- | 15.8 KB |
rmap.c | -rw-r--r-- | 22.6 KB |
shmem.c | -rw-r--r-- | 57.6 KB |
slab.c | -rw-r--r-- | 81.4 KB |
swap.c | -rw-r--r-- | 12.0 KB |
swap_state.c | -rw-r--r-- | 9.3 KB |
swapfile.c | -rw-r--r-- | 41.4 KB |
thrash.c | -rw-r--r-- | 2.6 KB |
tiny-shmem.c | -rw-r--r-- | 2.7 KB |
truncate.c | -rw-r--r-- | 8.9 KB |
vmalloc.c | -rw-r--r-- | 12.9 KB |
vmscan.c | -rw-r--r-- | 35.1 KB |
![swh spinner](/static/img/swh-spinner.gif)
Computing file changes ...