having looked at the oops below and studied the code a bit more i think
that the safest thing to do for 2.4.x to fix this race condition is to get
rid of the unsafe promote-to-front hash list traversal. i considered
other fixes (such as a separate spinlock for just the hash list, or a
small array of them to give some amount of concurrency) ... but in the
interests of stability the following patch seems the most appropriate.
besides... nobody has responded to my mail, so perhaps proposing this
patch to marcelo will cause folks to wake up and object :)
-dean
Signed-off-by: dean gaudet <dean@arctic.org>
--- linux-2.4.29/drivers/md/lvm-snap.c.orig 2005-03-25 22:03:43.000000000 -0800
+++ linux-2.4.29/drivers/md/lvm-snap.c 2005-03-30 01:46:17.000000000 -0800
@@ -119,7 +119,6 @@ static inline lv_block_exception_t *lvm_
unsigned long mask = lv->lv_snapshot_hash_mask;
int chunk_size = lv->lv_chunk_size;
lv_block_exception_t *ret;
- int i = 0;
hash_table =
&hash_table[hashfn(org_dev, org_start, mask, chunk_size)];
@@ -132,15 +131,9 @@ static inline lv_block_exception_t *lvm_
exception = list_entry(next, lv_block_exception_t, hash);
if (exception->rsector_org == org_start &&
exception->rdev_org == org_dev) {
- if (i) {
- /* fun, isn't it? :) */
- list_del(next);
- list_add(next, hash_table);
- }
ret = exception;
break;
}
- i++;
}
return ret;
}
On Thu, 17 Mar 2005, dean gaudet wrote:
> i seem to have run into an oops described in the archives -- in particular
> this thread from last year:
>
> https://www.redhat.com/archives/linux-lvm/2004-January/msg00110.html
>
> has there been any further work on this? any success stories for the #2
> patch mentioned in the above thread?
>
> my system is running 2.4.29 + linux-2.4.22-VFS-lock.patch ... it's a dual
> xeon w/hyperthreading. i've been using snapshots for over a year for
> nightly backups and this is the first time i've run into the oops.
>
> thanks
> -dean
>
> Mar 16 03:31:19 twinlark kernel: Unable to handle kernel NULL pointer dereference at virtual address 00000004
> Mar 16 03:31:19 twinlark kernel: printing eip:
> Mar 16 03:31:19 twinlark kernel: c021f30c
> Mar 16 03:31:19 twinlark kernel: *pde = 00000000
> Mar 16 03:31:19 twinlark kernel: Oops: 0002
> Mar 16 03:31:19 twinlark kernel: CPU: 3
> Mar 16 03:31:19 twinlark kernel: EIP: 0010:[lvm_snapshot_remap_block+220/256] Not tainted
> Mar 16 03:31:19 twinlark kernel: EFLAGS: 00010202
> Mar 16 03:31:19 twinlark kernel: eax: f8a3dde0 ebx: f8d36800 ecx: f8a31000 edx: 00000000
> Mar 16 03:31:19 twinlark kernel: esi: 00010180 edi: 00000001 ebp: 00000903 esp: e1483ce0
> Mar 16 03:31:19 twinlark kernel: ds: 0018 es: 0018 ss: 0018
> Mar 16 03:31:19 twinlark kernel: Process qmail-smtpd (pid: 18415, stackpage=e1483000)
> Mar 16 03:31:19 twinlark kernel: Stack: 00000038 00000903 00000000 f7ba9600 e94e5c00 e94e5d70 00010000 c021bb28
> Mar 16 03:31:19 twinlark kernel: e1483d34 e1483d2c 00010180 e94e5c00 00000001 c01c7662 00000001 f7ba9770
> Mar 16 03:31:19 twinlark kernel: f750c000 00010180 00000000 000101b8 000101b8 09030903 00003a00 f7535980
> Mar 16 03:31:19 twinlark kernel: Call Trace: [lvm_map+408/1136] [submit_bh+82/256] [lvm_make_request_fn+23/48] [generic_make_request+228/320] [submit_bh+82/256]
> Mar 16 03:31:19 twinlark kernel: [__refile_buffer+86/112] [sync_page_buffers+168/176] [try_to_free_buffers+331/368] [shrink_cache+827/1040] [shrink_caches+74/96] [try_to_free_pages_zone+98/240]
> Mar 16 03:31:19 twinlark kernel: [balance_classzone+77/496] [__alloc_pages+376/640] [filemap_nopage+494/560] [filemap_nopage+0/560] [do_no_page+330/640] [handle_mm_fault+117/256]
> Mar 16 03:31:19 twinlark kernel: [do_page_fault+968/1397] [do_mmap_pgoff+874/1488] [old_mmap+272/304] [filp_close+143/208] [do_page_fault+0/1397] [error_code+52/60]
> Mar 16 03:31:19 twinlark kernel:
> Mar 16 03:31:19 twinlark kernel: Code: 89 42 04 8b 03 89 48 04 89 01 89 59 04 89 0b 89 ca eb 9f b8
_______________________________________________
linux-lvm mailing list
linux-lvm@redhat.com
https://www.redhat.com/mailman/listinfo/linux-lvm
read the LVM HOW-TO at http://tldp.org/HOWTO/LVM-HOWTO/
