On Tue, 2012-07-31 at 05:43 +0100, Ben Hutchings wrote:
> This is the start of the stable review cycle for the 3.2.25 release.
> There are 73 patches in this series, which will be posted as responses
> to this one. If anyone has any issues with these being applied, please
> let me know.
>
> Responses should be made by Thu Aug 2 10:00:00 UTC 2012.
> Anything received after that time might be too late.
>
> A combined patch relative to 3.2.24 will be posted as an additional
> response to this, and the diffstat can be found below.
I tested this against configs I normally test against my own code, and I
hit this bug:
[ 37.030215]
=============================================================================
[ 37.031170] BUG blkdev_queue: Poison overwritten
[ 37.031170]
-----------------------------------------------------------------------------
[ 37.031170]
[ 37.031170] INFO: 0xffff8800757287b0-0xffff8800757287b0. First byte 0x6a
instead of 0x6b
[ 37.031170] INFO: Allocated in blk_alloc_queue_node+0x25/0x1fb age=3399
cpu=1 pid=1
[ 37.031170] __slab_alloc+0x2e4/0x365
[ 37.031170] kmem_cache_alloc_node+0x92/0x18d
[ 37.031170] blk_alloc_queue_node+0x25/0x1fb
[ 37.031170] blk_init_queue_node+0x24/0x5c
[ 37.031170] blk_init_queue+0x11/0x13
[ 37.031170] floppy_init+0x78/0x5d9^M
[ 37.031170] do_one_initcall+0x7f/0x140
[ 37.092043] kernel_init+0xc9/0x143
[ 37.092043] kernel_thread_helper+0x4/0x10
[ 37.092043] INFO: Freed in blk_release_queue+0x86/0x8b age=78 cpu=0 pid=1
[ 37.092043] __slab_free+0x38/0x377
[ 37.092043] kmem_cache_free+0xf7/0x155
[ 37.092043] blk_release_queue+0x86/0x8b
[ 37.092043] kobject_cleanup+0xc4/0xeb
[ 37.092043] kobject_release+0xd/0xf
[ 37.092043] kobject_put+0x4a/0x4f
[ 37.092043] blk_cleanup_queue+0x159/0x162
[ 37.092043] floppy_init+0x5b6/0x5d9
[ 37.092043] do_one_initcall+0x7f/0x140
[ 37.092043] kernel_init+0xc9/0x143
[ 37.092043] kernel_thread_helper+0x4/0x10
[ 37.092043] INFO: Slab 0xffffea0001d5ca00 objects=9 used=9 fp=0x
(null) flags=0x100000000004080
[ 37.092043] INFO: Object 0xffff880075728000 @offset=0 fp=0xffff88007572e700
[...]
[ 37.092043] Object ffff880075728b80: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b
6b 6b 6b kkkkkkkkkkkkkkkk
[ 37.092043] Object ffff880075728b90: 6b 6b 6b 6b 6b 6b 6b a5
kkkkkkk.
[ 37.092043] Redzone ffff880075728b98: bb bb bb bb bb bb bb bb
........
[ 37.092043] Padding ffff880075728cd8: 5a 5a 5a 5a 5a 5a 5a 5a
ZZZZZZZZ
[ 37.092043] Pid: 1, comm: swapper/0 Not tainted 3.2.0-test+ #1
[ 37.092043] Call Trace:
[ 37.092043] [<ffffffff8114d18b>] ? print_section+0x3d/0x3f
[ 37.092043] [<ffffffff8114db29>] print_trailer+0x10a/0x113
[ 37.092043] [<ffffffff8114df53>] check_bytes_and_report+0xb1/0xea
[ 37.092043] [<ffffffff8114e050>] check_object+0xc4/0x1fc
[ 37.092043] [<ffffffff813f7c63>] ? blk_alloc_queue_node+0x25/0x1fb
[ 37.092043] [<ffffffff813f7c63>] ? blk_alloc_queue_node+0x25/0x1fb
[ 37.092043] [<ffffffff819e465e>] alloc_debug_processing+0xa7/0x14a
[ 37.092043] [<ffffffff819e4fb1>] __slab_alloc+0x2e4/0x365
[ 37.092043] [<ffffffff813f7c63>] ? blk_alloc_queue_node+0x25/0x1fb
[ 37.092043] [<ffffffff810b113d>] ? trace_hardirqs_on+0xd/0xf
[ 37.092043] [<ffffffff8114fd71>] kmem_cache_alloc_node+0x92/0x18d
[ 37.092043] [<ffffffff810b10f9>] ? trace_hardirqs_on_caller+0x12f/0x166
[ 37.092043] [<ffffffff813f7c63>] ? blk_alloc_queue_node+0x25/0x1fb
[ 37.092043] [<ffffffff813f7c63>] blk_alloc_queue_node+0x25/0x1fb
[ 37.092043] [<ffffffff813f7e4a>] blk_alloc_queue+0x11/0x13
[ 37.092043] [<ffffffff81550489>] brd_alloc+0x79/0x185
[ 37.092043] [<ffffffff82264a24>] brd_init+0xc6/0x19c
[ 37.092043] [<ffffffff8226495e>] ? floppy_init+0x5d9/0x5d9
[ 37.092043] [<ffffffff8100020f>] do_one_initcall+0x7f/0x140
[ 37.092043] [<ffffffff82234c44>] kernel_init+0xc9/0x143
[ 37.092043] [<ffffffff81a14874>] kernel_thread_helper+0x4/0x10
[ 37.092043] [<ffffffff81a0c5f4>] ? retint_restore_args+0x13/0x13
[ 37.092043] [<ffffffff82234b7b>] ? start_kernel+0x3af/0x3af
[ 37.092043] [<ffffffff81a14870>] ? gs_change+0x13/0x13
[ 37.092043] FIX blkdev_queue: Restoring
0xffff8800757287b0-0xffff8800757287b0=0x6b
I then checked against 3.2.24 and it bugged too, as well as vanilla 3.2.
I then checked against 3.3 and it did not bug. I kicked off a 'reverse'
bisect with ktest (bad is good and good is bad) and found the fix:
commit 3f9a5aabd0a9fe0e0cd308506f48963d79169aa7
Author: Vivek Goyal <vgo...@redhat.com>
floppy: Cleanup disk->queue before caling put_disk() if add_disk() was never
called
I applied it against v3.2.24 and it solved the bug. I did not apply it
against 25-rc1, but I'm pretty sure it should work for that too.
I haven't checked 3.0 if that has an issue with my config.
Anyway, please apply this patch to stable 3.2. Either for 25 or for 26.
Thanks!
-- Steve
commit 3f9a5aabd0a9fe0e0cd308506f48963d79169aa7
Author: Vivek Goyal <vgo...@redhat.com>
Date: Wed Feb 8 20:03:38 2012 +0100
floppy: Cleanup disk->queue before caling put_disk() if add_disk() was
never called
add_disk() takes gendisk reference on request queue. If driver failed during
initialization and never called add_disk() then that extra reference is not
taken. That reference is put in put_disk(). floppy driver allocates the
disk, allocates queue, sets disk->queue and then relizes that floppy
controller is not present. It tries to tear down everything and tries to
put a reference down in put_disk() which was never taken.
In such error cases cleanup disk->queue before calling put_disk() so that
we never try to put down a reference which was never taken in first place.
Reported-and-tested-by: Suresh Jayaraman <sjayara...@suse.com>
Tested-by: Dirk Gouders <goud...@et.bocholt.fh-gelsenkirchen.de>
Signed-off-by: Vivek Goyal <vgo...@redhat.com>
Acked-by: Tejun Heo <t...@kernel.org>
Signed-off-by: Jens Axboe <ax...@kernel.dk>
diff --git a/drivers/block/floppy.c b/drivers/block/floppy.c
index 510fb10..401ba78 100644
--- a/drivers/block/floppy.c
+++ b/drivers/block/floppy.c
@@ -4368,8 +4368,14 @@ out_unreg_blkdev:
out_put_disk:
while (dr--) {
del_timer_sync(&motor_off_timer[dr]);
- if (disks[dr]->queue)
+ if (disks[dr]->queue) {
blk_cleanup_queue(disks[dr]->queue);
+ /*
+ * put_disk() is not paired with add_disk() and
+ * will put queue reference one extra time. fix it.
+ */
+ disks[dr]->queue = NULL;
+ }
put_disk(disks[dr]);
}
return err;
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
