lkml.org 
[lkml]   [2004]   [Mar]   [21]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
SubjectRe: Linux 2.6.5-rc1
From
Date
Linus Torvalds <torvalds@osdl.org> writes:

> o UDF filesystem update

For some reason I don't understand, this makes the UDF filesystem lock
up when I write a bunch of mp3 files to a CDRW using the packet
writing patch. Both "cp" and pdflush get stuck in __down. Reverting
the semaphore changes as in the patch below makes the problem go away,
but it's probably not the right solution to re-introduce lock_kernel()
calls.

diff -puN fs/udf/file.c~udf fs/udf/file.c
--- linux/fs/udf/file.c~udf 2004-03-21 23:25:26.000000000 +0100
+++ linux-petero/fs/udf/file.c 2004-03-21 23:26:46.000000000 +0100
@@ -247,9 +247,9 @@ static int udf_release_file(struct inode
{
if (filp->f_mode & FMODE_WRITE)
{
- down(&inode->i_sem);
+ lock_kernel();
udf_discard_prealloc(inode);
- up(&inode->i_sem);
+ unlock_kernel();
}
return 0;
}
diff -puN fs/udf/inode.c~udf fs/udf/inode.c
--- linux/fs/udf/inode.c~udf 2004-03-21 23:25:53.000000000 +0100
+++ linux-petero/fs/udf/inode.c 2004-03-21 23:26:21.000000000 +0100
@@ -84,9 +84,9 @@ void udf_put_inode(struct inode * inode)
{
if (!(inode->i_sb->s_flags & MS_RDONLY))
{
- down(&inode->i_sem);
+ lock_kernel();
udf_discard_prealloc(inode);
- up(&inode->i_sem);
+ unlock_kernel();
}
}

pdflush D CA33461B 5504 6 3 8 5 (L-TLB)
c5f2bd68 00000046 c3ee2740 ca33461b 0000009f c5f2be8c c5f2bd40 c5478c3c
c10822a8 c5478cf8 ca33461b 0000009f c3ee2740 00097523 ca33461b 0000009f
c5f2d8e0 c5f2dad0 c5478cbc c5f2a000 00000286 c5f2bdc4 c0106303 00000000
Call Trace:
[<c0106303>] __down+0x143/0x360
[<ca8d6710>] udf_writepage+0x0/0x30 [udf]
[<c011a360>] default_wake_function+0x0/0x20
[<c0106a27>] __down_failed+0xb/0x14
[<ca8db8b5>] .text.lock.inode+0x5/0x20 [udf]
[<c019cce6>] iput+0x76/0x90
[<c01a677f>] sync_sb_inodes+0x26f/0x420
[<ca8e0a98>] udf_write_super+0x178/0x1d0 [udf]
[<c01a6acb>] writeback_inodes+0x19b/0x4b0
[<c014e1eb>] wb_kupdate+0x10b/0x190
[<c014e0e0>] wb_kupdate+0x0/0x190
[<c014ebdc>] __pdflush+0x24c/0x660
[<c0119ddb>] schedule+0x3cb/0x900
[<c014f001>] pdflush+0x11/0x20
[<c014e0e0>] wb_kupdate+0x0/0x190
[<c013ceda>] kthread+0xaa/0xb0
[<c014eff0>] pdflush+0x0/0x20
[<c013ce30>] kthread+0x0/0xb0
[<c0105455>] kernel_thread_helper+0x5/0x10
cp D C10573C8 3988 1701 1083 (NOTLB)
c1783b54 00000082 4419442a c10573c8 07d4103c 12171503 0d390f23 c5478c3c
07d4103c 12171503 440f442a c1783b40 07d4103c 0001d27c e17f96d9 0000009a
c1856ca0 c1856e90 c5478cbc c1782000 00000286 c1783bb0 c0106303 c1783b90
Call Trace:
[<c0106303>] __down+0x143/0x360
[<ca8d9883>] udf_write_inode+0xa3/0x1c0 [udf]
[<ca8d6710>] udf_writepage+0x0/0x30 [udf]
[<c011a360>] default_wake_function+0x0/0x20
[<c0106a27>] __down_failed+0xb/0x14
[<ca8db8b5>] .text.lock.inode+0x5/0x20 [udf]
[<c019cce6>] iput+0x76/0x90
[<c01a677f>] sync_sb_inodes+0x26f/0x420
[<c0200000>] nfs_release+0xc0/0x200
[<c01a6acb>] writeback_inodes+0x19b/0x4b0
[<c014cfc9>] get_page_state+0x19/0x20
[<c014dcc4>] get_dirty_limits+0x14/0xd0
[<c014de56>] balance_dirty_pages+0xd6/0x190
[<c01498c4>] generic_file_aio_write_nolock+0x564/0xbf0
[<c0147fbf>] do_generic_mapping_read+0x1bf/0x3e0
[<c0148542>] generic_file_aio_read+0x52/0x70
[<c0149fc8>] generic_file_write_nolock+0x78/0x90
[<c01fd737>] nfs_file_read+0xb7/0x110
[<c0173657>] do_sync_read+0x87/0xc0
[<c014a0d5>] generic_file_write+0x55/0x70
[<ca8d5b44>] udf_file_write+0x44/0x180 [udf]
[<c017391f>] vfs_write+0xaf/0x120
[<c0173a2f>] sys_write+0x3f/0x60
[<c010835b>] syscall_call+0x7/0xb
--
Peter Osterlund - petero2@telia.com
http://w1.894.telia.com/~u89404340
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-03-22 14:01    [W:0.231 / U:0.136 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site