Jörn Engel wrote:
> And the last one is xfs_mapping_buftarg(). I am completely at a loss.
> As far as I can tell, inode allocated, partially initiated and -
> leaked. Am I missing something?
It's not leaked... the new inode's mapping is saved in the buftarg:
btp->bt_mapping = mapping;
and then eventually the inode is put/freed in xfs_free_buftarg():
iput(btp->bt_mapping->host);
The inode's mapping itself is used in several places:
File Function Line
0 xfs/linux-2.6/xfs_buf.h <global> 84 struct address_space *bt_mapping;
1 xfs/linux-2.6/xfs_buf.c _xfs_buf_lookup_pages 345 struct address_space *mapping = bp->b_target->bt_mapping;
2 xfs/linux-2.6/xfs_buf.c xfs_buf_readahead 671 bdi = target->bt_mapping->backing_dev_info;
3 xfs/linux-2.6/xfs_buf.c xfs_buf_lock 906 blk_run_address_space(bp->b_target->bt_mapping);
4 xfs/linux-2.6/xfs_buf.c xfs_buf_wait_unpin 978 blk_run_address_space(bp->b_target->bt_mapping);
5 xfs/linux-2.6/xfs_buf.c xfs_buf_iowait 1291 blk_run_address_space(bp->b_target->bt_mapping);
6 xfs/linux-2.6/xfs_buf.c xfs_free_buftarg 1451 iput(btp->bt_mapping->host);
7 xfs/linux-2.6/xfs_buf.c xfs_mapping_buftarg 1545 btp->bt_mapping = mapping;
8 xfs/linux-2.6/xfs_buf.c xfsbufd 1728 blk_run_address_space(target->bt_mapping);
9 xfs/linux-2.6/xfs_buf.c xfs_flush_buftarg 1801 blk_run_address_space(target->bt_mapping);
-Eric
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
- References:
- Re: [PATCH] make last_inode counter in new_inode 32-bit on kernels that offer x86 compatability
- From: Matthew Wilcox <[email protected]>
- Re: [PATCH] make last_inode counter in new_inode 32-bit on kernels that offer x86 compatability
- From: Jeff Layton <[email protected]>
- Re: [PATCH] make last_inode counter in new_inode 32-bit on kernels that offer x86 compatability
- From: Jörn Engel <[email protected]>
- Re: [PATCH] make last_inode counter in new_inode 32-bit on kernels that offer x86 compatability
- From: Eric Sandeen <[email protected]>
- Re: [PATCH] make last_inode counter in new_inode 32-bit on kernels that offer x86 compatability
- From: Jörn Engel <[email protected]>
- Re: [PATCH] make last_inode counter in new_inode 32-bit on kernels that offer x86 compatability
- From: Eric Sandeen <[email protected]>
- Re: [PATCH] make last_inode counter in new_inode 32-bit on kernels that offer x86 compatability
- From: Jeff Layton <[email protected]>
- Re: [PATCH] make last_inode counter in new_inode 32-bit on kernels that offer x86 compatability
- From: Jörn Engel <[email protected]>
- Re: [PATCH] make last_inode counter in new_inode 32-bit on kernels that offer x86 compatability
- From: Jörn Engel <[email protected]>
- Re: [PATCH] make last_inode counter in new_inode 32-bit on kernels that offer x86 compatability
- From: Jörn Engel <[email protected]>
- Re: [PATCH] make last_inode counter in new_inode 32-bit on kernels that offer x86 compatability
- From: Jörn Engel <[email protected]>
- Re: [PATCH] make last_inode counter in new_inode 32-bit on kernels that offer x86 compatability
- Prev by Date: Re: [PATCH] make last_inode counter in new_inode 32-bit on kernels that offer x86 compatability
- Next by Date: Re: + sched-use-tasklet-to-call-balancing.patch added to -mm tree
- Previous by thread: Re: [PATCH] make last_inode counter in new_inode 32-bit on kernels that offer x86 compatability
- Next by thread: Re: [PATCH] make last_inode counter in new_inode 32-bit on kernels that offer x86 compatability
- Index(es):
 |