lkml.org 
[lkml]   [2003]   [May]   [21]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
SubjectRe: [Ext2-devel] [RFC] probably bug in current ext3/jbd
From
Date
>>>>> Andrew Morton (AM) writes:

AM> Alex Tomas <bzzz@tmi.comex.ru> wrote:
>>
>> looks Andrew Morton should return BKL in ext3_get_block_handle() in -mm tree?
>> this BKL protects ext3_alloc_branch() -> ext3_alloc_block() -> ext3_new_block()
>> call chain. or we may implement new protection schema where each jh has some
>> reference alike 'used by transaction N'

AM> Can this be solved by spinlocking the relevant buffer_head, in a similar
AM> way to your recent changes to journal_add_journal_head()?

yes, we may protect it by such lock, but this lock have to be held all time
ext3_new_block() uses some b_committed_data because last one may be freed
during current ext3_new_block(). I don't think it's good.

probably, another solution is do not free b_committed_data if it's referenced
by new transaction and make copy from b_frozen instead of freeing it.

-
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 13:35    [W:0.046 / U:0.104 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site