[lkml]   [2008]   [Mar]   [11]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
Patch in this message
SubjectRe: [RFC, PATCH 4/6] jbd: refactor nested journal log recovery loop into separate functions
On Tue, Mar 11, 2008 at 03:35:50PM +0100, Jan Kara wrote:
> Hmm, if I read your patch correctly, previously we aborted journal
> replay when we found a block with unknown block magic but now we
> continue replaying. Why have you done such change? And similarly when
> some error happened when parsing revoke records block...

You are right, that was an error on my part. Thanks for your careful
review. Please find below an incremental patch. I'll roll this into the
main patch before posting the next version.


"I never could learn to drink that blood and call it wine" - Bob Dylan

diff --git a/fs/jbd/recovery.c b/fs/jbd/recovery.c
index 453c5fe..34db55a 100644
--- a/fs/jbd/recovery.c
+++ b/fs/jbd/recovery.c
@@ -673,16 +673,22 @@ static int do_one_pass(journal_t *journal,
/* If we aren't in the REVOKE pass, then we can
* just skip over this block. */
- if (pass == PASS_REVOKE) {
- err = scan_revoke_records(
- journal, bh, next_commit_ID, info);
+ if (pass != PASS_REVOKE)
+ break;
+ err = scan_revoke_records(journal, bh,
+ next_commit_ID, info);
+ if (err) {
+ brelse(bh);
+ goto failed;

jbd_debug(3, "Unrecognised magic %d, end of scan.\n",
- break;
+ brelse(bh);
+ goto done;

@@ -695,6 +701,7 @@ static int do_one_pass(journal_t *journal,
goto failed;

* We broke out of the log scan loop: either we came to the
* known end of the log or we found an unexpected block in the

 \ /
  Last update: 2008-03-12 02:05    [W:0.056 / U:0.932 seconds]
©2003-2018 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site