Hi!
On Fri, Oct 16, 2009 at 04:56:42PM +0200, Sergei Golubchik wrote:
> Hi, Sergey!
>
> On Oct 09, Sergey Vojtovich wrote:
> > #At file:///home/svoj/devel/bzr-mysql/mysql-5.1-bugteam-bug47073/ based on
> revid:mattias.jonsson@stripped
> >
> > 3146 Sergey Vojtovich 2009-10-09
> > BUG#47073 - valgrind errs, corruption,failed repair of partition,
> > low myisam_sort_buffer_size
> >
> > === modified file 'storage/myisam/sort.c'
> > --- a/storage/myisam/sort.c 2009-08-28 16:21:54 +0000
> > +++ b/storage/myisam/sort.c 2009-10-09 16:16:29 +0000
> > @@ -788,7 +788,11 @@ static int NEAR_F merge_many_buff(MI_SOR
> > cleanup:
> > close_cached_file(to_file); /* This holds old result */
> > if (to_file == t_file)
> > + {
>
> you forgot to add an assert here:
>
> DBUG_ASSERT(t_file2.type == WRITE_CACHE);
>
> as we discussed.
There are no excuses for this. Will fix in the follow-up patch.
> > *t_file=t_file2; /* Copy result file */
> > + t_file->current_pos= &t_file->write_pos;
> > + t_file->current_end= &t_file->write_end;
> > + }
> >
> > DBUG_RETURN(*maxbuffer >= MERGEBUFF2); /* Return 1 if interrupted
> */
> > } /* merge_many_buff */
>
> Why did you disable a part of the test case in the next patch ?
Was requested by Joro to make trees green. Will re-enable as soon as
a follow-up patch is approved, which for some reason wasn't handled by
the bugdb. The patch is:
http://lists.mysql.com/commits/86661
Regards,
Sergey
--
Sergey Vojtovich <svoj@stripped>
MySQL AB, Software Engineer
Izhevsk, Russia, www.mysql.com
