database/mariadb
1
0
Fork 0
mirror of https://github.com/MariaDB/server.git synced 2025-07-30 16:24:05 +03:00
Code Activity

Fixed bug #31310.

Browse Source 
Locked rows of the InnoDB storage was silently skipped in the read-committed
isolation level.

QUICK_RANGE_SELECT for unique ranges lacks second (blocking) read
of the record that was read semi-consistently and just skip it.

The handler::read_multi_range_next method has been modified
to retry previous unique range if the previous read was
semi-consistent.
This commit is contained in:
gshchepa/uchum@gleb.loc
2007-10-09 01:07:15 +05:00
parent 5df890e524
commit 09f6306ac3
3 changed files with 210 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
sql/handler.cc
View File

@ -3145,6 +3145,8 @@ int handler::read_multi_range_next(KEY_MULTI_RANGE **found_range_p)
}
else
{
if (was_semi_consistent_read())
goto scan_it_again;
/*
We need to set this for the last range only, but checking this
condition is more expensive than just setting the result code.
@ -3152,10 +3154,10 @@ int handler::read_multi_range_next(KEY_MULTI_RANGE **found_range_p)
result= HA_ERR_END_OF_FILE;
}
multi_range_curr++;
scan_it_again:
/* Try the next range(s) until one matches a record. */
for (multi_range_curr++;
multi_range_curr < multi_range_end;
multi_range_curr++)
for (; multi_range_curr < multi_range_end; multi_range_curr++)
{
result= read_range_first(multi_range_curr->start_key.keypart_map ?
&multi_range_curr->start_key : 0,