Fixed bug #11088: a crash for queries with GROUP BY a BLOB column
+ COUNT(DISTINCT...) due to an attempt to allocate a too large
buffer for the BLOB field.
Now the size of the buffer is limited by max_sort_length.
group_by.test, group_by.result:
Added a test case for bug #11088.
produces invalid query results
mi_key.c:
well_formed_length should be executed before space trimming, not after.
ctype_utf8.test:
ctype_utf8.result:
adding test.
when doing "make install" (they are not needed/useful) outside the build tree. Added
"mysql_client_test" to the "MySQL-bench" RPM.
- some small RPM spec file improvements (more variables than hard-coded values, cleanups)
Try kill all children to the mysql-test-run process when exiting
mtr_process.pl:
Added new function mtr_exit() that try kill all children to the mysql-test-run process when exiting
mysql-test-run.pl:
Try kill all children to the mysql-test-run process when exiting
New --vardir caused problems on Windows
Changed faulty calls to error() to mtr_error()
BUG#9826 Server crash on schema change ("drop table", "alter table") with NDB
See note on 10948 for detailed explanation.
struct a { void* a; long long b[1]; };
on PPC32 (and 32bit pa risc): 16
on x86: 12
so a malloc(sizeof(void*)+extra_bits) is wrong.
Assuming that the long long is 64 bit aligned as non-aligned 64bit accesses
are rather expensive on ppc.
Thanks to paulus for doing the PPC port of valgrind.
Without which I would no doubt still be trying to find this.
After review fix
Copy from internal state to share state only when in lock write
mode (happens only when lock table x write has been performed since
update_state_info is only called when holding a TL_READ_NO_INSERT
lock normally. Previous patch would have failed in combination with
delayed writes.
Ensure that 'null_value' is not accessed before val() is called in FIELD() functions
Fixed initialization of key maps. This fixes some problems with keys when you have more than 64 keys
Fixed that ROLLUP don't always create a temporary table. This fix ensures that func_gconcat.test results are now predictable
Analyze table corrupts the state on
data_file_length, records, index_file_length...
by writing the shared state when there is an updated internal
state due to inserts or deletes
Fixed by synching the shared state with the internal state before
writing it to disk
Added test cases of 2 error cases and a normal case in new
analyze test case
- support additional platforms (OpenBSD, Darwin)
- make use of /proc/cpuinfo CPU flag information
(needed to distinguish EMT64 Xeons from normal ones)
- chose the right CPU related options depending on
GCC version and target architecture
- check if the chosen CPU type is actually supported
by the GCC version , fallback to CPUs predecessor
until test compile succeeds
adding test
field.cc:
bug#10714 Inserting double value into utf8 column crashes server:
sprintf was executed with too big length, which caused
crash on some Windows platforms.
