« back to all changes in this revision
Viewing changes to tests/r/key_cache.result
- Committer: Stewart Smith
- Date: 2009-02-18 14:18:03 UTC
- mto: This revision was merged to the branch mainline in revision 889.
- Revision ID: stewart@flamingspork.com-20090218141803-o3nh9vi7s4jkq5rp
Fix key_cache test for 32bit.
The key cache block structure has pointers in it.
On 32 and 64 bit, these are different sizes.
There's even a comment in the test to indicate this.
The test file should replace all possible numbers with KEY_BLOCKS_UNUSED, but this was messed up in fixing the
test case for Drizzle. This cset fixes that.
Currently tested on x86 and x86-64. Pretty sure (99.9%) that this will cover all platforms involved.
Our numbers are different from current MySQL 6 for some random reason that I don't really care about.
Number of blocks * block_size roughly equals key cache size - so i'm happy.
The key cache block structure has pointers in it.
On 32 and 64 bit, these are different sizes.
There's even a comment in the test to indicate this.
The test file should replace all possible numbers with KEY_BLOCKS_UNUSED, but this was messed up in fixing the
test case for Drizzle. This cset fixes that.
Currently tested on x86 and x86-64. Pretty sure (99.9%) that this will cover all platforms involved.
Our numbers are different from current MySQL 6 for some random reason that I don't really care about.
Number of blocks * block_size roughly equals key cache size - so i'm happy.
- files modified:
- tests/r/key_cache.result
added
removed
tests/r/key_cache.result
