]> git.karo-electronics.de Git - karo-tx-linux.git/commit
md: do not compute parity unless it is on a failed drive
authorDan Williams <dan.j.williams@intel.com>
Fri, 6 Jun 2008 05:45:54 +0000 (22:45 -0700)
committerLinus Torvalds <torvalds@linux-foundation.org>
Fri, 6 Jun 2008 18:29:08 +0000 (11:29 -0700)
commitc337869d95011495fa181536786e74aa2d7ff031
tree786cdf119ce33641438cd841deb0c480dd92f900
parenta6d8113a986c66aeb379a26b6e0062488b3e59e1
md: do not compute parity unless it is on a failed drive

If a block is computed (rather than read) then a check/repair operation
may be lead to believe that the data on disk is correct, when infact it
isn't.  So only compute blocks for failed devices.

This issue has been around since at least 2.6.12, but has become harder to
hit in recent kernels since most reads bypass the cache.

echo repair > /sys/block/mdN/md/sync_action will set the parity blocks to the
correct state.

Cc: <stable@kernel.org>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
drivers/md/raid5.c