]> asedeno.scripts.mit.edu Git - linux.git/commit
inode: don't memset the inode address space twice
authorDave Chinner <dchinner@redhat.com>
Wed, 7 Mar 2018 01:30:34 +0000 (17:30 -0800)
committerDarrick J. Wong <darrick.wong@oracle.com>
Mon, 12 Mar 2018 03:27:56 +0000 (20:27 -0700)
commitae23395d8858a0c91de978a60b317ec8468b2aba
treee5fbc63f66da6f754dde9a57d8952b511190bffa
parenta78ee256c325ecfaec13cafc41b315bd4e1dd518
inode: don't memset the inode address space twice

Noticed when looking at why cycling 600k inodes/s through the inode
cache was taking a total of 8% cpu in memset() during inode
initialisation.  There is no need to zero the inode.i_data structure
twice.

This increases single threaded bulkstat throughput from ~200,000
inodes/s to ~220,000 inodes/s, so we save a substantial amount of
CPU time per inode init by doing this.

Signed-Off-By: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Carlos Maiolino <cmaiolino@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
fs/inode.c