projects / linux-2.6-block.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: d0a22a4) | patch
gfs2: read journal in large chunks
authorAbhi Das <adas@redhat.com>
Thu, 2 May 2019 19:17:40 +0000 (14:17 -0500)
committerAndreas Gruenbacher <agruenba@redhat.com>
Tue, 7 May 2019 21:39:15 +0000 (23:39 +0200)
commitf4686c26ecc34e8e458b8235f0af5198c9b13bfd
tree6768f35e9f263165a70e22b3f5d164e0b400e856tree | snapshot
parentd0a22a4b03b8475b7aa3fa41243c26c291407844commit | diff
gfs2: read journal in large chunks

Use bios to read in the journal into the address space of the journal inode
(jd_inode), sequentially and in large chunks.  This is faster for locating the
journal head that the previous binary search approach.  When performing
recovery, we keep the journal in the address space until recovery is done,
which further speeds up things.

Signed-off-by: Abhi Das <adas@redhat.com>
Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
fs/gfs2/glops.c diff | blob | blame | history
fs/gfs2/log.c diff | blob | blame | history
fs/gfs2/lops.c diff | blob | blame | history
fs/gfs2/lops.h diff | blob | blame | history
fs/gfs2/ops_fstype.c diff | blob | blame | history
fs/gfs2/recovery.c diff | blob | blame | history
fs/gfs2/recovery.h diff | blob | blame | history
fs/gfs2/super.c diff | blob | blame | history
Linux 6.x block layer and io_uring tree(s)