Message ID | 1353803794-11593-3-git-send-email-tytso@mit.edu |
---|---|
State | Superseded, archived |
Headers | show |
On Sat, 24 Nov 2012, Theodore Ts'o wrote: > Date: Sat, 24 Nov 2012 19:36:30 -0500 > From: Theodore Ts'o <tytso@mit.edu> > To: Ext4 Developers List <linux-ext4@vger.kernel.org> > Cc: Theodore Ts'o <tytso@mit.edu> > Subject: [PATCH 2/6] e2fsck: optimize pass1 for CPU time > > Optimize e2fsck pass 1 by marking entire extents as being in use at a > time, instead of block by block. This optimization only works for > non-bigalloc file systems for now (it's tricky to handle bigalloc file > systems since this code is also responsible for dealing with blocks > that are not correctly aligned within a cluster). When the > optimization works, the CPU savings can be significant: up to two or > three CPU minutes for a full 4T disk. > > Addresses-Google-Bug: #7534813 Looks good. Reviewed-by: Lukas Czerner <lczerner@redhat.com> > > Signed-off-by: "Theodore Ts'o" <tytso@mit.edu> > --- > e2fsck/pass1.c | 18 ++++++++++++++++-- > 1 file changed, 16 insertions(+), 2 deletions(-) > > diff --git a/e2fsck/pass1.c b/e2fsck/pass1.c > index 78fbe8d..cc00e0f 100644 > --- a/e2fsck/pass1.c > +++ b/e2fsck/pass1.c > @@ -1432,6 +1432,16 @@ static _INLINE_ void mark_block_used(e2fsck_t ctx, blk64_t block) > } > } > > +static _INLINE_ void mark_blocks_used(e2fsck_t ctx, blk64_t block, > + unsigned int num) > +{ > + if (ext2fs_test_block_bitmap_range2(ctx->block_found_map, block, num)) > + ext2fs_mark_block_bitmap_range2(ctx->block_found_map, block, num); > + else > + while (num--) > + mark_block_used(ctx, block++); > +} > + > /* > * Adjust the extended attribute block's reference counts at the end > * of pass 1, either by subtracting out references for EA blocks that > @@ -1867,11 +1877,15 @@ static void scan_extent_node(e2fsck_t ctx, struct problem_context *pctx, > goto failed_add_dir_block; > } > } > + if (!ctx->fs->cluster_ratio_bits) { > + mark_blocks_used(ctx, extent.e_pblk, extent.e_len); > + pb->num_blocks += extent.e_len; > + } > for (blk = extent.e_pblk, blockcnt = extent.e_lblk, i = 0; > i < extent.e_len; > blk++, blockcnt++, i++) { > - if (!(ctx->fs->cluster_ratio_bits && > - pb->previous_block && > + if (ctx->fs->cluster_ratio_bits && > + !(pb->previous_block && > (EXT2FS_B2C(ctx->fs, blk) == > EXT2FS_B2C(ctx->fs, pb->previous_block)) && > (blk & EXT2FS_CLUSTER_MASK(ctx->fs)) == > -- To unsubscribe from this list: send the line "unsubscribe linux-ext4" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
diff --git a/e2fsck/pass1.c b/e2fsck/pass1.c index 78fbe8d..cc00e0f 100644 --- a/e2fsck/pass1.c +++ b/e2fsck/pass1.c @@ -1432,6 +1432,16 @@ static _INLINE_ void mark_block_used(e2fsck_t ctx, blk64_t block) } } +static _INLINE_ void mark_blocks_used(e2fsck_t ctx, blk64_t block, + unsigned int num) +{ + if (ext2fs_test_block_bitmap_range2(ctx->block_found_map, block, num)) + ext2fs_mark_block_bitmap_range2(ctx->block_found_map, block, num); + else + while (num--) + mark_block_used(ctx, block++); +} + /* * Adjust the extended attribute block's reference counts at the end * of pass 1, either by subtracting out references for EA blocks that @@ -1867,11 +1877,15 @@ static void scan_extent_node(e2fsck_t ctx, struct problem_context *pctx, goto failed_add_dir_block; } } + if (!ctx->fs->cluster_ratio_bits) { + mark_blocks_used(ctx, extent.e_pblk, extent.e_len); + pb->num_blocks += extent.e_len; + } for (blk = extent.e_pblk, blockcnt = extent.e_lblk, i = 0; i < extent.e_len; blk++, blockcnt++, i++) { - if (!(ctx->fs->cluster_ratio_bits && - pb->previous_block && + if (ctx->fs->cluster_ratio_bits && + !(pb->previous_block && (EXT2FS_B2C(ctx->fs, blk) == EXT2FS_B2C(ctx->fs, pb->previous_block)) && (blk & EXT2FS_CLUSTER_MASK(ctx->fs)) ==
Optimize e2fsck pass 1 by marking entire extents as being in use at a time, instead of block by block. This optimization only works for non-bigalloc file systems for now (it's tricky to handle bigalloc file systems since this code is also responsible for dealing with blocks that are not correctly aligned within a cluster). When the optimization works, the CPU savings can be significant: up to two or three CPU minutes for a full 4T disk. Addresses-Google-Bug: #7534813 Signed-off-by: "Theodore Ts'o" <tytso@mit.edu> --- e2fsck/pass1.c | 18 ++++++++++++++++-- 1 file changed, 16 insertions(+), 2 deletions(-)