aboutsummaryrefslogtreecommitdiff
path: root/fs/btrfs/file-item.c
diff options
context:
space:
mode:
authorChris Mason <chris.mason@oracle.com>2008-07-31 15:42:53 -0400
committerChris Mason <chris.mason@oracle.com>2008-09-25 11:04:05 -0400
commit61b4944018449003ac5f9757f4d125dce519cf51 (patch)
tree553855996c641a945344db870b6dfd0d2d02086e /fs/btrfs/file-item.c
parent37d1aeee3990385e9bb436c50c2f7e120a668df6 (diff)
Btrfs: Fix streaming read performance with checksumming on
Large streaming reads make for large bios, which means each entry on the list async work queues represents a large amount of data. IO congestion throttling on the device was kicking in before the async worker threads decided a single thread was busy and needed some help. The end result was that a streaming read would result in a single CPU running at 100% instead of balancing the work off to other CPUs. This patch also changes the pre-IO checksum lookup done by reads to work on a per-bio basis instead of a per-page. This results in many extra btree lookups on large streaming reads. Doing the checksum lookup right before bio submit allows us to reuse searches while processing adjacent offsets. Signed-off-by: Chris Mason <chris.mason@oracle.com>
Diffstat (limited to 'fs/btrfs/file-item.c')
-rw-r--r--fs/btrfs/file-item.c77
1 files changed, 77 insertions, 0 deletions
diff --git a/fs/btrfs/file-item.c b/fs/btrfs/file-item.c
index 2311061f070..a5ff19b34b2 100644
--- a/fs/btrfs/file-item.c
+++ b/fs/btrfs/file-item.c
@@ -134,6 +134,83 @@ int btrfs_lookup_file_extent(struct btrfs_trans_handle *trans,
return ret;
}
+int btrfs_lookup_bio_sums(struct btrfs_root *root, struct inode *inode,
+ struct bio *bio)
+{
+ u32 sum;
+ struct bio_vec *bvec = bio->bi_io_vec;
+ int bio_index = 0;
+ u64 offset;
+ u64 item_start_offset = 0;
+ u64 item_last_offset = 0;
+ u32 diff;
+ int ret;
+ struct btrfs_path *path;
+ struct btrfs_csum_item *item = NULL;
+ struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
+
+ path = btrfs_alloc_path();
+ path->reada = 2;
+
+ WARN_ON(bio->bi_vcnt <= 0);
+
+ while(bio_index < bio->bi_vcnt) {
+ offset = page_offset(bvec->bv_page) + bvec->bv_offset;
+ ret = btrfs_find_ordered_sum(inode, offset, &sum);
+ if (ret == 0)
+ goto found;
+
+ if (!item || offset < item_start_offset ||
+ offset >= item_last_offset) {
+ struct btrfs_key found_key;
+ u32 item_size;
+
+ if (item)
+ btrfs_release_path(root, path);
+ item = btrfs_lookup_csum(NULL, root, path,
+ inode->i_ino, offset, 0);
+ if (IS_ERR(item)) {
+ ret = PTR_ERR(item);
+ if (ret == -ENOENT || ret == -EFBIG)
+ ret = 0;
+ sum = 0;
+ printk("no csum found for inode %lu start "
+ "%llu\n", inode->i_ino,
+ (unsigned long long)offset);
+ goto found;
+ }
+ btrfs_item_key_to_cpu(path->nodes[0], &found_key,
+ path->slots[0]);
+
+ item_start_offset = found_key.offset;
+ item_size = btrfs_item_size_nr(path->nodes[0],
+ path->slots[0]);
+ item_last_offset = item_start_offset +
+ (item_size / BTRFS_CRC32_SIZE) *
+ root->sectorsize;
+ item = btrfs_item_ptr(path->nodes[0], path->slots[0],
+ struct btrfs_csum_item);
+ }
+ /*
+ * this byte range must be able to fit inside
+ * a single leaf so it will also fit inside a u32
+ */
+ diff = offset - item_start_offset;
+ diff = diff / root->sectorsize;
+ diff = diff * BTRFS_CRC32_SIZE;
+
+ read_extent_buffer(path->nodes[0], &sum,
+ (unsigned long)item + diff,
+ BTRFS_CRC32_SIZE);
+found:
+ set_state_private(io_tree, offset, sum);
+ bio_index++;
+ bvec++;
+ }
+ btrfs_free_path(path);
+ return 0;
+}
+
int btrfs_csum_one_bio(struct btrfs_root *root, struct inode *inode,
struct bio *bio)
{