@@ -1501,6 +1501,22 @@ static int close_blocks(u64 blocknr, u64 other, u32 blocksize)
return 0;
}
+#ifdef __LITTLE_ENDIAN
+
+/*
+ * Compare two keys, on little-endian the disk order is same as CPU order and
+ * we can avoid the conversion.
+ */
+static int comp_keys(const struct btrfs_disk_key *disk_key,
+ const struct btrfs_key *k2)
+{
+ const struct btrfs_key *k1 = (const struct btrfs_key *)disk_key;
+
+ return btrfs_comp_cpu_keys(k1, k2);
+}
+
+#else
+
/*
* compare two keys in a memcmp fashion
*/
@@ -1513,6 +1529,7 @@ static int comp_keys(const struct btrfs_disk_key *disk,
return btrfs_comp_cpu_keys(&k1, k2);
}
+#endif
/*
* same as comp_keys only with two btrfs_key's
@@ -1894,6 +1894,52 @@ BTRFS_SETGET_STACK_FUNCS(disk_key_objectid, struct btrfs_disk_key,
BTRFS_SETGET_STACK_FUNCS(disk_key_offset, struct btrfs_disk_key, offset, 64);
BTRFS_SETGET_STACK_FUNCS(disk_key_type, struct btrfs_disk_key, type, 8);
+#ifdef __LITTLE_ENDIAN
+
+/*
+ * Optimized helpers for little-endian architectures where CPU and on-disk
+ * structures have the same endianity and we can skip conversions.
+ */
+
+static inline void btrfs_disk_key_to_cpu(struct btrfs_key *cpu_key,
+ const struct btrfs_disk_key *disk_key)
+{
+ memcpy(cpu_key, disk_key, sizeof(struct btrfs_key));
+}
+
+static inline void btrfs_cpu_key_to_disk(struct btrfs_disk_key *disk_key,
+ const struct btrfs_key *cpu_key)
+{
+ memcpy(disk_key, cpu_key, sizeof(struct btrfs_key));
+}
+
+static inline void btrfs_node_key_to_cpu(const struct extent_buffer *eb,
+ struct btrfs_key *cpu_key, int nr)
+{
+ struct btrfs_disk_key *disk_key = (struct btrfs_disk_key *)cpu_key;
+
+ btrfs_node_key(eb, disk_key, nr);
+}
+
+static inline void btrfs_item_key_to_cpu(const struct extent_buffer *eb,
+ struct btrfs_key *cpu_key, int nr)
+{
+ struct btrfs_disk_key *disk_key = (struct btrfs_disk_key *)cpu_key;
+
+ btrfs_item_key(eb, disk_key, nr);
+}
+
+static inline void btrfs_dir_item_key_to_cpu(const struct extent_buffer *eb,
+ const struct btrfs_dir_item *item,
+ struct btrfs_key *cpu_key)
+{
+ struct btrfs_disk_key *disk_key = (struct btrfs_disk_key *)cpu_key;
+
+ btrfs_dir_item_key(eb, item, disk_key);
+}
+
+#else
+
static inline void btrfs_disk_key_to_cpu(struct btrfs_key *cpu,
const struct btrfs_disk_key *disk)
{
@@ -1935,6 +1981,8 @@ static inline void btrfs_dir_item_key_to_cpu(const struct extent_buffer *eb,
btrfs_disk_key_to_cpu(key, &disk_key);
}
+#endif
+
/* struct btrfs_header */
BTRFS_SETGET_HEADER_FUNCS(header_bytenr, struct btrfs_header, bytenr, 64);
BTRFS_SETGET_HEADER_FUNCS(header_generation, struct btrfs_header,
The CPU and on-disk keys are mapped to two different structures because of the endianity. There's an intermediate buffer used to do the conversion, but this is not necessary when CPU and on-disk endianity matches. Add optimized versions of helpers that take disk_key and use the buffer directly for CPU keys or drop the intermediate buffer and conversion. This saves a lot of stack space accross many functions and removes about 6K of generated binary code: text data bss dec hex filename 1090439 17468 14912 1122819 112203 pre/btrfs.ko 1084613 17456 14912 1116981 110b35 post/btrfs.ko Delta: -5826 Signed-off-by: David Sterba <dsterba@suse.com> --- fs/btrfs/ctree.c | 17 +++++++++++++++++ fs/btrfs/ctree.h | 48 ++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 65 insertions(+)