| Message ID | 20200527081227.3408-1-johannes.thumshirn@wdc.com (mailing list archive) |
|---|---|
| Headers | show |
| Series | 3 small cleanups for find_first_block_group | expand |
On 27/05/2020 10:12, Johannes Thumshirn wrote: > While trying to learn the Block Group code I've found some cleanup > possibilities for find_first_block_group(). > > Here's a proposal to make $ffbg a bit more easier to read by untangling the > gotos and if statements. > > The patch set is based on misc-next from May 26 morning with > HEAD 3f4a266717ed ("btrfs: split btrfs_direct_IO to read and write part") > and xfstests showed no regressions to the base misc-next in my test setup. > > Changes to v1: > - Pass btrfs_path instead of leaf & slot to read_bg_from_eb (Nikolay) > - Don't comment about the size change (Nikolay) > - Add Nikolay's Reviewed-by's > > Johannes Thumshirn (3): > btrfs: remove pointless out label in find_first_block_group > btrfs: get mapping tree directly from fsinfo in find_first_block_group > btrfs: factor out reading of bg from find_frist_block_group > > fs/btrfs/block-group.c | 108 ++++++++++++++++++++++------------------- > 1 file changed, 58 insertions(+), 50 deletions(-) > David, any comment if you will consider this series? Patch 2/3 actually makes the function more complainant to the kernel's coding style.
On Fri, May 29, 2020 at 02:25:32PM +0000, Johannes Thumshirn wrote: > On 27/05/2020 10:12, Johannes Thumshirn wrote: > > While trying to learn the Block Group code I've found some cleanup > > possibilities for find_first_block_group(). > > > > Here's a proposal to make $ffbg a bit more easier to read by untangling the > > gotos and if statements. > > > > The patch set is based on misc-next from May 26 morning with > > HEAD 3f4a266717ed ("btrfs: split btrfs_direct_IO to read and write part") > > and xfstests showed no regressions to the base misc-next in my test setup. > > > > Changes to v1: > > - Pass btrfs_path instead of leaf & slot to read_bg_from_eb (Nikolay) > > - Don't comment about the size change (Nikolay) > > - Add Nikolay's Reviewed-by's > > > > Johannes Thumshirn (3): > > btrfs: remove pointless out label in find_first_block_group > > btrfs: get mapping tree directly from fsinfo in find_first_block_group > > btrfs: factor out reading of bg from find_frist_block_group > > > > fs/btrfs/block-group.c | 108 ++++++++++++++++++++++------------------- > > 1 file changed, 58 insertions(+), 50 deletions(-) > > David, any comment if you will consider this series? Patch 2/3 actually makes > the function more complainant to the kernel's coding style. 2/3 and 3/3 are good cleanups, please resend, thanks.
While trying to learn the Block Group code I've found some cleanup possibilities for find_first_block_group(). Here's a proposal to make $ffbg a bit more easier to read by untangling the gotos and if statements. The patch set is based on misc-next from May 26 morning with HEAD 3f4a266717ed ("btrfs: split btrfs_direct_IO to read and write part") and xfstests showed no regressions to the base misc-next in my test setup. Changes to v1: - Pass btrfs_path instead of leaf & slot to read_bg_from_eb (Nikolay) - Don't comment about the size change (Nikolay) - Add Nikolay's Reviewed-by's Johannes Thumshirn (3): btrfs: remove pointless out label in find_first_block_group btrfs: get mapping tree directly from fsinfo in find_first_block_group btrfs: factor out reading of bg from find_frist_block_group fs/btrfs/block-group.c | 108 ++++++++++++++++++++++------------------- 1 file changed, 58 insertions(+), 50 deletions(-)