diff options
| author | Kent Overstreet <koverstreet@google.com> | 2013-06-05 17:24:39 +0400 |
|---|---|---|
| committer | Kent Overstreet <koverstreet@google.com> | 2013-06-27 08:58:04 +0400 |
| commit | 72c270612bd33192fa836ad0f2939af1ca218292 (patch) | |
| tree | 344129d75f3b5c0abcf77dd4b6340783a126cde8 /drivers/md/bcache/sysfs.c | |
| parent | 279afbad4e54acbd61bf88a54a73af3bbfdeb5dd (diff) | |
| download | linux-72c270612bd33192fa836ad0f2939af1ca218292.tar.xz | |
bcache: Write out full stripes
Now that we're tracking dirty data per stripe, we can add two
optimizations for raid5/6:
* If a stripe is already dirty, force writes to that stripe to
writeback mode - to help build up full stripes of dirty data
* When flushing dirty data, preferentially write out full stripes first
if there are any.
Signed-off-by: Kent Overstreet <koverstreet@google.com>
Diffstat (limited to 'drivers/md/bcache/sysfs.c')
| -rw-r--r-- | drivers/md/bcache/sysfs.c | 8 |
1 files changed, 8 insertions, 0 deletions
diff --git a/drivers/md/bcache/sysfs.c b/drivers/md/bcache/sysfs.c index cf8d91ec3238..70c6dff0d0cd 100644 --- a/drivers/md/bcache/sysfs.c +++ b/drivers/md/bcache/sysfs.c @@ -81,6 +81,9 @@ rw_attribute(writeback_rate_p_term_inverse); rw_attribute(writeback_rate_d_smooth); read_attribute(writeback_rate_debug); +read_attribute(stripe_size); +read_attribute(partial_stripes_expensive); + rw_attribute(synchronous); rw_attribute(journal_delay_ms); rw_attribute(discard); @@ -147,6 +150,9 @@ SHOW(__bch_cached_dev) sysfs_hprint(dirty_data, bcache_dev_sectors_dirty(&dc->disk) << 9); + sysfs_hprint(stripe_size, (1 << dc->disk.stripe_size_bits) << 9); + var_printf(partial_stripes_expensive, "%u"); + var_printf(sequential_merge, "%i"); var_hprint(sequential_cutoff); var_hprint(readahead); @@ -286,6 +292,8 @@ static struct attribute *bch_cached_dev_files[] = { &sysfs_writeback_rate_d_smooth, &sysfs_writeback_rate_debug, &sysfs_dirty_data, + &sysfs_stripe_size, + &sysfs_partial_stripes_expensive, &sysfs_sequential_cutoff, &sysfs_sequential_merge, &sysfs_clear_stats, |