Hi all, This series is a repost of a handful of patches that happened to have fallen through the cracks. Patch 1 relates back to the discussion[1] on increasing the writeback bio size limit. We can't safely do that until the bounce path can be enhanced to handle larger sizes, but we can at least fix the writeback code to completely fill the bio. Patches 2-3 are a couple more byte order fixes that were uncovered by swapping an on-disk filesystem between big and little endian machines. As of these two, I've not encountered any further byte-order issues via this sort of test.
All three of these patches have been soaking in my test branch [2] for quite some time. Thoughts, reviews, flames appreciated. Brian [1] https://lore.kernel.org/linux-bcachefs/20230927112338.262207-3-bfos...@redhat.com/ [2] https://evilpiepirate.org/~testdashboard/ci?branch=bfoster Brian Foster (3): bcachefs: allow writeback to fill bio completely bcachefs: byte order swap bch_alloc_v4.fragmentation_lru field bcachefs: use swab40 for bch_backpointer.bucket_offset bitfield fs/bcachefs/alloc_background.c | 10 +--------- fs/bcachefs/backpointers.c | 2 +- fs/bcachefs/backpointers.h | 9 +++++++++ fs/bcachefs/fs-io-buffered.c | 19 ++++++++++++++++--- 4 files changed, 27 insertions(+), 13 deletions(-) -- 2.41.0
