On Thu, Sep 22, 2016 at 01:38:44PM +0200, Marcin Mirosław wrote: > W dniu 22.09.2016 o 13:34, Kent Overstreet pisze: > > On Thu, Sep 22, 2016 at 01:14:45PM +0200, Marcin Mirosław wrote: > >> W dniu 22.09.2016 o 12:58, Marcin Mirosław pisze: > >>> Hi! > >>> I'm at: > >>> commit b7cb7779de9609899fa53f7bab4e3cee69ac1f7e > >>> Author: Kent Overstreet <kent.overstreet@xxxxxxxxx> > >>> Date: Wed Sep 21 17:38:38 2016 -0800 > >>> > >>> bcache: fix an ordering issue in btree_interior_update code > >>> > >>> Also a bunch of refactoring > >>> > >>> > >>> Steps: > >>> # bcache format --discard --compression_type=lz4 /dev/system10/bcache > >>> (underlaying devices does not support discard) > >>> # mount & use fs > >>> # echo crc64 > data_checksum > >>> # echo crc64 > metadata_checksum > >>> # use filesystem > >>> After sometime I'm getting: > >>> [ 6444.192279] bcache (dm-11): IO error on dm-11 for checksum error > >> > >> After some trying to read files dmesg says: > >> [ 7651.598498] bcache (dm-11): IO error on dm-11 for checksum error > >> [ 7657.066482] bio_checksum_uncompress: 83 callbacks suppressed > > > > The crc64 fix turned out to be an incompatible change for crc32c, so I had to > > revert it. I'm just going to wait till I do all the other breaking changes for > > that one (which should be soon). > > > Even when files are written after changing to crc64? That was the original bug, so yeah. -- To unsubscribe from this list: send the line "unsubscribe linux-bcache" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
