On 6/15/22 08:54, Bart Van Assche wrote: > On 6/14/22 16:50, Damien Le Moal wrote: >> We try that "fix" with the work for zoned btrfs. It does not work. Even >> adding a delay to wait for out of order requests (if there is a hole in a >> write sequence) does not reliably work as FSes may sometimes take 10s of >> seconds to issue all write requests that can be all ordered into a nice >> write stream. Even with that delay increased to minutes, we were still >> seeing unaligned write errors. > > Please clarify. Doesn't BTRFS use REQ_OP_ZONE_APPEND for zoned storage? > REQ_OP_ZONE_APPEND requests do not trigger write errors if reordered. The trial I am mentioning above was before we moved to using zone append and implementing scsi emulation for it. Note that btrfs uses zone append write for data only. Metadata writes still require regular write commands. > > Additionally, our tests with F2FS on top of this patch series pass. Try on an AHCI SMR drive and I am 100% sure that you will see unaligned write errors. > > Thanks, > > Bart. -- Damien Le Moal Western Digital Research
