On Sun, Apr 15, 2012 at 9:08 AM, FUJITA Tomonori <fujita.tomonori@xxxxxxxxxxxxx> wrote: > On Sun, 15 Apr 2012 07:15:10 +1000 > ronnie sahlberg <ronniesahlberg@xxxxxxxxx> wrote: > >> Fair enough. > > xfs and ext4 handle it so I tend to blame btrfs. I need to investigate > the issue more though. Could have been early EXT4 tests too, havent used this that much in the last few weeks and things are still changing I guess in the kernel. But you are right, I shouldnt try to manually zero out unaligned UNMAPs. That is the job for the filesystem, or actually, what I think we should do is to tell the initiator what the mapping is, by automatically setting the LOGICAL BLOCKS PER PHYSICAL BLOCK EXPONENT in the READCAPACITY16 parameters. Then, if an initiator tries to use UNMAP that is not aligned properly it can be classed an initiator bug and then we dont have to worry about how the underlying filesystem to TGTD handles unaligned unmaps. Let me update the patch and resend it to do this automatic setting of LBPPBE, that should make sure that initiators never try unaligned unmaps. regards ronnie sahlberg -- To unsubscribe from this list: send the line "unsubscribe stgt" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
