Hi Martin,
On 11/09/2018 03:07 AM, Martin K. Petersen wrote:
There are too many unresolved issues with DIX outside of zfcp such as
wrong protection data on writesame/discard (over device-mapper)
We don't configure protected transfers for anything but read and write
commands. There is currently no protection information generated for
WRITE SAME.
So if you guys are seeing failures, it must be due to zfcp
not handling the scsi_cmnd prot_op/prot_flags or the command PROTECT bit
correctly.
I think we're good since
("scsi: zfcp: fix queuecommand for scsi_eh commands when DIX enabled")
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/drivers/s390/scsi?id=71b8e45da51a7b64a23378221c0a5868bd79da4f.
Previously, at least regular (non-recovery) I/O should have been good by
having checked at least scsi_prot_sg_count().
or due to unstable page writes.
BDI_CAP_STABLE_WRITES should take care of this. What's the configuration
that fails?
Apologies, if the commit description sounds unfair. I did not mean to
blame anyone. It's just the collection of issues we saw in distros over
the years. Some of the old issues might be fixed with above zfcp patch
or common code changes. Unfortunately, I could not handle the DIX things
we saw. I think, DIF by itself provides a lot of the protection benefit
and was not affected by the encountered issues. We would like to give
users an easy way to operate in such setup.
--
Mit freundlichen Gruessen / Kind regards
Steffen Maier
Linux on IBM Z Development
IBM Deutschland Research & Development GmbH
Vorsitzende des Aufsichtsrats: Martina Koederitz
Geschaeftsfuehrung: Dirk Wittkopp
Sitz der Gesellschaft: Boeblingen
Registergericht: Amtsgericht Stuttgart, HRB 243294
