On 06/12/2018 04:22, Martin K. Petersen wrote:
Hi Martin,
John,
I'll have to go revisit the archives for your exact fio bits.
Your fio test case ran to completion on all my test systems.
But I successfully ran our DIX/T10 PI qualification tooling without any
errors on the latest SCSI tree. On both scsi_debug and FC hardware.
I'm not fimilar with this tooling. Is it internal to your company?
Sadly, yes :( We have not had much luck upstreaming the necessary aio
integrity passthrough enhancements.
Unfortunately it seems that no other SAS HBA kernel driver supports
DIX, so we have nothing to test against for this transport.
We did a preliminary qual of LSISAS3008 many moons ago. But we had no
immediate use for it in our offerings so it never went beyond that. I
know other companies are using it, though.
It seems from https://www.spinics.net/lists/linux-scsi/msg97818.html
that 3008 mainline driver does support DIX:
"DIX is supported in mpt3sas , but not by default and hence while
loading mpt3sas driver set module parameter "prot_mask=0x7F" to enable
DIX support. "
Maybe we can try it.
BTW, on a loosely related topic, in drivers/scsi/sd.h, I noticed that
the arrays in sd_prot_op() and sd_prot_flag_mask() could be made static.
In doing so I found sd.o object code shrunk by ~100B.
Thanks,
John
