On Mon, 2011-07-18 at 10:16 -0700, Andy Grover wrote: > On 07/17/2011 12:57 PM, Nicholas A. Bellinger wrote: > > Although this patch did not directly remove the memsets for newly > > allocated pages used with transport_generic_map_mem_to_cmd(), the issue > > as below of reproducable strangeness with control CDB payloads without > > the explict memset of allocated pages. As much as I would have liked to > > leave the memset out, I don't think it's safe to do. Added the > > following: > > > > commit 7c7649750eb47035ac4fb0123f2e49c1c47c8a15 > > Author: Nicholas Bellinger <nab@xxxxxxxxxxxxxxx> > > Date: Sun Jul 17 00:20:07 2011 -0700 > > > > target/iscsi: Make iscsit_alloc_buffs explict memset allocated pages > > Hi nab, obviously if it's breaking things the memsets need to go back > for now, but surely it's a sign that somewhere there's a bug. You only > see corruption with control CDBs you say? Any more info on the > strangeness you're seeing? > The immediate case that I ran into with the very latest iscsi-target code was bogus output with 'sg_inq -i' (INQUIRY EVPD=0x83) in the 'Designation descriptor number 5' field. -- To unsubscribe from this list: send the line "unsubscribe linux-scsi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html