[DRBD-user] [dm-devel] [PATCH v5 01/11] block: make generic_make_request handle arbitrarily sized bios

Martin K. Petersen martin.petersen at oracle.com
Tue Aug 11 20:01:45 CEST 2015

Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.

>>>>> "Mike" == Mike Snitzer <snitzer at redhat.com> writes:

Mike> That is the benefit.  And when coupled with the new default
Mike> max_discard of 64K (pending change from Jens for 4.3) this 2GB
Mike> upper limit really isn't such a big deal.  Unless I'm missing
Mike> something...

2GB is fine for current SATA due to the stupid range descriptors. But
there are some changes in the pipeline to support descriptors with
bigger ranges so that will change.

However, we currently do 4GB discards on SCSI so the proposed cap will
cut that in half. I'm OK with that as an interim solution. though.

But I'm a bit concerned about what might be lurking in dm thinp if you
trip over partial blocks like in Ming's example...

Martin K. Petersen	Oracle Linux Engineering

More information about the drbd-user mailing list