On Wed, Jul 16, 2014 at 11:44:18AM -0400, Martin K. Petersen wrote: > There are lots of devices out there that support WRITE SAME(10) or (16) > without the UNMAP bit. And there are devices that support WRITE SAME w/ > UNMAP functionality but not "regular" WRITE SAME. Oh, we actually have devices that support WRITE SAME with unmap, but not without? That's defintively a little strange. > no_write_same is there to prevent the REQ_WRITE_SAME use case (for which > we have really weak heuristics). Your patch overloads no_write_same so > it also governs a REQ_DISCARD use case. Yes, and it did this intentionally. I really wouldn't expect devices to support WRITE SAME with UNMAP but blow up on a WRITE SAME without it (and not just simple fail it in an orderly way). > My proposed black list patch fixes the hyperv discard issue. So I don't > see why we'd need to overload no_write_same which was meant for an > entirely different purpose. It definitively seems odd to default to trying WRITE SAME for unmap for a device that explicitly tells us that it doesn't support WRITE SAME. Note that I'm not against your patch - I suspect forcing us to read EVPD pages even for devices that claim to be SPC-2 will come in useful in various scenarios. -- 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
