>From Roland Drier: > I looked through the DMA paths on the hardware drivers to ensure they > could take advantage of the SG chaining, and it seems that every one > except iPath uses the system's DMA routines, which have been converted > to handle chaining. iPath looks like it should be OK, but I have no way > to test it.
set srp_sg_tablesize=256 Add SRP target with max_sects=8192 or more echo 4096 > /sys/block/sda/queue/max_sectors_kb echo noop > /sys/block/sda/queue/scheduler dd if=/dev/zero of=/dev/sda oflag=direct bs=4096k count=10 dd if=/dev/sda iflag=direct of=/dev/sda bs=4096k count=10 You may not get 4MB I/Os because of memory fragmentation, but you should see 1MB or better. You can use a real file as a data source/sink to verify against corruption. Dave _______________________________________________ general mailing list [email protected] http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
