On Sun, Jul 10, 2011 at 03:41:51PM -0700, Andy Grover wrote: > We can assume each se_cmd only needs 1 se_task??? > > I couldn't see where the max sectors per se_cmd was limited, therefore > we needed to support cmds being broken up into > se_sub_dev.max_sectors-size chunks, and therefore being >1 se_task per > se_cmd, for all backstores not just pscsi? The most important backend, iblock already handles arbitrarily sized payloads by using multiple bios internally, which is much lower overhead than using the multiple task complexity, and the ramdisk code has no inherent limitation either. The file code has a limitation on the number of iovecs accepted by the kernel, and only the pscsi backend really needs multiple "tasks". Given that the performance of iblock is the important one streamline the path for this backend as much as possible sounds like a good idea. -- 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
