Messages in this thread Patch in this message | | | Date | Wed, 4 Mar 1998 13:54:42 +0100 (MET) | From | Rik van Riel <> | Subject | [PATCH] blk.h new sorting (completely untested!!) |
| |
Hi,
I've implemented the 'new' sorting scheme for blk.h I just made up while writing the previous message.
Beware though, that this patch is completely untested. Worse yet, I haven't even proofread it :-)
succes...
Rik. +-----------------------------+------------------------------+ | For Linux mm-patches, go to | "I'm busy managing memory.." | | my homepage (via LinuxHQ). | H.H.vanRiel@fys.ruu.nl | | ...submissions welcome... | http://www.fys.ruu.nl/~riel/ | +-----------------------------+------------------------------+
--- linux/include/linux/blk.h.orig Wed Mar 4 13:47:35 1998 +++ linux/include/linux/blk.h Wed Mar 4 13:50:28 1998 @@ -26,10 +26,14 @@ * over writes any more --- although reads are more time-critical than * writes, by treating them equally we increase filesystem throughput. * This turns out to give better overall performance. -- sct + * We ignore the major number and the first bit of the minor number, + * this way we hope to achieve some parrallelism <sp?> when flushing + * the disk. We ignore the SCSI minor-number scheme, since SCSI doesn't + * use this queue anyway -- Rik. */ #define IN_ORDER(s1,s2) \ -((s1)->rq_dev < (s2)->rq_dev || (((s1)->rq_dev == (s2)->rq_dev && \ -(s1)->sector < (s2)->sector))) +(((s1)->rq_dev & 0x7f) < ((s2)->rq_dev & 0x7f) || ((((s1)->rq_dev & 0x7f) \ +== ((s2)->rq_dev & 0x7f) && (s1)->sector < (s2)->sector))) /* * These will have to be changed to be aware of different buffer - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu
| |