[patch] qlogic "this should not happen"

• Next message: Ingo Molnar: "Re: Problem with the O(1) scheduler in 2.4.19"
• Previous message: Thomas Koller: "RE: Kernel Oops after X-Server shutdown"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

Patch tested on 2.5.31, 2.5.31-mm1, 2.5.32-mm1, 2.5.32-mm2.

Without patch, 2.5.x during heavy benchmark/stress testing
eventually locks up with these final messages:

kernel: qlogicfc0 : no handle slots, this should not happen.
kernel: hostdata->queued is 6, in_ptr: 7d

Derived from Doug Ledford's patch:
http://marc.theaimsgroup.com/?l=linux-kernel&m=103005703808312&w=2
and Eric Weigle's patch:
http://marc.theaimsgroup.com/?l=linux-kernel&m=103005790509079&w=2

2.5.33 locked up without the patch.

diff -ruN linux-2.5.33/drivers/scsi/qlogicfc.c linux/drivers/scsi/qlogicfc.c
--- linux-2.5.33/drivers/scsi/qlogicfc.c 2002-07-24 19:09:03.000000000 -0400
+++ linux/drivers/scsi/qlogicfc.c 2002-09-01 16:44:33.000000000 -0400
@@ -1342,18 +1342,11 @@

        num_free = QLOGICFC_REQ_QUEUE_LEN - REQ_QUEUE_DEPTH(in_ptr, out_ptr);
        num_free = (num_free > 2) ? num_free - 2 : 0;
- host->can_queue = hostdata->queued + num_free;
+ host->can_queue = host->host_busy + num_free;
        if (host->can_queue > QLOGICFC_REQ_QUEUE_LEN)
                host->can_queue = QLOGICFC_REQ_QUEUE_LEN;
        host->sg_tablesize = QLOGICFC_MAX_SG(num_free);

- /* this is really gross */
- if (host->can_queue <= host->host_busy){
- if (host->can_queue+2 < host->host_busy)
- DEBUG(printk("qlogicfc%d.c crosses its fingers.\n", hostdata->host_id));
- host->can_queue = host->host_busy + 1;
- }
-
        LEAVE("isp2x00_queuecommand");

        return 0;
@@ -1623,17 +1616,11 @@

        num_free = QLOGICFC_REQ_QUEUE_LEN - REQ_QUEUE_DEPTH(in_ptr, out_ptr);
        num_free = (num_free > 2) ? num_free - 2 : 0;
- host->can_queue = hostdata->queued + num_free;
+ host->can_queue = host->host_busy + num_free;
        if (host->can_queue > QLOGICFC_REQ_QUEUE_LEN)
                host->can_queue = QLOGICFC_REQ_QUEUE_LEN;
        host->sg_tablesize = QLOGICFC_MAX_SG(num_free);

- if (host->can_queue <= host->host_busy){
- if (host->can_queue+2 < host->host_busy)
- DEBUG(printk("qlogicfc%d : crosses its fingers.\n", hostdata->host_id));
- host->can_queue = host->host_busy + 1;
- }
-
        outw(HCCR_CLEAR_RISC_INTR, host->io_port + HOST_HCCR);
        LEAVE_INTR("isp2x00_intr_handler");
 }
diff -ruN linux-2.5.33/drivers/scsi/qlogicfc.h linux/drivers/scsi/qlogicfc.h
--- linux-2.5.33/drivers/scsi/qlogicfc.h 2002-07-24 19:09:03.000000000 -0400
+++ linux/drivers/scsi/qlogicfc.h 2002-09-01 16:41:47.000000000 -0400
@@ -65,7 +65,7 @@
 #define DATASEGS_PER_COMMAND 2
 #define DATASEGS_PER_CONT 5

-#define QLOGICFC_REQ_QUEUE_LEN 127 /* must be power of two - 1 */
+#define QLOGICFC_REQ_QUEUE_LEN 255 /* must be power of two - 1 */
 #define QLOGICFC_MAX_SG(ql) (DATASEGS_PER_COMMAND + (((ql) > 0) ? DATASEGS_PER_CONT*((ql) - 1) : 0))
 #define QLOGICFC_CMD_PER_LUN 8

-- 
Randy Hron
http://home.earthlink.net/~rwhron/kernel/bigbox.html
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

• Next message: Ingo Molnar: "Re: Problem with the O(1) scheduler in 2.4.19"
• Previous message: Thomas Koller: "RE: Kernel Oops after X-Server shutdown"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

This archive was generated by hypermail 2b29 : Sat Sep 07 2002 - 22:00:18 EST