[PATCH] AHA152x driver hangs on PCMCIA card eject, kernel 2.4.22-pre6

["Bhavesh P. Davda" <bhavesh@avaya.com>]

[-- Attachment #1: Type: text/plain, Size: 1346 bytes --]

Juergen, David; please review this patch, and recommend that Marcelo apply
this patch if you think it is okay... Thanks!

Attached is a patch against linux-2.4.22-pre6, to fix a hang problem when an
Adaptec SlimSCSI (PCMCIA) adapter is ejected from a PCMCIA card reader (e.g.
yenta/TI1225 based).

The fix involves:

1. A change to the common aha152x driver to ignore an interrupt in the
top-half handler if it cannot read valid data from the I/O ports (possibly
due to a bad host-adapter chip or an ejected PCMCIA card). This way, a
shared interrupt handler (e.g. yenta) can pick up the interrupt if the IRQ
is really meant for it. This is where the original hang was taking place;
the aha152x bottom half was getting into an infinite loop, though the
SlimSCSI card had been ejected, and the actual IRQ was meant for yenta.

2. A change to the aha152x_cs stub driver to not use the SCSI error-handling
thread code. The aha152x_cs driver calls scsi_unregister_module() as a
queued timer task when it gets a CS_EVENT_CARD_REMOVAL event, which causes
scsi_unregister_host() to do a down() on a semaphore, calling schedule(),
when executing the timer_bh for the timer.

Thanks!

- Bhavesh
--
Bhavesh P. Davda     E-mail    : bhavesh@avaya.com
Avaya Inc.           Phone/Fax : (303) 538-4438
Room B3-B03, 1300 West 120th Avenue
Westminster, CO 80234

[-- Attachment #2: linux-2.4.22-aha152x.patch --]
[-- Type: application/octet-stream, Size: 1814 bytes --]

diff -Naur linux-2.4.22-pre6/drivers/scsi/aha152x.c linux-2.4.22-bpd/drivers/scsi/aha152x.c
--- linux-2.4.22-pre6/drivers/scsi/aha152x.c	2003-06-13 08:51:36.000000000 -0600
+++ linux-2.4.22-bpd/drivers/scsi/aha152x.c	2003-07-15 14:38:27.000000000 -0600
@@ -1936,6 +1936,23 @@
 		return;
 	}
 
+	/*
+	 * Read a couple of registers that are known to not be all 1's. If
+	 * we read all 1's (-1), that means that either:
+	 * a. The host adapter chip has gone bad, and we cannot control it,
+	 * OR
+	 * b. The host adapter is a PCMCIA card that has been ejected
+	 * In either case, we cannot do anything with the host adapter at
+	 * this point in time. So just ignore the interrupt and return.
+	 * In the latter case, the interrupt might actually be meant for
+	 * someone else sharing this IRQ, and that driver will handle it
+	 */
+	rev = GETPORT(REV);
+	dmacntrl0 = GETPORT(DMACNTRL0);
+	if ((rev == 0xFF) && (dmacntrl0 == 0xFF)) {
+		return;
+	}
+
 	/* no more interrupts from the controller, while we're busy.
 	   INTEN is restored by the BH handler */
 	CLRBITS(DMACNTRL0, INTEN);
diff -Naur linux-2.4.22-pre6/drivers/scsi/pcmcia/aha152x_stub.c linux-2.4.22-bpd/drivers/scsi/pcmcia/aha152x_stub.c
--- linux-2.4.22-pre6/drivers/scsi/pcmcia/aha152x_stub.c	2001-12-21 10:41:55.000000000 -0700
+++ linux-2.4.22-bpd/drivers/scsi/pcmcia/aha152x_stub.c	2003-07-15 14:41:20.000000000 -0600
@@ -244,6 +244,11 @@
 
     /* Configure card */
     driver_template.module = &__this_module;
+    /* 
+     * Don't use the SCSI error handling thread. This causes schedule() to
+     * be called from the timer_bh when the AHA152x card is ejected
+     */
+    driver_template.use_new_eh_code = 0;
     link->state |= DEV_CONFIG;
 
     tuple.DesiredTuple = CISTPL_CFTABLE_ENTRY;