[PATCH] PCI probe status cleanup

[Tim Hockin <thockin@sun.com>]

[-- Attachment #1: Type: text/plain, Size: 402 bytes --]

All,

Attached is a very small patch against 2.4.8, which I have been sending for
a while.  It merely clears the Master Abort bit on a PCI bridge when a
failed probe occurs.

If there is any reason this can't go into the next 2.4.x release, please
let me know.  It has been in use here for months.

Tim
-- 
Tim Hockin
Systems Software Engineer
Sun Microsystems, Cobalt Server Appliances
thockin@sun.com

[-- Attachment #2: pci_bridge_cleanup.diff --]
[-- Type: text/plain, Size: 883 bytes --]

diff -ruN dist+patches-2.4.8/drivers/pci/pci.c cobalt-2.4.8/drivers/pci/pci.c
--- dist+patches-2.4.8/drivers/pci/pci.c	Wed Jul  4 09:41:34 2001
+++ cobalt-2.4.8/drivers/pci/pci.c	Mon Aug 13 16:42:12 2001
@@ -1232,8 +1236,19 @@
 		return NULL;
 
 	/* some broken boards return 0 or ~0 if a slot is empty: */
-	if (l == 0xffffffff || l == 0x00000000 || l == 0x0000ffff || l == 0xffff0000)
+	if (l == 0xffffffff || l == 0x00000000 
+	 || l == 0x0000ffff || l == 0xffff0000) {
+		/*
+		 * host/pci and pci/pci bridges will set Received Master Abort
+		 * (bit 13) on failed configuration access (happens when
+		 * searching for devices).  To be safe, clear the status
+		 * register.
+		 */
+		unsigned short st;
+		pci_read_config_word(temp, PCI_STATUS, &st);
+		pci_write_config_word(temp, PCI_STATUS, st);
 		return NULL;
+	}
 
 	dev = kmalloc(sizeof(*dev), GFP_KERNEL);
 	if (!dev)