[PATCH] Disable (false?) IRQ check on 2.6 in orinoco_cs (was: yenta_socket.c:yenta_get_status returns bad value compared to 2.4)

[Stefan Jones]

Ok here is the report,

I had 

CONFIG_ISA=y
in my .config which in turn set
CONFIG_PCMCIA_PROBE=y

which enabled some invalid (for me) code in
linux/drivers/pcmcia/rsrc_mgr.c

as the machine does not need ISA 

( I always enable it by default as a few years ago an ISA free machine
wouldn't boot without it! )

This caused the hangs. With CONFIG_ISA unset it's fine.

Now when the kernel loaded the orinoco_cs driver I got:

Jul 27 17:15:37 [kernel] orinoco.c 0.13e (David Gibson
<hermes at gibson.dropbear.id.au> and others)
Jul 27 17:15:37 [kernel] orinoco_cs.c 0.13e (David Gibson
<hermes at gibson.dropbear.id.au> and others)
Jul 27 17:15:37 [kernel] orinoco_cs: RequestIRQ: Resource in use

Now I guessed that was an artifact of irq sharing + new kernel internals
so I used the patch: ( or just luck )

--- orinoco_cs.c.jj	2003-07-27 17:20:28.000000000 +0100
+++ orinoco_cs.c	2003-07-27 17:20:16.000000000 +0100
@@ -469,7 +469,7 @@
   		link->irq.Handler = orinoco_interrupt; 
   		link->irq.Instance = dev; 
 		
-		CS_CHECK(RequestIRQ, link->handle, &link->irq);
+		/*CS_CHECK(RequestIRQ, link->handle, &link->irq);*/
 	}
 
 	/* We initialize the hermes structure before completing PCMCIA

And the wireless card loaded with the normal blurb.

Jul 27 17:18:24 [cardmgr] socket 0: Netgear MA401RA
Jul 27 17:18:24 [cardmgr] executing: 'modprobe hermes'
Jul 27 17:18:24 [cardmgr] executing: 'modprobe orinoco'
Jul 27 17:18:24 [cardmgr] executing: 'modprobe orinoco_cs'

Now this fix is needed for me, your millage will vary.

Sorry for the spam

Stefan