lkml.org 
[lkml]   [1998]   [Apr]   [18]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
From
Date
SubjectSolution for hang in SMC setup (alpha/2.1.9[67])
Recent kernels hung on my alpha PC164/433Mhz, just in SMC init.
Don't know why, this started after I tried aic7xxx-5.0.12 in 2.1.95,
continued in 2.1.96/97 which has that aic driver activated.

The strange thing was that the console messages did not indicate to a problem
in aic driver, actually, the hang happened long before the driver init.

The last messages were:
PCI: Probing PCI hardware.
SMC FDC37C93X Ultra I/O Controller found @ 0x370

Then I tried to enable debug (define DEBUG) in drivers/pci/pci.c, and (magic!)
the kernel booted ok. So I started to suspect compiler/binutils.

I turned off the debug in pci.c and started to insert debug messages into
arch/alpha/kernel/smc37c93x.c and it showed that the hang happens on a
particular outb. This seemed strange, and I started to suspect hardware.

The debug in pci.c caused several messages to be printed and I thought it can
be the delay of printk's that prevented the hang. I inserted udelay(500) at
top of SMC93x_Init and (magic again!) the kernel booted ok.

So the questions are: does anyone experience the hang on boot too? and does
anybody know if pci bus scan can cause consequential io to hang? Maybe
isa/pci bridge is unstable for some time after bus scan?

Ok, here is what I did:
(maybe the delay is more appropriate in pci.c/pci_scan_bus)

--- smc37c93x.c.~1~ Tue Apr 7 14:45:04 1998
+++ smc37c93x.c Sat Apr 18 16:59:30 1998
@@ -12,6 +12,7 @@
#include <asm/hwrpb.h>
#include <asm/io.h>
#include <asm/segment.h>
+#include <asm/delay.h>

#if 0
# define DBG_DEVS(args) printk args
@@ -233,6 +234,8 @@
int __init SMC93x_Init(void)
{
unsigned long SMCUltraBase;
+
+ udelay(500);

if ((SMCUltraBase = SMCDetectUltraIO()) != 0UL) {
printk("SMC FDC37C93X Ultra I/O Controller found @ 0x%lx\n",
Alexander.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu

\
 
 \ /
  Last update: 2005-03-22 13:42    [W:0.038 / U:0.128 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site