From: ao@morpork.shnet.org (A. Ott)
Subject: Re: 1.0.9b-pre2 uploaded (2.2.13 SMP testing)
Date: 27 Feb 2000 11:28:00 +0100
Next Article (by Date): Re: 1.0.9b-pre2 uploaded (2.2.13 SMP testing) ao@morpork.shnet.org (A. Ott)
Previous Article (by Date): Re: 1.0.9b-pre2 uploaded (2.2.13 SMP testing) Jesse Pollard
Top of Thread: Re: 1.0.9b-pre2 uploaded (2.2.13 SMP testing) Jesse Pollard
Next in Thread: Re: 1.0.9b-pre2 uploaded (2.2.13 SMP testing) ao@morpork.shnet.org (A. Ott)
Articles sorted by: [Date]
[Author]
[Subject]
********* ***************** ********** **** ***** ***** ************
To subject Re: 1.0.9b-pre2 uploaded (2.2.13 SMP testing)
pollard@cats-chateau.net (Jesse Pollard) wrote:
********** ******************** ****** ******** ******* *************
> I rebuilt the kernel starting from a base 2.2.13 SMP configuration, added
> the tarball and configured the maintenance kernel.
>
> It did get farther in the boot, but it crashed (rather than hung). I also
> tried a uniprocessor version to see if there were any differences between
> them. It crashed too.
That's strange, because it says locking cannot be at fault there.
> This may have something to do with my processor type since the uniprocessor
> version works for you. I'm using a dual PPro (tyan S1662D). Compilers are
> gcc 2.7.2.3. There are a few other custom restructuring done:
> 1. All disk controllers are modules except BT958 and BT948
> 2. my /var directory tree is on IDE, so I have to install modules before
> anything else runs.
> 3. I'm not doing the devpts file system.
>
> The dump that occured was minimal. All of the RSBAC init modules completed:
>
> rsbac_init(): setting RSBAC auto timer <---previous hang occurred before
> this Freeing unused kernel memory: 68k freed
> INIT: version 2.76 booting
> <usual stuff here> scsi0 reports, scsi1 reports, adding swap
> then:
>
> Updating module dependancies for Linux 2.2.13.smp-rsbac-maint.pre2
>
> wait_on_bh, CPU 0:
> irq: 0 [0 0]
> 1 [0 1]
> <[c010b270]><[c010b278]><[c01a12bd]><[c01a1dee]><[c01a7a71]>
>
> This repeated after approx 15 seconds(this is rough estimate, made after I
> wrote down the first line, by then there were 4 more) with trace lines
> with the same addresses.
>
> If I have decoded the System.map file correctly the addresses are in the
> routines of:
>
> c010b230 T synchronize_bh (symbol preceeding address c010b270)
> c010a100 t common_interrupt
> c01a11a4 t do_con_write
> c01a1dd0 t con_write
> c01a78fc t opost_block
I never touched those functions, but it might be a console lock problem.
Please retry with the following patch:
--- linux-2.3.46-rsbac-1.0.9b/rsbac/help/debug.c Wed Feb 23 14:02:24 2000
+++ linux-2.3.47-rsbac-1.0.9b/rsbac/help/debug.c Sun Feb 27 11:22:19 2000
@@ -18,6 +18,7 @@
#include <linux/init.h>
#include <linux/config.h>
#include <linux/module.h>
+#include <linux/console.h>
/* Boolean debug switch for data structures */
int rsbac_debug_ds = 0;
@@ -705,7 +706,7 @@
return error;
}
-static spinlock_t rsbac_console_lock;
+/* static spinlock_t rsbac_console_lock; */
#ifdef CONFIG_RSBAC_REG
EXPORT_SYMBOL(rsbac_printk);
@@ -719,7 +720,7 @@
static signed char msg_level = -1;
long flags;
- spin_lock_irqsave(&rsbac_console_lock, flags);
+ spin_lock_irqsave(&console_lock, flags);
va_start(args, fmt);
i = vsprintf(buf + 3, fmt, args); /* hopefully i < sizeof(buf)-4 */
buf_end = buf + 3 + i;
@@ -759,7 +760,7 @@
if (line_feed)
msg_level = -1;
}
- spin_unlock_irqrestore(&rsbac_console_lock, flags);
+ spin_unlock_irqrestore(&console_lock, flags);
wake_up_interruptible(&rlog_wait);
return i;
}
@@ -1894,7 +1895,7 @@
if (!rsbac_is_initialized())
{
-#if defined(CONFIG_RSBAC_RMSG)
+#if 0 && defined(CONFIG_RSBAC_RMSG)
/* init console lock */
spin_lock_init(&rsbac_console_lock);
#endif
--
Please remove second ao for E-Mail reply - no spam please!
## CrossPoint v3.11 ##
-
To unsubscribe from the rsbac list, send a mail to
majordomo@morpork.shnet.org with
unsubscribe rsbac
as single line in the body.
Next Article (by Date): Re: 1.0.9b-pre2 uploaded (2.2.13 SMP testing) ao@morpork.shnet.org (A. Ott)
Previous Article (by Date): Re: 1.0.9b-pre2 uploaded (2.2.13 SMP testing) Jesse Pollard
Top of Thread: Re: 1.0.9b-pre2 uploaded (2.2.13 SMP testing) Jesse Pollard
Next in Thread: Re: 1.0.9b-pre2 uploaded (2.2.13 SMP testing) ao@morpork.shnet.org (A. Ott)
Articles sorted by: [Date]
[Author]
[Subject]