Re: [PATCH] 2.4.16 kernel/printk.c (per processor initializationcheck)

[j-nomura@ce.jp.nec.com]

Hi,
excuse me for not prompt response. I've been off-line for 2 days.

> > The reason I put it in release_console_sem() is that release_console_sem()
> > can be called from other functions than printk(), e.g. console_unblank().
> > I agree with you that it is clearer but I think it is not sufficient.
> 
> I really doubt if any of those paths could be called before
> even the MMU is set up.

I didn't have any intention to say that console_unblank() is called so early.

OK. Here is revised patch which checks if cpu initialization is done
just before down_trylock(). This works for me.

diff -u -r1.1.1.8 printk.c
--- kernel/printk.c	2001/11/27 04:41:49	1.1.1.8
+++ kernel/printk.c	2001/12/06 04:54:50
@@ -438,7 +438,13 @@
 			log_level_unknown = 1;
 	}
 
-	if (!down_trylock(&console_sem)) {
+	if (!(cpu_online_map & 1UL << smp_processor_id())) {
+		/*
+		 * The cpu has not been initialized completely
+		 * enough to call console drivers.
+		 */
+		spin_unlock_irqrestore(&logbuf_lock, flags);
+	} else if (!down_trylock(&console_sem)) {
 		/*
 		 * We own the drivers.  We can drop the spinlock and let
 		 * release_console_sem() print the text



Best regards.
--
NOMURA, Jun'ichi <j-nomura@ce.jp.nec.com, nomura@hpc.bs1.fc.nec.co.jp>
HPC Operating System Group, 1st Computers Software Division,
Computers Software Operations Unit, NEC Solutions.