kdb
[Top] [All Lists]

Re: [PATCH] KDB x86_64 should use register_die_notifier

To: minyard@xxxxxxx
Subject: Re: [PATCH] KDB x86_64 should use register_die_notifier
From: Keith Owens <kaos@xxxxxxx>
Date: Sat, 02 Jun 2007 21:28:41 +1000
Cc: kdb@xxxxxxxxxxx, cminyard@xxxxxxxxxx
In-reply-to: Your message of "Fri, 01 Jun 2007 14:23:12 EST." <20070601192312.GA21882@localdomain>
Sender: kdb-bounce@xxxxxxxxxxx
Corey Minyard (on Fri, 01 Jun 2007 14:23:12 -0500) wrote:
>I was working on KDB and ran into a problem on initialization on x86_64.
>There's some magic that happens with register_die_notifier to make it
>work correctly.  die_chain should not be directly accessed.  KDB should
>go through register_die_notifier.
>
>Signed-off-by: Corey Minyard <cminyard@xxxxxxxxxx>
>
>Index: linux-2.6.21/arch/x86_64/kdb/kdbasupport.c
>===================================================================
>--- linux-2.6.21.orig/arch/x86_64/kdb/kdbasupport.c
>+++ linux-2.6.21/arch/x86_64/kdb/kdbasupport.c
>@@ -915,7 +915,7 @@ kdba_init(void)
> {
>       kdb_register("pt_regs", kdba_pt_regs, "address", "Format struct 
> pt_regs", 0);
>       kdb_register("cpu_pda", kdba_cpu_pda, "<cpu>", "Format struct cpu_pda", 
> 0);
>-      atomic_notifier_chain_register(&die_chain, &kdba_notifier);
>+      register_die_notifier(&kdba_notifier);
>       return;
> }

Thanks, but kdb-v4.4-2.6.22-rc1-x86_64-1 (2007-05-22) already has that
change.

---------------------------
Use http://oss.sgi.com/ecartis to modify your settings or to unsubscribe.
<Prev in Thread] Current Thread [Next in Thread>