[PATCH] x86, mce: Print warning if MCE handler fails to register /dev/mcelog

[PATCH] x86, mce: Print warning if MCE handler fails to register /dev/mcelog

[Khalid Aziz]

Print a warning if machine check handler fails to register
/dev/mcelog device.


Signed-off-by: Khalid Aziz <khalid.aziz@oracle.com>
---
 arch/x86/kernel/cpu/mcheck/mce.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c
index 7bc1263..7b4c28c 100644
--- a/arch/x86/kernel/cpu/mcheck/mce.c
+++ b/arch/x86/kernel/cpu/mcheck/mce.c
@@ -2412,7 +2412,8 @@ static __init int mcheck_init_device(void)
 	register_hotcpu_notifier(&mce_cpu_notifier);
 
 	/* register character device /dev/mcelog */
-	misc_register(&mce_chrdev_device);
+	if (misc_register(&mce_chrdev_device) != 0)
+		pr_warn("Failed to register mcelog device\n");
 
 	return err;
 }
-- 
1.7.10.4

RE: [PATCH] x86, mce: Print warning if MCE handler fails to register /dev/mcelog

[Luck, Tony]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset="utf-8", Size: 599 bytes --]

-	misc_register(&mce_chrdev_device);
+	if (misc_register(&mce_chrdev_device) != 0)
+		pr_warn("Failed to register mcelog device\n");

Did this actually happen to you?  Or is this just "good practice" to check the
return value from misc_register?

If this can really happen, we should remember that it failed and
just drop out of mce_log() early (no point in queueing up records
when there is no way for a user to ever access them).

-Tony
ÿôèº{.nÇ+‰·Ÿ®‰­†+%ŠËÿ±éݶ\x17¥Šwÿº{.nÇ+‰·¥Š{±þG«éÿŠ{ayº\x1dʇڙë,j\a­¢f£¢·hšïêÿ‘êçz_è®\x03(­éšŽŠÝ¢j"ú\x1a¶^[m§ÿÿ¾\a«þG«éÿ¢¸?™¨è­Ú&£ø§~á¶iO•æ¬z·švØ^\x14\x04\x1a¶^[m§ÿÿÃ\fÿ¶ìÿ¢¸?–I¥

Re: [PATCH] x86, mce: Print warning if MCE handler fails to register /dev/mcelog

[Borislav Petkov]

On Tue, Apr 09, 2013 at 06:30:58PM +0000, Luck, Tony wrote:
> -	misc_register(&mce_chrdev_device);
> +	if (misc_register(&mce_chrdev_device) != 0)
> +		pr_warn("Failed to register mcelog device\n");
> 
> Did this actually happen to you?  Or is this just "good practice" to check the
> return value from misc_register?

I'm very sceptical this actually happens - we would've otherwise seen
reports about mcelog not starting correctly.

Hmm.

-- 
Regards/Gruss,
    Boris.

Sent from a fat crate under my desk. Formatting is fine.
--

Re: [PATCH] x86, mce: Print warning if MCE handler fails to register /dev/mcelog

[Khalid Aziz]

On 04/09/2013 12:30 PM, Luck, Tony wrote:
> -	misc_register(&mce_chrdev_device);
> +	if (misc_register(&mce_chrdev_device) != 0)
> +		pr_warn("Failed to register mcelog device\n");
>
> Did this actually happen to you?  Or is this just "good practice" to check the
> return value from misc_register?
>

No, I did not see this happen. It is just "good practice". If it were to 
happen, mcelog daemon would fail to get any mce records and it will not 
be easy to figure out what happened. Most other places in kernel where 
misc_register() is called, do check for return code and take action if 
it fails.

--
Khalid