From mboxrd@z Thu Jan  1 00:00:00 1970
From: Albert ARIBAUD <albert.u.boot@aribaud.net>
Date: Mon, 20 Feb 2012 11:01:24 +0100
Subject: [U-Boot] [PATCH] UBIFS: Improve error message when reading
 superblock failed
In-Reply-To: <4F420E33.3000002@corscience.de>
References: <20120217143110.GA32414@agamemnon.er.corscience.de>
	<1329727489-31423-1-git-send-email-walle@corscience.de>
	<4F420B82.2070908@aribaud.net> <4F420E33.3000002@corscience.de>
Message-ID: <4F4219F4.3080005@aribaud.net>
List-Id: <u-boot.lists.denx.de>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
To: u-boot@lists.denx.de

Hi Bernhard,

Le 20/02/2012 10:11, Bernhard Walle a ?crit :
> Am 20.02.2012 09:59, schrieb Albert ARIBAUD:
>>
>> Dry numbers as error messages are better than no error messages but only
>> marginally IMO. Isn't there a way to emit a readable message re malloc
>> instead of emitting an int value?
>
> Well, I'm not familiar with the u-boot codebase. Does u-boot have a
> strerror table? How is it handled on other places? Should an error
> message printed directly before returning -ENOMEM?

I don't think there is an strerror() API. But at least, if the newly 
printed int uses errno values, then instead of '... %d!", you could 
print "... errno=%d", which would give some hint to the reader.

> Regards,
> Bernhard

Amicalement,
-- 
Albert.