linux-kernel - Re: [Lsf-pc] [LSF/MM/BPF TOPIC] tracing the source of errors

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <871q9mvg2d.fsf@mailhost.krisman.be>
Date: Thu, 08 Feb 2024 15:47:38 -0500
From: Gabriel Krisman Bertazi <krisman@...e.de>
To: Matthew Wilcox <willy@...radead.org>
Cc: Miklos Szeredi <miklos@...redi.hu>,  Jan Kara <jack@...e.cz>,  lsf-pc
 <lsf-pc@...ts.linux-foundation.org>,  linux-fsdevel@...r.kernel.org,
  linux-kernel@...r.kernel.org
Subject: Re: [Lsf-pc] [LSF/MM/BPF TOPIC] tracing the source of errors

Matthew Wilcox <willy@...radead.org> writes:

> Option (b)
>
> -#define EINVAL		22
> +#define E_INVAL	22
> +#define EINVAL		ERR(E_INVAL)

Note there will surely be cases where EINVAL is used as a soft failure
and the kernel will just try something else, instead of propagating the
error up the stack.  In this case, there is no point in logging the
first case of error, as it will just be expected behavior.

So there's really no way around explicitly annotating (ERR (EINVAL))
in place where it really matters, instead of changing the definition
of -EINVAL itself or automatically converting check sites.

>
> and then change all code that does something like:
>
> 	if (err == -EINVAL)
> to
> 	if (err == -E_INVAL)
>
> Or have I misunderstood?

-- 
Gabriel Krisman Bertazi