From: Junio C Hamano <gitster@pobox.com>
Date: Tue, 04 Aug 2009 06:07:30 GMT
Subject: Re: [PATCH] Fix compiler warning by properly initialize failed_errno
Message-ID: <7vmy6g6rj1.fsf@alter.siamese.dyndns.org>
In-Reply-To: <1249241675-77329-1-git-send-email-sn_@gmx.net>

David Soria Parra <sn_@gmx.net> writes:

> From: David Soria Parra <dsp@php.net>
>
> Initilize failed_error in start_command to avoid compiler warnings
>
> Signed-off-by: David Soria Parra <dsp@php.net>
> ---
>  run-command.c |    2 +-
>  1 files changed, 1 insertions(+), 1 deletions(-)
>
> diff --git a/run-command.c b/run-command.c
> index dc09433..510349b 100644
> --- a/run-command.c
> +++ b/run-command.c
> @@ -19,7 +19,7 @@ int start_command(struct child_process *cmd)
>  {
>  	int need_in, need_out, need_err;
>  	int fdin[2], fdout[2], fderr[2];
> -	int failed_errno;
> +	int failed_errno = 0;
>  
>  	/*
>  	 * In case of errors we must keep the promise to close FDs

We would want to be able to distinguish between a workaround for a
compiler that is not clever/careful enough, and a necessary
initialization.  In this particular case, it is the former, and we should
say

	int failed_errno = failed_errno;

instead.

The potentially uninitialized use your compiler is worried about is inside
if (cmd->pid < 0) after #ifdef/#else/#endif.

 (1) if not on MINGW32, we would have already assigned to failed_errno
     after fork() returns negative value to cmd->pid;

 (2) if on MINGW32, we would have assigned to failed_errno unconditionally
     after calling mingw_spawnvpe().

so its worry is unfounded.