Re: [PATCH 2/5] common-main: call git_extract_argv0_path()

From: Johannes Schindelin <johannes.schindelin@gmx.de>
Date: Jul 1, 2016, 08:05 UTC
Message-ID: <alpine.DEB.2.20.1607011000320.12947@virtualbox>
In-Reply-To: <20160701060404.GB4593@sigill.intra.peff.net>

Hi Peff,

On Fri, 1 Jul 2016, Jeff King wrote:

Show 20 quoted lines

> This has been an accident-waiting-to-happen for a long time,
> but wasn't triggered until recently because it involves one
> of those programs actually calling system_path(). That
> happened with git-credential-store in v2.8.0 with ae5f677
> (lazily load core.sharedrepository, 2016-03-11). The
> program:
> 
>   - takes a lock file, which...
> 
>   - opens a tempfile, which...
> 
>   - calls adjust_shared_perm to fix permissions, which...
> 
>   - lazy-loads the config (as of ae5f677), which...
> 
>   - calls system_path() to find the location of
>     /etc/gitconfig
> 
> On systems with RUNTIME_PREFIX, this means credential-store
> reliably hits that assert() and cannot be used.

Thank you for that thorough write-up. I am now even more upset that we had to go through the same steps (it took me an hour to figure out what was going wrong, mostly due to abort() *not* spitting out a stack trace, so I had to wield some gdb magic).

I am partly to blame here, of course, because I did not report what I did to this mailing list. But then: 1) I considered this a Windows-only problem, and 2) I was really already swamped, as it were.

The patch is good, of course, as are the rest of the patches (I did not really look at them very thoroughly, but then, they are pretty obvious improvements).

Ciao, Dscho