git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Antonio Ospite <ao2@ao2.it>
To: Elijah Newren <newren@gmail.com>
Cc: Git Mailing List <git@vger.kernel.org>,
	Johannes Schindelin <johannes.schindelin@gmx.de>,
	Jeff King <peff@peff.net>
Subject: Re: [PATCH] wrap-for-bin.sh: facilitate running Git executables under valgrind
Date: Wed, 9 May 2018 17:56:47 +0200	[thread overview]
Message-ID: <20180509175647.0961d469e01367783090e764@ao2.it> (raw)
In-Reply-To: <CABPp-BEvNOBkq0-v_Uq0CHkvRixCKmUhYPMeH-MHHZGb0x9NkA@mail.gmail.com>

On Wed, 9 May 2018 08:25:21 -0700
Elijah Newren <newren@gmail.com> wrote:

> Hi Antonio,
> 

Hi Elijah,

> On Wed, May 9, 2018 at 6:28 AM, Antonio Ospite <ao2@ao2.it> wrote:
> > Testing locally built git executables under valgrind is not immediate.
> >
> > Something like the following does not work:
> >
> >   $ valgrind ./bin-wrappers/git
> >
> > because the wrapper script forks and execs the command and valgrind does
> > not track children processes by default.
> >
> > Something like the following may work:
> >
> >   $ valgrind --trace-children=yes ./bin-wrappers/git
> >
> > However it's counterintuitive and not ideal anyways because valgrind is
> > supposed to be called on the actual executable, not on wrapper scripts.
> >
> > So, following the idea from commit 6a94088cc ("test: facilitate
> > debugging Git executables in tests with gdb", 2015-10-30) provide
> > a mechanism in the wrapper script to call valgrind directly on the
> > actual executable.
> >
> > This mechanism could even be used by the test infrastructure in the
> > future, but it is already useful by its own on the command line:
> >
> >   $ GIT_TEST_VALGRIND=1 \
> >     GIT_VALGRIND_OPTIONS="--leak-check=full" \
> >     ./bin-wrappers/git
> >
> 
> Wow, timing; nice to see someone else finds this kind of thing useful.
> 
> I submitted something very similar recently; see commit 842436466aa5
> ("Make running git under other debugger-like programs easy",
> 2018-04-24) from next, or the discussion at
> https://public-inbox.org/git/20180424234645.8735-1-newren@gmail.com/.
> That other patch has the advantage of enabling the user to run git
> under other debugger-like programs besides just gdb and valgrind.
> 

Thanks Elijah, I am not subscribed to the list so I didn't see your
change and I usually only track the master branch.

Obviously your changes work for me, so I am dropping my patch.

As the changes in 842436466aa5 ("Make running git under other
debugger-like programs easy", 2018-04-24) are not specific to valgrind
they should also address Jeff's concerns in the sense that it's up to
the particular GIT_DEBUGGER how it handles sub-processes.

In valgrind case one may still want to pass "--trace-children=yes" in
GIT_DEBUGGER after all for better coverage. Thank you Jeff for the
remark.

Ciao,
   Antonio

-- 
Antonio Ospite
https://ao2.it
https://twitter.com/ao2it

A: Because it messes up the order in which people normally read text.
   See http://en.wikipedia.org/wiki/Posting_style
Q: Why is top-posting such a bad thing?

      reply	other threads:[~2018-05-09 15:56 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-05-09 13:28 [PATCH] wrap-for-bin.sh: facilitate running Git executables under valgrind Antonio Ospite
2018-05-09 14:49 ` Jeff King
2018-05-09 15:25 ` Elijah Newren
2018-05-09 15:56   ` Antonio Ospite [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20180509175647.0961d469e01367783090e764@ao2.it \
    --to=ao2@ao2.it \
    --cc=git@vger.kernel.org \
    --cc=johannes.schindelin@gmx.de \
    --cc=newren@gmail.com \
    --cc=peff@peff.net \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).