archive mirror
 help / color / mirror / Atom feed
From: Jeff King <>
To: Junio C Hamano <>
Cc: Jonathan Tan <>,, Jeff Hostetler <>
Subject: Re: [PATCH] usage: trace2 BUG() invocations
Date: Fri, 5 Feb 2021 04:01:20 -0500	[thread overview]
Message-ID: <> (raw)
In-Reply-To: <>

On Thu, Feb 04, 2021 at 10:17:29PM -0800, Junio C Hamano wrote:

> Jonathan Tan <> writes:
> > die() messages are traced in trace2, but BUG() messages are not. Anyone
> > tracking die() messages would have even more reason to track BUG().
> > Therefore, write to trace2 when BUG() is invoked.
> >
> > Signed-off-by: Jonathan Tan <>
> > ---
> > This was noticed when we observed at $DAYJOB that a certain BUG()
> > invocation [1] wasn't written to traces.
> >
> > [1]
> > ---
> >  t/helper/test-trace2.c   |  9 +++++++++
> >  t/ | 19 +++++++++++++++++++
> >  usage.c                  |  6 ++++++
> >  3 files changed, 34 insertions(+)
> Sounds like a good idea.  Expert opinions?

I like the overall idea, but it does open the possibility of a BUG() in
the trace2 code looping infinitely.

For instance, injecting this bug:

  diff --git a/trace2/tr2_tgt_event.c b/trace2/tr2_tgt_event.c
  index 6353e8ad91..b386bae402 100644
  --- a/trace2/tr2_tgt_event.c
  +++ b/trace2/tr2_tgt_event.c
  @@ -242,6 +242,7 @@ static void fn_error_va_fl(const char *file, int line, const char *fmt,
          if (fmt && *fmt)
                  jw_object_string(&jw, "fmt", fmt);
  +       jw_end(&jw);
          tr2_dst_write_line(&tr2dst_event, &jw.json);

and running something like:

  GIT_TRACE2_EVENT=1 ./git config --file=does.not.exist --list

currently yields:

  [...a bunch of trace2 lines...]
  BUG: json-writer.c:397: json-writer: too many jw_end(): '{"event":"error",[...etc...]

With this patch it loops forever (well, probably until it runs out of
memory, but I stopped it after one minute and 10G of RAM). ;)

Presumably triggering this in practice would be pretty rare.  But then,
the point of BUG() is to find things we do not expect to happen.

We've had a similar problem on the die() side in the past, and solved it
with a recursion flag. But note it gets a bit non-trivial in the face of
threads. There's some discussion in 1ece66bc9e (run-command: use
thread-aware die_is_recursing routine, 2013-04-16).

That commit talks about a case where "die()" in a thread takes down the
thread but not the whole process. That wouldn't be true here (we'd
expect BUG() to take everything down). So a single counter might be OK
in practice, though I suspect we could trigger the problem racily
Likewise this is probably a lurking problem when other threaded code
calls die(), but we just don't do that often enough for anybody to have


  reply	other threads:[~2021-02-05  9:02 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-02-05  5:49 [PATCH] usage: trace2 BUG() invocations Jonathan Tan
2021-02-05  6:17 ` Junio C Hamano
2021-02-05  9:01   ` Jeff King [this message]
2021-02-05 12:51     ` Derrick Stolee
2021-02-05 13:44       ` Jeff King
2021-02-05 16:34       ` Jeff Hostetler
2021-02-05 20:09 ` [PATCH v2] " Jonathan Tan
2021-02-09 12:03   ` Jeff King
2021-02-09 19:34     ` Jonathan Tan
2021-02-09 21:18       ` Junio C Hamano
2021-02-09 22:15       ` Junio C Hamano
2021-03-27 17:56   ` Ævar Arnfjörð Bjarmason

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:

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

  git send-email \ \ \ \ \ \ \

* 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).