git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Junio C Hamano <gitster@pobox.com>
To: "Ævar Arnfjörð Bjarmason" <avarab@gmail.com>
Cc: git@vger.kernel.org, Bagas Sanjaya <bagasdotme@gmail.com>,
	Abhradeep Chakraborty <chakrabortyabhradeep79@gmail.com>
Subject: Re: [PATCH 1/5] usage.c: add a non-fatal bug() function to go with BUG()
Date: Wed, 25 May 2022 13:55:05 -0700	[thread overview]
Message-ID: <xmqqpmk15o46.fsf@gitster.g> (raw)
In-Reply-To: patch-1.5-faa1c708a79-20220521T170939Z-avarab@gmail.com

Ævar Arnfjörð Bjarmason  <avarab@gmail.com> writes:

> Any caller to bug() should follow up such calls with BUG_if_bug(),
> which will BUG() out (i.e. abort()) if there were any preceding calls
> to bug(). As the tests and documentation here show we'll catch missing
> BUG_if_bug() invocations in our exit() wrapper.

...

> +- `bug` (lower-case, not `BUG`) is supposed to be used like `BUG` but
> +  prints a "BUG" message instead of calling `abort()`. We then expect
> +  `BUG_if_bug()` to be called to `abort()` if there were any calls to
> +  `bug()`. If there weren't any a call to `BUG_if_bug()` is a NOOP.

OK.  So the expected pattern would be a series of calls to bug(),
each guarded by its own condition, concluded by a call to BUG_if_bug()

	if (condition1)
		bug(...);
	if (condition2)
		bug(...);
	...
	BUG_if_bug();

and when none of the guard conditions fired, BUG_if_bug() will
become a no-op.

> +/* usage.c: if bug() is called we should have a BUG_if_bug() afterwards */
> +extern int bug_called_must_BUG;

I am not sure about the name, but in essense, each call to bug()
ensures that this becomes true, so BUG_if_bug() can use it to see
if it should abort(), right?

>  __attribute__((format (printf, 3, 4))) NORETURN
>  void BUG_fl(const char *file, int line, const char *fmt, ...);
>  #define BUG(...) BUG_fl(__FILE__, __LINE__, __VA_ARGS__)
> +__attribute__((format (printf, 3, 4)))
> +void bug_fl(const char *file, int line, const char *fmt, ...);
> +#define bug(...) bug_fl(__FILE__, __LINE__, __VA_ARGS__)
> +#define BUG_if_bug() do { \
> +	if (bug_called_must_BUG) { \
> +		bug_called_must_BUG = 0; \
> +		BUG_fl(__FILE__, __LINE__, "see bug() output above"); \
> +	} \
> +} while (0)

I have a feeling that "see bug() output above" should come from the
caller of BUG_if_bug().  These bug() calls that are grouped together
must have a shared overall theme, which may want to be stated in
that final message.

Other than these two small points, this does not look too bad ;-)

Thanks.

  reply	other threads:[~2022-05-25 20:55 UTC|newest]

Thread overview: 49+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-05-21 17:14 [PATCH 0/5] usage API: add and use a bug() + BUG_if_bug() Ævar Arnfjörð Bjarmason
2022-05-21 17:14 ` [PATCH 1/5] usage.c: add a non-fatal bug() function to go with BUG() Ævar Arnfjörð Bjarmason
2022-05-25 20:55   ` Junio C Hamano [this message]
2022-05-26  4:17     ` Junio C Hamano
2022-05-26  7:59       ` Ævar Arnfjörð Bjarmason
2022-05-26 16:55         ` Junio C Hamano
2022-05-26 18:29           ` Ævar Arnfjörð Bjarmason
2022-05-26 18:55             ` Junio C Hamano
2022-05-31 16:52           ` Ævar Arnfjörð Bjarmason
2022-05-27 17:03   ` Josh Steadmon
2022-05-27 19:05     ` Junio C Hamano
2022-05-21 17:14 ` [PATCH 2/5] parse-options.c: use new bug() API for optbug() Ævar Arnfjörð Bjarmason
2022-05-27 17:03   ` Josh Steadmon
2022-05-21 17:14 ` [PATCH 3/5] parse-options.c: use optbug() instead of BUG() "opts" check Ævar Arnfjörð Bjarmason
2022-05-25 21:05   ` Junio C Hamano
2022-05-21 17:14 ` [PATCH 4/5] receive-pack: use bug() and BUG_if_bug() Ævar Arnfjörð Bjarmason
2022-05-25 21:15   ` Junio C Hamano
2022-05-27 17:04   ` Josh Steadmon
2022-05-27 22:53   ` Andrei Rybak
2022-05-21 17:14 ` [PATCH 5/5] cache-tree.c: " Ævar Arnfjörð Bjarmason
2022-05-27 21:29   ` Glen Choo
2022-05-27 17:02 ` [PATCH 0/5] usage API: add and use a bug() + BUG_if_bug() Josh Steadmon
2022-05-31 16:58 ` [PATCH v2 0/6] " Ævar Arnfjörð Bjarmason
2022-05-31 16:58   ` [PATCH v2 1/6] common-main.o: move non-trace2 exit() behavior out of trace2.c Ævar Arnfjörð Bjarmason
2022-06-01 18:37     ` Junio C Hamano
2022-05-31 16:58   ` [PATCH v2 2/6] usage.c: add a non-fatal bug() function to go with BUG() Ævar Arnfjörð Bjarmason
2022-05-31 18:18     ` Glen Choo
2022-06-01 18:50     ` Junio C Hamano
2022-05-31 16:58   ` [PATCH v2 3/6] parse-options.c: use new bug() API for optbug() Ævar Arnfjörð Bjarmason
2022-05-31 16:58   ` [PATCH v2 4/6] parse-options.c: use optbug() instead of BUG() "opts" check Ævar Arnfjörð Bjarmason
2022-05-31 17:38     ` Josh Steadmon
2022-06-01 18:55       ` Junio C Hamano
2022-05-31 16:58   ` [PATCH v2 5/6] receive-pack: use bug() and BUG_if_bug() Ævar Arnfjörð Bjarmason
2022-05-31 16:58   ` [PATCH v2 6/6] cache-tree.c: " Ævar Arnfjörð Bjarmason
2022-06-01 18:52     ` Junio C Hamano
2022-05-31 17:39   ` [PATCH v2 0/6] usage API: add and use a bug() + BUG_if_bug() Josh Steadmon
2022-06-02 12:25   ` [PATCH v3 " Ævar Arnfjörð Bjarmason
2022-06-02 12:25     ` [PATCH v3 1/6] common-main.o: move non-trace2 exit() behavior out of trace2.c Ævar Arnfjörð Bjarmason
2022-06-03  1:19       ` Junio C Hamano
2022-06-07 20:09       ` Josh Steadmon
2022-06-07 21:12         ` Ævar Arnfjörð Bjarmason
2022-06-07 22:05           ` Josh Steadmon
2022-06-02 12:25     ` [PATCH v3 2/6] usage.c: add a non-fatal bug() function to go with BUG() Ævar Arnfjörð Bjarmason
2022-06-02 12:25     ` [PATCH v3 3/6] parse-options.c: use new bug() API for optbug() Ævar Arnfjörð Bjarmason
2022-06-02 12:25     ` [PATCH v3 4/6] parse-options.c: use optbug() instead of BUG() "opts" check Ævar Arnfjörð Bjarmason
2022-06-02 12:25     ` [PATCH v3 5/6] receive-pack: use bug() and BUG_if_bug() Ævar Arnfjörð Bjarmason
2022-06-02 12:25     ` [PATCH v3 6/6] cache-tree.c: " Ævar Arnfjörð Bjarmason
2022-06-02 19:54       ` Junio C Hamano
2022-06-02 16:56     ` [PATCH v3 0/6] usage API: add and use a bug() + BUG_if_bug() Glen Choo

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=xmqqpmk15o46.fsf@gitster.g \
    --to=gitster@pobox.com \
    --cc=avarab@gmail.com \
    --cc=bagasdotme@gmail.com \
    --cc=chakrabortyabhradeep79@gmail.com \
    --cc=git@vger.kernel.org \
    /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).