git@vger.kernel.org mailing list mirror (one of many)
 help / color / mirror / code / Atom feed
From: "René Scharfe" <l.s.r@web.de>
To: Taylor Blau <me@ttaylorr.com>, git@vger.kernel.org
Cc: Jeff King <peff@peff.net>, Junio C Hamano <gitster@pobox.com>
Subject: Re: [PATCH] git-compat-util.h: introduce CALLOC(x)
Date: Mon, 5 Dec 2022 22:01:11 +0100	[thread overview]
Message-ID: <a8e33b1e-1056-5f75-55b5-65c0bceef3ca@web.de> (raw)
In-Reply-To: <6694c52b38674859eb0390c7f62da1209a8d8ec3.1670266373.git.me@ttaylorr.com>

Am 05.12.22 um 19:54 schrieb Taylor Blau:
> When zero-initializing a struct without the use of static initializers
> like "{ 0 }", it is common to write one of:
>
>     T *ptr = xcalloc(1, sizeof(T));
>     T *ptr = xcalloc(1, sizeof(*ptr));
>
> These correctly initialize "*ptr" to the all-zeros value, but have a
> couple of drawbacks. Notably, both initializations are verbose, but the
> former is a foot-gun. If "*ptr"'s type changes to something other than
> "T", the programmer has to remember to update not only the type of the
> variable, but the right-hand side of its initialization, too.
>
> In git.git, it is sometimes common to write something like:
>
>     T *ptr;
>     CALLOC_ARRAY(ptr, 1);
>
> ...but that is confusing, since we're not initializing an array. Rather,
> we're using the CALLOC_ARRAY() macro to pretend like we are. But since
> the array length is 1, the effect is zero initializing a single element.

An object and a single-element array of objects allocated on the heap are
indistinguishable.  In that sense there is no confusion -- we are indeed
allocating a single-element array.  But if the intent is to only get one
thing then having to fill in 1 in the bulk order form is a bit tedious,
especially since this is the most common kind of request.  A shortcut for
the frequent case makes sense.

> Introduce a new variant, CALLOC(x), which initializes "x" to the
> all-zeros value, without exposing the confusing use of CALLOC_ARRAY(),
> or the foot-guns available when using xcalloc() directly.

AFAIK the first "c" in "calloc" is for "continuous", not "zeroed".  A
single object is always contiguous, so the "C" in "CALLOC" is
tautologic if read in that way.  It fits our naming scheme for
ALLOC_ARRAY and CALLOC_ARRAY, though, so that might not be much of a
problem.

And there are lots of occurrences of xmalloc(sizeof(T)) and
xmalloc(sizeof(*ptr)) that would benefit from the automatic size
inference of ALLOC_ARRAY -- an ALLOC macro would complement
CALLOC nicely.

> Write a Coccinelle patch which codifies these rules, but mark it as
> "pending" since the resulting diff is too large to include in a single
> patch:
>
>     $ git apply .build/contrib/coccinelle/xcalloc.pending.cocci.patch
>     $ git diff --shortstat
>      89 files changed, 221 insertions(+), 178 deletions(-)
>
> Signed-off-by: Taylor Blau <me@ttaylorr.com>
> ---
> This is a follow-up on [1], where introducing CALLOC(x) as the preferred
> alternative to CALLOC_ARRAY(x, 1) was first suggested.
>
> The patch is straightforward, and I am pretty sure that the Coccinelle
> rules are right, too ;-). But as a first-time Coccinelle user, any extra
> scrutiny on those bits would be appreciated.
>
> The main point of discussion I think is whether we should consider
> adopting this rule. I am biased, of course, but I think that we should.
>
> In any case, we should focus our efforts on polishing v2.39.0, but I
> wanted to send this out to the list before I forgot about it.
>
> [1]: https://lore.kernel.org/git/Y1MrXoobkVKngYL1@coredump.intra.peff.net/
>
>  contrib/coccinelle/xcalloc.pending.cocci | 21 +++++++++++++++++++++
>  git-compat-util.h                        |  8 ++++++++
>  2 files changed, 29 insertions(+)
>  create mode 100644 contrib/coccinelle/xcalloc.pending.cocci
>
> diff --git a/contrib/coccinelle/xcalloc.pending.cocci b/contrib/coccinelle/xcalloc.pending.cocci
> new file mode 100644
> index 0000000000..83e4ca1a68
> --- /dev/null
> +++ b/contrib/coccinelle/xcalloc.pending.cocci
> @@ -0,0 +1,21 @@
> +@@
> +type T;
> +T *ptr;
> +@@
> +- ptr = xcalloc(1, \( sizeof(T) \| sizeof(*ptr) \) )
> ++ CALLOC(ptr)
> +
> +@@
> +type T;
> +identifier ptr;
> +@@
> +- T ptr = xcalloc(1, \( sizeof(T) \| sizeof(*ptr) \) );
> ++ T ptr;
> ++ CALLOC(ptr);

This rule would turn this code:

	struct foo *bar = xcalloc(1, sizeof(*bar));
	int i;

... into:

	struct foo *bar;
	CALLOC(bar);
	int i;

... which violates the coding guideline to not mix declarations and
statements (-Wdeclaration-after-statement).

> +
> +@@
> +type T;
> +T *ptr;
> +@@
> +- CALLOC_ARRAY(ptr, 1)
> ++ CALLOC(ptr)
> diff --git a/git-compat-util.h b/git-compat-util.h
> index a76d0526f7..827e5be89c 100644
> --- a/git-compat-util.h
> +++ b/git-compat-util.h
> @@ -1107,6 +1107,14 @@ static inline void move_array(void *dst, const void *src, size_t n, size_t size)
>  		memmove(dst, src, st_mult(size, n));
>  }
>
> +/*
> + * Like CALLOC_ARRAY, but the argument is treated as a pointer to a
> + * single struct.
> + *
> + * Preferable over xcalloc(1, sizeof(...)), or CALLOC_ARRAY(..., 1).
> + */
> +#define CALLOC(x) (x) = CALLOC_ARRAY((x), 1)
> +
>  /*
>   * These functions help you allocate structs with flex arrays, and copy
>   * the data directly into the array. For example, if you had:
> --
> 2.38.0.16.g393fd4c6db


  reply	other threads:[~2022-12-05 21:03 UTC|newest]

Thread overview: 21+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-12-05 18:54 [PATCH] git-compat-util.h: introduce CALLOC(x) Taylor Blau
2022-12-05 21:01 ` René Scharfe [this message]
2022-12-05 22:36   ` Taylor Blau
2022-12-05 23:12     ` Ævar Arnfjörð Bjarmason
2022-12-06  1:47       ` Jeff King
2022-12-06  1:58         ` Ævar Arnfjörð Bjarmason
2022-12-07  6:02           ` Jeff King
2022-12-07  2:36         ` Taylor Blau
2022-12-07  2:34       ` Taylor Blau
2022-12-07  3:17         ` Ævar Arnfjörð Bjarmason
2022-12-06  1:43     ` Jeff King
2022-12-07  2:29       ` Taylor Blau
2022-12-07  3:51         ` Ævar Arnfjörð Bjarmason
2022-12-05 23:57 ` Junio C Hamano
2022-12-06  0:29   ` Taylor Blau
2022-12-06  1:21     ` Jeff King
2022-12-06  1:35       ` Junio C Hamano
2022-12-07  2:38         ` Taylor Blau
2022-12-07  6:08           ` Jeff King
2022-12-06  2:12     ` Ævar Arnfjörð Bjarmason
2022-12-07  6:06       ` Jeff King

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

  List information: http://vger.kernel.org/majordomo-info.html

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

  git send-email \
    --in-reply-to=a8e33b1e-1056-5f75-55b5-65c0bceef3ca@web.de \
    --to=l.s.r@web.de \
    --cc=git@vger.kernel.org \
    --cc=gitster@pobox.com \
    --cc=me@ttaylorr.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.
Code repositories for project(s) associated with this public inbox

	https://80x24.org/mirrors/git.git

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 read-only IMAP folder(s) and NNTP newsgroup(s).