git@vger.kernel.org mailing list mirror (one of many)
 help / color / mirror / code / Atom feed
From: Jeff King <peff@peff.net>
To: "Nguyễn Thái Ngọc Duy" <pclouds@gmail.com>
Cc: git@vger.kernel.org, larsxschneider@gmail.com,
	Junio C Hamano <gitster@pobox.com>
Subject: Re: [PATCH] Makefile: detect compiler and enable more warnings in DEVELOPER=1
Date: Sun, 18 Mar 2018 05:26:30 -0400	[thread overview]
Message-ID: <20180318092629.GB29208@sigill.intra.peff.net> (raw)
In-Reply-To: <20180318081834.16081-1-pclouds@gmail.com>

On Sun, Mar 18, 2018 at 09:18:34AM +0100, Nguyễn Thái Ngọc Duy wrote:

> The set of extra warnings we enable when DEVELOPER has to be
> conservative because we can't assume any compiler version the
> developer may use. Detect the compiler version so we know when it's
> safe to enable -Wextra and maybe more.

Hrm, I was planning to expand my patch into a series. In particular, in
my experiments some of those warnings do not need to be disabled under
-Wextra (I tested going back to gcc 5 and clang 4.0).

I think it's worth splitting up the boilerplate changes from the
decisions on individual warnings, because I suspect we'll need to revise
the latter going forward.

I also noticed that all of those options seem to go back at least to
gcc-5. It makes me wonder how much we even need this version-detecting
framework. There are one or two "clang vs gcc" things I've found, but I
haven't found a case where old versions of gcc don't support a
particular option.

IOW, could we get away with just adding these to the DEVELOPER knob and
assuming that people who use it have a reasonably modern gcc or clang
setup? That's more or less what the existing knobs do.

But I didn't follow whether there was any earlier discussion on specific
problems.

>  -dumpversion does not work correctly for clang. So I use "-v" instead
>  which seems to produce the same output for both gcc and clang (with a
>  minor difference in freebsd where it says "FreeBSD clang" instead of
>  just "clang"). Not sure if it helps your "cc on debian" case though

Heh. At first I was confused, as it seems to work for me:

  $ clang-4.0 -dumpversion
  4.2.1

But then I tried this:

  $ clang-5.0 -dumpversion
  4.2.1

Whoops. It returns the same value up through clang 7.

-Peff

  parent reply	other threads:[~2018-03-18  9:26 UTC|newest]

Thread overview: 47+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-03-03  1:46 [PATCH] travis-ci: enable more warnings on travis linux-gcc job Nguyễn Thái Ngọc Duy
2018-03-16 19:33 ` [PATCH v2] " Nguyễn Thái Ngọc Duy
2018-03-16 21:22   ` Jeff King
2018-03-17  8:01     ` Duy Nguyen
2018-03-17 14:29       ` Lars Schneider
2018-03-17 14:59         ` Duy Nguyen
2018-03-17 16:08           ` Jeff King
2018-03-17 16:12             ` Jeff King
2018-03-17 23:50             ` Junio C Hamano
2018-03-18  8:18             ` [PATCH] Makefile: detect compiler and enable more warnings in DEVELOPER=1 Nguyễn Thái Ngọc Duy
2018-03-18  9:06               ` Eric Sunshine
2018-03-18  9:17                 ` Duy Nguyen
2018-03-18  9:20                   ` Jeff King
2018-03-18  9:24                   ` Eric Sunshine
2018-03-18  9:28                 ` Jeff King
2018-03-18  9:37                   ` Eric Sunshine
2018-03-18  9:26               ` Jeff King [this message]
2018-03-18  9:45                 ` Duy Nguyen
2018-03-18 15:55               ` Duy Nguyen
2018-03-18 18:56                 ` Ramsay Jones
2018-03-19 16:30                   ` Duy Nguyen
2018-03-20  5:32                 ` Jeff King
2018-03-24 12:53               ` [PATCH v2] " Nguyễn Thái Ngọc Duy
2018-03-25  0:40                 ` Eric Sunshine
2018-03-26 22:02                 ` Junio C Hamano
2018-03-27 15:03                   ` Duy Nguyen
2018-03-27 16:52                     ` Junio C Hamano
2018-03-29 15:03                 ` [PATCH v3 0/3] Enable more compiler warnings for devs Nguyễn Thái Ngọc Duy
2018-03-29 15:03                   ` [PATCH v3 1/3] connect.c: mark die_initial_contact() NORETURN Nguyễn Thái Ngọc Duy
2018-03-29 15:03                   ` [PATCH v3 2/3] Makefile: detect compiler and enable more warnings in DEVELOPER=1 Nguyễn Thái Ngọc Duy
2018-03-29 15:03                   ` [PATCH v3 3/3] Makefile: add EAGER_DEVELOPER mode Nguyễn Thái Ngọc Duy
2018-03-31 16:40                   ` [PATCH 4/3] Makefile: untangle DEVELOPER and -Werror Ævar Arnfjörð Bjarmason
2018-03-31 18:36                     ` Duy Nguyen
2018-04-03  9:19                       ` Ævar Arnfjörð Bjarmason
2018-04-03 15:17                         ` Duy Nguyen
2018-04-06 21:42                           ` Jeff King
2018-04-07  9:52                             ` Duy Nguyen
2018-04-07 12:36                               ` Ævar Arnfjörð Bjarmason
2018-04-07 17:03                                 ` Duy Nguyen
2018-04-07 18:38                                   ` Ævar Arnfjörð Bjarmason
2018-04-14 19:19                   ` [PATCH v4 0/4] Make DEVELOPER more more flexible with DEVOPTS Ævar Arnfjörð Bjarmason
2018-04-16  4:57                     ` Junio C Hamano
2018-04-14 19:19                   ` [PATCH v4 1/4] connect.c: mark die_initial_contact() NORETURN Ævar Arnfjörð Bjarmason
2018-04-14 19:19                   ` [PATCH v4 2/4] Makefile: detect compiler and enable more warnings in DEVELOPER=1 Ævar Arnfjörð Bjarmason
2018-04-14 19:19                   ` [PATCH v4 3/4] Makefile: add a DEVOPTS to suppress -Werror under DEVELOPER Ævar Arnfjörð Bjarmason
2018-04-14 19:19                   ` [PATCH v4 4/4] Makefile: add a DEVOPTS to get all of -Wextra Ævar Arnfjörð Bjarmason
2018-03-17 15:16         ` [PATCH v2] travis-ci: enable more warnings on travis linux-gcc job 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=20180318092629.GB29208@sigill.intra.peff.net \
    --to=peff@peff.net \
    --cc=git@vger.kernel.org \
    --cc=gitster@pobox.com \
    --cc=larsxschneider@gmail.com \
    --cc=pclouds@gmail.com \
    /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).