From: Simon Marchi <simark@simark.ca>
To: "H.J. Lu" <hjl.tools@gmail.com>
Cc: GCC Patches <gcc-patches@gcc.gnu.org>,
"H.J. Lu via Binutils" <binutils@sourceware.org>,
Aaron Merey <amerey@redhat.com>,
Andreas Schwab <schwab@linux-m68k.org>,
Tom Tromey <tom@tromey.com>, GDB <gdb-patches@sourceware.org>
Subject: Re: V2 [PATCH] PKG_CHECK_MODULES: Check if $pkg_cv_[]$1[]_LIBS works
Date: Tue, 28 Jul 2020 11:13:35 -0400 [thread overview]
Message-ID: <3209078a-429a-4be7-b151-93c3f4a53655@simark.ca> (raw)
In-Reply-To: <CAMe9rOobJcEydFKYdj8Eqp17oOtVSg7T2sUgQegYHK-D2ec1Pg@mail.gmail.com>
On 2020-07-28 11:05 a.m., H.J. Lu via Gdb-patches wrote:
>> Can you clarify how this magic works, is this standard autoconf? Because I am trying this
>> on Fedora, so pretty much the same setup as you, and I don't see this behavior:
>>
>> $ /home/simark/src/binutils-gdb/configure CC="gcc -m32" CXX="g++ -m32"
>> checking build system type... x86_64-pc-linux-gnu
>> checking host system type... x86_64-pc-linux-gnu
>> checking target system type... x86_64-pc-linux-gnu
>> ...
>>
>
> I checked it again. I also passed i686-linux, not --host=, to configure.
Ok I see, the configure line you pasted was wrapped by your email client so was not very readable.
Doing `./configure <triplet>` looks like a deprecated way to set all build/host/target, as the
warning message it shows implies:
configure: WARNING: you should use --build, --host, --target
Anyway, my point still stands: the problem is you not using a pkg-config configured properly for
the cross compilation you are attempting, not pkg.m4.
>>
>> And even if it worked, why would it set "build" to i686, it doesn't make sense. The gcc you
>> compile with, and its environment, is still x86_64, not i686.
>>
>> So *if* it works, it would be a shortcut for setting --host=i686-something, maybe. So
>> regardless of how that above works, that doesn't remove the need to configure pkg-config
>> correctly for the host system.
>>
>> If you don't want to learn about with pkg-config and deal with it, then please say "I think
>> we should not use pkg-config", and ideally give supporting points. Please don't unilaterally
>> push patches just to paper over your own problems. You just make it so that somebody will
>> need to untangle more mess later.
>
> I am OK to remove pkg.m4.
Well I am not. I find it quite handy to avoid having to hardcode necessary CFLAGS and LDFLAGS
required to build against a library, so I think it's better to use pkg-config if the libraries
we want to use provide a .pc file.
I propose that we revert the patch for now to go back to the pristing pkg.m4 version.
Simon
next prev parent reply other threads:[~2020-07-28 15:13 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <20200502022903.175852-1-amerey@redhat.com>
2020-05-05 16:41 ` [PATCH] config/debuginfod.m4: Use PKG_CHECK_MODULES Jeff Law
2020-05-05 21:23 ` Aaron Merey
2020-05-08 20:56 ` Tom Tromey
2020-07-16 22:17 ` Aaron Merey
2020-07-16 22:37 ` Aaron Merey
2020-07-19 15:43 ` Simon Marchi
2020-07-21 15:20 ` Tom Tromey
2020-07-21 18:11 ` Aaron Merey
2020-07-24 20:03 ` Aaron Merey
2020-07-25 16:01 ` H.J. Lu
2020-07-27 15:31 ` [PATCH] config/debuginfod.m4: Restore AC_CHECK_LIB check H.J. Lu
2020-07-27 16:11 ` Aaron Merey
2020-07-27 19:14 ` [PATCH] PKG_CHECK_MODULES: Check if $pkg_cv_[]$1[]_LIBS works H.J. Lu
2020-07-27 19:32 ` V2 " H.J. Lu
2020-07-28 10:45 ` H.J. Lu
2020-07-28 12:46 ` Simon Marchi
2020-07-28 13:33 ` H.J. Lu
2020-07-28 13:51 ` Andreas Schwab
2020-07-28 13:56 ` H.J. Lu
2020-07-28 14:01 ` Simon Marchi
2020-07-28 14:11 ` H.J. Lu
2020-07-28 14:34 ` Simon Marchi
2020-07-28 15:05 ` H.J. Lu
2020-07-28 15:13 ` Simon Marchi [this message]
2020-07-28 16:07 ` H.J. Lu
2020-07-28 16:28 ` Simon Marchi
2020-07-28 17:26 ` H.J. Lu
2020-07-28 17:43 ` Simon Marchi
2020-07-28 18:31 ` H.J. Lu
2020-07-28 18:47 ` Simon Marchi
2020-07-28 18:57 ` H.J. Lu
2020-07-28 14:54 ` Andreas Schwab
2020-07-28 13:53 ` H.J. Lu
2020-07-28 13:54 ` Simon Marchi
2020-07-28 12:54 ` Mark Wielaard
2020-07-28 13:27 ` [PATCH] PKG_CHECK_MODULES: Properly check " H.J. Lu
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=3209078a-429a-4be7-b151-93c3f4a53655@simark.ca \
--to=simark@simark.ca \
--cc=amerey@redhat.com \
--cc=binutils@sourceware.org \
--cc=gcc-patches@gcc.gnu.org \
--cc=gdb-patches@sourceware.org \
--cc=hjl.tools@gmail.com \
--cc=schwab@linux-m68k.org \
--cc=tom@tromey.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.
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).