From: Pedro Alves <pedro@codesourcery.com>
To: gdb-patches@sourceware.org, Tom Tromey <tromey@redhat.com>
Subject: Re: RFA: automatic dependency tracking
Date: Thu, 31 Jul 2008 18:51:00 -0000 [thread overview]
Message-ID: <200807311950.46853.pedro@codesourcery.com> (raw)
In-Reply-To: <m34p6742hu.fsf@fleche.redhat.com>
On Wednesday 30 July 2008 20:05:01, Tom Tromey wrote:
> Pedro> I trust Tom. :-)
>
> Thanks. Though... we all make mistakes :)
I trust you to be around to fix them.
> One way you could help move this forward would be to test it on a
> machine that does not have GNU make. I don't have one of these.
As you know, I tested building GDB with BSD make. I used an x86
FreeBSD 6.0-RELEASE.
src/opcodes needs a little tweak to build,
http://sourceware.org/ml/binutils/2008-07/msg00403.html
but other than that, pristine GDB builds fine with BSD make.
> The patch probably doesn't apply cleanly any more. Unfortunately this
> sort of patch requires constant care to keep it up-to-date.
I tested with a checkout of the date of your patch to avoid that issue.
GDB built perfectly with your patch, with BSD make with and
without -j4 and with gmake also {,-j4}, always from a clean build dir.
I used:
>gcc -v
Using built-in specs.
Configured with: FreeBSD/i386 system compiler
Thread model: posix
gcc version 3.4.4 [FreeBSD] 20050518
In addition to doing a couple of touch,make tests, to check if
dependencies are being respected with gmake, I tested building gdb
in a clean build dir, waiting a couple of minutes, and then
touching defs.h, and issuing gmake again.
By checking timestamps, I found that these are the files weren't
rebuilt by touching defs.h:
18:19 ..
18:22 .gdbinit
18:22 Makefile
18:22 ada-exp.c
18:22 ada-lex.c
18:22 c-exp.c
18:22 config.cache
18:22 config.h
18:22 config.log
18:22 config.status
18:22 cp-name-parser.c
18:22 doc
18:22 f-exp.c
18:22 gnulib
18:22 jv-exp.c
18:22 m2-exp.c
18:22 nm.h
18:22 objc-exp.c
18:22 observer.h
18:22 observer.inc
18:22 p-exp.c
18:22 stamp-h
18:22 stamp-h1
18:22 stamp-xml
18:22 testsuite
18:22 version.c
18:22 xml-builtin.c
18:23 cp-name-parser.o
18:23 version.o
18:24 xml-builtin.o
With BSD make, touching header files doesn't trigger recompiles, as
expected.
--
Pedro Alves
next prev parent reply other threads:[~2008-07-31 18:51 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-07-21 22:14 Tom Tromey
2008-07-30 17:53 ` Pedro Alves
2008-07-30 19:12 ` Tom Tromey
2008-07-31 18:51 ` Pedro Alves [this message]
2008-07-31 10:48 ` Andrew STUBBS
2008-07-31 12:16 ` Tom Tromey
2008-07-31 14:43 ` Tom Tromey
2008-08-02 20:34 ` Thiago Jung Bauermann
2008-08-07 5:50 ` Ralf Wildenhues
2008-08-07 13:47 ` Tom Tromey
2008-08-07 13:53 ` Daniel Jacobowitz
2008-08-07 19:17 ` Ralf Wildenhues
2008-08-07 20:07 ` Tom Tromey
2008-08-07 20:46 ` Ralf Wildenhues
2008-08-07 20:51 ` Tom Tromey
2008-08-07 20:17 ` Daniel Jacobowitz
2008-08-07 20:22 ` Tom Tromey
2008-08-07 20:46 ` Ralf Wildenhues
2008-08-07 20:49 ` Tom Tromey
2008-08-07 20:59 ` Daniel Jacobowitz
2008-08-07 22:51 ` Ralf Wildenhues
2008-08-08 14:37 ` Thiago Jung Bauermann
2008-08-08 15:18 ` Tom Tromey
2008-08-08 15:24 ` Ralf Wildenhues
2008-08-08 17:27 ` Tom Tromey
2008-08-08 17:35 ` Daniel Jacobowitz
2008-08-07 17:09 ` Tom Tromey
2008-08-07 19:23 ` Ralf Wildenhues
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=200807311950.46853.pedro@codesourcery.com \
--to=pedro@codesourcery.com \
--cc=gdb-patches@sourceware.org \
--cc=tromey@redhat.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