Mirror of the gdb-patches mailing list
 help / color / mirror / Atom feed
From: Michael Hope <michael.hope@linaro.org>
To: Mark Kettenis <mark.kettenis@xs4all.nl>
Cc: gdb-patches@sourceware.org, patches@linaro.org
Subject: Re: [PATCH] Check for POSIX_MADV_WILLNEED to support building against the LSB
Date: Mon, 11 Jun 2012 21:40:00 -0000	[thread overview]
Message-ID: <CANLjY-nz=V7bzWViShNo24RPtX6wB_2AsXTDPh0GppeqGLeekQ@mail.gmail.com> (raw)
In-Reply-To: <201206110818.q5B8ImK9025818@glazunov.sibelius.xs4all.nl>

On 11 June 2012 20:18, Mark Kettenis <mark.kettenis@xs4all.nl> wrote:
>> From: Michael Hope <michael.hope@linaro.org>
>> Date: Mon, 11 Jun 2012 18:16:51 +1200
>>
>> On 11 June 2012 17:59, Mark Kettenis <mark.kettenis@xs4all.nl> wrote:
>> >> Date: Mon, 11 Jun 2012 16:25:22 +1200
>> >> From: Michael Hope <michael.hope@linaro.org>
>> >>
>> >> The Linux Standard Base APIs include posix_madvise() but don't define
>> >> values for the 'advice' argument.  Check to see if POSIX_MADV_WILLNEED
>> >> is defined before using.
>> >
>> > What real-world problem does this fix?
>>
>> Being able to build GDB using the LSB tools so that you can make a
>> binary release that works on any Linux distro from RHEL4 and above.
>> We use it for the linaro-toolchain-binaries.
>
> I fear you're using a broken toolchain, because...
>
>> > It makes no sense to have
>> > posix_madvise() but not define any values for the 'advice' argument.
>> > I'd say this is a "bug" in LSB.  Either they should drop
>> > posix_madvise() from the standard, or include at least the 'advice'
>> > values required by POSIX.
>>
>> The LSB has a policy of not dropping functions.  I can suggest that
>> they add the advise values for a later version but that will take some
>> time and won't cover existing LSB releases.
>
> Both posix_madvise():
>
> http://www.linuxbase.org/navigator/browse/int_single.php?cmd=list-by-name&Iname=posix_madvise&Ilibrary=libc
>
> and POSIX_MADV_WILLNEED:
>
> http://www.linuxbase.org/navigator/browse/headgroup.php?cmd=list-byheadgroup&HGid=56
>
> were added in LSB 3.2.

I've dug a bit deeper and there seems to be a bug in the LSB 3.2
release.  posix_madvise() was defined but the advice constants weren't
added to the header files.  See sys/mman.h in:
 http://ftp.linuxbase.org/pub/lsb/lsbdev/released-3.2.0/binary/amd64/lsb-build-base-3.2.2-1.x86_64.rpm

Ubuntu ships with this version.  I'll change our build system to use
4.1.3 in the 3.2 compatibility mode instead.

Ta,

-- Michael


      reply	other threads:[~2012-06-11 21:40 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-06-11  4:25 Michael Hope
2012-06-11  5:59 ` Mark Kettenis
2012-06-11  6:17   ` Michael Hope
2012-06-11  8:19     ` Mark Kettenis
2012-06-11 21:40       ` Michael Hope [this message]

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='CANLjY-nz=V7bzWViShNo24RPtX6wB_2AsXTDPh0GppeqGLeekQ@mail.gmail.com' \
    --to=michael.hope@linaro.org \
    --cc=gdb-patches@sourceware.org \
    --cc=mark.kettenis@xs4all.nl \
    --cc=patches@linaro.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