Buildroot Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
To: buildroot@busybox.net
Subject: [Buildroot] [PATCH v2] libdrm: fix static build (tests/nouveau)
Date: Sun, 11 Aug 2019 15:38:46 +0200	[thread overview]
Message-ID: <20190811153846.68028925@windsurf.home> (raw)
In-Reply-To: <20190809231908.22209-1-ps.report@gmx.net>

Hello Peter,

Thanks for working on this topic. Some questions/comments below.

On Sat, 10 Aug 2019 01:19:08 +0200
Peter Seiderer <ps.report@gmx.net> wrote:

> Some toolchains (e.g. br-arm-cortex-m4-full) provide empty libdl
> libraries. This fools the dynamic/static detection for tests/nouveau,
> so explicit check for library type instead.
> 
> Fixes [1]:
> 
>   ../tests/nouveau/threaded.c:24:10: fatal error: dlfcn.h: No such file or directory

So the error is about not finding a header.

> + # Among others FreeBSD does not have a separate dl library.
> + if not cc.has_function('dlsym')
> ++  # fooled in case empty libdl provided, e.g. toolchain br-arm-cortex-m4-full
> +   dep_dl = cc.find_library('dl', required : with_nouveau)

What about instead checking for the availability of dlfcn.h, in
addition to checking the availability of the libdl library ?

Also, how come the cc.has_function('dlsym') checks returns true ? I
guess in a static library configuration, this check should return false.

Also, could you drop the references to br-arm-cortex-m4-full, so that
the patch can be upstreamed ?

BTW, with uClibc-ng and musl, the libdl library is always empty,
regardless of whether it has dynamic library support or not. Here is a
uClibc-ng toolchain, with dynamic library support:

-rw-r--r-- 1 thomas thomas 8  1 ao?t  22:23 output/host/usr/arm-buildroot-linux-uclibcgnueabi/sysroot/usr/lib/libdl.a

Indeed, since quite a while, uClibc-ng is all in a single file,
libc.so, including symbols that used to be in libdl.so:

$ ./output/host/bin/arm-linux-readelf -s output/host/usr/arm-buildroot-linux-uclibcgnueabi/sysroot/lib/libc.so.1  | grep "FUNC.*dlsym"
   884: 00058fe0   468 FUNC    GLOBAL DEFAULT    8 dlsym
  7356: 00058fe0   468 FUNC    GLOBAL DEFAULT    8 dlsym

Best regards,

Thomas
-- 
Thomas Petazzoni, CTO, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com

  reply	other threads:[~2019-08-11 13:38 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-08-09 23:19 [Buildroot] [PATCH v2] libdrm: fix static build (tests/nouveau) Peter Seiderer
2019-08-11 13:38 ` Thomas Petazzoni [this message]
2019-08-11 21:55   ` Peter Seiderer

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=20190811153846.68028925@windsurf.home \
    --to=thomas.petazzoni@bootlin.com \
    --cc=buildroot@busybox.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.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox