Buildroot Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Nathaniel Roach <nroach44@gmail.com>
To: buildroot@busybox.net
Subject: [Buildroot] [PATCH] bandwidthd: fix static build
Date: Sun, 05 Oct 2014 21:01:43 +0800	[thread overview]
Message-ID: <54314137.7020007@gmail.com> (raw)
In-Reply-To: <20141005144043.7b3aaac0@free-electrons.com>

On 05/10/14 20:40, Thomas Petazzoni wrote:
> Baruch, Nathaniel,
> 
> On Sun, 5 Oct 2014 07:50:37 +0300, Baruch Siach wrote:
> 
>>> Hi Baruch, I didn't end up using pcap-config, but could you check git
>>> version 4b07a0b3d3a280cdde582060cb29f3333ba4bf6e for me? It seems to
>>> work in the config you pasted above and I haven't had any issues in my
>>> other tests.
>>
>> I can confirm that upgrading bandwidthd to 
>> 4b07a0b3d3a280cdde582060cb29f3333ba4bf6e fixes the build of the config above. 
>> I still think that holding the complete knowledge of all your indirect 
>> dependencies, mandatory and optional, is not robust. A better future proof 
>> solution IMO is to use tools like pkg-config (or pcap-config in the case of 
>> libpcap) to list all dependencies that are in actual use for this specific 
>> build. But this is your call as upstream.
> 
> I definitely agree with Baruch here. Using pkg-config is much more
> robust, as it figures out the indirect dependencies automatically,
> without hardcoding them in bandwidthd and/or Buildroot.
> 
> However, Baruch patch doesn't seem to be fully correct: it does both a
> PKG_CHECK_MODULES on libpng, and an AC_CHECK_LIB, which seems a bit
> redundant.
> 
> Therefore, Nathaniel, since you're now the upstream developer, could
> you make the necessary changes in configure.ac to use pkg-config
> instead, and then send a patch updating Buildroot to a new version of
> bandwidthd ?
> 
> In the mean time, I'll mark:
> 
>   http://patchwork.ozlabs.org/patch/395809/
>   http://patchwork.ozlabs.org/patch/381899/
> 
> as 'Changes requested'.
> 
> Thanks,
> 
> Thomas
> 

Thomas, Baruch:

Yeah, I'm starting to lean that way too. Although I was a little
hesitant adding another build dependency, I now realise that it's
probably getting built anyway, and the version that's currently up on
github was more a case of "getting it to work".

I've had a look at pcap-config and it doesn't seem to do -lpthreads (or
from what I could see anything useful in this case), so I'll likely
adapt Baruch's original patch into upstream for both lpcap and lpng.

It might take me a few days as my study load has increased, but I'll get
it done.

Thanks,

Nathaniel.

  reply	other threads:[~2014-10-05 13:01 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-10-02  6:49 [Buildroot] [PATCH] bandwidthd: fix static build Baruch Siach
     [not found] ` <542CFECE.9020103@gmail.com>
2014-10-02  7:52   ` Baruch Siach
2014-10-02  7:56     ` Nathaniel Roach
2014-10-02  7:58       ` Baruch Siach
2014-10-02  8:24         ` Nathaniel Roach
2014-10-02  8:33           ` Baruch Siach
2014-10-03  6:35             ` Nathaniel Roach
2014-10-05  4:50               ` Baruch Siach
2014-10-05 12:40                 ` Thomas Petazzoni
2014-10-05 13:01                   ` Nathaniel Roach [this message]
2014-10-05 13:12                     ` Thomas Petazzoni

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=54314137.7020007@gmail.com \
    --to=nroach44@gmail.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