All of lore.kernel.org
 help / color / mirror / Atom feed
From: Peter Urbanec <openembedded-devel@urbanec.net>
To: Otavio Salvador <otavio@ossystems.com.br>,
	Richard Purdie <richard.purdie@linuxfoundation.org>
Cc: Patches and discussions about the oe-core layer
	<openembedded-core@lists.openembedded.org>
Subject: Re: [PATCH] gstreamer1.0: Shorten __FILE__ in gst_debug_log output on all platforms.
Date: Fri, 27 Feb 2015 14:28:05 +1100	[thread overview]
Message-ID: <54EFE445.7030209@urbanec.net> (raw)
In-Reply-To: <CAP9ODKoZO1m-2BomHz6h_YtmwKsoEURsCxAMXPmYN3X6E1xqfQ@mail.gmail.com>

On 27/02/15 09:00, Otavio Salvador wrote:
> On Thu, Feb 26, 2015 at 6:47 PM, Richard Purdie
> <richard.purdie@linuxfoundation.org> wrote:
>> On Fri, 2015-02-27 at 04:31 +1100, Peter Urbanec wrote:
>>> On WIN32 the file argument to gst_debug_log_valist is shortened to just
>>> the filename. This is useful not only for MSVC, but also with gcc/Linux
>>> when doing cross-compilation builds and out-of-tree builds.
>>
>> Ultimately I think we need to address this issue in gcc itself, probably
>> setting some kind of base path in the environment which it removes from
>> __FILE__ (set to ${S}). There were more complex discussions about using
>> the same mapping code as can be used with the debug symbols code too.
>
> Agreed; this should indeed be a global fix as this affects virtually
> all packages.

I also agree, but this patch can go ahead regardless of what the global 
solution is. The patch enables a mechanism that already exists in 
gstreamer (but is only enabled on WIN32) and solves an immediate need in 
that context. Whatever the global solution is, this patch is very likely 
to continue working correctly.

Having a proper solution at the gcc level will significantly reduce the 
size of the generated files, so there is certainly a win to be had 
there. I had a look at both C99 and C++98 standards and the semantics of 
the __FILE__ macro are implementation defined and do not stipulate any 
particular interpretation. In particular, there's no requirement to have 
path information preserved.

I think a simple patch to gcc to transform a full pathname to basename 
is probably sufficient for oe-core - the less complexity is introduced 
the better.


      parent reply	other threads:[~2015-02-27  3:28 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-02-26 17:31 [PATCH] gstreamer1.0: Shorten __FILE__ in gst_debug_log output on all platforms Peter Urbanec
2015-02-26 21:47 ` Richard Purdie
2015-02-26 21:56   ` Andreas Oberritter
2015-02-26 22:11     ` Otavio Salvador
2015-02-26 22:27       ` Andreas Oberritter
2015-02-26 22:00   ` Otavio Salvador
2015-02-26 22:37     ` Randy Witt
2015-02-26 22:40       ` Otavio Salvador
2015-02-27  3:28     ` Peter Urbanec [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=54EFE445.7030209@urbanec.net \
    --to=openembedded-devel@urbanec.net \
    --cc=openembedded-core@lists.openembedded.org \
    --cc=otavio@ossystems.com.br \
    --cc=richard.purdie@linuxfoundation.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.