Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 3 Feb 2026 08:46:49 +0900
From:      Tatsuki Makino <tatsuki_makino@hotmail.com>
To:        Charlie Li <vishwin@freebsd.org>, "freebsd-ports@FreeBSD.org" <freebsd-ports@FreeBSD.org>
Subject:   Re: graphics/gegl: gegl_find_relocatable_exe: assertion 'file != NULL' failed
Message-ID:  <SI2PR01MB5036CC01EB4718F387D04CB9FA9AA@SI2PR01MB5036.apcprd01.prod.exchangelabs.com>
In-Reply-To: <4c9a0c95-427a-4cab-aa2d-3279f8f2c710@freebsd.org>
References:  <SI2PR01MB503630C89930B873EB21C78AFA9AA@SI2PR01MB5036.apcprd01.prod.exchangelabs.com> <4c9a0c95-427a-4cab-aa2d-3279f8f2c710@freebsd.org>

index | next in thread | previous in thread | raw e-mail

Thank you very much.

On 2026/02/03 2:50, Charlie Li wrote:
> Currently dealing with upstream over the meaning of "relocatable". Turns out this does not mean ELF relocation. When the feature is enabled, since there is no FreeBSD-specific preprocessor guard, the Linux path is taken, which includes reading stuff from Linux's procfs. In theory we should take whatever approach works for macOS, but I'm still working with upstream on how best to support this feature even though it should be disabled by default.
> 
> For now, I will disable the feature for this port too.
> 

This time, in my environment, linprocfs started being mounted constantly :), so it seems that I should have looked inside it.
The function gegl_find_relocatable_exe doesn't seem to be doing anything particularly fancy.
It seems to me that we can handle it simply by making the following replacements.

/proc/self -> /proc/curproc
/proc/self/exe -> /proc/curproc/file
/proc/self/maps -> /proc/curproc/map and, change the column to be read.

Well, I don't know how the values obtained in that way will be used later, so that's just my guess :)

Regards.



home | help

Want to link to this message? Use this
URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?SI2PR01MB5036CC01EB4718F387D04CB9FA9AA>