X-Authentication-Warning: delorie.com: mail set sender to geda-user-bounces using -f
X-Recipient: geda-user AT delorie DOT com
Date: Mon, 14 Dec 2020 22:29:20 +0100 (CET)
From: Roland Lutz <rlutz AT hedmen DOT org>
To: "Glenn (glimrick AT epilitimus DOT com) [via geda-user AT delorie DOT com]" <geda-user AT delorie DOT com>
Subject: Re: [geda-user] Memory leak patches
In-Reply-To: <4df42e11-5211-8177-4afc-e3aaaf9a61f4@epilitimus.com>
Message-ID: <alpine.DEB.2.21.2012142220380.1180@nimbus>
References: <e4ff3c96-939b-a93e-a32f-5e938b6daa63 AT epilitimus DOT com> <alpine DOT DEB DOT 2 DOT 21 DOT 2011302045040 DOT 2894 AT nimbus> <20201130220505 DOT 0AE4282C54FD AT turkos DOT aspodata DOT se> <7c75ed03-456c-b408-8b50-0448f6b3a04f AT epilitimus DOT com> <alpine DOT DEB DOT 2 DOT 21 DOT 2012011606400 DOT 6785 AT nimbus>
 <1b2c64b3-6a36-c1f3-dd54-bb583c6bea17 AT epilitimus DOT com> <alpine DOT DEB DOT 2 DOT 21 DOT 2012021253440 DOT 1262 AT nimbus> <475f980e-fddd-60d1-9a02-a5bc5fb5805b AT epilitimus DOT com> <alpine DOT DEB DOT 2 DOT 21 DOT 2012071333290 DOT 3907 AT nimbus> <e6c5135f-b127-fd55-91b2-79aa357a5a07 AT epilitimus DOT com>
 <alpine DOT DEB DOT 2 DOT 21 DOT 2012081644480 DOT 3649 AT nimbus> <8706c896-e01a-9e6e-2bab-fe3607064093 AT epilitimus DOT com> <alpine DOT DEB DOT 2 DOT 21 DOT 2012111645180 DOT 13096 AT nimbus> <63273f86-b0fd-8835-a13e-6ab408f06dfc AT epilitimus DOT com> <alpine DOT DEB DOT 2 DOT 21 DOT 2012121301180 DOT 1221 AT nimbus>
 <4a1fbfca-3921-f674-1499-f11df8d5a58e AT epilitimus DOT com> <alpine DOT DEB DOT 2 DOT 21 DOT 2012141510100 DOT 22156 AT nimbus> <4df42e11-5211-8177-4afc-e3aaaf9a61f4 AT epilitimus DOT com>
User-Agent: Alpine 2.21 (DEB 202 2017-01-01)
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="8323329-270082969-1607981360=:1180"
Reply-To: geda-user AT delorie DOT com
Errors-To: nobody AT delorie DOT com
X-Mailing-List: geda-user AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk

  This message is in MIME format.  The first part should be readable text,
  while the remaining parts are likely unreadable without MIME-aware tools.

--8323329-270082969-1607981360=:1180
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8BIT

On Mon, 14 Dec 2020, Glenn (glimrick AT epilitimus DOT com) [via 
geda-user AT delorie DOT com] wrote:
> Something is definitely leaking memory. I couldn't figure out what which 
> is why I just marked it, It may be a leak in the library and so not 
> fixable.
>
> There are actually three leaks listed. The first two look to be the same
> leak triggered twice:
>
>> ==16700== 11,175 (3,840 direct, 7,335 indirect) bytes in 5 blocks are
>> definitely lost in loss record 13,377 of 13,440
>> ==16700==    at 0x4839D7B: realloc (vg_replace_malloc.c:836)
>> ==16700==    by 0x5318299: ??? (in /usr/lib64/libfontconfig.so.1.8.0)
>> ==16700==    by 0x5318921: ??? (in /usr/lib64/libfontconfig.so.1.8.0)
>> ==16700==    by 0x5315C6A: FcFontRenderPrepare (in
>> /usr/lib64/libfontconfig.so.1.8.0)
>> ==16700==    by 0x531614F: FcFontMatch (in
>> /usr/lib64/libfontconfig.so.1.8.0)
>> ==16700==    by 0x50F3285: ??? (in /usr/lib64/libpangoft2-1.0.so.0.3800.1)
>> ==16700==    by 0x50F342C: ??? (in /usr/lib64/libpangoft2-1.0.so.0.3800.1)
>> ==16700==    by 0x5FE34ED: pango_context_get_metrics (in
>> /usr/lib64/libpango-1.0.so.0.3800.1)

This appears to be current_fonts->patterns->match.  As far as I can tell, 
the last part of the stack trace would read (static functions in braces):

FcFontMatch
[pango_fc_patterns_get_font_pattern]
[pango_fc_fontset_load_next_font]
[pango_fc_fontset_get_font_at]
[pango_fc_fontset_foreach]
[pango_fontset_foreach]
[get_base_metrics]
pango_context_get_metrics

It *may* be that this is not a library leak as much as a usage error, like 
a Pango context which hasn't been freed somewhere.  In that case, I'd 
expect there to be another leak, though; this is most certainly not an 
error in eda_renderer_get_font_descent.

The other two leaks are much harder to trace (I gave up around 
pango_font_map_load_fontset), but I think it's safe to assume that they 
are of a similar kind.

Roland

--8323329-270082969-1607981360=:1180--