X-Authentication-Warning: delorie.com: mail set sender to geda-user-bounces using -f
X-Recipient: geda-user AT delorie DOT com
X-Original-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed;
        d=gmail.com; s=20161025;
        h=mime-version:references:in-reply-to:from:date:message-id:subject:to;
        bh=Yri1FkluL3epVV68Fq3UQfyPKY0brZDtFdemJ2lnVt4=;
        b=J34VROx3eYQyi92HS3CIkwvUTwrFfyZdxNJ9Sed6GskdAQrukuN+IiXsOtHM7iY5Lp
         wlduCFB/eDskXOXjp3iaGFwsvvYLH2o6pDBc6D60g0fH+06uQVOofc+gEroQwzSV65RS
         8DxXrNYnXTF58V/c43oNKL0gQZmVKKUepgE2eK6NHgHD9jMjdK6vy0Kxs9lsD2rP0Y9j
         UneOWCPKcP/Eqq6lYIKzAnqXj9gVUfUjKzEZ9B8nbAhnf+sDP8O2xUo8qlD6dmhMK+sS
         n+wSHDOFISuJu/bF14VBDWh6v1eX6/mpBIVgtjfRpX/AQGo1ibk1ltGT9r3zNKRAatT4
         xYqg==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed;
        d=1e100.net; s=20161025;
        h=x-gm-message-state:mime-version:references:in-reply-to:from:date
         :message-id:subject:to;
        bh=Yri1FkluL3epVV68Fq3UQfyPKY0brZDtFdemJ2lnVt4=;
        b=PID8449Z+o8UIga8M/dSRFA2ilG/4dkKmAUvvLegrT/c3qAkyXqXhnzdGNa0FADzye
         Uu0yh0iGafZP5J5uyVUaZaflMOoaiGOpVJIsOf4t0eLlTTblNNIOOoWhwC0qWBX4cQi7
         WgBC7g9TScnRy1aoBHVG7xV146wafm0Ni++dExW1pIxljRHDeNWTmcLlekL/gbMS1EId
         3y/FTpMHJ+RljKCG++frPEoTiIaynByuYgOqkqqgs+pMXSehwgwlCgZmq+wyX6dIY6l+
         FTKzPhYxzsj5CpLo/awUHFJ9bPhyR6JzgY9ImPT5muu6ySnumEH+zNiqdgojihb67JcP
         0iWQ==
X-Gm-Message-State: AOAM530WvBgczxKVPlaNBqlloetGDe+4O9bazv9iZVxcu0t9p6s5DV4m
	LCZzpdmvMbcz2SU9ABKPoaTNtA59qyxaQMwxM24otbkh8vjJEg==
X-Google-Smtp-Source: ABdhPJy2DTmvg0BYeEEBbG0RUSFh2qBl3ACmiQqfS2BPoiYGXLksA6hf1zCWJPvzQq3CDoPWIsDFO7pVzKGYeyHOJj8=
X-Received: by 2002:a67:7347:: with SMTP id o68mr10280976vsc.36.1629129491769;
 Mon, 16 Aug 2021 08:58:11 -0700 (PDT)
MIME-Version: 1.0
References: <611a1807 DOT 1c69fb81 DOT d149c DOT 293bSMTPIN_ADDED_BROKEN AT mx DOT google DOT com>
 <CAJZxidA+gjtR6kM4Wo4==xSsMD9u0Cz=xkY0Cms7o7fvpdG+bQ AT mail DOT gmail DOT com>
 <CAJZxidBF-csCQTfiUW1ZCxT5yWvXRa40QPh4E2=aGxKFPJqqbQ AT mail DOT gmail DOT com> <611a8823 DOT 1c69fb81 DOT eb245 DOT e8eeSMTPIN_ADDED_BROKEN AT mx DOT google DOT com>
In-Reply-To: <611a8823.1c69fb81.eb245.e8eeSMTPIN_ADDED_BROKEN@mx.google.com>
From: "Chad Parker (parker DOT charles AT gmail DOT com) [via geda-user AT delorie DOT com]" <geda-user AT delorie DOT com>
Date: Mon, 16 Aug 2021 11:58:00 -0400
Message-ID: <CAJZxidDKEOCigYrbAdiLhwbKMyvACNrMSP=_aBqmdfY6J=J6cg@mail.gmail.com>
Subject: Re: [geda-user] pcb modules - are they actually used ?
To: geda-user AT delorie DOT com
Content-Type: multipart/alternative; boundary="000000000000a66cb805c9af4452"
Reply-To: geda-user AT delorie DOT com

--000000000000a66cb805c9af4452
Content-Type: text/plain; charset="UTF-8"

 I'm not familiar with LTO, so I don't have a ready answer for related
issues. I've never had trouble with pcb taking more than a few minutes to
compile, even on older hardware. So, something is definitely wrong. If you
provide the relevant information, I can try to reproduce the problem.

It's perfectly possible to compile outside of the source tree, or in a
dedicated build directory inside the source tree. I do this routinely. I
believe you just have to run the configure script from wherever you want
the build to be located. I have an example script that I can send when I
get home if for some reason that doesn't work. It's possible that there's a
flag or something that you have to specify. I haven't looked at it for a
while.

The build system has been around for a very long time and there are likely
a lot of opportunities for improvement there. It's one of the items on my
very long list of things to do.

It's autotools that configures the use of environment variables such as LD
when it generates the Makefiles. I don't believe that pcb makes any
modifications to such variables. So, if the system is failing to respect
their values, I would suspect that it's an issue with how something outside
of pcb is set up. I haven't tried specifying an alternate linker before.

On Mon, Aug 16, 2021 at 11:45 AM Branko Badrljica (brankob AT s5tehnika DOT net)
[via geda-user AT delorie DOT com] <geda-user AT delorie DOT com> wrote:

> On Mon, 16 Aug 2021 15:09:27 +0000
> "Branko Badrljica (brankob AT s5tehnika DOT net) [via geda-user AT delorie DOT com]"
> <geda-user AT delorie DOT com> wrote:
> >
> > For example, fist LTO attempts failed for me UNTIL the first time
> > I did full non-LTO compile and install. After that, LTO compiles and
> > installs finish fine...
> >
>
> Errrm. scratch that. I've borked CFLAGS /LDFLAG setting on later
> attempts. LTO build never finishes fine.
>
> It looks like configure re-compiles ssconfig with LTO flags and
> something gets badly broken in the process.
> It also doesn't allow me to set different ld, with for example:
> "LD=/usr/bin/ld.lld"
>
> ssconfig gets built in the same context as the application and library.
>
>

--000000000000a66cb805c9af4452
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">
<div>I&#39;m not familiar with LTO, so I don&#39;t have a ready answer for =
related issues. I&#39;ve never had trouble with pcb taking more than a few=
=20
minutes to compile, even on older hardware. So, something is definitely=20
wrong. If you provide the relevant information, I can try to reproduce the =
problem.<br></div><div><br></div><div>It&#39;s perfectly possible to compil=
e outside of the source tree, or in a dedicated build directory inside the =
source tree. I do this routinely. I believe you just have to run the config=
ure script from wherever you want the build to be located. I have an exampl=
e script that I can send when I get home if for some reason that doesn&#39;=
t work. It&#39;s possible that there&#39;s a flag or something that you hav=
e to specify. I haven&#39;t looked at it for a while.<br></div><div><br></d=
iv><div>The build system has been around for a very long time and there are=
 likely a lot of opportunities for improvement there. It&#39;s one of the i=
tems on my very long list of things to do.<br></div><div><br></div><div>It&=
#39;s autotools that configures the use of environment variables such as LD=
 when it generates the Makefiles. I don&#39;t believe that pcb makes any mo=
difications to such variables. So, if the system is failing to respect thei=
r values, I would suspect that it&#39;s an issue with how something outside=
 of pcb is set up. I haven&#39;t tried specifying an alternate linker befor=
e. <br></div>

</div><br><div class=3D"gmail_quote"><div dir=3D"ltr" class=3D"gmail_attr">=
On Mon, Aug 16, 2021 at 11:45 AM Branko Badrljica (<a href=3D"mailto:branko=
b AT s5tehnika DOT net">brankob AT s5tehnika DOT net</a>) [via <a href=3D"mailto:geda-use=
r AT delorie DOT com">geda-user AT delorie DOT com</a>] &lt;<a href=3D"mailto:geda-user AT d=
elorie.com">geda-user AT delorie DOT com</a>&gt; wrote:<br></div><blockquote class=
=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-left:1px solid rg=
b(204,204,204);padding-left:1ex">On Mon, 16 Aug 2021 15:09:27 +0000<br>
&quot;Branko Badrljica (<a href=3D"mailto:brankob AT s5tehnika DOT net" target=3D"=
_blank">brankob AT s5tehnika DOT net</a>) [via <a href=3D"mailto:geda-user AT delorie=
.com" target=3D"_blank">geda-user AT delorie DOT com</a>]&quot;<br>
&lt;<a href=3D"mailto:geda-user AT delorie DOT com" target=3D"_blank">geda-user AT de=
lorie.com</a>&gt; wrote:<br>
&gt; <br>
&gt; For example, fist LTO attempts failed for me UNTIL the first time<br>
&gt; I did full non-LTO compile and install. After that, LTO compiles and<b=
r>
&gt; installs finish fine...<br>
&gt; <br>
<br>
Errrm. scratch that. I&#39;ve borked CFLAGS /LDFLAG setting on later<br>
attempts. LTO build never finishes fine.<br>
<br>
It looks like configure re-compiles ssconfig with LTO flags and<br>
something gets badly broken in the process.<br>
It also doesn&#39;t allow me to set different ld, with for example:<br>
&quot;LD=3D/usr/bin/ld.lld&quot;<br>
<br>
ssconfig gets built in the same context as the application and library.<br>
<br>
</blockquote></div>

--000000000000a66cb805c9af4452--