[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAFPCfQw+3M60kH7W821dUV5FcjW=edTEkuZH=RmbJ9hq9uCurg@mail.gmail.com>
Date: Sun, 16 Nov 2014 15:50:23 -0600
From: will cunningham <willpanther@...il.com>
To: sabotage@...ts.openwall.com
Subject: Re: changing/fixing the tarballs in the butch
How ironic it's these two I'm having a go at on the one yard line. Big
picture I'd wait for the changeover to qt5 everywhere being done, and just
skip qt4, and the Lord hath spoken that it's time for me to move on from
firefox. However, for tangible reasons I think you'd spend your time
better working on other things for a while before delving into those black
holes:
QT is weird even for C++ and QT5 is a pretty significant shift, so by the
time the base of Sabotage is cleaned up the move to it'll be
done(technically I understand you could say that's already happened, but I
mean practically QT5 worked through the entire app community including
smoothing out problems.)
As for firefox, I can do raw photo editing with less of a RAM overhead;
well almost. They also aren't the easiest app to go into the boiler room
of if there are problems compiling, so the value for musl testing would be
firefoxish specific methinks.
I'm in a real world, long term project that has its focus down to a small
set of apps, and one of my main reasons for using Sabotage is to clean some
of those apps up while taking control of my OS.
After I get this base system up, I'm writing butch files for the rest of my
core and then hacking on a few of them, so I'll be enough of a pain in the
ass to y'all until the entire set is up that I'd say focus on the libs/apps
that are tighter to begin with(qt is wonderful, but I mean natural to work
with lean optimized code) until all the quirks are worked out before
messing with the idiosyncratic apps/libs.
There are some really nice lean apps out there the main Linux distros are
just falling behind for versioning, and my focus is on getting them working
as the support set for apps with your VERY cool set of libs without all the
bull***** to use the CPU/RAM for things that need them like dsp audio and
raw photo editing.
I think that journey will clear up actual quirks/bugs/missing code for
Sabotage and musl with busybox etc a great deal as I've already narrowed
our world here to really good apps that likely won't be the problem, and my
pdf tracing the journey will be a cherry on the sundae.
This was a long way of saying bye bye Firefox, hello Webkit2! This is an
exciting moment for me.
Peace out,
Will
On Sun, Nov 16, 2014 at 2:48 PM, John Spencer <maillist-musl@...fooze.de>
wrote:
> will cunningham wrote:
>
>>
>> I have two things not compiling ever, qt4 and firefox. There are no
>> errors
>> to the console(except of course 'failed to build!'} Here are the last few
>> lines from their respective /src/logs/build_*** files:
>>
>> qt4-
>> In file included from ../../include/QtGui/private/qcups_p.h:1:0,
>> from painting/qpdf.cpp:47:
>> ../../include/QtGui/private/../../../src/gui/painting/qcups_p.h:78:11:
>> error: 'ppd_file_t' does not name a type
>>
>
> did you have cups installed before building this ?
> maybe qt4 needs something along the lines of --disable-cups
> (often configure scripts detect and activate optional features if some
> library was detected).
> also note that qt4 wasnt tested/used for a long time (there's nothing
> depending on it in sabotage)...
>
> firefox-
>>
>> `/src/build/firefox/mozilla-release/firefox-build-dir/ipc/ipdl'
>> make[5]: *** .deps/UnifiedProtocols9.o.pp: I/O error. Stop.
>>
>
> I/O error ? maybe your harddrive is full ?
>
>
>> What should I try next?
>>
>
> are you trying to build those packages because you actually need them or
> just out of curiosity ?
>
> --JS
>
>
Content of type "text/html" skipped
Powered by blists - more mailing lists
Confused about mailing lists and their use?
Read about mailing lists on Wikipedia
and check out these
guidelines on proper formatting of your messages.
