I think it might help if someone could get the debugging symbols from the Pale Moon team.

http://en.zimagez.com/zimage/screens...2022-22-11.php
Thats a Screen shot of me running palemoon which I compiled with the build script I posted in the post above.

I'll try them. But are you doing it in a clean environment?

I have had it, like I said, where it will all of a sudden pop and begin working. But I don't think that that is the quality of package/scripts that should be submitted to SBo.

I compiled it on a clean fresh install of slackware and I had no issues. I'm not getting any errors, its not complaining at all about anything whatsoever.

Did you try the already built package I uploaded?

https://docs.google.com/uc?id=0B__WG...xport=download

Yep, I tried that package. I am compiling from that SlackBuild that you posted next to it now.

*No dice on the latest compile.
I might just drop back to 14.1. Then I wont have to worry about it. I've got other things going on that I need to worry about.

The sad thing is that the Pale Moon "devs" are going to run into issues sooner or later. One would think that with all the Slackers that I just seen with threads/posts on their forums, they would think that maybe they need to look at what Slackware is doing because we are obviously using newer packages than what they are building with. Apparently they build on CentOS 7, Which is using a patched glibc-2.17 (even older than what I previously thought they were using, Debian/patched glibc-2.19).

I hate to switch browsers, I don't really like Chrome/Chrome-based browsers, but it might be inevitable. Firefox just drags a$$, comes with completely useless features built in and has to be customized for an hour before it's usable. Firefox has outgrown itself IMO.

wild I wonder why it works for me tho?

Did it compile but just segfault? I mean this latest attempt from you? What kind of cpu do you have? I'm trying to imagine why it would work for me, but not you?

It compiled, but segfaults on launching.

I've tried packages/compiling on both:

HP Pavilion p6130y, AMD Phenom X4 9750 @ 2.4 GHz, 8 GB RAM, nVidia, 750 GB HDD -my HTPC
and
Dell Inspiron 1520, Intel Core 2 Duo T7250 @ 2 GHz, 4 GB RAM, Intel, 320 GB HDD -my test/build rig

I have a ASUS M2NPV-VM rig (AMD Athlon X2 6000+ @ 3.0 GHz, 4 GB RAM, nVidia, 320 GB HDD), but it is currently being used as my server and cannot be disturbed right now. Plus about 4 other PC's that are in pieces/missing pieces.

I've rebuilt the x86_64 package there with debug symbols (I added a build flag ENABLE_DEBUG), try that: just be sure that you don't have any previous versions of palemoon lying in the filesystem, install that package, launch it from a terminal with the command "palemoon-26.4.1" and watch the output, or go into /usr/lib$LIBDIRSUFFIX/palemoon-26.4.1 and launch "gdb palemoon".
for those of you that are on 32bit I'm rebuilding the i586 one too and when the build will be complete I'll copy it in the same spot, you will recognize the versions with debug info because they will be tagged "debug" and because they're huge (nearly ten times bigger than the stripped ones)!

the sources you can find there are the ones that khronosschoty submitted to SBo, that I just adapted for committing in the repository.

I just tried khronosschoty's latest SlackBuild on a clean Slackware64-14.2 VM, and it works fine. I see no warnings at all when running from a terminal. So it's very strange that some people are getting the segfault. Maybe those people could try in a VM, in case it's related to a hardware issue?

khronosschoty, FYI the latest palemoon-26.4.1.desktop file won't work, because the executable is now named palemoon-26.4.1, not palemoon. I'm also uneasy about some of the things in the script, for example in these lines:

Why are these files already installed in the root /usr directory before the package is even created? Anyway, if you submitted it to SBo, I guess the admins will take care of any issues like this.

I'll do a clean install and try khronosschoty's SlackBuild again before I try ponce's -symbols build, which is appreciated still.

The last few times I didn't have a clean environment so there could be something hanging around.

*Negative on 2 different machines from terminal with the binary package from here: https://docs.google.com/uc?id=0B__WG...xport=download. segfaulting on both.
1 HP/AMD/nVidia(nouveau) and 1 Dell/Intel/Intel

Full clean installs.

Attached Thumbnails

 

Just tried it

bash-4.3# palemoon-26.4.1 %u
Segmentation fault
bash-4.3#

Oh thanks for catching that montagude. I had to change the naming because it was in conflict with anther package on SBo. As far as the symlinks go, nothing is being installed to the root directory; that wont happen until you run installpkg. The symlinks get removed when makepkg is run and then automatically added to a doinst.sh which then makes them. If you were for example to look inside the actual package it creates you'd notice there is nothing in /usr/bin.

Also fyi the git does not reflect what I submitted to SBo. I will likely update it soon to what I have out to SBo.

Ok, I see. I forgot that ln -s will let you make a symbolic link to a non-existent file, hence I thought that /usr/lib${LIBDIRSUFFIX}/palemoon-$VERSION/palemoon (for example) must have already been there. Thanks for clearing that up.

https://notabug.org/khronosschoty/Sl...aster/PaleMoon

Here are the SlackBuilds as they will be on SBo. Ponce helped by making the script conform more closely to other SlackBuild scripts like icecat, firefox etc and by cleaning up the .desktop file and adding a debug option.

I would be interested in seeing the results that come from those having issues (Gordie, Skaendo), compiling this build with the ENABLE_DEBUG=yes option.

I downloaded and compiled your latest (PaleMoon not palemoon) and it is still segfaulting here. I didn't build with debug, my fast(er) PC is busy right now. I will try it tom.

As usual, full clean install (minus kdei) Slackware64-14.2 then compiled.
Nothing added, no settings changed, etc. Just installed and compiled.