[Slackbuilds-users] webkit2gtk build problems

[Habs]

On Mon, 18 Sep 2017 17:20:21 +0100
Habs <gen-bch at useyouresp.org.uk> wrote:

> On Mon, 18 Sep 2017 11:02:48 -0400
> Jason Graham <jgraha8 at gmail.com> wrote:
> 
> > On Mon, Sep 18, 2017 at 10:56 AM, Ythogtha <yth at ythogtha.org> wrote:
> >   
> > > > > I got the same error as Habs on a stable install.    
> > > >
> > > > I was convinced by Matteo and David that webkit2gtk build fine
> > > > on stable and current for both arch, BUT it requires a LOT of
> > > > resources, so i cancelled my revert and update README instead to
> > > > make additional SWAP space if you have small amount of memory
> > > > installed.    
> > >
> > >         After the package is created, the temporary build
> > > directory takes 1.1Gb.
> > > It's not that big, but it is also true that the build takes up a
> > > lot of resources during compilation.
> > >         My computer only has 8Gb of RAM. Maybe using multiple jobs
> > > may be a problem, I use -j2, and my computer is on its knees (or
> > > chips, or circuits, don't know... ^^) when building, but it builds
> > > fine.
> > >
> > >
> > >    
> > 
> > I had a similar experience. I first tried with -j8 on a VM and the
> > compilation finished (or at least the cmake progress showed 100%)
> > but it didn't proceed from there. There were 4Gb of ram allocated
> > to this VM so I suppose it was resource constrained based on other
> > folks' findings. It stalled for quite some time (~30min) before I
> > killed the job.
> > 
> > So I tried again with -j1 and it failed in the same location has
> > Habs.
> > 
> > Best,
> > Jason  
> 
> Ive started the build off again after clearing it out.
> As I said, I'm simple minded ;-)  so please forgive me, but it puzzles
> me that the error message is related to space - is that disk resource
> and,or ram resources  ...and more over, why would the previous version
> build fine on the exact same machine _without issue_ but not this
> updated version...what is it doing different I wonder (all said
> rhetorically speaking).
> 
> I will see what happens in the early hours  :-)
> 
> Thanks again everyone
> Habs

Well, it failed again - both machines (one Intel one AMD tower) - as
detailed previously.  They were doing nothing else other than building
this.

The machines are by no means gutsy, but still...4Gb RAM, 500Gb disk
space etc. Swap wasnt overly affected from what I could see, which
just leaves memory etc ...why would that cause it to conk I dunno.

If it is  a resources issue for building, producing the error at the
same place each time, then it seems like its getting to the point of
needing a Cray computer to do this. Ridiculous.  :-)

From reading other posts, I will have a go with the -j switch and see
what gives, but it seems remote.  Not sure what else to do, or I can
do.  The previous version builds fine, so I can use that and not
upgrade any more, until I understand what's going on (hmmm).

Thank you, good night (as it is for me) to all.
Habs

---
[ 76%] Building CXX object
Source/WebCore/CMakeFiles/WebCore.dir/platform/graphics/cairo/BackingStoreBackendCairoImpl.cpp.o
In file included
from /tmp/SBo/webkitgtk-2.18.0/Source/WebCore/platform/graphics/cairo/BackingStoreBackendCairoImpl.cpp:21:0:
/tmp/SBo/webkitgtk-2.18.0/Source/WebCore/platform/graphics/cairo/BackingStoreBackendCairoImpl.h:23:40:
fatal error: WebCore/HysteresisActivity.h: No such file or directory
compilation terminated.
Source/WebCore/CMakeFiles/WebCore.dir/build.make:51125: recipe for
target
'Source/WebCore/CMakeFiles/WebCore.dir/platform/graphics/cairo/BackingStoreBackendCairoImpl.cpp.o' failed
make[2]: ***
[Source/WebCore/CMakeFiles/WebCore.dir/platform/graphics/cairo/BackingStoreBackendCairoImpl.cpp.o]
Error 1
CMakeFiles/Makefile2:1065: recipe for target
'Source/WebCore/CMakeFiles/WebCore.dir/all' failed
make[1]: *** [Source/WebCore/CMakeFiles/WebCore.dir/all] Error 2
Makefile:149: recipe for target 'all' failed
make: *** [all] Error 2

---