On 2018-10-24 19:09, Bruce Dubbs via blfs-dev wrote:
On 10/24/2018 05:44 PM, Ken Moffat via blfs-dev wrote:
On Thu, Oct 25, 2018 at 12:21:37AM +0200, Tim Tassonis via blfs-dev
wrote:
Hi all
Just tried to build firefox 63.0 and got the folllowing error:
-12:10.16 /lgl-bld/firefox-63.0/js/src/vm/JSContext-inl.h:180:73:
error: no
‘void JSContext::checkImpl_63(int, const Head&, const Tail& ...)’
member
function declared in class ‘JSContext’
12:10.16 JSContext::checkImpl(int argIndex, const Head& head, const
Tail&... tail)
12:10.16
I got a similar error in the same place:
/mnt/tmp/firefox/firefox-63.0/js/src/vm/JSContext-inl.h:180:1: error:
no declaration matches ‘void JSContext::checkImpl_63(int, const Head&,
const Tail& ...)’
0:06.42 JSContext::checkImpl(int argIndex, const Head& head, const
Tail&... tail)
I do have node-v9.11.2 installed.
Using gcc 8.2.
grepping for JSContext::checkImpl only finds a match in
js/src/vm/JSContext-inl.h
I could find nothing similar to JSContext::checkImpl in /usr/include
-- Bruce
In my case, I can build it fine if I run at -j1 (96 SBU!), but I cannot
install it. When I attempt to install it, it bombs out about "elfhack".
Something throws an "iostream" error:
BLFS Start INSTALL
0:00.72 /usr/bin/make -C . -j1 -s -w install
0:00.73 make: Entering directory
'/sources/firefox-63.0/firefox-63.0/firefox-build-dir'
0:00.79 make[1]: Entering directory
'/sources/firefox-63.0/firefox-63.0/firefox-build-dir/browser/installer'
0:45.74 terminate called after throwing an instance of
'std::__ios_failure'
0:45.76 what(): basic_ios::clear: iostream error
0:57.35 Traceback (most recent call last):
0:57.35 File
"/sources/firefox-63.0/firefox-63.0/toolkit/mozapps/installer/packager.py",
line 339, in <module>
0:57.35 main()
0:57.35 File
"/sources/firefox-63.0/firefox-63.0/toolkit/mozapps/installer/packager.py",
line 333, in main
0:57.35 copier.copy(args.destination)
0:57.35 File
"/sources/firefox-63.0/firefox-63.0/python/mozbuild/mozpack/copier.py",
line 431, in copy
0:58.01 copy_results.append((destfile, f.copy(destfile,
skip_if_older)))
0:58.01 File
"/sources/firefox-63.0/firefox-63.0/python/mozbuild/mozpack/files.py",
line 296, in copy
0:58.01 elfhack(dest)
0:58.01 File
"/sources/firefox-63.0/firefox-63.0/python/mozbuild/mozpack/executables.py",
line 124, in elfhack
0:58.29 errors.fatal('Error executing ' + ' '.join(cmd))
0:58.29 File
"/sources/firefox-63.0/firefox-63.0/python/mozbuild/mozpack/errors.py",
line 103, in fatal
0:58.33 self._handle(self.FATAL, msg)
0:58.33 File
"/sources/firefox-63.0/firefox-63.0/python/mozbuild/mozpack/errors.py",
line 98, in _handle
0:58.33 raise ErrorMessage(msg)
0:58.33 mozpack.errors.ErrorMessage: Error: Error executing
/sources/firefox-63.0/firefox-63.0/firefox-build-dir/build/unix/elfhack/elfhack
../../dist/firefox/libxul.so
0:58.35 make[1]: ***
[/sources/firefox-63.0/firefox-63.0/toolkit/mozapps/installer/packager.mk:22:
stage-package] Error 1
0:58.35 make[1]: Leaving directory
'/sources/firefox-63.0/firefox-63.0/firefox-build-dir/browser/installer'
0:58.35 make: ***
[/sources/firefox-63.0/firefox-63.0/browser/build.mk:15: install] Error
2
0:58.35 make: Leaving directory
'/sources/firefox-63.0/firefox-63.0/firefox-build-dir'
The only reason why I used -j1 (and shut off all of my other cores) is
because I found a resource leak in rust that leads to an OOM when
compiling Firefox (I have 64GB of RAM in this system):
[608008.254835] Out of memory: Kill process 28009 (rustc) score 346 or
sacrifice child
[608008.254843] Killed process 28009 (rustc) total-vm:3734704kB,
anon-rss:2777716kB, file-rss:0kB, shmem-rss:0kB
[608008.383204] oom_reaper: reaped process 28009 (rustc), now
anon-rss:0kB, file-rss:0kB, shmem-rss:0kB
It repeated that (with a call trace) for all 12 "CPUs" in the system
(6c/12t CPU). Even at -j4, I'm leaking and OOMing. If I feel adventurous
(or if anyone suggests it), I can try a newer version of rustc to see if
the leaking problem is gone (the new librsvg that JUST came out needs
1.27 anyway
[https://mail.gnome.org/archives/ftp-release-list/2018-October/msg00087.html]).
I have 504 coredumps on this system (and it's not more than a couple
weeks old). A majority of them were due to ICE's (CPU was too new for
it's quirks to be covered in the kernel, as well as some Illegal
Operation errors), but I have one from elfhack here:
$ coredumpctl info
/sources/firefox-63.0/firefox-63.0/firefox-build-dir/build/unix/elfhack/elfhack
PID: 7047 (elfhack)
UID: 0 (root)
GID: 0 (root)
Signal: 6 (ABRT)
Timestamp: Wed 2018-10-24 17:52:09 CDT (2h 47min ago)
Command Line:
/sources/firefox-63.0/firefox-63.0/firefox-build-dir/build/unix/elfhack/elfhack
../../dist/firefox/libxul.so
Executable:
/sources/firefox-63.0/firefox-63.0/firefox-build-dir/build/unix/elfhack/elfhack
Control Group: /user.slice/user-1000.slice/session-c1.scope
Unit: session-c1.scope
Slice: user-1000.slice
Session: c1
Owner UID: 1000 (renodr)
Boot ID: 3ec1bf24fdbd4caf8556e29527e93458
Machine ID: 43c995037fc0492a843f1f7d6c9d14ed
Hostname: RENODR-LFSWKST
Storage:
/var/lib/systemd/coredump/core.elfhack.0.3ec1bf24fdbd4caf8556e29527e93458.7047.1540421529000000.xz
(inaccessible)
Message: Process 7047 (elfhack) of user 0 dumped core.
Off-topic:
As for the rest of my crashes, I've had three in lxpanel, several in
COGL/Clutter (I'm aware of those, have to update Mesa), and several in
wayland, GTK2, LLVM, ld, and ruby. This system hasn't been kind to me
for it's initial software load. I suppose I could do another build if
absolutely necessary, but I'd rather not, and they aren't related to
Firefox-63.
--
http://lists.linuxfromscratch.org/listinfo/blfs-dev
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page
