Commit Graph

69 Commits

Author SHA1 Message Date
Kevin Kofler
ed26f4f5e2 Rebase the no-sse2 patch
Now all patches should be rebased.
2016-07-17 21:06:36 +02:00
Kevin Kofler
c44c0d51a8 Rebase patches (TODO: no-sse2 patch)
* Sun Jul 17 2016 Kevin Kofler <Kevin@tigcc.ticalc.org> - 5.7.0-1
- Drop no-format patch, fixed upstream (they stopped passing -Wno-format)
- Rebase linux-pri patch (use_system_protobuf is now a qmake flag)
- Rebase system-nspr-prtime and system-icu-utf patches
2016-07-17 17:54:52 +02:00
Kevin Kofler
778be18d72 Update to 5.7.0
* Sat Jul 16 2016 Kevin Kofler <Kevin@tigcc.ticalc.org> - 5.7.0-1
- Update to 5.7.0
- Update version numbers of bundled stuff
- Update system libvpx/libwebp version requirements (now F24+ only)

TODO: rebase patches
2016-07-17 01:38:49 +02:00
Rex Dieter
3eab3a62ce rebuild (glibc) 2016-06-14 08:44:28 -05:00
Rex Dieter
87609ad33f add .spec comment to qtbase versioned dep 2016-06-12 08:25:23 -05:00
Rex Dieter
3ae0724acb add versioned qt5-qtbase runtime dep
User reported after recent rawhide 5.6.1 update:
Cannot mix incompatible Qt library (version 0x50600) with this library (version 0x50601)
2016-06-12 08:24:03 -05:00
Kevin Kofler
6d2db9a7a4 Add the new designer/libqwebengineview.so plugin to the file list 2016-06-11 03:40:44 +02:00
Kevin Kofler
d8f3edfd1d Update to 5.6.1
* Fri Jun 10 2016 Kevin Kofler <Kevin@tigcc.ticalc.org> - 5.6.1-1
- Update to 5.6.1
- Rebase linux-pri patch (drop the parts already fixed upstream)
- Drop backported chimera-nss-init patch, already applied upstream
- Rebase no-sse2 patch (the core_module.pro change)
2016-06-11 01:32:18 +02:00
Rex Dieter
c95fe8489c Release++ 2016-06-06 13:03:15 -05:00
Rex Dieter
efad6cc70a workaround stackmashing runtime errors in re2-related bundled headers (#1337585) 2016-06-06 13:02:56 -05:00
Rex Dieter
463136e0f9 rebuild (pciutuils)
See also:
https://bugzilla.redhat.com/show_bug.cgi?id=1337735#c5
2016-05-21 10:43:20 -05:00
Rex Dieter
1c39d15ba2 BR: git-core
several places in build.log contain:
sh: git: command not found
so include it to be on the safe side.
2016-05-18 08:50:48 -05:00
David Tardon
6a3240cd13 rebuild for ICU 57.1 2016-04-15 20:15:22 +02:00
Than Ngo
66e732407f drop ppc ppc64 ppc64le from ExclusiveArch, it's not supported yet 2016-04-08 18:52:37 +02:00
Kevin Kofler
9659c061c0 Fix // in #include in content/renderer/gpu to avoid debugedit failure
This actually shows up as a trailing slash in the directory table.
I hope this is the last offender.
2016-03-24 03:39:15 +01:00
Kevin Kofler
855dc1312c Fix -g1 workaround
I was missing the -i flag to sed.
2016-03-22 19:34:52 +01:00
Kevin Kofler
94d8130832 Work around debugedit failure by removing "./" from #line commands
At least I hope that that's the problem. I don't know what other path
would be the offender. There are no double slashes "//".
2016-03-22 19:19:24 +01:00
Kevin Kofler
29f484a302 Revert "Add debugging for debugedit failure"
This reverts commit a90ae4b2bb.
2016-03-22 19:12:41 +01:00
Kevin Kofler
cea79e8095 Revert "Fix readelf command for debugedit debugging"
This reverts commit 9a8e5f9ba2.
2016-03-22 19:12:29 +01:00
Kevin Kofler
4fe7f8982c Use -g1 instead of -g on non-x86_64 to avoid memory exhaustion
At least i686 runs out of memory otherwise, but ARM would probably crash
too if it got that far, so be conservative.
2016-03-21 18:36:00 +01:00
Kevin Kofler
070252022a Revert "Use CONFIG-=use_gold_linker because ld.gold keeps running out of memory"
This reverts commit 6fbf35079b.
2016-03-21 18:27:03 +01:00
Kevin Kofler
6fbf35079b Use CONFIG-=use_gold_linker because ld.gold keeps running out of memory
Let's see whether the good old ld.bfd works any better.
2016-03-21 06:19:01 +01:00
Kevin Kofler
9a8e5f9ba2 Fix readelf command for debugedit debugging 2016-03-21 02:37:39 +01:00
Kevin Kofler
a90ae4b2bb Add debugging for debugedit failure
(Yet another instance of the venerable old bug #304121.)
2016-03-21 02:16:42 +01:00
Kevin Kofler
1b4283ef9d Build with CONFIG+=force_debug_info so that -g gets added for C files too. 2016-03-21 02:13:41 +01:00
Kevin Kofler
2b1c85b7d4 Revert "Also set CFLAGS so that -g gets picked up for C files too."
This reverts commit bd99caed44.
2016-03-21 02:11:45 +01:00
Kevin Kofler
9673c10a27 Revert "skip added debuginfo for now, hits rpm bugs"
This reverts commit ee59b253e9.
2016-03-21 02:10:41 +01:00
Rex Dieter
ee59b253e9 skip added debuginfo for now, hits rpm bugs 2016-03-20 19:10:06 -05:00
Kevin Kofler
bd99caed44 Also set CFLAGS so that -g gets picked up for C files too. 2016-03-19 01:44:43 +01:00
Kevin Kofler
c13613a7be CONFIG+="webcore_debug v8base_debug", -fno-delete-null-pointer-checks
* Fri Mar 18 2016 Kevin Kofler <Kevin@tigcc.ticalc.org> - 5.6.0-3
- Build with CONFIG+="webcore_debug v8base_debug"
- Force -fno-delete-null-pointer-checks through CXXFLAGS, Qt flags not used here
2016-03-19 00:34:32 +01:00
Kevin Kofler
4f844a7b2b Avoid checking for the nonexistent icudtl.dat and silence the warnings 2016-03-18 04:59:54 +01:00
Kevin Kofler
4596e73fc6 Update to 5.6.0 (final)
* Thu Mar 17 2016 Kevin Kofler <Kevin@tigcc.ticalc.org> - 5.6.0-1
- Update to 5.6.0 (final)
- Drop system-icu54 patch, fixed upstream
2016-03-18 00:41:34 +01:00
Kevin Kofler
b6bc15418f Rebase no-sse2 patch, fix file list
- Rebase no-sse2 patch
- Update file list for 5.6.0 RC (resources now in resources/ subdirectory)
- Tag translations with correct %lang tags
2016-02-25 01:24:47 +01:00
Kevin Kofler
7f40671ad8 Rebase linux-pri patch, rm BR pkgconfig(flac), pkgconfig(speex)
- Rebase linux-pri patch
- Remove BuildRequires pkgconfig(flac), pkgconfig(speex), no longer needed
2016-02-24 23:28:14 +01:00
Kevin Kofler
26c2ef598c Update to 5.6.0 RC
* Wed Feb 24 2016 Kevin Kofler <Kevin@tigcc.ticalc.org> - 5.6.0-0.19.rc
- Update to 5.6.0 RC
2016-02-24 16:15:14 +01:00
Kevin Kofler
58b406a343 no-sse2 patch: Fix FFT (RealFourier) in webrtc on non-SSE2 x86
* Tue Feb 23 2016 Kevin Kofler <Kevin@tigcc.ticalc.org> - 5.6.0-0.18.beta
- no-sse2 patch: Fix FFT (RealFourier) in webrtc on non-SSE2 x86
2016-02-24 04:50:18 +01:00
Kevin Kofler
57c1fcecf2 Revert "- Uploaded rc"
This reverts commit 65fd4d09a1.

The tarball is not redistributable, the EVR bump was done wrong (goes
backwards), and the package does not even build (because the uploaded
tarball does not have "-clean" in the file name), so there is absolutely
nothing worth keeping in that commit.
2016-02-24 03:08:04 +01:00
Helio Chissini de Castro
65fd4d09a1 - Uploaded rc 2016-02-23 17:07:31 +01:00
Kevin Kofler
9ac4c0c992 clean_ffmpeg.sh: Add missing aarch64 files. (Fixes #1310753.)
* Tue Feb 23 2016 Kevin Kofler <Kevin@tigcc.ticalc.org> - 5.6.0-0.17.beta
- Fix FTBFS on aarch64: Respin tarball with fixed clean_ffmpeg.sh (#1310753).
2016-02-23 01:47:14 +01:00
Fedora Release Engineering
1968dfff52 - Rebuilt for https://fedoraproject.org/wiki/Fedora_24_Mass_Rebuild 2016-02-04 21:48:01 +00:00
Kevin Kofler
a41ab0a661 Update comment on no-sse2.patch
V8 is already built shared and twice on i686 now (no longer TODO).
2016-01-22 02:14:50 +01:00
Kevin Kofler
7c8a25cacb Bump Release for official Rawhide build 2016-01-19 20:05:23 +01:00
Kevin Kofler
fa42c47889 Add Provides/Requires filtering for libv8.so (i686) and for plugins 2016-01-19 17:37:35 +01:00
Kevin Kofler
24bef19c77 Add the private library directory to the file list on i686 2016-01-19 15:03:04 +01:00
Kevin Kofler
306e720cfb no-sse2.patch: Install the shared libraries, don't link the sse2 one 2016-01-19 14:55:13 +01:00
Kevin Kofler
e670d513af no-sse2.patch: Build both the x87 version and the SSE2 version of V8 on i686
Note that almost all the patch size increase is from the v8_sse2.gyp
file that is automatically generated from v8.gyp using a small script I
wrote, which is also included in the patch.
2016-01-19 03:32:24 +01:00
Kevin Kofler
90225e7e77 no-sse2.patch: Build V8 as a shared library on i686
* Mon Jan 18 2016 Kevin Kofler <Kevin@tigcc.ticalc.org> - 5.6.0-0.14.beta.nosse2.1
- Build V8 as a shared library on i686 to allow for swappable backends

This is the first prerequisite step towards building both the x87 and
SSE2 backends.
2016-01-18 01:10:42 +01:00
Kevin Kofler
2684898152 Bump Release now that the no-sse2 patch builds and looks good
Swappable V8 is still TODO though.
2016-01-18 00:36:47 +01:00
Kevin Kofler
42f26ca168 no-sse2.patch: remove -ffloat-store, fix DirectConvolver FTBFS
* Remove the -ffloat-store workaround: The media player issue that
  caused problems with x87 floating point (and ultimately made upstream
  require SSE2) should already be fixed / worked around by:
  https://crrev.com/d2c745b13c93ecff5108bed57d8e052126715492
  So I will only readd -ffloat-store if people request it, and even
  then, only to specific parts of code and not everything.
* Fix src/3rdparty/chromium/third_party/WebKit/Source/platform/audio/
  DirectConvolver.cpp to not require SSE2 on x86. (No runtime detection,
  sorry.)
* Update the changelog date in the specfile.
2016-01-17 02:34:08 +01:00
Kevin Kofler
6d0400baa5 Do not require SSE2 on i686
See https://bugzilla.redhat.com/show_bug.cgi?id=1244196#c22 for the
history of the patch. Compared to the last revision there (v5):
* I fixed WebGLImageConversion.cpp to not assume SSE2 on x86.
  Unfortunately, it would be really hard to make runtime detection work
  there, because it uses intrinsics in inline templates, and GCC only
  supports intrinsics when building the whole code for the vector
  instruction set. So I had to turn off that SSE2 code entirely when
  __SSE2__ is not defined at compile time.
* I fixed webrtc to check for SSE2 at runtime (using the runtime
  checking code that webrtc already has and uses in several places)
  before attempting to use OpenMAX on x86. OpenMAX requires SSE2.
  So we just fall back to the Ooura implementation of FFTs when needed.
TODO: build V8 shared and twice (once for x87, once for SSE2)
2016-01-17 00:17:04 +01:00