Package Details: chromium-vaapi 75.0.3770.142-1

Git Clone URL: (read-only)
Package Base: chromium-vaapi
Description: Chromium with VA-API support to enable hardware acceleration
Upstream URL:
Keywords: browser web
Licenses: BSD
Conflicts: chromium
Provides: chromium
Submitter: samcv
Maintainer: OneObsession (maximbaz)
Last Packager: maximbaz
Votes: 76
Popularity: 8.502114
First Submitted: 2016-07-09 09:44
Last Updated: 2019-07-16 16:02

Dependencies (48)

Required by (47)

Sources (11)

Latest Comments

« First ‹ Previous ... 37 38 39 40 41 42 43 44 45 46 47 ... Next › Last »

OneObsession commented on 2017-06-26 12:12

@maximbaz How many cores are you using for the build? Maybe you are using only
one core. You can specify the number of jobs for compilation in the makepkg.conf file, e.g MAKEFLAGS="-j$(nproc)" which uses all the available processors. I believe such remote server should compile much faster than my old laptop.

For more information have a look at

maximbaz commented on 2017-06-26 10:51

Codeship managed to build approximately half of the project before the build was terminated after 3 hours. In total I checked around 7 different CI projects like Travis, there's nothing else that is free and allows build times longer than 3 hours.

I guess that's the end of idea of having automated builds for now, if someone finds a host to setup this - let me know! It doesn't have to be an Arch machine, any 64bit Linux where we can run a docker container is fine.

OneObsession commented on 2017-06-26 05:47

@maximbaz about 2h 30min with 4 cores with hyperthreading on a notebook.

maximbaz commented on 2017-06-25 20:18

Attempted, there is a 1 hour limit too :(

@OneObsession how long does the build usually take for you? I found, they have 3 hour limit, I'll try it next.

At least I have a Docker image now, which is reliably building this package now :)

maximbaz commented on 2017-06-25 16:45

Sync is indeed what was missing, however now it seems I hit the real blocker :(

Travis is terminating the build after 50 minutes, that's the limit according to their documentation [1]. Chromium clearly takes more than an hour to build...

Will try to see if I can find some other approach to run a build, I don't think we can decrease the build time.


OneObsession commented on 2017-06-25 12:10

@maximbaz the build definitely compiles on my side.
It seems that your packages for Arch are out of date. You are running an old
kernel 4.10.xx and your gcc version is 6.3.1. Can you sync your packages with
the current Arch repo? Gcc 7 is a must otherwise you have to remove the patch
specifically for gcc v7.

maximbaz commented on 2017-06-25 10:51

Thank you, that was the trick - next question :)
The build was running for 45 minutes when it failed with the following error. Could you please confirm that the latest version of this package is successfully building on your side?

[5915/27653] CXX obj/third_party/libxml/libxml/libxml_utils.o
FAILED: obj/third_party/libxml/libxml/libxml_utils.o
g++ -MMD -MF obj/third_party/libxml/libxml/libxml_utils.o.d -DV8_DEPRECATION_WARNINGS -DUSE_UDEV -DUSE_AURA=1 -DUSE_PANGO=1 -DUSE_CAIRO=1 -DUSE_GLIB=1 -DUSE_NSS_CERTS=1 -DUSE_X11=1 -DDISABLE_NACL -DFULL_SAFE_BROWSING -DSAFE_BROWSING_CSD -DSAFE_BROWSING_DB_LOCAL -DCHROMIUM_BUILD -DENABLE_MEDIA_ROUTER=1 -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -D_FORTIFY_SOURCE=2 -DNDEBUG -DNVALGRIND -DDYNAMIC_ANNOTATIONS_ENABLED=0 -I../.. -Igen -I/usr/include/libxml2 -fno-strict-aliasing --param=ssp-buffer-size=4 -fstack-protector -Wno-builtin-macro-redefined -D__DATE__= -D__TIME__= -D__TIMESTAMP__= -funwind-tables -fPIC -pipe -m64 -march=x86-64 -pthread -Wall -Wno-unused-local-typedefs -Wno-maybe-uninitialized -Wno-missing-field-initializers -Wno-unused-parameter -O2 -fno-ident -fdata-sections -ffunction-sections -fomit-frame-pointer -g0 -fvisibility=hidden -fvisibility-inlines-hidden -std=gnu++11 -Wno-narrowing -fno-rtti -fno-exceptions -c ../../third_party/libxml/chromium/ -o obj/third_party/libxml/libxml/libxml_utils.o
../../third_party/libxml/chromium/ In member function 'bool XmlReader::Load(const string&)':
../../third_party/libxml/chromium/ error: 'XML_PARSE_NOXXE' was not declared in this scope
XML_PARSE_NOXXE; // no external entities
../../third_party/libxml/chromium/ In member function 'bool XmlReader::LoadFile(const string&)':
../../third_party/libxml/chromium/ error: 'XML_PARSE_NOXXE' was not declared in this scope
XML_PARSE_NOXXE; // no external entities
[5916/27653] CXX obj/third_party/libwebm/libwebm/mkvmuxer.o
[5917/27653] ACTION //third_party/libvpx:libvpx_yasm_action(//build/toolchain/linux:x64)
ninja: build stopped: subcommand failed.
==> ERROR: A failure occurred in build().

Full log file:

OneObsession commented on 2017-06-25 06:18

@maximbaz Something is odd about your make. It should patch several files before compiling the source. The entire prepare() statement is missing.
Can you try the command "makepkg -s" in the directory of the PKGBUILD?

maximbaz commented on 2017-06-24 22:45

@OneObsession I'm trying to get an automated build working, but I'm stuck on this compilation failure - maybe you could give me a hint?

I'm building this package on a clean Archlinux docker image, tried different images and this same error happens constantly on the same build stage:

[502/502] LINK ./gn
Building gn manually in a temporary directory for bootstrapping...
Building gn using itself to out/Release...
Done. Made 5568 targets from 1254 files in 3918ms
ninja: Entering directory `out/Release'
ninja: error: '../../third_party/yasm/source/patched-yasm/modules/arch/x86/x86cpu.gperf', needed by 'gen/third_party/yasm/x86cpu.c', missing and no known rule to make it
==> ERROR: A failure occurred in build().

Could it be that some dependency is missing? The Archlinux image is clean, I install base and base-devel before starting the chromium-vaapi build.

Full logs are available here (different images):

agm28011997 commented on 2017-06-22 09:40

i am warning about use the last patch for chromium vaapi other, the patch fix the problem of chromium disabling the hardware acceleration but not the vaapi use problem, in chrome : media-internals not show gpuvidedecoder like in my intel graphics card.. for that reason, the patch is not fixing the problem and build a normal chromium without accel(tested by me in a few emails with samcv and confirmated with the link of chrome)
Sorry for this, but i have to tell that the users of amd( and maybe nvidia too) like me in one of my computers, have to wait for someone to apply a good patch or it never have hardware accel.. a shame for chrome and firefox to not to have but it is the reality, the only browser hat have accel with intel and a patch is chromium, out of this there isn't any browser more, you can only use mpv with smplayer for view videos..