diff options
| author | scherkus@chromium.org <scherkus@chromium.org@0039d316-1c4b-4281-b951-d872f2087c98> | 2009-04-10 21:53:00 +0000 |
|---|---|---|
| committer | scherkus@chromium.org <scherkus@chromium.org@0039d316-1c4b-4281-b951-d872f2087c98> | 2009-04-10 21:53:00 +0000 |
| commit | 8fead9b6d8ec131f11d5fe638340cda8b6e1bb48 (patch) | |
| tree | 638e5f7c5e9c683e7d3dccba52c6b04e326d9d7d /chrome/renderer | |
| parent | 10f92ad326ee8b42f2cdba4f4b471c516c18e281 (diff) | |
| download | chromium_src-8fead9b6d8ec131f11d5fe638340cda8b6e1bb48.zip chromium_src-8fead9b6d8ec131f11d5fe638340cda8b6e1bb48.tar.gz chromium_src-8fead9b6d8ec131f11d5fe638340cda8b6e1bb48.tar.bz2 | |
Removing --enable-video and instead relying on having DLLs present.
I measured a ~10ms startup regression when the DLLs are not present and a ~20ms
regression when the DLLs are present and loaded in render_process.cc.
Not sure how else to tackle this problem since we need these DLLs loaded before
the sandbox is engaged.
Review URL: http://codereview.chromium.org/63101
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@13542 0039d316-1c4b-4281-b951-d872f2087c98
Diffstat (limited to 'chrome/renderer')
| -rw-r--r-- | chrome/renderer/render_process.cc | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/chrome/renderer/render_process.cc b/chrome/renderer/render_process.cc index e8af233..52043db 100644 --- a/chrome/renderer/render_process.cc +++ b/chrome/renderer/render_process.cc @@ -140,7 +140,7 @@ void RenderProcess::Init() { StatisticsRecorder::set_dump_on_exit(true); } - if (command_line.HasSwitch(switches::kEnableVideo) && LoadFFmpeg()) { + if (LoadFFmpeg()) { webkit_glue::SetMediaPlayerAvailable(true); } } |