Linux: Add more selective plugin blacklisting code. Blacklist a plugin that fails to check for SSE2…

Linux: Add more selective plugin blacklisting code. Blacklist a plugin that fails to check for SSE2 on platforms without SSE2 support.

BUG=140086
TEST=none

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=150446

[Lei Zhang, 2012-08-02 00:15:49]

I still have a computer at home that was built in 1999. I can probably test and
verify this works.

[Bernhard Bauer, 2012-08-02 03:18:19]

https://chromiumcodereview.appspot.com/10823132/diff/1/webkit/plugins/npapi/p...
File webkit/plugins/npapi/plugin_list_posix.cc (right):

https://chromiumcodereview.appspot.com/10823132/diff/1/webkit/plugins/npapi/p...
webkit/plugins/npapi/plugin_list_posix.cc:27: enum PluginQuark {
Do you mean quirk? Or am I missing something and we're support quantum plugins
now? ;-)

https://chromiumcodereview.appspot.com/10823132/diff/1/webkit/plugins/npapi/p...
webkit/plugins/npapi/plugin_list_posix.cc:50: if (quarks == PLUGIN_QUARK_NONE)
Can you convert this to a switch statement so that the compiler makes sure we
catch all cases?

https://chromiumcodereview.appspot.com/10823132/diff/1/webkit/plugins/npapi/p...
webkit/plugins/npapi/plugin_list_posix.cc:55: if (!cpu.has_sse2()) {
Nit: no braces?

[Lei Zhang, 2012-08-02 03:25:12]

https://chromiumcodereview.appspot.com/10823132/diff/1/webkit/plugins/npapi/p...
File webkit/plugins/npapi/plugin_list_posix.cc (right):

https://chromiumcodereview.appspot.com/10823132/diff/1/webkit/plugins/npapi/p...
webkit/plugins/npapi/plugin_list_posix.cc:27: enum PluginQuark {
On 2012/08/02 03:18:19, Bernhard Bauer wrote:
> Do you mean quirk? Or am I missing something and we're support quantum plugins
> now? ;-)

Done.

https://chromiumcodereview.appspot.com/10823132/diff/1/webkit/plugins/npapi/p...
webkit/plugins/npapi/plugin_list_posix.cc:50: if (quarks == PLUGIN_QUARK_NONE)
On 2012/08/02 03:18:19, Bernhard Bauer wrote:
> Can you convert this to a switch statement so that the compiler makes sure we
> catch all cases?

I hope we don't have to do this very often, but potentially a plugin can have
multiple quirks. e.g. QUIRK_FOO|QUIRK_BAR. We'd want to handle that.

https://chromiumcodereview.appspot.com/10823132/diff/1/webkit/plugins/npapi/p...
webkit/plugins/npapi/plugin_list_posix.cc:55: if (!cpu.has_sse2()) {
On 2012/08/02 03:18:19, Bernhard Bauer wrote:
> Nit: no braces?

Done.

[Bernhard Bauer, 2012-08-02 04:10:31]

LGTM.

https://chromiumcodereview.appspot.com/10823132/diff/1/webkit/plugins/npapi/p...
File webkit/plugins/npapi/plugin_list_posix.cc (right):

https://chromiumcodereview.appspot.com/10823132/diff/1/webkit/plugins/npapi/p...
webkit/plugins/npapi/plugin_list_posix.cc:50: if (quarks == PLUGIN_QUARK_NONE)
On 2012/08/02 03:25:12, Lei Zhang wrote:
> On 2012/08/02 03:18:19, Bernhard Bauer wrote:
> > Can you convert this to a switch statement so that the compiler makes sure
we
> > catch all cases?
> 
> I hope we don't have to do this very often, but potentially a plugin can have
> multiple quirks. e.g. QUIRK_FOO|QUIRK_BAR. We'd want to handle that.

Oh right, it's a bit field.

[Lei Zhang, 2012-08-07 21:04:58]

Tested with kvm -cpu pentium2.

[commit-bot: I haz the power, 2012-08-07 21:05:16]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/thestig@chromium.org/10823132/5001

[commit-bot: I haz the power, 2012-08-07 23:44:21]

Change committed as 150446