Enable the out of process var test. It seems to work.

Enable the out of process var test. It seems to work.

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=112775

[bbudge, 2011-12-02 01:24:05]

LGTM.
By the way, I'm breaking up the FileIO and URLLoader tests into individual cases
(as dmichael started with )http://codereview.chromium.org/8477015/, with
PostMessage), since some of those test cases are still failing or flaky. If you
supect that flakiness might remain, we might want to do the same with Var to
allow sheriffs to disable only the failing cases.

[dmichael(do not use this one), 2011-12-02 01:33:51]

iirc, it was failing because of the utf8 check, which got fixed for oop
when you refactored the var stuff.

Lgtm too
On Dec 1, 2011 6:06 PM, <brettw@chromium.org> wrote:

> Reviewers: bbudge1,
>
> Description:
> Enable the out of process var test. It seems to work.
>
>
> Please review this at
http://codereview.chromium.**org/8771026/<http://codereview.chromium.org/8771...
>
> SVN Base:
svn://svn.chromium.org/chrome/**trunk/src<http://svn.chromium.org/chrome/trunk/src>
>
> Affected files:
>  M chrome/test/ui/ppapi_uitest.cc
>
>
> Index: chrome/test/ui/ppapi_uitest.cc
> diff --git a/chrome/test/ui/ppapi_uitest.**cc
> b/chrome/test/ui/ppapi_uitest.**cc
> index 8446fb300596e7f4f06b33033a1cba**f35f274cc2..**
> d9a1071d7b38b7de61700d62b0dfd2**edad9141d9 100644
> --- a/chrome/test/ui/ppapi_uitest.**cc
> +++ b/chrome/test/ui/ppapi_uitest.**cc
> @@ -316,10 +316,7 @@ TEST_PPAPI_IN_PROCESS(Crypto)
>  TEST_PPAPI_OUT_OF_PROCESS(**Crypto)
>
>  TEST_PPAPI_IN_PROCESS(Var)
> -// http://crbug.com/89961
> -TEST_F(OutOfProcessPPAPITest, FAILS_Var) {
> -  RunTest("Var");
> -}
> +TEST_PPAPI_OUT_OF_PROCESS(**Var)
>  TEST_PPAPI_NACL_VIA_HTTP(Var)
>
>  TEST_PPAPI_IN_PROCESS(**VarDeprecated)
>
>
>

[brettw, 2011-12-02 04:38:19]

On Thu, Dec 1, 2011 at 5:24 PM,  <bbudge@chromium.org> wrote:
> LGTM.
> By the way, I'm breaking up the FileIO and URLLoader tests into individual
> cases
> (as dmichael started with )http://codereview.chromium.org/8477015/, with
> PostMessage), since some of those test cases are still failing or flaky. If
> you
> supect that flakiness might remain, we might want to do the same with Var to
> allow sheriffs to disable only the failing cases.
>
> http://codereview.chromium.org/8771026/

I should have all the FileIO tests passing out of process now. The
patch is ready but not checked in yet (I have to update another test).

Brett

[bbudge, 2011-12-02 17:54:06]

On 2011/12/02 04:38:19, brettw wrote:
> On Thu, Dec 1, 2011 at 5:24 PM,  <mailto:bbudge@chromium.org> wrote:
> > LGTM.
> > By the way, I'm breaking up the FileIO and URLLoader tests into individual
> > cases
> > (as dmichael started with )http://codereview.chromium.org/8477015/, with
> > PostMessage), since some of those test cases are still failing or flaky. If
> > you
> > supect that flakiness might remain, we might want to do the same with Var to
> > allow sheriffs to disable only the failing cases.
> >
> > http://codereview.chromium.org/8771026/
> 
> I should have all the FileIO tests passing out of process now. The
> patch is ready but not checked in yet (I have to update another test).
> 
> Brett

I will update those tests when you land the FileIO proxy. But we still need to
break up FileIO at least, because Touch and SetLength fail on Mac and on Linux
in debug (more restrictive sandboxing).