GM: signal error if the total number of tests run was less than expected

GM: signal error if the total number of tests run was less than expected
Committed: https://code.google.com/p/skia/source/detail?r=8614

[epoger, 2013-04-10 19:09:29]

I'm not aware of any cases when this would happen, but it would be good to know
if/when it DOES happen...

[borenet, 2013-04-10 19:12:56]

LGTM. We'll see what happens!

[epoger, 2013-04-11 15:08:49]

I looked at the most recent GenerateGMs run on every builder, and saw that this
change would have made *all* of the Android builders turn red.

Filed to myself as https://code.google.com/p/skia/issues/detail?id=1222 ('GM is
unexpectedly skipping tests on Android'); I won't commit this change until we
think that is resolved.

[borenet, 2013-04-11 15:10:40]

On 2013/04/11 15:08:49, epoger wrote:
> I looked at the most recent GenerateGMs run on every builder, and saw that
this
> change would have made *all* of the Android builders turn red.
> 
> Filed to myself as https://code.google.com/p/skia/issues/detail?id=1222 ('GM
is
> unexpectedly skipping tests on Android'); I won't commit this change until we
> think that is resolved.

SGTM.  Thanks for verifying first.

[epoger, 2013-04-11 15:21:17]

https://codereview.chromium.org/13863009/diff/5001/gm/gmmain.cpp
File gm/gmmain.cpp (right):

https://codereview.chromium.org/13863009/diff/5001/gm/gmmain.cpp#newcode1831
gm/gmmain.cpp:1831: #ifdef SK_BUILD_FOR_ANDROID
Actually, Eric, whaddaya think of this for now?  At least we'll catch any
*regressions* on this point, on the non-Android platforms...

[epoger, 2013-04-11 15:22:44]

On 2013/04/11 15:21:17, epoger wrote:
> https://codereview.chromium.org/13863009/diff/5001/gm/gmmain.cpp
> File gm/gmmain.cpp (right):
> 
> https://codereview.chromium.org/13863009/diff/5001/gm/gmmain.cpp#newcode1831
> gm/gmmain.cpp:1831: #ifdef SK_BUILD_FOR_ANDROID
> Actually, Eric, whaddaya think of this for now?  At least we'll catch any
> *regressions* on this point, on the non-Android platforms...

Whoops, I got the ifdef backwards, hold on...

[epoger, 2013-04-11 15:24:04]

https://codereview.chromium.org/13863009/diff/9001/gm/gmmain.cpp
File gm/gmmain.cpp (right):

https://codereview.chromium.org/13863009/diff/9001/gm/gmmain.cpp#newcode1831
gm/gmmain.cpp:1831: #ifndef SK_BUILD_FOR_ANDROID
That's better.  Eric, you OK with me committing this?

[borenet, 2013-04-11 15:38:41]

LGTM at patch set 3.

https://codereview.chromium.org/13863009/diff/9001/gm/gmmain.cpp
File gm/gmmain.cpp (right):

https://codereview.chromium.org/13863009/diff/9001/gm/gmmain.cpp#newcode1831
gm/gmmain.cpp:1831: #ifndef SK_BUILD_FOR_ANDROID
On 2013/04/11 15:24:04, epoger wrote:
> That's better.  Eric, you OK with me committing this?

Yes, assuming it looks like all is well on other platforms.

[epoger, 2013-04-11 15:41:06]

Committed patchset #3 manually as r8614 (presubmit successful).