v8_external_snapshot target visible only when using v8_use_external_startup_data

v8_external_snapshot target visible only when using v8_use_external_startup_data

Prevent snapshot_blob.bin from being unnecessarily generated when building all targets.
When such file exists, chrome_tests assumes we will load V8's initial snapshot from
external files, even when such feature is turned off, which prevents the tests from
being successfully compiled.

BUG=421063
LOG=N
R=rmcilroy@chromium.org

Committed: https://code.google.com/p/v8/source/detail?r=24550

[baixo, 2014-10-10 16:21:04]

baixo@chromium.org changed reviewers:
+ rmcilroy@chromium.org, vogelheim@chromium.org

[baixo, 2014-10-10 16:21:04]

Please take a look.

[rmcilroy, 2014-10-10 16:31:33]

Please update the description to mention the reason why this is a problem
currently.  Also please add the tracking bug in a BUG= line.

https://codereview.chromium.org/643163002/diff/1/tools/gyp/v8.gyp
File tools/gyp/v8.gyp (right):

https://codereview.chromium.org/643163002/diff/1/tools/gyp/v8.gyp#newcode1405
tools/gyp/v8.gyp:1405: [ 'v8_use_external_startup_data==1', {
I just had a look at how natives_blob does this, and it looks like it has the
target unconditionally, but puts the actions in a conditions block which depends
on 'v8_use_external_startup_data==1'.  I think the same thing would work on
v8_external_snapshot - could we do that instead to keep things consistent?

[baixo, 2014-10-10 17:49:40]

Addressed Ross' comment.

https://codereview.chromium.org/643163002/diff/1/tools/gyp/v8.gyp
File tools/gyp/v8.gyp (right):

https://codereview.chromium.org/643163002/diff/1/tools/gyp/v8.gyp#newcode1405
tools/gyp/v8.gyp:1405: [ 'v8_use_external_startup_data==1', {
On 2014/10/10 16:31:33, rmcilroy wrote:
> I just had a look at how natives_blob does this, and it looks like it has the
> target unconditionally, but puts the actions in a conditions block which
depends
> on 'v8_use_external_startup_data==1'.  I think the same thing would work on
> v8_external_snapshot - could we do that instead to keep things consistent?

Done.

[rmcilroy, 2014-10-11 12:05:59]

LGTM, thanks Andre.

[rmcilroy, 2014-10-13 09:36:25]

Committed patchset #2 (id:60001) manually as 24550 (presubmit successful).