Slightly faster variant of Array.push.

Slightly faster variant of Array.push.

Committed: http://code.google.com/p/v8/source/detail?r=4279

[antonm, 2010-03-25 07:32:46]

And another small review for you, Mads.

Most important point is change from undefined to holes which is crucial.

[Mads Ager (chromium), 2010-03-25 09:56:27]

LGTM

http://codereview.chromium.org/1265003/diff/1/2
File src/ia32/stub-cache-ia32.cc (right):

http://codereview.chromium.org/1265003/diff/1/2#newcode1317
src/ia32/stub-cache-ia32.cc:1317: __ mov(ecx, Operand(esp, argc *
kPointerSize));
Could you add a comment here:

// Push the element.

http://codereview.chromium.org/1265003/diff/1/2#newcode1321
src/ia32/stub-cache-ia32.cc:1321: Immediate(Factory::the_hole_value()));
This looks like the right thing in any case.  I remember commenting that this
should probably be the_hole on the first review.  ;)

[antonm, 2010-03-25 14:19:12]

Thanks a lot for review, Mads.  Running through tests and submitting.

http://codereview.chromium.org/1265003/diff/1/2
File src/ia32/stub-cache-ia32.cc (right):

http://codereview.chromium.org/1265003/diff/1/2#newcode1317
src/ia32/stub-cache-ia32.cc:1317: __ mov(ecx, Operand(esp, argc *
kPointerSize));
On 2010/03/25 09:56:28, Mads Ager wrote:
> Could you add a comment here:
> 
> // Push the element.

Done.

http://codereview.chromium.org/1265003/diff/1/2#newcode1321
src/ia32/stub-cache-ia32.cc:1321: Immediate(Factory::the_hole_value()));
On 2010/03/25 09:56:28, Mads Ager wrote:
> This looks like the right thing in any case.  I remember commenting that this
> should probably be the_hole on the first review.  ;) 

I am sorry, Mads.