src: Use simple/fast macro version of MinMax in JS

src: Use simple/fast macro version of MinMax in JS

Use the simple macro version of {Min, Max} where possible to
improve performance
Little cleanup of variable names 

Follow-up to CR: https://codereview.chromium.org/1331993004

BUG=

[skomski, 2015-09-16 21:26:36]

karl@skomski.com changed reviewers:
+ jkummerow@chromium.org, littledan@chromium.org

[skomski, 2015-09-16 21:26:51]

PTAL

[Dan Ehrenberg, 2015-09-16 21:34:22]

One nit. Generally, do you think you could separate this into two patches, one
to start using MIN_SIMPLE/MAX_SIMPLE, and another to do the other assorted
cleanups? That would make the code easier to review, and if we have to back one
of them out, we'd have a better chance letting the other land.

https://codereview.chromium.org/1345333002/diff/20001/src/arraybuffer.js
File src/arraybuffer.js (right):

https://codereview.chromium.org/1345333002/diff/20001/src/arraybuffer.js#newc...
src/arraybuffer.js:54: var byteLength = %_ArrayBufferGetByteLength(this);
Why did you move this inside of the conditional rather than before it?

[skomski, 2015-09-16 21:57:39]

https://codereview.chromium.org/1345333002/diff/20001/src/arraybuffer.js
File src/arraybuffer.js (right):

https://codereview.chromium.org/1345333002/diff/20001/src/arraybuffer.js#newc...
src/arraybuffer.js:54: var byteLength = %_ArrayBufferGetByteLength(this);
On 2015/09/16 21:34:21, Dan Ehrenberg wrote:
> Why did you move this inside of the conditional rather than before it?

Same code as in SubArray; related to https://codereview.chromium.org/201873005.

[Jakob Kummerow, 2015-09-17 07:36:34]

https://codereview.chromium.org/1345333002/diff/20001/src/arraybuffer.js
File src/arraybuffer.js (right):

https://codereview.chromium.org/1345333002/diff/20001/src/arraybuffer.js#newc...
src/arraybuffer.js:54: var byteLength = %_ArrayBufferGetByteLength(this);
On 2015/09/16 21:57:39, skomski wrote:
> On 2015/09/16 21:34:21, Dan Ehrenberg wrote:
> > Why did you move this inside of the conditional rather than before it?
> 
> Same code as in SubArray; related to
https://codereview.chromium.org/201873005.

That's not a good reason. It didn't make sense there, it doesn't make sense here
either.

[skomski, 2015-09-17 08:07:03]

https://codereview.chromium.org/1345333002/diff/20001/src/arraybuffer.js
File src/arraybuffer.js (right):

https://codereview.chromium.org/1345333002/diff/20001/src/arraybuffer.js#newc...
src/arraybuffer.js:54: var byteLength = %_ArrayBufferGetByteLength(this);
On 2015/09/17 07:36:34, Jakob wrote:
> On 2015/09/16 21:57:39, skomski wrote:
> > On 2015/09/16 21:34:21, Dan Ehrenberg wrote:
> > > Why did you move this inside of the conditional rather than before it?
> > 
> > Same code as in SubArray; related to
> https://codereview.chromium.org/201873005.
> 
> That's not a good reason. It didn't make sense there, it doesn't make sense
here
> either.

First one was not a reason rather I thought he missed that it's the same
construct but the main reason was the related bit.

1) TO_INTEGER needs to be called always before GetByteLength
(https://codereview.chromium.org/201873005)
2) IS_UNDEFINED does not need to be called twice

results into this construct.

[Dan Ehrenberg, 2015-09-21 22:09:46]

On 2015/09/17 at 08:07:03, karl wrote:
> https://codereview.chromium.org/1345333002/diff/20001/src/arraybuffer.js
> File src/arraybuffer.js (right):
> 
>
https://codereview.chromium.org/1345333002/diff/20001/src/arraybuffer.js#newc...
> src/arraybuffer.js:54: var byteLength = %_ArrayBufferGetByteLength(this);
> On 2015/09/17 07:36:34, Jakob wrote:
> > On 2015/09/16 21:57:39, skomski wrote:
> > > On 2015/09/16 21:34:21, Dan Ehrenberg wrote:
> > > > Why did you move this inside of the conditional rather than before it?
> > > 
> > > Same code as in SubArray; related to
> > https://codereview.chromium.org/201873005.
> > 
> > That's not a good reason. It didn't make sense there, it doesn't make sense
here
> > either.
> 
> First one was not a reason rather I thought he missed that it's the same
construct but the main reason was the related bit.
> 
> 1) TO_INTEGER needs to be called always before GetByteLength
(https://codereview.chromium.org/201873005)
> 2) IS_UNDEFINED does not need to be called twice
> 
> results into this construct.

Oh, I see, the constraint exists because valueOf() may have a side effect. A way
to avoid calling it twice is to store the result of IS_UNDEFINED, and use that
to determine whether to run the later code.

Could you separate out your patch for using MAX_SIMPLE and MIN_SIMPLE into a
separate patch from this code cleanup that you're doing? I feel more comfortable
approving that change than this larger one.