MIPS: Pretenure ASCII cons string in high promotion mode.

MIPS: Pretenure ASCII cons string in high promotion mode.

Port r14540 (c6093a6e)

BUG=

Committed: https://code.google.com/p/v8/source/detail?r=14549

[Paul Lind, 2013-05-03 23:24:27]

There is one register naming problem here that should be fixed. THis issue
exists in the arm part of 14540 also, Hannes, PTAL at that.

https://codereview.chromium.org/14678006/diff/1/src/mips/macro-assembler-mips.cc
File src/mips/macro-assembler-mips.cc (right):

https://codereview.chromium.org/14678006/diff/1/src/mips/macro-assembler-mips...
src/mips/macro-assembler-mips.cc:3144: lw(scratch1, MemOperand(t0, 0));
t0 here should be scratch1. (currently, there is only one caller, and t0 is
passed for scratch1, but if this function gets called differently in the future,
it would be a bug.)

Hannes, this would be true for your original r14540 also, you might want to look
at that.

[palfia, 2013-05-03 23:50:39]

Thanks, the fixed patchset is uploaded.

https://codereview.chromium.org/14678006/diff/1/src/mips/macro-assembler-mips.cc
File src/mips/macro-assembler-mips.cc (right):

https://codereview.chromium.org/14678006/diff/1/src/mips/macro-assembler-mips...
src/mips/macro-assembler-mips.cc:3144: lw(scratch1, MemOperand(t0, 0));
On 2013/05/03 23:24:27, Paul Lind wrote:
> t0 here should be scratch1. (currently, there is only one caller, and t0 is
> passed for scratch1, but if this function gets called differently in the
future,
> it would be a bug.)
> 
> Hannes, this would be true for your original r14540 also, you might want to
look
> at that.
> 

Good catch, thanks! Fixed.

[Paul Lind, 2013-05-03 23:54:27]

lgtm, thanks!

[palfia, 2013-05-03 23:55:15]

Committed manually as r14549 (presubmit successful).