[turbofan] Add support for ARM64 Ubfx

[turbofan] Add support for ARM64 Ubfx

Support selecting Ubfx for shift-mask and mask-shift operations. Also, rename
the shifts to match the instruction names.

BUG=
R=bmeurer@chromium.org

Committed: https://code.google.com/p/v8/source/detail?r=24482

[m.m.capewell, 2014-10-08 09:25:06]

m.m.capewell@googlemail.com changed reviewers:
+ bmeurer@chromium.org, ulan@chromium.org

[Benedikt Meurer, 2014-10-08 11:40:12]

https://codereview.chromium.org/633123002/diff/1/src/base/bits.h
File src/base/bits.h (right):

https://codereview.chromium.org/633123002/diff/1/src/base/bits.h#newcode107
src/base/bits.h:107: inline uint64_t CountTrailingZeros64(uint64_t value) {
I think the result type of all these Count* functions should be unsigned. Can
you fix that for the 32-bit versions as well?

https://codereview.chromium.org/633123002/diff/1/src/base/utils/random-number...
File src/base/utils/random-number-generator.cc (right):

https://codereview.chromium.org/633123002/diff/1/src/base/utils/random-number...
src/base/utils/random-number-generator.cc:109: }
This is undefined behaviour in C++. Should be:

uint64_t lo = bit_cast<unsigned>(Next(32));
uint64_t hi = bit_cast<unsigned>(Next(32));
return lo | (hi << 32);

[m.m.capewell, 2014-10-08 17:35:02]

https://codereview.chromium.org/633123002/diff/1/src/base/bits.h
File src/base/bits.h (right):

https://codereview.chromium.org/633123002/diff/1/src/base/bits.h#newcode107
src/base/bits.h:107: inline uint64_t CountTrailingZeros64(uint64_t value) {
On 2014/10/08 11:40:12, Benedikt Meurer wrote:
> I think the result type of all these Count* functions should be unsigned. Can
> you fix that for the 32-bit versions as well?

Done.

https://codereview.chromium.org/633123002/diff/1/src/base/utils/random-number...
File src/base/utils/random-number-generator.cc (right):

https://codereview.chromium.org/633123002/diff/1/src/base/utils/random-number...
src/base/utils/random-number-generator.cc:109: }
On 2014/10/08 11:40:12, Benedikt Meurer wrote:
> This is undefined behaviour in C++. Should be:
> 
> uint64_t lo = bit_cast<unsigned>(Next(32));
> uint64_t hi = bit_cast<unsigned>(Next(32));
> return lo | (hi << 32);

Done.

[Benedikt Meurer, 2014-10-09 06:11:43]

lgtm

[Benedikt Meurer, 2014-10-09 09:18:47]

Committed patchset #2 (id:20001) manually as 24482 (presubmit successful).