[turbofan] Changed TruncateFloat64ToInt64 to TryTruncateFloat64ToInt64.

[turbofan] Changed TruncateFloat64ToInt64 to TryTruncateFloat64ToInt64.

The new operator provides a second output which indicates whether the
conversion from float64 to int64 was successful or not. The second
output returns 0 if the conversion fails. If the conversion succeeds,
then the second output is differs from 0.

The second output can be ignored, which means that the operator can be
used the same way as the original operator.

I implemented the new operator on x64 and arm64. @v8-mips-ports and
@v8-ppc-ports, can you please take care of the mips64 and ppc64
implementation of the second output?

R=titzer@chromium.org, v8-arm-ports@googlegroups.com

Committed: https://crrev.com/95844d94f3336489403c7f2d70c6ea01a0cf3002
Cr-Commit-Position: refs/heads/master@{#32653}

[ahaas, 2015-12-04 15:08:21]

Description was changed from

==========
[turbofan] Changed TruncateFloat64ToInt64 to TryTruncateFloat64ToInt64.

The new operator provides a second output which indicates whether the
conversion from float64 to int64 was successful or not. The second
output returns 0 if the conversion fails. If the conversion succeeds,
then the second output is differs from 0.

The second output can be ignored, which means that the operator can be
used the same way as the original operator.

I implemented the new operator on x64 and arm64. @v8-mips-ports and
@v8-ppc-ports, can you please take care of the mips64 and ppc64
implementation of the second output?

R=titzer@chromium.org, v8-arm-ports@googlegroups.com
CC=v8-mips-ports@googlegroups.com, v8-ppc-ports@googlegroups.com
==========

to

==========
[turbofan] Changed TruncateFloat64ToInt64 to TryTruncateFloat64ToInt64.

The new operator provides a second output which indicates whether the
conversion from float64 to int64 was successful or not. The second
output returns 0 if the conversion fails. If the conversion succeeds,
then the second output is differs from 0.

The second output can be ignored, which means that the operator can be
used the same way as the original operator.

I implemented the new operator on x64 and arm64. @v8-mips-ports and
@v8-ppc-ports, can you please take care of the mips64 and ppc64
implementation of the second output?

R=titzer@chromium.org, v8-arm-ports@googlegroups.com
==========

[Rodolph Perfetta (ARM), 2015-12-04 16:15:45]

rodolph.perfetta@arm.com changed reviewers:
+ rodolph.perfetta@arm.com

[Rodolph Perfetta (ARM), 2015-12-04 16:15:45]

https://codereview.chromium.org/1495213003/diff/1/src/compiler/arm64/code-gen...
File src/compiler/arm64/code-generator-arm64.cc (right):

https://codereview.chromium.org/1495213003/diff/1/src/compiler/arm64/code-gen...
src/compiler/arm64/code-generator-arm64.cc:1049: __ Cmp(i.OutputRegister(0),
i.InputInt64(1));
Have a look at MacroAssembler::TryConvertDoubleToInt64. It detects overflow and
underflow by comparing to 1 and -1 and checking for overflow. The benefit being
1 and -1 can be encoded in the cmp instructions while INT64_[MIN|MAX] cannot.
the code there also use conditional cmp to avoid some branches. fcmp also has a
coditional variant so all the cmp can be chained without branches.

https://codereview.chromium.org/1495213003/diff/1/src/compiler/arm64/code-gen...
src/compiler/arm64/code-generator-arm64.cc:1053: __ Mov(i.OutputRegister(1), 1);
using cset will give you more compact code.

[ahaas, 2015-12-07 11:00:34]

https://codereview.chromium.org/1495213003/diff/1/src/compiler/arm64/code-gen...
File src/compiler/arm64/code-generator-arm64.cc (right):

https://codereview.chromium.org/1495213003/diff/1/src/compiler/arm64/code-gen...
src/compiler/arm64/code-generator-arm64.cc:1049: __ Cmp(i.OutputRegister(0),
i.InputInt64(1));
On 2015/12/04 at 16:15:45, Rodolph Perfetta (ARM) wrote:
> Have a look at MacroAssembler::TryConvertDoubleToInt64. It detects overflow
and underflow by comparing to 1 and -1 and checking for overflow. The benefit
being 1 and -1 can be encoded in the cmp instructions while INT64_[MIN|MAX]
cannot. the code there also use conditional cmp to avoid some branches. fcmp
also has a coditional variant so all the cmp can be chained without branches.

Done.

https://codereview.chromium.org/1495213003/diff/1/src/compiler/arm64/code-gen...
src/compiler/arm64/code-generator-arm64.cc:1053: __ Mov(i.OutputRegister(1), 1);
On 2015/12/04 at 16:15:45, Rodolph Perfetta (ARM) wrote:
> using cset will give you more compact code.

Done.

[Rodolph Perfetta (ARM), 2015-12-07 11:19:47]

arm64 part lgtm with nit addressed.

Thanks

https://codereview.chromium.org/1495213003/diff/20001/src/compiler/arm64/inst...
File src/compiler/arm64/instruction-selector-arm64.cc (right):

https://codereview.chromium.org/1495213003/diff/20001/src/compiler/arm64/inst...
src/compiler/arm64/instruction-selector-arm64.cc:1248: Constant
int64min(INT64_MIN);
those are not needed anymore

[ahaas, 2015-12-07 11:54:51]

https://codereview.chromium.org/1495213003/diff/20001/src/compiler/arm64/inst...
File src/compiler/arm64/instruction-selector-arm64.cc (right):

https://codereview.chromium.org/1495213003/diff/20001/src/compiler/arm64/inst...
src/compiler/arm64/instruction-selector-arm64.cc:1248: Constant
int64min(INT64_MIN);
On 2015/12/07 at 11:19:47, Rodolph Perfetta (ARM) wrote:
> those are not needed anymore

Done, thanks for looking at the code so carefully.

[titzer, 2015-12-07 11:56:50]

lgtm

[ahaas, 2015-12-07 11:59:58]

The CQ bit was checked by ahaas@chromium.org

[ahaas, 2015-12-07 11:59:59]

The patchset sent to the CQ was uploaded after l-g-t-m from
rodolph.perfetta@arm.com
Link to the patchset: https://codereview.chromium.org/1495213003/#ps40001
(title: "Removed unused code.")

[commit-bot: I haz the power, 2015-12-07 12:00:03]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1495213003/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1495213003/40001

[commit-bot: I haz the power, 2015-12-07 12:12:24]

Committed patchset #3 (id:40001)

[commit-bot: I haz the power, 2015-12-07 12:13:14]

Description was changed from

==========
[turbofan] Changed TruncateFloat64ToInt64 to TryTruncateFloat64ToInt64.

The new operator provides a second output which indicates whether the
conversion from float64 to int64 was successful or not. The second
output returns 0 if the conversion fails. If the conversion succeeds,
then the second output is differs from 0.

The second output can be ignored, which means that the operator can be
used the same way as the original operator.

I implemented the new operator on x64 and arm64. @v8-mips-ports and
@v8-ppc-ports, can you please take care of the mips64 and ppc64
implementation of the second output?

R=titzer@chromium.org, v8-arm-ports@googlegroups.com
==========

to

==========
[turbofan] Changed TruncateFloat64ToInt64 to TryTruncateFloat64ToInt64.

The new operator provides a second output which indicates whether the
conversion from float64 to int64 was successful or not. The second
output returns 0 if the conversion fails. If the conversion succeeds,
then the second output is differs from 0.

The second output can be ignored, which means that the operator can be
used the same way as the original operator.

I implemented the new operator on x64 and arm64. @v8-mips-ports and
@v8-ppc-ports, can you please take care of the mips64 and ppc64
implementation of the second output?

R=titzer@chromium.org, v8-arm-ports@googlegroups.com

Committed: https://crrev.com/95844d94f3336489403c7f2d70c6ea01a0cf3002
Cr-Commit-Position: refs/heads/master@{#32653}
==========

[commit-bot: I haz the power, 2015-12-07 12:13:15]

Patchset 3 (id:??) landed as
https://crrev.com/95844d94f3336489403c7f2d70c6ea01a0cf3002
Cr-Commit-Position: refs/heads/master@{#32653}