[not for landing] Diff between a64 and r19234 for no-a64 files.

[not for landing] Diff between a64 and r19234 for no-a64 files.

BUG=

[ulan, 2014-02-11 16:40:58]

Jochen, Ross: fyi.

[ulan, 2014-02-11 17:24:25]

Rodolph, Jacob, could you please take a look at questions below?

https://codereview.chromium.org/131383004/diff/1/build/toolchain.gypi
File build/toolchain.gypi (right):

https://codereview.chromium.org/131383004/diff/1/build/toolchain.gypi#newcode523
build/toolchain.gypi:523: # Hide some GCC 4.8 warnings:
Remove before merge?

https://codereview.chromium.org/131383004/diff/1/build/toolchain.gypi#newcode628
build/toolchain.gypi:628: # Hide some GCC 4.8 warnings:
Remove before merge?

https://codereview.chromium.org/131383004/diff/1/src/compiler.cc
File src/compiler.cc (right):

https://codereview.chromium.org/131383004/diff/1/src/compiler.cc#newcode310
src/compiler.cc:310: if (FLAG_trace_opt) {
Remove before landing?

https://codereview.chromium.org/131383004/diff/1/test/mjsunit/div-mod.js
File test/mjsunit/div-mod.js (right):

https://codereview.chromium.org/131383004/diff/1/test/mjsunit/div-mod.js#newc...
test/mjsunit/div-mod.js:130: // TODO(all): due to a bug in fmod(), modulos
involving denormals
Remove before merge?

[Rodolph Perfetta, 2014-02-11 17:58:28]

https://codereview.chromium.org/131383004/diff/1/build/toolchain.gypi
File build/toolchain.gypi (right):

https://codereview.chromium.org/131383004/diff/1/build/toolchain.gypi#newcode523
build/toolchain.gypi:523: # Hide some GCC 4.8 warnings:
Yes remove.

On 2014/02/11 17:24:26, ulan wrote:
> Remove before merge?

https://codereview.chromium.org/131383004/diff/1/build/toolchain.gypi#newcode628
build/toolchain.gypi:628: # Hide some GCC 4.8 warnings:
Yes remove

On 2014/02/11 17:24:26, ulan wrote:
> Remove before merge?

https://codereview.chromium.org/131383004/diff/1/src/compiler.cc
File src/compiler.cc (right):

https://codereview.chromium.org/131383004/diff/1/src/compiler.cc#newcode310
src/compiler.cc:310: if (FLAG_trace_opt) {
Yes remove.

On 2014/02/11 17:24:26, ulan wrote:
> Remove before landing?

https://codereview.chromium.org/131383004/diff/1/test/mjsunit/div-mod.js
File test/mjsunit/div-mod.js (right):

https://codereview.chromium.org/131383004/diff/1/test/mjsunit/div-mod.js#newc...
test/mjsunit/div-mod.js:130: // TODO(all): due to a bug in fmod(), modulos
involving denormals
Yes but maybe leave a comment that those tests may fail on older libc?


On 2014/02/11 17:24:26, ulan wrote:
> Remove before merge?

[rmcilroy, 2014-02-11 18:31:05]

A couple more questions for the Arm folks:

https://codereview.chromium.org/131383004/diff/1/src/bootstrapper.cc
File src/bootstrapper.cc (right):

https://codereview.chromium.org/131383004/diff/1/src/bootstrapper.cc#newcode2640
src/bootstrapper.cc:2640: if (!FLAG_disable_native_files) {
Is this a general bug-fix or specific for A64 or something left over from
testing?

https://codereview.chromium.org/131383004/diff/1/src/builtins.cc
File src/builtins.cc (right):

https://codereview.chromium.org/131383004/diff/1/src/builtins.cc#newcode1631
src/builtins.cc:1631: // we can generate a lot of debug code on A64.
Is this still required?

https://codereview.chromium.org/131383004/diff/1/src/checks.h
File src/checks.h (right):

https://codereview.chromium.org/131383004/diff/1/src/checks.h#newcode58
src/checks.h:58: 
Could we remove these now?  There is only one TODO_UNIMPLEMENTED left in the
simulator-a64 and one ASM_UNIMPLEMENTED in lithium-codegen-a64.cc.  Otherwise,
could we make them non-a64 specific?

https://codereview.chromium.org/131383004/diff/1/src/code-stubs.h
File src/code-stubs.h (right):

https://codereview.chromium.org/131383004/diff/1/src/code-stubs.h#newcode110
src/code-stubs.h:110: #elif V8_TARGET_ARCH_A64
These seem to be the same, could we merge into the previous ifdef and make it
#if defined(V8_TARGET_ARCH_ARM) || defined(V8_TARGET_ARCH_A64).

https://codereview.chromium.org/131383004/diff/1/test/message/message.status
File test/message/message.status (right):

https://codereview.chromium.org/131383004/diff/1/test/message/message.status#...
test/message/message.status:35: ['arch == a64', {
Can we remove this?

[Rodolph Perfetta, 2014-02-12 01:08:23]

https://codereview.chromium.org/131383004/diff/1/src/bootstrapper.cc
File src/bootstrapper.cc (right):

https://codereview.chromium.org/131383004/diff/1/src/bootstrapper.cc#newcode2640
src/bootstrapper.cc:2640: if (!FLAG_disable_native_files) {
I believe this was to make it easier to start the port. You can remove.

On 2014/02/11 18:31:06, rmcilroy wrote:
> Is this a general bug-fix or specific for A64 or something left over from
> testing?

https://codereview.chromium.org/131383004/diff/1/src/builtins.cc
File src/builtins.cc (right):

https://codereview.chromium.org/131383004/diff/1/src/builtins.cc#newcode1631
src/builtins.cc:1631: // we can generate a lot of debug code on A64.
Yes, when in debug we generate a fair amount of extra code.

On 2014/02/11 18:31:06, rmcilroy wrote:
> Is this still required?

https://codereview.chromium.org/131383004/diff/1/src/checks.h
File src/checks.h (right):

https://codereview.chromium.org/131383004/diff/1/src/checks.h#newcode58
src/checks.h:58: 
making them non a64 specific sounds good to me.

The simulator enabled ones (ASM_LOCATION and so on) are particularly useful when
developing so would be good to keep them.

On 2014/02/11 18:31:06, rmcilroy wrote:
> Could we remove these now?  There is only one TODO_UNIMPLEMENTED left in the
> simulator-a64 and one ASM_UNIMPLEMENTED in lithium-codegen-a64.cc.  Otherwise,
> could we make them non-a64 specific?

https://codereview.chromium.org/131383004/diff/1/src/code-stubs.h
File src/code-stubs.h (right):

https://codereview.chromium.org/131383004/diff/1/src/code-stubs.h#newcode110
src/code-stubs.h:110: #elif V8_TARGET_ARCH_A64
Yes they can be consolidated.

On 2014/02/11 18:31:06, rmcilroy wrote:
> These seem to be the same, could we merge into the previous ifdef and make it
> #if defined(V8_TARGET_ARCH_ARM) || defined(V8_TARGET_ARCH_A64).

https://codereview.chromium.org/131383004/diff/1/test/message/message.status
File test/message/message.status (right):

https://codereview.chromium.org/131383004/diff/1/test/message/message.status#...
test/message/message.status:35: ['arch == a64', {
Yes.

On 2014/02/11 18:31:06, rmcilroy wrote:
> Can we remove this?