PPC/s390: [build] fix target_arch for ppc/s390 native builds

PPC/s390: [build] fix target_arch for ppc/s390 native builds

    target_arch is incorrectly set for the ppc/s390 native builds.
    The target name returned by g++ is not a subset of the target name
    passed as Make argument and therefore defaults to ia32. Adding
    a few more swap rules to set the target_arch correctly.
    This change shouldn't affect the simulator builds.

    The change to set target_arch to ia32 was added as part of:
    https://codereview.chromium.org/2045173003

R=machenbach@chromium.org, joransiu@ca.ibm.com, jyan@ca.ibm.com, michael_dawson@ca.ibm.com, mbrandy@us.ibm.com

BUG=
LOG=N

Committed: https://crrev.com/1298aef51fc0e7569e1b9c5905ad8d29a2903246
Cr-Commit-Position: refs/heads/master@{#37071}

[JaideepBajwa, 2016-06-16 22:44:36]

PTAL

[Michael Achenbach, 2016-06-17 06:35:01]

lgtm if this works for you...

https://codereview.chromium.org/2077733002/diff/1/Makefile
File Makefile (right):

https://codereview.chromium.org/2077733002/diff/1/Makefile#newcode453
Makefile:453: $(eval CXX_TARGET_ARCH:=$(subst ppc64,ppc,$(CXX_TARGET_ARCH)))
Not sure I understand the replacement of ppc64 with ppc. So if ppc64 is compiled
native, isn't the v8_target_arch == ppc64? In this case the condition below
would not set target_arch, but on the other hand it also wouldn't set ia32 as
also that condition wouldn't hold for ppc64. Maybe doesn't matter...

[john.yan, 2016-06-17 13:06:09]

lgtm

https://codereview.chromium.org/2077733002/diff/1/Makefile
File Makefile (right):

https://codereview.chromium.org/2077733002/diff/1/Makefile#newcode453
Makefile:453: $(eval CXX_TARGET_ARCH:=$(subst ppc64,ppc,$(CXX_TARGET_ARCH)))
On 2016/06/17 06:35:01, Michael Achenbach wrote:
> Not sure I understand the replacement of ppc64 with ppc. So if ppc64 is
compiled
> native, isn't the v8_target_arch == ppc64? In this case the condition below
> would not set target_arch, but on the other hand it also wouldn't set ia32 as
> also that condition wouldn't hold for ppc64. Maybe doesn't matter...

Hello Michael,

I found that when v8_target_arch == ppc64 and CXX_TARGET_ARCH == ppc,
"$(findstring $(CXX_TARGET_ARCH),$(V8_TARGET_ARCH))" will actually return TRUE.
So target_arch will actually be set correctly.

[JaideepBajwa, 2016-06-17 13:24:56]

On 2016/06/17 06:35:01, Michael Achenbach wrote:
> lgtm if this works for you...
> 
> https://codereview.chromium.org/2077733002/diff/1/Makefile
> File Makefile (right):
> 
> https://codereview.chromium.org/2077733002/diff/1/Makefile#newcode453
> Makefile:453: $(eval CXX_TARGET_ARCH:=$(subst ppc64,ppc,$(CXX_TARGET_ARCH)))
> Not sure I understand the replacement of ppc64 with ppc. So if ppc64 is
compiled
> native, isn't the v8_target_arch == ppc64? In this case the condition below
> would not set target_arch, but on the other hand it also wouldn't set ia32 as
> also that condition wouldn't hold for ppc64. Maybe doesn't matter...

Hi Michael,
For ppc64 native build it sets target_arch correctly (to ppc64) as well.
Since findstring (ppc, ppc64) would return true and target_arch would be
set to v8_target_arch. I added that case mainly for ppc native builds, where
CXX_targert is ppc64 but v8_target_arch is ppc.
Thanks.

[JaideepBajwa, 2016-06-17 13:26:06]

On 2016/06/17 13:06:09, john.yan wrote:
> lgtm
> 
> https://codereview.chromium.org/2077733002/diff/1/Makefile
> File Makefile (right):
> 
> https://codereview.chromium.org/2077733002/diff/1/Makefile#newcode453
> Makefile:453: $(eval CXX_TARGET_ARCH:=$(subst ppc64,ppc,$(CXX_TARGET_ARCH)))
> On 2016/06/17 06:35:01, Michael Achenbach wrote:
> > Not sure I understand the replacement of ppc64 with ppc. So if ppc64 is
> compiled
> > native, isn't the v8_target_arch == ppc64? In this case the condition below
> > would not set target_arch, but on the other hand it also wouldn't set ia32
as
> > also that condition wouldn't hold for ppc64. Maybe doesn't matter...
> 
> Hello Michael,
> 
> I found that when v8_target_arch == ppc64 and CXX_TARGET_ARCH == ppc,
> "$(findstring $(CXX_TARGET_ARCH),$(V8_TARGET_ARCH))" will actually return
TRUE.
> So target_arch will actually be set correctly.

Thanks John, my bad, didn't see your reply before.

[JaideepBajwa, 2016-06-17 13:26:42]

The CQ bit was checked by bjaideep@ca.ibm.com

[commit-bot: I haz the power, 2016-06-17 13:26:51]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/2077733002/1

[commit-bot: I haz the power, 2016-06-17 13:55:24]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2016-06-17 13:57:18]

Description was changed from

==========
PPC/s390: [build] fix target_arch for ppc/s390 native builds

    target_arch is incorrectly set for the ppc/s390 native builds.
    The target name returned by g++ is not a subset of the target name
    passed as Make argument and therefore defaults to ia32. Adding
    a few more swap rules to set the target_arch correctly.
    This change shouldn't affect the simulator builds.

    The change to set target_arch to ia32 was added as part of:
    https://codereview.chromium.org/2045173003

R=machenbach@chromium.org, joransiu@ca.ibm.com, jyan@ca.ibm.com,
michael_dawson@ca.ibm.com, mbrandy@us.ibm.com

BUG=
LOG=N
==========

to

==========
PPC/s390: [build] fix target_arch for ppc/s390 native builds

    target_arch is incorrectly set for the ppc/s390 native builds.
    The target name returned by g++ is not a subset of the target name
    passed as Make argument and therefore defaults to ia32. Adding
    a few more swap rules to set the target_arch correctly.
    This change shouldn't affect the simulator builds.

    The change to set target_arch to ia32 was added as part of:
    https://codereview.chromium.org/2045173003

R=machenbach@chromium.org, joransiu@ca.ibm.com, jyan@ca.ibm.com,
michael_dawson@ca.ibm.com, mbrandy@us.ibm.com

BUG=
LOG=N

Committed: https://crrev.com/1298aef51fc0e7569e1b9c5905ad8d29a2903246
Cr-Commit-Position: refs/heads/master@{#37071}
==========

[commit-bot: I haz the power, 2016-06-17 13:57:20]

Patchset 1 (id:??) landed as
https://crrev.com/1298aef51fc0e7569e1b9c5905ad8d29a2903246
Cr-Commit-Position: refs/heads/master@{#37071}