[gn] Fix setting v8_target_arch default

[gn] Fix setting v8_target_arch default

This moves setting the v8_target_arch default to the shared
v8.gni, so that it's available in build_overrides for
stand-alone mode.

BUG=chromium:605732, chromium:474921
NOTRY=true

Committed: https://crrev.com/52c4010ff79707fe76129a73f3930886ff2049b9
Cr-Commit-Position: refs/heads/master@{#36465}

[Michael Achenbach, 2016-05-24 09:56:49]

Description was changed from

==========
[gn] Fix setting v8_target_arch default

BUG=
==========

to

==========
[gn] Fix setting v8_target_arch default

BUG=chromium:605732, chromium:474921
NOTRY=true
==========

[Michael Achenbach, 2016-05-24 09:58:24]

Description was changed from

==========
[gn] Fix setting v8_target_arch default

BUG=chromium:605732, chromium:474921
NOTRY=true
==========

to

==========
[gn] Fix setting v8_target_arch default

This moves setting the v8_target_arch default to the shared
v8.gni, so that it's available in build_overrides for
stand-alone mode.

BUG=chromium:605732, chromium:474921
NOTRY=true
==========

[Michael Achenbach, 2016-05-24 09:58:24]

machenbach@chromium.org changed reviewers:
+ dpranke@chromium.org, vogelheim@chromium.org

[Michael Achenbach, 2016-05-24 09:58:48]

PTAL

[vogelheim, 2016-05-24 10:03:41]

lgtm

[Michael Achenbach, 2016-05-24 10:38:17]

The CQ bit was checked by machenbach@chromium.org

[commit-bot: I haz the power, 2016-05-24 10:38:21]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/2006213002/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/2006213002/1

[commit-bot: I haz the power, 2016-05-24 10:40:04]

Description was changed from

==========
[gn] Fix setting v8_target_arch default

This moves setting the v8_target_arch default to the shared
v8.gni, so that it's available in build_overrides for
stand-alone mode.

BUG=chromium:605732, chromium:474921
NOTRY=true
==========

to

==========
[gn] Fix setting v8_target_arch default

This moves setting the v8_target_arch default to the shared
v8.gni, so that it's available in build_overrides for
stand-alone mode.

BUG=chromium:605732, chromium:474921
NOTRY=true
==========

[commit-bot: I haz the power, 2016-05-24 10:40:05]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2016-05-24 10:44:59]

Description was changed from

==========
[gn] Fix setting v8_target_arch default

This moves setting the v8_target_arch default to the shared
v8.gni, so that it's available in build_overrides for
stand-alone mode.

BUG=chromium:605732, chromium:474921
NOTRY=true
==========

to

==========
[gn] Fix setting v8_target_arch default

This moves setting the v8_target_arch default to the shared
v8.gni, so that it's available in build_overrides for
stand-alone mode.

BUG=chromium:605732, chromium:474921
NOTRY=true

Committed: https://crrev.com/52c4010ff79707fe76129a73f3930886ff2049b9
Cr-Commit-Position: refs/heads/master@{#36465}
==========

[commit-bot: I haz the power, 2016-05-24 10:44:59]

Patchset 1 (id:??) landed as
https://crrev.com/52c4010ff79707fe76129a73f3930886ff2049b9
Cr-Commit-Position: refs/heads/master@{#36465}

[Dirk Pranke, 2016-05-24 18:49:05]

So, if I understand the effects of these two CLs correctly,

you've moved v8_target_arch to gni/v8.gni so that you can reference it from
//build_overrides/v8.gni , and then you use the value of v8_target_arch to
determine how to set v8_enable_gdbjit, with the assumptiont that other projects
may wish to override the logic for setting v8_enable_gdbjit, but don't need to
specify it as a build arg.

Does that sound right? lgtm if so.

Also, on a tangential note, what is "x87" supposed to indicate as an
architecture as opposed to "ia32" ?

And, we should probably actually call this "v8_target_cpu" to be more consistent
with GN's naming of "target_cpu", "host_cpu", etc.

[Michael Achenbach, 2016-05-24 19:09:05]

On 2016/05/24 18:49:05, Dirk Pranke wrote:
> So, if I understand the effects of these two CLs correctly,
> 
> you've moved v8_target_arch to gni/v8.gni so that you can reference it from
> //build_overrides/v8.gni , and then you use the value of v8_target_arch to
> determine how to set v8_enable_gdbjit, with the assumptiont that other
projects
> may wish to override the logic for setting v8_enable_gdbjit, but don't need to
> specify it as a build arg.
> 
> Does that sound right? lgtm if so.
> 

Sounds right. Not entirely sure about "other projects wishing to override the
logic for setting v8_enable_gdbjit". I wasn't able to use it for v8 stand-alone
without having it shared like this. Without this move, v8_enable_gdbjit was set
up wrong for stand-alone.

> Also, on a tangential note, what is "x87" supposed to indicate as an
> architecture as opposed to "ia32" ?

I'm not an expert. It's basically ia32 with a different floating point
instruction set. More in the internet: https://en.wikipedia.org/wiki/X87

> 
> And, we should probably actually call this "v8_target_cpu" to be more
consistent
> with GN's naming of "target_cpu", "host_cpu", etc.

Agreed. Will keep that in mind (with lower prio).

[Dirk Pranke, 2016-05-24 19:40:51]

On 2016/05/24 19:09:05, Michael Achenbach (slow) wrote:
> On 2016/05/24 18:49:05, Dirk Pranke wrote:
> > Also, on a tangential note, what is "x87" supposed to indicate as an
> > architecture as opposed to "ia32" ?
> 
> I'm not an expert. It's basically ia32 with a different floating point
> instruction set. More in the internet: https://en.wikipedia.org/wiki/X87

My understanding is that the x87 instructions have been part of the core x86 
instruction set since pre-Pentium days, and so not actually different from
"ia32",
so I'm puzzled by why you'd have them separate in v8 ... Looks like the x87
code is largely being maintained by someone from Intel, so who knows ...