Add gdb server variable for ant

Add gdb server variable for ant

This is the first CL to replace the ant property toolchain.dir with
android.gdbserver.

After this submitted
- Add android_gdbserver into WebKit gypis.
- Retire the toolchain.dir from ant by using android.gdbserver.
- Remove the ANDROID_TOOLCHAIN from both WebKit and chromium's gyp/gypi
- Remove the android_toolchain from envsetup_function.sh

TBR=jam
BUG=


Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=154298

[Yaron, 2012-08-29 23:22:26]

lgtm

cc: peter

http://codereview.chromium.org/10886046/diff/1/build/android/envsetup_functio...
File build/android/envsetup_functions.sh (right):

http://codereview.chromium.org/10886046/diff/1/build/android/envsetup_functio...
build/android/envsetup_functions.sh:122: case "${TARGET_ARCH}" in
Couldn't you simplify to: DEFINES+="target_arch=${TARGET_ARCH}"?

[michaelbai, 2012-08-29 23:40:07]

http://codereview.chromium.org/10886046/diff/1/build/android/envsetup_functio...
File build/android/envsetup_functions.sh (right):

http://codereview.chromium.org/10886046/diff/1/build/android/envsetup_functio...
build/android/envsetup_functions.sh:122: case "${TARGET_ARCH}" in
I tried, got some errors, The ia32 seemed be used before it change to x86, I
will try to replace ia32 with x86 in another CL.

On 2012/08/29 23:22:26, Yaron wrote:
> Couldn't you simplify to: DEFINES+="target_arch=${TARGET_ARCH}"?

[commit-bot: I haz the power, 2012-08-29 23:40:41]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/michaelbai@chromium.org/10886046/1

[commit-bot: I haz the power, 2012-08-29 23:40:47]

Presubmit check for 10886046-1 failed and returned exit status 1.


Running presubmit commit checks ...

** Presubmit Messages **
If this change has an associated bug, add BUG=[bug number].

** Presubmit ERRORS **
Missing LGTM from an OWNER for files in these directories:
    content

[commit-bot: I haz the power, 2012-08-29 23:44:50]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/michaelbai@chromium.org/10886046/1

[commit-bot: I haz the power, 2012-08-29 23:44:57]

Presubmit check for 10886046-1 failed and returned exit status 1.


Running presubmit commit checks ...

** Presubmit Messages **
If this change has an associated bug, add BUG=[bug number].

** Presubmit ERRORS **
Missing LGTM from an OWNER for files in these directories:
    content

[commit-bot: I haz the power, 2012-08-29 23:49:30]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/michaelbai@chromium.org/10886046/1

[commit-bot: I haz the power, 2012-08-30 00:42:20]

Try job failure for 10886046-1 (retry) on android for steps "compile, build"
(clobber build).
It's a second try, previously, steps "compile, build" failed.
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=android&nu...

[Isaac (away), 2012-08-30 03:08:12]

http://codereview.chromium.org/10886046/diff/1/build/android/envsetup_functio...
File build/android/envsetup_functions.sh (right):

http://codereview.chromium.org/10886046/diff/1/build/android/envsetup_functio...
build/android/envsetup_functions.sh:122: case "${TARGET_ARCH}" in
how about:

# TODO(michaelbai)
DEFINES+="target_arch=${TARGET_ARCH/x86/ia32}"

This will do the same thing as case statement.  Downside is its a little hard to
read though...

[Peter Beverloo, 2012-08-30 11:08:17]

http://codereview.chromium.org/10886046/diff/1/build/android/envsetup_functio...
File build/android/envsetup_functions.sh (right):

http://codereview.chromium.org/10886046/diff/1/build/android/envsetup_functio...
build/android/envsetup_functions.sh:122: case "${TARGET_ARCH}" in
Why was this block moved out of the switch statement above? I don't think the
few extra lines in the TARGET_PRODUCT switch will matter that much. We could
default it to "arm" outside the switch and only override it for x86 builds.

http://codereview.chromium.org/10886046/diff/1/build/android/envsetup_functio...
build/android/envsetup_functions.sh:132: esac
It's impossible to reach the default case here, considering every allowed
TARGET_PRODUCT defines a TARGET_ARCH and the function returns when an invalid
one is ran in to.

[michaelbai, 2012-08-30 14:59:39]

http://codereview.chromium.org/10886046/diff/1/build/android/envsetup_functio...
File build/android/envsetup_functions.sh (right):

http://codereview.chromium.org/10886046/diff/1/build/android/envsetup_functio...
build/android/envsetup_functions.sh:122: case "${TARGET_ARCH}" in
On 2012/08/30 11:08:18, Peter Beverloo wrote:
> Why was this block moved out of the switch statement above? I don't think the
> few extra lines in the TARGET_PRODUCT switch will matter that much. We could
> default it to "arm" outside the switch and only override it for x86 builds.

Finally, we will use DEFINES+="target_arch=${TARGET_ARCH}" as long as ia32 could
be completely replaced with x86. Let's use this for now.

http://codereview.chromium.org/10886046/diff/1/build/android/envsetup_functio...
build/android/envsetup_functions.sh:132: esac
It could be if TARGET_ARCH set to something other than x86 and arm, but the
author forget to update this block.

On 2012/08/30 11:08:18, Peter Beverloo wrote:
> It's impossible to reach the default case here, considering every allowed
> TARGET_PRODUCT defines a TARGET_ARCH and the function returns when an invalid
> one is ran in to.

[commit-bot: I haz the power, 2012-08-30 15:00:34]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/michaelbai@chromium.org/10886046/13001

[commit-bot: I haz the power, 2012-08-30 15:52:35]

Try job failure for 10886046-13001 (retry) on android for steps "compile, build"
(clobber build).
It's a second try, previously, steps "compile, build" failed.
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=android&nu...

[commit-bot: I haz the power, 2012-08-30 15:53:04]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/michaelbai@chromium.org/10886046/22001

[Peter Beverloo, 2012-08-30 16:07:33]

https://chromiumcodereview.appspot.com/10886046/diff/1/build/android/envsetup...
File build/android/envsetup_functions.sh (right):

https://chromiumcodereview.appspot.com/10886046/diff/1/build/android/envsetup...
build/android/envsetup_functions.sh:122: case "${TARGET_ARCH}" in
On 2012/08/30 14:59:39, michaelbai wrote:
> On 2012/08/30 11:08:18, Peter Beverloo wrote:
> > Why was this block moved out of the switch statement above? I don't think
the
> > few extra lines in the TARGET_PRODUCT switch will matter that much. We could
> > default it to "arm" outside the switch and only override it for x86 builds.
> 
> Finally, we will use DEFINES+="target_arch=${TARGET_ARCH}" as long as ia32
could
> be completely replaced with x86. Let's use this for now.

To illustrate my suggestion (not tested, has a typo too):
https://chromiumcodereview.appspot.com/10916013

I think that's a lot cleaner and introduces less new code.

https://chromiumcodereview.appspot.com/10886046/diff/1/build/android/envsetup...
build/android/envsetup_functions.sh:132: esac
On 2012/08/30 14:59:39, michaelbai wrote:
> It could be if TARGET_ARCH set to something other than x86 and arm, but the
> author forget to update this block.
> 
> On 2012/08/30 11:08:18, Peter Beverloo wrote:
> > It's impossible to reach the default case here, considering every allowed
> > TARGET_PRODUCT defines a TARGET_ARCH and the function returns when an
invalid
> > one is ran in to.
> 

That seems very unlikely in the near future.

[Peter Beverloo, 2012-08-30 16:15:48]

For the record, please address this in a follow-up CL.

[michaelbai, 2012-08-30 16:20:25]

On 2012/08/30 16:15:48, Peter Beverloo wrote:
> For the record, please address this in a follow-up CL.

Will do

[commit-bot: I haz the power, 2012-08-30 16:53:13]

Try job failure for 10886046-22001 (retry) on android for steps "compile, build"
(clobber build).
It's a second try, previously, steps "compile, build" failed.
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=android&nu...

[michaelbai, 2012-08-30 17:10:17]

Can't reproduce the compile error locally, fallback to original implementation.

[commit-bot: I haz the power, 2012-08-30 17:10:27]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/michaelbai@chromium.org/10886046/13002

[commit-bot: I haz the power, 2012-08-30 19:15:30]

Try job failure for 10886046-13002 (retry) on android for steps "compile, build"
(clobber build).
It's a second try, previously, steps "compile, build" failed.
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=android&nu...

[commit-bot: I haz the power, 2012-08-30 20:14:31]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/michaelbai@chromium.org/10886046/24003

[commit-bot: I haz the power, 2012-08-30 22:54:11]

Change committed as 154298