Fix ProcessUtilTest.FDRemapping on Android.

Fix ProcessUtilTest.FDRemapping on Android.

This provides a simple implementation of file descriptors remapping on Android.

BUG=136720
Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=169652

[nilesh, 2012-11-26 05:39:32]

LGTM

[Philippe, 2012-11-26 11:34:09]

Thanks Nilesh. Pawel, can you please take a look?

[Paweł Hajdan Jr., 2012-11-26 18:18:41]

https://chromiumcodereview.appspot.com/11417115/diff/5006/base/test/multiproc...
File base/test/multiprocess_test_android.cc (right):

https://chromiumcodereview.appspot.com/11417115/diff/5006/base/test/multiproc...
base/test/multiprocess_test_android.cc:32: // Parent process.
nit: If you lay it out like that use braces {} for the "if".

https://chromiumcodereview.appspot.com/11417115/diff/5006/base/test/multiproc...
base/test/multiprocess_test_android.cc:45: HANDLE_EINTR(close(fd));
nit: This is a macro, so use {} around it.

https://chromiumcodereview.appspot.com/11417115/diff/5006/base/test/multiproc...
base/test/multiprocess_test_android.cc:48: const int old_fd = it->first;
nit: No need for "const int" if this is not really a constant.

https://chromiumcodereview.appspot.com/11417115/diff/5006/base/test/multiproc...
base/test/multiprocess_test_android.cc:51: PLOG(ERROR) << "dup2";
Why do you make that non-fatal?

[Philippe, 2012-11-26 18:30:44]

Thanks Pawel.

https://chromiumcodereview.appspot.com/11417115/diff/5006/base/test/multiproc...
File base/test/multiprocess_test_android.cc (right):

https://chromiumcodereview.appspot.com/11417115/diff/5006/base/test/multiproc...
base/test/multiprocess_test_android.cc:32: // Parent process.
On 2012/11/26 18:18:41, Paweł Hajdan Jr. wrote:
> nit: If you lay it out like that use braces {} for the "if".

Done.

https://chromiumcodereview.appspot.com/11417115/diff/5006/base/test/multiproc...
base/test/multiprocess_test_android.cc:45: HANDLE_EINTR(close(fd));
On 2012/11/26 18:18:41, Paweł Hajdan Jr. wrote:
> nit: This is a macro, so use {} around it.

I think that such macros should be written in a safe way (e.g. by using do {}
while (0)) so that they don't cause subtle bugs but I agree that it's better not
to take any risk.

https://chromiumcodereview.appspot.com/11417115/diff/5006/base/test/multiproc...
base/test/multiprocess_test_android.cc:48: const int old_fd = it->first;
On 2012/11/26 18:18:41, Paweł Hajdan Jr. wrote:
> nit: No need for "const int" if this is not really a constant.

Done.

https://chromiumcodereview.appspot.com/11417115/diff/5006/base/test/multiproc...
base/test/multiprocess_test_android.cc:51: PLOG(ERROR) << "dup2";
On 2012/11/26 18:18:41, Paweł Hajdan Jr. wrote:
> Why do you make that non-fatal?

Done.

[Paweł Hajdan Jr., 2012-11-26 18:39:24]

LGTM

[Philippe, 2012-11-26 18:51:40]

On 2012/11/26 18:39:24, Paweł Hajdan Jr. wrote:
> LGTM

Thanks Pawel!

[commit-bot: I haz the power, 2012-11-27 10:53:41]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/pliard@chromium.org/11417115/11007

[commit-bot: I haz the power, 2012-11-27 13:45:00]

Change committed as 169652