go/system: adding two-phase data pipe methods

go/system: adding two-phase methods of data pipe

R=gautham.dorai@gmail.com, jamesr@chromium.org

Committed: https://chromium.googlesource.com/external/mojo/+/bca91b1411ab309b44404fa7fda20fe54ccd07e5

[rogulenko, 2015-01-21 01:15:42]

Patchset #1 (id:1) has been deleted

[rogulenko, 2015-01-21 01:17:48]

Patchset #1 (id:20001) has been deleted

[rogulenko, 2015-01-21 01:32:10]

rogulenko@google.com changed reviewers:
+ gauthamt@google.com, jamesr@google.com

[gauthamt, 2015-01-21 18:52:34]

gautham.dorai@gmail.com changed reviewers:
+ gautham.dorai@gmail.com

[gauthamt, 2015-01-21 18:52:35]

lgtm

[gauthamt, 2015-01-21 18:52:36]

lgtm

[jamesr, 2015-01-21 18:56:14]

jamesr@chromium.org changed reviewers:
+ jamesr@chromium.org

[jamesr, 2015-01-21 18:56:15]

lgtm

https://codereview.chromium.org/860193002/diff/40001/mojo/go/tests/system_tes...
File mojo/go/tests/system_test.go (right):

https://codereview.chromium.org/860193002/diff/40001/mojo/go/tests/system_tes...
mojo/go/tests/system_test.go:234: if r, _ =
hc.Wait(system.MOJO_HANDLE_SIGNAL_READABLE, 1000); r != system.MOJO_RESULT_OK {
sorry I missed this earlier, but what does the constant "1000" signify here?  If
this is the deadline then we should probably pass MOJO_DEADLINE_INDEFINITE to
avoid flaky behavior when the system is unexpected slow or preempted at
unfortunate times. alternately if we never want to wait we should pass 0 here

[rogulenko, 2015-01-21 20:00:00]

https://codereview.chromium.org/860193002/diff/40001/mojo/go/tests/system_tes...
File mojo/go/tests/system_test.go (right):

https://codereview.chromium.org/860193002/diff/40001/mojo/go/tests/system_tes...
mojo/go/tests/system_test.go:234: if r, _ =
hc.Wait(system.MOJO_HANDLE_SIGNAL_READABLE, 1000); r != system.MOJO_RESULT_OK {
On 2015/01/21 18:56:15, jamesr wrote:
> sorry I missed this earlier, but what does the constant "1000" signify here? 
If
> this is the deadline then we should probably pass MOJO_DEADLINE_INDEFINITE to
> avoid flaky behavior when the system is unexpected slow or preempted at
> unfortunate times. alternately if we never want to wait we should pass 0 here

Changed all deadlines to either MOJO_DEADLINE_INDEFINITE or 0.

[rogulenko, 2015-01-21 20:07:28]

Committed patchset #2 (id:60001) manually as
bca91b1411ab309b44404fa7fda20fe54ccd07e5 (presubmit successful).