pool:wificell or pool:stress-wifi?

> - Harpreet pointed out that network_WiFi_SuspendStress.stress_24HT40 and network_WiFi_SuspendStress.stress_WPA2 run 690 iterations each.
> Can we shorten this to 100?

Those are only run in 'pool:stress-wifi' (suite:wifi_stress). We can shorten them if we have a reason, but it does seem fitting that the "stress" pool would do more iterations...

And more importantly: shortening those won't free you up any time on pool:wificell

> - network_WiFi_SuspendStress.11a/b : 802.11a and 802.11b together

Sure, as long as we still keep some coverage on this test for more popular configurations.

> Those are only run in 'pool:stress-wifi' (suite:wifi_stress). We can shorten them if we have a reason, but it does seem fitting that the "stress" pool would do more iterations...

Correct. There are only 3 devices in this pool and we haven't been tracking these results. So in its current state its not really useful. I could use this space for other 'new' device installs if we are ok with removing the stress pool / suite. But if we see value in keeping it, it'll be best to expand the pool to at least cover a decent subset of kernels / wifi modules and start tracking test results. Note - I did not account for expanding this pool so we'll need to wait till early 2019 to get additional RF enclosures to expand this pool.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromiumos/third_party/autotest/+/0edd4fefffd73192552d0208e201fa1d20922cb7

commit 0edd4fefffd73192552d0208e201fa1d20922cb7
Author: Kirtika Ruchandani <kirtika@google.com>
Date: Wed Aug 29 18:15:13 2018

network_WiFi_SuspendStress: Remove dead code

This test doesn't rely on servo anymore, and doesn't need a reboot
to clean up state.

BUG= chromium:877652 
TEST=Ran network_WiFi_SuspendStress on eve after change.

Change-Id: I39912766590754007c5157bf7d7ce79233835411
Signed-off-by: Kirtika Ruchandani <kirtika@google.com>
Reviewed-on: https://chromium-review.googlesource.com/1188993
Commit-Ready: Kirtika Ruchandani <kirtika@chromium.org>
Tested-by: Kirtika Ruchandani <kirtika@chromium.org>
Reviewed-by: Harpreet Grewal <harpreet@chromium.org>

[modify] https://crrev.com/0edd4fefffd73192552d0208e201fa1d20922cb7/server/site_tests/network_WiFi_SuspendStress/network_WiFi_SuspendStress.py

The following revision refers to this bug:
  https://chromium.googlesource.com/chromiumos/third_party/autotest/+/c00409f990d28b116a2612c8643b474bf0ea8789

commit c00409f990d28b116a2612c8643b474bf0ea8789
Author: Kirtika Ruchandani <kirtika@google.com>
Date: Wed Aug 29 18:15:14 2018

network_WiFi_SuspendStress: Kill 802.11a/b variants

Based on observing autoserv.DEBUG timestamps from these test runs,
they have highly variable runtime. Each one can take from 10 mins to 60
mins, the suspend uses a random number for time to suspend/stay awake.
Both 802.11a and 80211b are old protocols that see very little usage
in the field (less than 0.1% combined of total users). It makes sense
to kill these tests to allow pool:wificell to spend time on more
relevant tests (prioritizing HT40, VHT80, adding reboot tests etc).
802.11a and 802.11b will continue getting some coverage via
network_WiFi_SimpleConnect.

BUG= chromium:877652 
TEST=Ran network_WiFi_SuspendStress on eve after change.

Change-Id: Ia3f6393cc8134d2a0dc1737bcbc3b3d11e624a50
Signed-off-by: Kirtika Ruchandani <kirtika@google.com>
Reviewed-on: https://chromium-review.googlesource.com/1188994
Commit-Ready: Kirtika Ruchandani <kirtika@chromium.org>
Tested-by: Kirtika Ruchandani <kirtika@chromium.org>
Reviewed-by: Harpreet Grewal <harpreet@chromium.org>

[delete] https://crrev.com/0edd4fefffd73192552d0208e201fa1d20922cb7/server/site_tests/network_WiFi_SuspendStress/control.11b
[delete] https://crrev.com/0edd4fefffd73192552d0208e201fa1d20922cb7/server/site_tests/network_WiFi_SuspendStress/control.11a