GitLab

* commit '0cc49e8d':
  Add BOARD_WLAN_DEVICE=mrvl check for ANDROID_P2P support

Change-Id: I5cfe2a1b66c0bb4fe5266e1bdd3fd929905ef520
Signed-off-by: Mike J. Chen <mjchen@google.com>

The build fails when the include order is reversed.
On the other hand, the build system doesn't guarantee the file order
returned by all-subdir-makefiles.

Change-Id: I398e536fc8189d0f26ebc0b0342c778d234532ad

Conflicts:

	Android.mk

Change-Id: Idfcff5f2799863075160cc6c2565f2fe774c2552

* commit 'c0aec062':

* commit '930540de':
  P2P: Fix network removal to select correct block
  wpa_supplicant: Postpone P2P scan if usual scan is in progress

Change-Id: Ibb893211cd723e60db60b80a3da19c58575461e2

Change-Id: Id8e9b62ec7d117c1b6bdbeb739be7148f15f31cb

* commit 'efb9743f':
  Increase timeouts for p2p

* commit '86c93df1':
  Increase timeouts for p2p

Change-Id: I99d44c2e57eba3b98e5d217591968559fa3d1fe3

Dongles can be slow to respond and a quick turn around time on packets does not help.
Allow for upto 350ms on retry attempts for provision discovery, negotiation and invitation

This will catch slow responses within the first attempt and prevent any kind of issues
with sequence number handling

Bug: 7445415
Change-Id: I88a849d400b10f42ac298bad6d01f49803fcc8ba

* commit 'a27072d9':
  Add parameter checking and error reporting to wpa_blacklist API.

* commit 'b8fdecbf':
  Add parameter checking and error reporting to wpa_blacklist API.

Bug: 7423119
Change-Id: Iae85361aa0dc94a75bffd07deef451c85e3d9d96
Signed-off-by: Sasha Levitskiy <sanek@google.com>

* commit 'd73deeec':
  wpa_supplicant: Try to reject proper BSSID even if not set by the driver

* commit '54cb0f6f':
  wpa_supplicant: Try to reject proper BSSID even if not set by the driver

Bug: 7423119

Change-Id: Id0266274c9c60f8eea872ca3cfa5ad40d9180789
Signed-off-by: Dmitry Shmidt <dimitrysh@google.com>

* commit '03df168d':
  Add better handling of CTRL-EVENT-ASSOC-REJECT

* commit 'b485b188':
  Add better handling of CTRL-EVENT-ASSOC-REJECT

In an enterprise environment a given Access Point (AP) may reject an
association request due to load balancing.  In an enterprise with a
congested Wi-Fi network we may have to connect to many APs before we
find one that will accept the connection.  Currently when the
wpa_supplicant receives a CTRL-EVENT-ASSOC-REJECT it will continue to
count down the 10 second authentication timer, and doesn't realize
that it should now attempt to find a different AP.  Fix this issue in
multiple ways.  First, we increase the number of association rejects
we handle before we disable the network.  This will allow us more
opportunity to authenticate with other APs which are sharing the same
SSID.  Second, when we are rejected we immediately blacklist the AP
and rescan so that we can immediately attempt to connect to other APs.

Bug: 7329568
Change-Id: I0ff66a0e05e6d4a9dec3dea98eccd850ecd5e343

* commit 'a44422be':
  Support conc_pref when use_p2p_group_interface=0

* commit '12b1cd9e':
  Support conc_pref when use_p2p_group_interface=0

The original implementation of wpas_p2p_handle_frequency_conflicts()
only works correctly when P2P groups use a separate interface
(i.e. use_p2p_group_interface=1).  Update the implementation so
that it also works when the device interfaces is used
(i.e. use_p2p_group_interface=0).

Bug: 7290509
Change-Id: Icbc489125c5b7bd6e174d6aecd53555cdc9ad99d

* commit 'f1b71881':
  Increase channel delay to 100ms

* commit '81931b8b':
  Increase channel delay to 100ms

There is an inherent flaw in the p2p protocol design where an ACK loss right
before channel switch leads to peers being out of sync.

A work around of 50ms was added persistent case, but it turns out in heavy lossy
conditions that is not enough. Increase it to 100ms.

Experimental evaluation showed 100ms improves the reliability of p2p persistence
reinvocation.

Bug: 7359500
Change-Id: I73ab1b64e32b87101e95b0e487c90818c2c0aaf3

Change-Id: I1253d2b70d2ba5be1771d5824386c8c42bdbdf0f

* commit '532a21e6':
  P2P: Improve robustness against lost ctrl::ack

* commit '2c5b17df':
  P2P: Improve robustness against lost ctrl::ack