diff options
author | Scott Murray <scott.murray@konsulko.com> | 2021-09-13 18:13:42 -0400 |
---|---|---|
committer | Scott Murray <scott.murray@konsulko.com> | 2021-09-15 12:00:09 +0000 |
commit | 0102bc389bf53c0c7c6faa637a5a1f26a2f020b4 (patch) | |
tree | 5cfc833ae538a15dd11dfdb2cf2643b07a2197ef /lib_agent.c | |
parent | c80bb1626628b7d1eb10c1ac6f655ba8ed5c3f34 (diff) |
Rework adapter detection logiclamprey_12.1.9lamprey_12.1.8lamprey_12.1.7lamprey_12.1.6lamprey_12.1.5lamprey_12.1.4lamprey_12.1.3lamprey_12.1.20lamprey_12.1.2lamprey_12.1.19lamprey_12.1.18lamprey_12.1.17lamprey_12.1.16lamprey_12.1.15lamprey_12.1.14lamprey_12.1.13lamprey_12.1.12lamprey_12.1.11lamprey_12.1.10lamprey_12.1.1lamprey_12.1.0lamprey/12.1.9lamprey/12.1.8lamprey/12.1.7lamprey/12.1.6lamprey/12.1.5lamprey/12.1.4lamprey/12.1.3lamprey/12.1.20lamprey/12.1.2lamprey/12.1.19lamprey/12.1.18lamprey/12.1.17lamprey/12.1.16lamprey/12.1.15lamprey/12.1.14lamprey/12.1.13lamprey/12.1.12lamprey/12.1.11lamprey/12.1.10lamprey/12.1.1lamprey/12.1.012.1.912.1.812.1.712.1.612.1.512.1.412.1.312.1.2012.1.212.1.1912.1.1812.1.1712.1.1612.1.1512.1.1412.1.1312.1.1212.1.1112.1.1012.1.112.1.0lamprey
Rework the adapter detection logic in the helper thread to use the
initialization done signal to the main process to indicate that basic
BlueZ D-Bus initialization has completed, then loop forever looking
for an adapter as opposed to just trying once. This improves the
behavior with hardware such as the TI Wilink chipset or the Broadcom
chipset on Raspberry Pi 4, where the firmware loading delay may
result in the HCI device not actually being available yet when the
binding is started.
Bug-AGL: SPEC-4060
Signed-off-by: Scott Murray <scott.murray@konsulko.com>
Change-Id: I1da5c2e3168acd0f8f55dc75c2408647d03c0a4e
(cherry picked from commit c301f67772c2a2c643dd47b2291f0c971f8006ce)
Diffstat (limited to 'lib_agent.c')
0 files changed, 0 insertions, 0 deletions