tree b73372074f6b9079f11382c426f061f586334863
parent 7c3204dc644deda4e3ee661e1ab9fb9ea541d24c
author Brian Norris <briannorris@chromium.org> 1493748106 -0700
committer chrome-bot <chrome-bot@chromium.org> 1494125546 -0700

shill: udev: remove device "reloader"

This was implemented with the expectation that we'd add custom uevents
into the kernel (e.g., CL:187492). But those were never merged [1]. In
the meantime, mwifiex has picked up automatic reset support entirely
within the kernel (it already had it for SDIO [2]; it has now been
implemented for PCIe via Function Level Reset [3]).

(These scripts were seemingly added only for mwifiex + PCIe, but
incidentally, I believe all other Wifi drivers have similar automatic
reset mechanisms that don't require the help of user-space like this.)

Drop these scripts.

[1] See for example, the TEST= line for CL:187474, which implies that
the test required local driver hacks. Grepping src/third_party/kernel/v*
yields no instance of the "device_hang" event.

[2] Upstream Linux commit, in kernel v3.8:
    d31ab3577eca mwifiex: add support for SDIO card reset

[3] CL:475016

CQ-DEPEND=CL:493794
BUG=none
TEST=Wifi (including 'reset' functionality) on Kevin; e.g., force a
     command timeout like the following, and see mwifiex recover:
     `iw mlan0 scan > /dev/null & cat /sys/kernel/debug/mwifiex/mlan0/device_dump`
TEST=`echo 1 > /sys/bus/pci/devices/0000\:01\:00.0/reset`

Change-Id: I88b603d2afdb61f59f9dde71ae1f543748c04660
Signed-off-by: Brian Norris <briannorris@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/493888
Reviewed-by: Paul Stewart <pstew@chromium.org>
Reviewed-by: Ben Chan <benchan@chromium.org>