Looks like the USB3 driver was made a built-in for precise which is
causing some users a headache for S3 workarounds. For a bunch of
machines it seems that removing and reloading the driver over a
suspend/resume cycle is the only way forward to making S3 work reliably.
I'm kinda confused by your comment in
I'm really not sure what that relates to since having the module built
in causes us more S3 issues from what I understand... can you explain
the comment further?
Danged if I can remember what it meant. I clearly didn't do my job and
get enough context into the bug so that my comment makes sense.
The issue with the USB host drivers not being built in is that devices
can get assigned the wrong speed if the host drivers don't load (or
register) in the right order.
Tim Gardner email@example.com
kernel-team mailing list