[Bug 98505] [radeon, amdgpu] Regression introduced in 4.8-rc3

bugzilla-daemon at freedesktop.org bugzilla-daemon at freedesktop.org
Sun Nov 13 11:24:51 UTC 2016


https://bugs.freedesktop.org/show_bug.cgi?id=98505

--- Comment #25 from Peter Wu <peter at lekensteyn.nl> ---
"Resume" means "runtime resume" as opposed to "resume from system sleep" right
(based on the logs).

When this happens, what is the output of "sudo lspci -s4: -nnvv"? (is the "rev"
field "ff" or something else? Does the Power Management register report D0 or
D3?)

Can you reproduce the problem in this way:

 1. Blacklist amdgpu (and radeon if enabled)
 2. Boot system
 3. Load amdgpu
 4. Wait for runtime PM to kick in (5 seconds)
 5. Wake up (e.g. "lspci")

If so, can you build your kernel with CONFIG_ACPI_DEBUG=y and either boot with
acpi.trace_state=enable? If that gives too many messages, you can also try
setting the option between step 2 and 3:

 echo enable > /sys/module/acpi/parameters/trace_state

What is the dmesg for this? (At this moment it is not clear for me what the
port PM is doing versus what amdgpu is doing, hopefully some logs will clarify
this).

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/dri-devel/attachments/20161113/0ea1718d/attachment.html>


More information about the dri-devel mailing list