<div dir="ltr"><div>Thx for the suggestions.</div><div><br></div>No, it does not respond to ping.<div><br></div><div>radeon.dpm=0 does not help. But it only tells to use the old power management right ?</div><div>So I tried: low, mid and high for /sys/class/drm/card0/device/<wbr>prower_profile (and setting profile for power_mode)</div><div>With radeon.dpm=1 I tried all values for power_dpm_state / power_dpm_force_performance_<wbr>level. Same results.</div><div><br></div><div>I also tried today amd-staging-4.9 branch and same result.</div><div><br></div><div>Note that this also happens on W600 (verde), W9000 (tahiti) and W9100 (hawaii), with radeonsi driver.</div><div><br></div><div>I have open a bug here: <a href="https://bugs.freedesktop.org/show_bug.cgi?id=100465">https://bugs.freedesktop.org/show_bug.cgi?id=100465</a> . It contains a test to reproduce the freeze.</div><div><br></div><div>Cheers</div><div>Julien<br><div class="gmail_extra"><br><div class="gmail_quote">On 29 March 2017 at 18:26, Nicolai Hähnle <span dir="ltr"><<a href="mailto:nhaehnle@gmail.com" target="_blank">nhaehnle@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div class="gmail-m_3361228430822729419gmail-HOEnZb"><div class="gmail-m_3361228430822729419gmail-h5">On 29.03.2017 11:36, Michel Dänzer wrote:<br>
<blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
On 29/03/17 06:07 PM, Christian König wrote:<br>
<blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
Am 29.03.2017 um 10:59 schrieb Michel Dänzer:<br>
<blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
On 28/03/17 08:00 PM, Julien Isorce wrote:<br>
<blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
On 28 March 2017 at 10:36, Michel Dänzer <<a href="mailto:michel@daenzer.net" target="_blank">michel@daenzer.net</a><br>
<mailto:<a href="mailto:michel@daenzer.net" target="_blank">michel@daenzer.net</a>>> wrote:<br>
<br>
On 28/03/17 05:24 PM, Julien Isorce wrote:<br>
> Hi Michel,<br>
><br>
> About the hard lockup, I noticed that I cannot have it with the<br>
> following conditions:<br>
><br>
> 1. soft lockup fix (the 0->i change which avoids infinite loop)<br>
> 2. Your suggestion: (!(rbo->flags & RADEON_GEM_CPU_ACCESS)<br>
> 3. radeon.gartsize=512 radeon.vramlimit=1024 (any other values<br>
above do<br>
> not help, for example (1024, 1024) or (1024, 2048))<br>
><br>
> Without 1 and 2, but with 3, our test reproduces the soft<br>
lockup (just<br>
> discovered few days ago).<br>
> Without 3 (and with or without 1., 2.), our test reproduces<br>
the hard<br>
> lockup which one does not give any info in kern.log (sometimes<br>
some NUL<br>
> ^@ characters but not always).<br>
<br>
What exactly does "hard lockup" mean? What are the symptoms?<br>
<br>
<br>
Screens are frozen, cannot ssh, no mouse/keyboard, no kernel panic.<br>
Requires hard reboot.<br>
After reboot, nothing in /var/crash, nothing in /sys/fs/pstore, nothing<br>
in kern.log except sometimes some nul characters ^@.<br>
</blockquote>
Does it still respond to ping when it's hung?<br>
<br>
<br>
<blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
Using a serial console did not show additional debug messages. kgdb was<br>
not useful but probably worth another attempt.<br>
</blockquote>
Right.<br>
<br>
Anyway, I'm afraid it sounds like it's probably not directly related to<br>
the issue I was thinking of for my previous test patch or other similar<br>
ones I was thinking of writing.<br>
</blockquote>
<br>
Yeah, agree.<br>
<br>
Additional to that a complete crash where you don't even get anything<br>
over serial console is rather unlikely to be cause by something an<br>
application can do directly.<br>
<br>
Possible causes are more likely power management or completely messing<br>
up a bus system. Have you tried disabling dpm as well?<br>
</blockquote>
<br>
Might also be worth trying the amdgpu kernel driver instead of radeon,<br>
not sure how well the former currently works with Cape Verde though.<br>
</blockquote>
<br></div></div>
I've recently used it to experiment with the sparse buffer support. It worked well enough for that :)<br>
<br>
Cheers,<br>
Nicolai<span class="gmail-m_3361228430822729419gmail-HOEnZb"><font color="#888888"><br>
-- <br>
Lerne, wie die Welt wirklich ist,<br>
Aber vergiss niemals, wie sie sein sollte.<br>
</font></span></blockquote></div><br></div></div></div>