dri-devel Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Daniel Mota Leite <daniel@motaleite.net>
To: "Bridgman, John" <John.Bridgman@amd.com>
Cc: DRI Development <dri-devel@lists.freedesktop.org>
Subject: low mclock after some suspend cycles (was Re: Slower 3D with kernel 4.11.x)
Date: Mon, 10 Jul 2017 05:27:25 +0100	[thread overview]
Message-ID: <20170710052725.463aeeb4.daniel@motaleite.net> (raw)
In-Reply-To: <20170603063140.13636311.daniel@motaleite.net>


[-- Attachment #1.1: Type: text/plain, Size: 1825 bytes --]

Hi again

> > OK, no powerplay messages that time. Was performance back to what you
> > expected ?  
> Yes, performance on 4.10.12 is what i call normal

	I was bisect the 4.11 kernel when 4.12 was released, and guess
what? The problem is fixed in 4.12, i can get normal performance in
vulkan again... but not all is perfect...

	After the kernel update to 4.12 and also a firmware update from
the distro, i see no errors on the dmesg (probably the most of the errors
were firmware related?), but i found that after several suspend cycled,
the GPU is locked in the lowest mclck and even the sclck seems to be
harder to increase.

	As per radeon-profile, the performance is still auto and even
trying to switch to high or manual makes no difference, mclck is locked on
the lower value.

	So something in the suspend *sometimes* lock GPU the power profile
on a lower state.

	I also found that even after boot, and switching to manual, i
can't decrease the mclock, at least it by default on the highest
setting, but i may point to some problem switching the mclck.
sclck i can manually change, so no problem there.

Thanks for all the mesa and amdgpu work
higuita
-- 
Naturally the common people don't want war... but after all it is the
leaders of a country who determine the policy, and it is always a 
simple matter to drag the people along, whether it is a democracy, or
a fascist dictatorship, or a parliament, or a communist dictatorship.
Voice or no voice, the people can always be brought to the bidding of
the leaders. That is easy. All you have to do is tell them they are 
being attacked, and denounce the pacifists for lack of patriotism and
exposing the country to danger.  It works the same in every country.
           -- Hermann Goering, Nazi and war criminal, 1883-1946

[-- Attachment #1.2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 195 bytes --]

[-- Attachment #2: Type: text/plain, Size: 160 bytes --]

_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

      parent reply	other threads:[~2017-07-10  4:27 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-05-31 23:47 Slower 3D with kernel 4.11.x Daniel Mota Leite
2017-06-01 12:44 ` Alex Deucher
2017-06-01 20:18   ` Daniel Mota Leite
2017-06-01 21:09     ` Bridgman, John
2017-06-03  1:47       ` Daniel Mota Leite
2017-06-03  1:53         ` Bridgman, John
2017-06-03  5:31           ` Daniel Mota Leite
2017-06-10 15:28             ` Daniel Mota Leite
2017-07-10  4:27             ` Daniel Mota Leite [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20170710052725.463aeeb4.daniel@motaleite.net \
    --to=daniel@motaleite.net \
    --cc=John.Bridgman@amd.com \
    --cc=dri-devel@lists.freedesktop.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox