From mboxrd@z Thu Jan 1 00:00:00 1970 From: bugzilla-daemon@freedesktop.org Subject: [Bug 77673] New: [bisected] moving moire starting with "drm/radeon: improve PLL params if we don't match exactly v2" Date: Sat, 19 Apr 2014 15:38:22 +0000 Message-ID: Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="===============1886320494==" Return-path: Received: from culpepper.freedesktop.org (unknown [131.252.210.165]) by gabe.freedesktop.org (Postfix) with ESMTP id 297DF6E068 for ; Sat, 19 Apr 2014 08:38:23 -0700 (PDT) List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dri-devel-bounces@lists.freedesktop.org Sender: "dri-devel" To: dri-devel@lists.freedesktop.org List-Id: dri-devel@lists.freedesktop.org --===============1886320494== Content-Type: multipart/alternative; boundary="1397921902.335c1B0.15207"; charset="us-ascii" --1397921902.335c1B0.15207 Date: Sat, 19 Apr 2014 15:38:22 +0000 MIME-Version: 1.0 Content-Type: text/plain; charset="UTF-8" https://bugs.freedesktop.org/show_bug.cgi?id=77673 Priority: medium Bug ID: 77673 Assignee: dri-devel@lists.freedesktop.org Summary: [bisected] moving moire starting with "drm/radeon: improve PLL params if we don't match exactly v2" Severity: normal Classification: Unclassified OS: Linux (All) Reporter: jsagarribay@gmail.com Hardware: x86-64 (AMD64) Status: NEW Version: DRI CVS Component: DRM/Radeon Product: DRI Created attachment 97606 --> https://bugs.freedesktop.org/attachment.cgi?id=97606&action=edit Output of 'xrandr --prop' Hello, I'm testing drm-next and have found this problem. The display shows some kind of moving wavy pattern, noticeable in midtones and grandients. I've bisected it to the mentioned commit. The display is the internal LVDS panel, I haven't tested external displays. $ git bisect log git bisect start # good: [aa3a38dd40b610de206fb04777cc5469ce057de8] drm/radeon: update CI DPM powertune settings git bisect good aa3a38dd40b610de206fb04777cc5469ce057de8 # bad: [bc00aa0ebfed7edc1ca91ec2d9b0d5b04c5e06a8] drm: bochs: drop unused struct fields git bisect bad bc00aa0ebfed7edc1ca91ec2d9b0d5b04c5e06a8 # bad: [8bf5dda3777f2a477e3c448b38e18f5e1b502eae] drm/radeon/si: make sure mc ucode is loaded before checking the size git bisect bad 8bf5dda3777f2a477e3c448b38e18f5e1b502eae # good: [8a6a70662fc93c2c43ba48de3e46bf7e860e18cc] drm/radeon: re-enable mclk dpm on R7 260X asics git bisect good 8a6a70662fc93c2c43ba48de3e46bf7e860e18cc # good: [aeb0c2d8d3fc60ea931f807275166ea8abe1feeb] drm/radeon: memory leak on bo reservation failure. v2 git bisect good aeb0c2d8d3fc60ea931f807275166ea8abe1feeb # bad: [9b17c2afd74b5f56497ffacbcac85ee2cf0a1c1b] drm/radeon: improve PLL params if we don't match exactly v2 git bisect bad 9b17c2afd74b5f56497ffacbcac85ee2cf0a1c1b # first bad commit: [9b17c2afd74b5f56497ffacbcac85ee2cf0a1c1b] drm/radeon: improve PLL params if we don't match exactly v2 (the commit ID are local, as I've rebased drm-next and drm-fixes on v3.14). I attach two kernel logs booting and doing: echo 0xe >/sys/module/drm/parameters/debug xrand --rate 50 xrand --rate 60 I think the relevant difference is (first line is good, next is buggy): - [drm:radeon_compute_pll_avivo] 122000 - 13725, pll dividers - fb: 122.0 ref: 6, post 4 + [drm:radeon_compute_pll_avivo] 122000 - 12198, pll dividers - fb: 126.5 ref: 7, post 4 -- You are receiving this mail because: You are the assignee for the bug. --1397921902.335c1B0.15207 Date: Sat, 19 Apr 2014 15:38:22 +0000 MIME-Version: 1.0 Content-Type: text/html; charset="UTF-8"
Priority medium
Bug ID 77673
Assignee dri-devel@lists.freedesktop.org
Summary [bisected] moving moire starting with "drm/radeon: improve PLL params if we don't match exactly v2"
Severity normal
Classification Unclassified
OS Linux (All)
Reporter jsagarribay@gmail.com
Hardware x86-64 (AMD64)
Status NEW
Version DRI CVS
Component DRM/Radeon
Product DRI

Created attachment 97606 [details]
Output of 'xrandr --prop'

Hello, I'm testing drm-next and have found this problem. The display shows some
kind of moving wavy pattern, noticeable in midtones and grandients. I've
bisected it to the mentioned commit. The display is the internal LVDS panel, I
haven't tested external displays.


$ git bisect log
git bisect start
# good: [aa3a38dd40b610de206fb04777cc5469ce057de8] drm/radeon: update CI DPM
powertune settings
git bisect good aa3a38dd40b610de206fb04777cc5469ce057de8
# bad: [bc00aa0ebfed7edc1ca91ec2d9b0d5b04c5e06a8] drm: bochs: drop unused
struct fields
git bisect bad bc00aa0ebfed7edc1ca91ec2d9b0d5b04c5e06a8
# bad: [8bf5dda3777f2a477e3c448b38e18f5e1b502eae] drm/radeon/si: make sure mc
ucode is loaded before checking the size
git bisect bad 8bf5dda3777f2a477e3c448b38e18f5e1b502eae
# good: [8a6a70662fc93c2c43ba48de3e46bf7e860e18cc] drm/radeon: re-enable mclk
dpm on R7 260X asics
git bisect good 8a6a70662fc93c2c43ba48de3e46bf7e860e18cc
# good: [aeb0c2d8d3fc60ea931f807275166ea8abe1feeb] drm/radeon: memory leak on
bo reservation failure. v2
git bisect good aeb0c2d8d3fc60ea931f807275166ea8abe1feeb
# bad: [9b17c2afd74b5f56497ffacbcac85ee2cf0a1c1b] drm/radeon: improve PLL
params if we don't match exactly v2
git bisect bad 9b17c2afd74b5f56497ffacbcac85ee2cf0a1c1b
# first bad commit: [9b17c2afd74b5f56497ffacbcac85ee2cf0a1c1b] drm/radeon:
improve PLL params if we don't match exactly v2

(the commit ID are local, as I've rebased drm-next and drm-fixes on v3.14).

I attach two kernel logs booting and doing:
echo 0xe >/sys/module/drm/parameters/debug
xrand --rate 50
xrand --rate 60

I think the relevant difference is (first line is good, next is buggy):
- [drm:radeon_compute_pll_avivo] 122000 - 13725, pll dividers - fb: 122.0 ref:
6, post 4
+ [drm:radeon_compute_pll_avivo] 122000 - 12198, pll dividers - fb: 126.5 ref:
7, post 4


You are receiving this mail because:
  • You are the assignee for the bug.
--1397921902.335c1B0.15207-- --===============1886320494== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ dri-devel mailing list dri-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/dri-devel --===============1886320494==--