AMD-GFX Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Leo Li <sunpeng.li@amd.com>
To: Shengyu Qu <wiagn233@outlook.com>, Michele Palazzi <sysdadmin@m1k.cloud>
Cc: amd-gfx@lists.freedesktop.org, harry.wentland@amd.com,
	alexander.deucher@amd.com, christian.koenig@amd.com,
	siqueira@igalia.com, "Michel Dänzer" <michel.daenzer@mailbox.org>
Subject: Re: [PATCH 1/1] drm/amd/display: complete cursor vblank events immediately
Date: Wed, 22 Apr 2026 10:31:03 -0400	[thread overview]
Message-ID: <426dc28a-5351-4a15-a80c-ed6dea9c910e@amd.com> (raw)
In-Reply-To: <TY4PR01MB144321DF5415AC90AC7C4E34898212@TY4PR01MB14432.jpnprd01.prod.outlook.com>


On 2026-04-18 15:37, Shengyu Qu wrote:
> Hi Leo,
> 
> Could you merge this patch first and do more discovery into this bug
> later? Since this solve some problems for users.
> 
> Best regards,
> Shengyu

Hi Shengyu, Michele

I vaguely recall that restoring the old 5s vblank off delay helped alleviate this by allowing idle less frequently.
That may be a better bandaid for now.

Let me know if the below patch helps, we'll merge it if it does.

Thanks,
Leo
---
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index 3fa4dbda4517c..ce5063928413c 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -9511,9 +9511,21 @@ static void manage_dm_interrupts(struct amdgpu_device *adev,
 	if (acrtc_state) {
 		timing = &acrtc_state->stream->timing;
 
-		if (amdgpu_ip_version(adev, DCE_HWIP, 0) <
-			   IP_VERSION(3, 5, 0) ||
-			   !(adev->flags & AMD_IS_APU)) {
+		if (amdgpu_ip_version(adev, DCE_HWIP, 0) >=
+		      IP_VERSION(3, 2, 0) &&
+		      !(adev->flags & AMD_IS_APU)) {
+			/*
+			 * DGPUs NV3x and newer that support idle optimizations
+			 * experience intermittent flip-done timeouts on cursor
+			 * updates. Restore 5s offdelay behavior for now.
+			 *
+			 * Discussion on the issue:
+			 * https://lore.kernel.org/amd-gfx/20260217191632.1243826-1-sysdadmin@m1k.cloud/
+			 */
+			config.offdelay_ms = 5000;
+			config.disable_immediate = false;
+		} else if (amdgpu_ip_version(adev, DCE_HWIP, 0) <
+			     IP_VERSION(3, 5, 0)) {
 			/*
 			 * Older HW and DGPU have issues with instant off;
 			 * use a 2 frame offdelay.
-- 
2.53.0


  reply	other threads:[~2026-04-22 14:31 UTC|newest]

Thread overview: 44+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-02-17 19:16 [PATCH 1/1] drm/amd/display: complete cursor vblank events immediately Michele Palazzi
2026-02-23 15:27 ` Leo Li
2026-02-27  8:53   ` Michele Palazzi
2026-02-27  8:58     ` Michele Palazzi
2026-03-02 22:13       ` Leo Li
2026-03-03  8:17         ` Shengyu Qu
2026-03-03 19:07           ` Leo Li
2026-03-04 14:00             ` Michele Palazzi
2026-03-04 14:20               ` Leo Li
2026-03-05 22:30                 ` Leo Li
2026-03-06  8:37                   ` Michele Palazzi
2026-03-09 16:49                     ` Michele Palazzi
2026-03-10 23:50                       ` Leo Li
2026-03-11 10:16                         ` Shengyu Qu
2026-03-11 10:38                         ` Michele Palazzi
2026-03-11 17:56                           ` Leo Li
2026-03-16 14:55                             ` Michele Palazzi
2026-03-16 15:17                               ` Michele Palazzi
2026-03-16 18:39                                 ` Leo Li
2026-03-16 18:48                                   ` Leo Li
2026-03-18 11:36                                     ` Michele Palazzi
2026-03-20  0:52                                       ` Leo Li
2026-03-20  1:33                                         ` Michele Palazzi
2026-03-31 12:57                                           ` Michele Palazzi
2026-05-04 18:48                                             ` Leo Li
2026-05-05  7:27                                               ` Michele Palazzi
2026-05-07 20:19                                                 ` Leo Li
2026-05-08  7:37                                                   ` Michele Palazzi
2026-05-05  7:31                                               ` Shengyu Qu
2026-04-18 19:37                                         ` Shengyu Qu
2026-04-22 14:31                                           ` Leo Li [this message]
2026-04-22 14:57                                             ` Michele Palazzi
2026-02-27 19:43   ` Alex Deucher
2026-03-02  8:53   ` Michel Dänzer
2026-03-02 22:14     ` Leo Li
  -- strict thread matches above, loose matches on Subject: below --
2026-02-18  0:31 Michele Palazzi
2026-02-18  9:41 ` Michel Dänzer
2026-02-18 10:09   ` Michele Palazzi
2026-02-19 11:09     ` Michel Dänzer
2026-02-19 13:08       ` Michele Palazzi
2026-02-19 13:59         ` Michel Dänzer
2026-02-19 15:56           ` Michele Palazzi
2026-02-19 16:02             ` Michel Dänzer
2026-02-20 11:10               ` Michele Palazzi

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=426dc28a-5351-4a15-a80c-ed6dea9c910e@amd.com \
    --to=sunpeng.li@amd.com \
    --cc=alexander.deucher@amd.com \
    --cc=amd-gfx@lists.freedesktop.org \
    --cc=christian.koenig@amd.com \
    --cc=harry.wentland@amd.com \
    --cc=michel.daenzer@mailbox.org \
    --cc=siqueira@igalia.com \
    --cc=sysdadmin@m1k.cloud \
    --cc=wiagn233@outlook.com \
    /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