From: "Timur Kristóf" <timur.kristof@gmail.com>
To: amd-gfx@lists.freedesktop.org,
Alex Deucher <alexander.deucher@amd.com>,
christian.koenig@amd.com, Tvrtko Ursulin <tursulin@ursulin.net>,
pierre-eric.pelloux-prayer@amd.com,
Natalie Vock <natalie.vock@gmx.de>
Cc: "Timur Kristóf" <timur.kristof@gmail.com>
Subject: [PATCH 00/11] drm/amdgpu/gfx6: Use GFX IP block soft reset on GFX6
Date: Mon, 13 Jul 2026 15:06:58 +0200 [thread overview]
Message-ID: <20260713130709.34262-1-timur.kristof@gmail.com> (raw)
Fix up various things in the GFX6 code to prepare it
to work with IP block soft reset. The main challenge
was figuring out how to properly initialize the
compute queues so that they work well after a reset,
as well as realizing that on GFX6 the compute queues
also have a PFP (as opposed to GFX7+ where PFP is
a graphics only thing). There are also other minor
fixes.
Implement GFX6 IP block soft reset: just reset everything
in the GFX IP block (instead of taking a guess at which
parts need to be reset) to make it consistent, and
also handle clock and power gating to make sure not to
degrade GPU functionality after a GFX IP block soft reset.
This improves current user experience on all GFX6 chips:
On GFX6 dGPUs, the current GPU recovery method
always clears the contents of VRAM, which means that
a buggy (hanging) app can crash the whole graphical
session, which is less than ideal.
Using GFX IP block soft reset means that we can now
move on from GFX hangs on dGPUs without crashing the
whole system.
Tested on the following chips:
Tahiti (FirePro W9000, Radeon HD 7870 XT)
Cape Verde (Radeon R7 450)
Pitcairn (Radeon R9 270X)
Oland (Radeon 430)
Timur Kristóf (11):
drm/amdgpu/gfx6: Improve emit_cntxcntl()
drm/amdgpu/gfx6: Fixup emitting SWITCH_BUFFER packets
drm/amdgpu/gfx6: Use PFP on the compute queues too
drm/amdgpu/gfx6: Initialize compute rings before CP start
drm/amdgpu/gfx6: Clean up rings during reset
drm/amdgpu/gfx6: Execute CLEAR_STATE when initializing compute rings
drm/amdgpu/gfx6: Properly enable/disable priv_req and priv_inst
interrupts
drm/amdgpu/gfx6: Adjust how harvested TCCs are set up
drm/amdgpu/gfx6: Use COND_EXEC
drm/amdgpu/gfx6: Add IP block soft reset implementation
drm/amdgpu/gfx6: Enable IP block soft reset as a GPU recovery method
drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c | 403 +++++++++++++-----
.../include/asic_reg/gca/gfx_6_0_sh_mask.h | 32 +-
2 files changed, 316 insertions(+), 119 deletions(-)
--
2.55.0
next reply other threads:[~2026-07-13 13:07 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-07-13 13:06 Timur Kristóf [this message]
2026-07-13 13:06 ` [PATCH 01/11] drm/amdgpu/gfx6: Improve emit_cntxcntl() Timur Kristóf
2026-07-13 13:07 ` [PATCH 02/11] drm/amdgpu/gfx6: Fixup emitting SWITCH_BUFFER packets Timur Kristóf
2026-07-13 13:07 ` [PATCH 03/11] drm/amdgpu/gfx6: Use PFP on the compute queues too Timur Kristóf
2026-07-13 13:07 ` [PATCH 04/11] drm/amdgpu/gfx6: Initialize compute rings before CP start Timur Kristóf
2026-07-13 13:07 ` [PATCH 05/11] drm/amdgpu/gfx6: Clean up rings during reset Timur Kristóf
2026-07-13 13:07 ` [PATCH 06/11] drm/amdgpu/gfx6: Execute CLEAR_STATE when initializing compute rings Timur Kristóf
2026-07-13 13:07 ` [PATCH 07/11] drm/amdgpu/gfx6: Properly enable/disable priv_req and priv_inst interrupts Timur Kristóf
2026-07-15 10:19 ` Tvrtko Ursulin
2026-07-15 10:53 ` Timur Kristóf
2026-07-15 11:22 ` Tvrtko Ursulin
2026-07-15 19:23 ` Timur Kristóf
2026-07-16 11:05 ` Tvrtko Ursulin
2026-07-13 13:07 ` [PATCH 08/11] drm/amdgpu/gfx6: Adjust how harvested TCCs are set up Timur Kristóf
2026-07-13 13:07 ` [PATCH 09/11] drm/amdgpu/gfx6: Use COND_EXEC Timur Kristóf
2026-07-13 13:07 ` [PATCH 10/11] drm/amdgpu/gfx6: Add IP block soft reset implementation Timur Kristóf
2026-07-13 13:07 ` [PATCH 11/11] drm/amdgpu/gfx6: Enable IP block soft reset as a GPU recovery method Timur Kristóf
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=20260713130709.34262-1-timur.kristof@gmail.com \
--to=timur.kristof@gmail.com \
--cc=alexander.deucher@amd.com \
--cc=amd-gfx@lists.freedesktop.org \
--cc=christian.koenig@amd.com \
--cc=natalie.vock@gmx.de \
--cc=pierre-eric.pelloux-prayer@amd.com \
--cc=tursulin@ursulin.net \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.