From: Maarten Lankhorst <dev@lankhorst.se>
To: intel-gfx@lists.freedesktop.org, intel-xe@lists.freedesktop.org
Cc: linux-rt-devel@lists.linux.dev,
Maarten Lankhorst <dev@lankhorst.se>,
Mario Kleiner <mario.kleiner.de@gmail.com>,
Mike Galbraith <umgwanakikbuti@gmail.com>,
Thomas Gleixner <tglx@linutronix.de>,
Sebastian Andrzej Siewior <bigeasy@linutronix.de>,
Clark Williams <clrkwllms@kernel.org>,
Steven Rostedt <rostedt@goodmis.org>
Subject: [PATCH v2 0/7] drm/i915/display: Handle vblank evasion with CONFIG_PREEMPT_RT
Date: Tue, 4 Nov 2025 09:36:24 +0100 [thread overview]
Message-ID: <20251104083634.670753-1-dev@lankhorst.se> (raw)
There is a critical section between intel_pipe_update_start() and
intel_pipe_update_end() where we only program hardware registers,
should not take any lock and complete as fast as possible.
The previous approach used to remove the local_irq_enable/disable()
in this critical, but that increases the probability that the time
sensitive section does not complete in 100 µs, potentially causing
the hardware to hang.
I went through all the lockdep splats that occurred in CI, and fixed
them 1 by 1 until there were none left. This additionally improves
latency by not removing any locks in the fastpath.
In intel_de.h the implicit DMC wakelock was added, ideally we can get rid
of it, but for now we can simply use the same _fw variants as are
needed on i915.
I believe this series is enough to make xe and perhaps good enough to make
i915's display RT safe.
Changes since v1:
- Fix compilation on i915.
- Submit to intel-gfx ml too.s
Cc: Mario Kleiner <mario.kleiner.de@gmail.com>
Cc: Mike Galbraith <umgwanakikbuti@gmail.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Cc: Clark Williams <clrkwllms@kernel.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Maarten Lankhorst (6):
drm/i915/display: Make get_vblank_counter use intel_de_read_fw()
drm/i915/display: Use intel_de_write_fw in intel_pipe_fastset
drm/i915/display: Move vblank put until after critical section
drm/i915/display: Remove locking from intel_vblank_evade critical
section
drm/i915/display: Make icl_dsi_frame_update use _fw too
drm/i915/display: Enable interrupts earlier on PREEMPT_RT
Mike Galbraith (1):
drm/i915: Use preempt_disable/enable_rt() where recommended
drivers/gpu/drm/i915/display/icl_dsi.c | 4 +-
drivers/gpu/drm/i915/display/intel_crtc.c | 10 +++
drivers/gpu/drm/i915/display/intel_cursor.c | 8 +-
drivers/gpu/drm/i915/display/intel_de.h | 6 ++
drivers/gpu/drm/i915/display/intel_display.c | 36 ++++-----
drivers/gpu/drm/i915/display/intel_vblank.c | 80 ++++++++++++-------
drivers/gpu/drm/i915/display/intel_vrr.c | 16 ++--
.../drm/xe/compat-i915-headers/intel_uncore.h | 2 +
8 files changed, 103 insertions(+), 59 deletions(-)
--
2.51.0
next reply other threads:[~2025-11-04 8:45 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-11-04 8:36 Maarten Lankhorst [this message]
2025-11-04 8:36 ` [PATCH v2 1/7] drm/i915/display: Make get_vblank_counter use intel_de_read_fw() Maarten Lankhorst
2025-11-04 8:36 ` [PATCH v2 2/7] drm/i915/display: Use intel_de_write_fw in intel_pipe_fastset Maarten Lankhorst
2025-11-26 19:19 ` Shankar, Uma
2025-11-26 19:42 ` Ville Syrjälä
2025-11-26 19:56 ` Shankar, Uma
2025-12-01 17:43 ` Maarten Lankhorst
2025-12-11 14:35 ` Maarten Lankhorst
2025-11-04 8:36 ` [PATCH v2 3/7] drm/i915/display: Move vblank put until after critical section Maarten Lankhorst
2025-11-26 19:25 ` Shankar, Uma
2025-11-04 8:36 ` [PATCH v2 4/7] drm/i915/display: Remove locking from intel_vblank_evade " Maarten Lankhorst
2025-11-26 20:03 ` Shankar, Uma
2025-12-01 14:13 ` Maarten Lankhorst
2025-11-26 21:17 ` Ville Syrjälä
2025-12-11 9:59 ` Maarten Lankhorst
2025-11-04 8:36 ` [PATCH v2 5/7] drm/i915/display: Make icl_dsi_frame_update use _fw too Maarten Lankhorst
2025-11-04 8:36 ` [PATCH v2 6/7] drm/i915/display: Enable interrupts earlier on PREEMPT_RT Maarten Lankhorst
2025-11-26 20:45 ` Shankar, Uma
2025-11-26 22:57 ` Ville Syrjälä
2025-12-01 17:39 ` Maarten Lankhorst
2025-11-04 8:36 ` [PATCH v2 7/7] drm/i915: Use preempt_disable/enable_rt() where recommended Maarten Lankhorst
2025-11-05 13:47 ` [PATCH v2 0/7] drm/i915/display: Handle vblank evasion with CONFIG_PREEMPT_RT Sebastian Andrzej Siewior
2025-11-05 20:42 ` Maarten Lankhorst
2025-11-10 16:09 ` Sebastian Andrzej Siewior
2025-11-10 18:17 ` Maarten Lankhorst
2025-11-11 9:47 ` Sebastian Andrzej Siewior
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=20251104083634.670753-1-dev@lankhorst.se \
--to=dev@lankhorst.se \
--cc=bigeasy@linutronix.de \
--cc=clrkwllms@kernel.org \
--cc=intel-gfx@lists.freedesktop.org \
--cc=intel-xe@lists.freedesktop.org \
--cc=linux-rt-devel@lists.linux.dev \
--cc=mario.kleiner.de@gmail.com \
--cc=rostedt@goodmis.org \
--cc=tglx@linutronix.de \
--cc=umgwanakikbuti@gmail.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;
as well as URLs for NNTP newsgroup(s).