From: Jani Nikula <jani.nikula@linux.intel.com>
To: "Jouni Högander" <jouni.hogander@intel.com>,
intel-gfx@lists.freedesktop.org, intel-xe@lists.freedesktop.org
Cc: "Jouni Högander" <jouni.hogander@intel.com>
Subject: Re: [PATCH 5/6] drm/i915/display: Use display parent interface for i915 runtime pm
Date: Wed, 22 Oct 2025 12:35:31 +0300 [thread overview]
Message-ID: <1a12b894e6e003710bac9a69b4a3f8c3c25c9540@intel.com> (raw)
In-Reply-To: <20251022085548.876150-6-jouni.hogander@intel.com>
On Wed, 22 Oct 2025, Jouni Högander <jouni.hogander@intel.com> wrote:
> Start using display parent interface for i915 runtime pm. Doing the same
> for xe is done in coming changes.
>
> Signed-off-by: Jouni Högander <jouni.hogander@intel.com>
> ---
> .../gpu/drm/i915/display/intel_display_rpm.c | 43 +++++++++++--------
> 1 file changed, 24 insertions(+), 19 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/display/intel_display_rpm.c b/drivers/gpu/drm/i915/display/intel_display_rpm.c
> index 56c4024201c16..622646814e0bf 100644
> --- a/drivers/gpu/drm/i915/display/intel_display_rpm.c
> +++ b/drivers/gpu/drm/i915/display/intel_display_rpm.c
> @@ -4,66 +4,71 @@
> #include "i915_drv.h"
> #include "intel_display_core.h"
> #include "intel_display_rpm.h"
> -#include "intel_runtime_pm.h"
> -
> -static struct intel_runtime_pm *display_to_rpm(struct intel_display *display)
> -{
> - struct drm_i915_private *i915 = to_i915(display->drm);
> -
> - return &i915->runtime_pm;
> -}
> +#include "drm/intel/display_parent_interface.h"
This should use <> for include, as it's in include/drm, and placed
appropriately in the include list.
>
> struct ref_tracker *intel_display_rpm_get_raw(struct intel_display *display)
> {
> - return intel_runtime_pm_get_raw(display_to_rpm(display));
> + return display->parent->rpm && display->parent->rpm->get_raw ?
> + display->parent->rpm->get_raw(display->drm) : NULL;
I think we should require and assume ->rpm is set, and also all the
function pointers are set. It should be a rare documented exception if
something can be NULL.
I'm a bit divided about keeping the wrappers for calling the
interface. If we assume the pointers are valid, we could switch to
calling e.g.
- wakeref = intel_display_rpm_get_raw();
+ wakeref = display->parent->rpm->get_raw(display->drm);
but since that's a bunch of churn, it's okay to leave that be for
now. We can decide on it later. But let's not have all those NULL
checks.
BR,
Jani.
> }
>
> void intel_display_rpm_put_raw(struct intel_display *display, struct ref_tracker *wakeref)
> {
> - intel_runtime_pm_put_raw(display_to_rpm(display), wakeref);
> + if (display->parent->rpm && display->parent->rpm->put_raw)
> + display->parent->rpm->put_raw(display->drm, wakeref);
> }
>
> struct ref_tracker *intel_display_rpm_get(struct intel_display *display)
> {
> - return intel_runtime_pm_get(display_to_rpm(display));
> + return display->parent->rpm && display->parent->rpm->get ?
> + display->parent->rpm->get(display->drm) : NULL;
> +
> }
>
> struct ref_tracker *intel_display_rpm_get_if_in_use(struct intel_display *display)
> {
> - return intel_runtime_pm_get_if_in_use(display_to_rpm(display));
> + return display->parent->rpm && display->parent->rpm->get_if_in_use ?
> + display->parent->rpm->get_if_in_use(display->drm) : NULL;
> }
>
> struct ref_tracker *intel_display_rpm_get_noresume(struct intel_display *display)
> {
> - return intel_runtime_pm_get_noresume(display_to_rpm(display));
> + return display->parent->rpm && display->parent->rpm->get_noresume ?
> + display->parent->rpm->get_noresume(display->drm) : NULL;
> }
>
> void intel_display_rpm_put(struct intel_display *display, struct ref_tracker *wakeref)
> {
> - intel_runtime_pm_put(display_to_rpm(display), wakeref);
> + if (display->parent->rpm && display->parent->rpm->put)
> + display->parent->rpm->put(display->drm, wakeref);
> }
>
> void intel_display_rpm_put_unchecked(struct intel_display *display)
> {
> - intel_runtime_pm_put_unchecked(display_to_rpm(display));
> + if (display->parent->rpm && display->parent->rpm->put_unchecked)
> + display->parent->rpm->put_unchecked(display->drm);
> }
>
> bool intel_display_rpm_suspended(struct intel_display *display)
> {
> - return intel_runtime_pm_suspended(display_to_rpm(display));
> + return display->parent->rpm && display->parent->rpm->suspended ?
> + display->parent->rpm->suspended(display->drm) : false;
> }
>
> void assert_display_rpm_held(struct intel_display *display)
> {
> - assert_rpm_wakelock_held(display_to_rpm(display));
> + if (display->parent->rpm && display->parent->rpm->assert_held)
> + display->parent->rpm->assert_held(display->drm);
> }
>
> void intel_display_rpm_assert_block(struct intel_display *display)
> {
> - disable_rpm_wakeref_asserts(display_to_rpm(display));
> + if (display->parent->rpm && display->parent->rpm->assert_block)
> + display->parent->rpm->assert_block(display->drm);
> }
>
> void intel_display_rpm_assert_unblock(struct intel_display *display)
> {
> - enable_rpm_wakeref_asserts(display_to_rpm(display));
> + if (display->parent->rpm && display->parent->rpm->assert_block)
> + display->parent->rpm->assert_unblock(display->drm);
> }
--
Jani Nikula, Intel
next prev parent reply other threads:[~2025-10-22 9:35 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-10-22 8:55 [PATCH 0/6] Use display parent interface for runtime pm Jouni Högander
2025-10-22 8:55 ` [PATCH 1/6] drm/{i915, xe}/display: pass parent interface to display probe Jouni Högander
2025-10-22 9:17 ` [PATCH 1/6] drm/{i915,xe}/display: " Jani Nikula
2025-10-23 7:18 ` Jani Nikula
2025-10-22 8:55 ` [PATCH 2/6] drm/{i915, xe}/display: Add display runtime pm parent interface Jouni Högander
2025-10-22 9:22 ` Jani Nikula
2025-10-22 8:55 ` [PATCH 3/6] drm/i915/display: Runtime pm wrappers for display " Jouni Högander
2025-10-22 9:27 ` Jani Nikula
2025-10-22 8:55 ` [PATCH 4/6] drm/xe/display: " Jouni Högander
2025-10-22 9:29 ` Jani Nikula
2025-10-22 8:55 ` [PATCH 5/6] drm/i915/display: Use display parent interface for i915 runtime pm Jouni Högander
2025-10-22 9:35 ` Jani Nikula [this message]
2025-10-22 8:55 ` [PATCH 6/6] drm/xe/display: Use display parent interface for xe " Jouni Högander
2025-10-22 9:36 ` Jani Nikula
2025-10-22 9:41 ` ✗ CI.checkpatch: warning for Use display parent interface for " Patchwork
2025-10-22 9:42 ` ✓ CI.KUnit: success " Patchwork
2025-10-22 10:00 ` ✗ CI.checksparse: warning " Patchwork
2025-10-22 11:02 ` ✓ Xe.CI.BAT: success " Patchwork
2025-10-22 13:13 ` ✗ Xe.CI.Full: failure " Patchwork
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=1a12b894e6e003710bac9a69b4a3f8c3c25c9540@intel.com \
--to=jani.nikula@linux.intel.com \
--cc=intel-gfx@lists.freedesktop.org \
--cc=intel-xe@lists.freedesktop.org \
--cc=jouni.hogander@intel.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