Intel-GFX Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Rodrigo Vivi <rodrigo.vivi@intel.com>
To: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: <intel-gfx@lists.freedesktop.org>,
	<intel-xe@lists.freedesktop.org>,
	"Jani Nikula" <jani.nikula@intel.com>
Subject: Re: [RFC] drm/xe/display: Merge xe's display info probe and i915 HAS_DISPLAY checks
Date: Thu, 5 Sep 2024 09:44:43 -0400	[thread overview]
Message-ID: <Ztm1yy9JxTUgQMyd@intel.com> (raw)
In-Reply-To: <ituy73yzuwpjxjdwm53a6etoanj3y7brwk5ag57bhw4hlhf6xs@hs6iaclhkvt7>

On Thu, Sep 05, 2024 at 08:29:22AM -0500, Lucas De Marchi wrote:
> On Wed, Sep 04, 2024 at 01:37:13PM GMT, Rodrigo Vivi wrote:
> > Instead of having multiple checks and the has_display in the middle
> > of the functions, only execute the Xe display functions if
> > Xe probed display and pipe_masks still have something valid
> > after i915's runtime init.
> > 
> > Cc: Jani Nikula <jani.nikula@intel.com>
> > Cc: Lucas De Marchi <lucas.demarchi@intel.com>
> > Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
> > ---
> > drivers/gpu/drm/xe/display/xe_display.c | 65 +++++++++++++------------
> > 1 file changed, 35 insertions(+), 30 deletions(-)
> > 
> > diff --git a/drivers/gpu/drm/xe/display/xe_display.c b/drivers/gpu/drm/xe/display/xe_display.c
> > index 75736faf2a80..1e248c7aaff0 100644
> > --- a/drivers/gpu/drm/xe/display/xe_display.c
> > +++ b/drivers/gpu/drm/xe/display/xe_display.c
> > @@ -13,7 +13,6 @@
> > #include <uapi/drm/xe_drm.h>
> > 
> > #include "soc/intel_dram.h"
> > -#include "i915_drv.h"		/* FIXME: HAS_DISPLAY() depends on this */
> > #include "intel_acpi.h"
> > #include "intel_audio.h"
> > #include "intel_bw.h"
> > @@ -34,7 +33,12 @@
> > 
> > static bool has_display(struct xe_device *xe)
> 
> I think has_display is already conflated enough. From the xe side I
> wouldn't call this has_display if it means something else than "the
> hardware is present or we know how to drive it". That is the meaning of
> this flag in drivers/gpu/drm/xe/xe_pci.c and this function here later
> changed to mean something else :-/
> 
> > {
> > -	return HAS_DISPLAY(xe);
> > +	/*
> > +	 * Xe has probed and configured the display,
> > +	 * and some pipes remains enable after
> > +	 * __intel_display_device_info_runtime_init()
> > +	 */
> > +	return xe->info.probe_display && HAS_DISPLAY(&xe->display);
> 
> I'm not following the motivation here... Shouldn't we remove the
> HAS_DISPLAY() from here and rather change the display side to do a
> return-early?
> 
> if probe_display == 1, from the xe perspective we probed display, we
> shouldn't be looking at the internal state of display to know what to do
> on this side of the fence.

Right, but so far we have other if (has_display()) in the middle of our code.
Like Jani reminded me yesterday, we still need to check the state of
pipe_mask after runtime init function before many calls.

So, the quick goal here was to just consolidate the checks instead of
the current sprinkled 'if (has_display()) calls that we have around.

So we can proceed with a further clean-up and split on the display
calls and then move them out to intel_display and minimize this
file as much as we can...

https://lore.kernel.org/intel-xe/Zth56C3s8lPQMEBB@intel.com

But another question that I'm making myself since yesterday is:
Shouldn't we simply kill the Xe's display modprobe option at this point?

> 
> Lucas De Marchi
> 
> > }
> > 
> > /**
> > @@ -104,7 +108,7 @@ static void xe_display_fini_nommio(struct drm_device *dev, void *dummy)
> > {
> > 	struct xe_device *xe = to_xe_device(dev);
> > 
> > -	if (!xe->info.probe_display)
> > +	if (!has_display(xe))
> > 		return;
> > 
> > 	intel_power_domains_cleanup(xe);
> > @@ -112,7 +116,7 @@ static void xe_display_fini_nommio(struct drm_device *dev, void *dummy)
> > 
> > int xe_display_init_nommio(struct xe_device *xe)
> > {
> > -	if (!xe->info.probe_display)
> > +	if (!has_display(xe))
> > 		return 0;
> > 
> > 	/* Fake uncore lock */
> > @@ -129,7 +133,7 @@ static void xe_display_fini_noirq(void *arg)
> > 	struct xe_device *xe = arg;
> > 	struct intel_display *display = &xe->display;
> > 
> > -	if (!xe->info.probe_display)
> > +	if (!has_display(xe))
> > 		return;
> > 
> > 	intel_display_driver_remove_noirq(xe);
> > @@ -141,7 +145,7 @@ int xe_display_init_noirq(struct xe_device *xe)
> > 	struct intel_display *display = &xe->display;
> > 	int err;
> > 
> > -	if (!xe->info.probe_display)
> > +	if (!has_display(xe))
> > 		return 0;
> > 
> > 	intel_display_driver_early_probe(xe);
> > @@ -172,7 +176,7 @@ static void xe_display_fini_noaccel(void *arg)
> > {
> > 	struct xe_device *xe = arg;
> > 
> > -	if (!xe->info.probe_display)
> > +	if (!has_display(xe))
> > 		return;
> > 
> > 	intel_display_driver_remove_nogem(xe);
> > @@ -182,7 +186,7 @@ int xe_display_init_noaccel(struct xe_device *xe)
> > {
> > 	int err;
> > 
> > -	if (!xe->info.probe_display)
> > +	if (!has_display(xe))
> > 		return 0;
> > 
> > 	err = intel_display_driver_probe_nogem(xe);
> > @@ -194,7 +198,7 @@ int xe_display_init_noaccel(struct xe_device *xe)
> > 
> > int xe_display_init(struct xe_device *xe)
> > {
> > -	if (!xe->info.probe_display)
> > +	if (!has_display(xe))
> > 		return 0;
> > 
> > 	return intel_display_driver_probe(xe);
> > @@ -202,7 +206,7 @@ int xe_display_init(struct xe_device *xe)
> > 
> > void xe_display_fini(struct xe_device *xe)
> > {
> > -	if (!xe->info.probe_display)
> > +	if (!has_display(xe))
> > 		return;
> > 
> > 	intel_hpd_poll_fini(xe);
> > @@ -213,7 +217,7 @@ void xe_display_fini(struct xe_device *xe)
> > 
> > void xe_display_register(struct xe_device *xe)
> > {
> > -	if (!xe->info.probe_display)
> > +	if (!has_display(xe))
> > 		return;
> > 
> > 	intel_display_driver_register(xe);
> > @@ -223,7 +227,7 @@ void xe_display_register(struct xe_device *xe)
> > 
> > void xe_display_unregister(struct xe_device *xe)
> > {
> > -	if (!xe->info.probe_display)
> > +	if (!has_display(xe))
> > 		return;
> > 
> > 	intel_unregister_dsm_handler();
> > @@ -233,7 +237,7 @@ void xe_display_unregister(struct xe_device *xe)
> > 
> > void xe_display_driver_remove(struct xe_device *xe)
> > {
> > -	if (!xe->info.probe_display)
> > +	if (!has_display(xe))
> > 		return;
> > 
> > 	intel_display_driver_remove(xe);
> > @@ -243,7 +247,7 @@ void xe_display_driver_remove(struct xe_device *xe)
> > 
> > void xe_display_irq_handler(struct xe_device *xe, u32 master_ctl)
> > {
> > -	if (!xe->info.probe_display)
> > +	if (!has_display(xe))
> > 		return;
> > 
> > 	if (master_ctl & DISPLAY_IRQ)
> > @@ -254,7 +258,7 @@ void xe_display_irq_enable(struct xe_device *xe, u32 gu_misc_iir)
> > {
> > 	struct intel_display *display = &xe->display;
> > 
> > -	if (!xe->info.probe_display)
> > +	if (!has_display(xe))
> > 		return;
> > 
> > 	if (gu_misc_iir & GU_MISC_GSE)
> > @@ -263,7 +267,7 @@ void xe_display_irq_enable(struct xe_device *xe, u32 gu_misc_iir)
> > 
> > void xe_display_irq_reset(struct xe_device *xe)
> > {
> > -	if (!xe->info.probe_display)
> > +	if (!has_display(xe))
> > 		return;
> > 
> > 	gen11_display_irq_reset(xe);
> > @@ -271,7 +275,7 @@ void xe_display_irq_reset(struct xe_device *xe)
> > 
> > void xe_display_irq_postinstall(struct xe_device *xe, struct xe_gt *gt)
> > {
> > -	if (!xe->info.probe_display)
> > +	if (!has_display(xe))
> > 		return;
> > 
> > 	if (gt->info.id == XE_GT0)
> > @@ -311,7 +315,7 @@ static void xe_display_flush_cleanup_work(struct xe_device *xe)
> > /* TODO: System and runtime suspend/resume sequences will be sanitized as a follow-up. */
> > void xe_display_pm_runtime_suspend(struct xe_device *xe)
> > {
> > -	if (!xe->info.probe_display)
> > +	if (!has_display(xe))
> > 		return;
> > 
> > 	if (xe->d3cold.allowed)
> > @@ -324,7 +328,8 @@ void xe_display_pm_suspend(struct xe_device *xe, bool runtime)
> > {
> > 	struct intel_display *display = &xe->display;
> > 	bool s2idle = suspend_to_idle();
> > -	if (!xe->info.probe_display)
> > +
> > +	if (!has_display(xe))
> > 		return;
> > 
> > 	/*
> > @@ -333,7 +338,7 @@ void xe_display_pm_suspend(struct xe_device *xe, bool runtime)
> > 	 */
> > 	intel_power_domains_disable(xe);
> > 	intel_fbdev_set_suspend(&xe->drm, FBINFO_STATE_SUSPENDED, true);
> > -	if (!runtime && has_display(xe)) {
> > +	if (!runtime) {
> > 		drm_kms_helper_poll_disable(&xe->drm);
> > 		intel_display_driver_disable_user_access(xe);
> > 		intel_display_driver_suspend(xe);
> > @@ -345,7 +350,7 @@ void xe_display_pm_suspend(struct xe_device *xe, bool runtime)
> > 
> > 	intel_hpd_cancel_work(xe);
> > 
> > -	if (!runtime && has_display(xe)) {
> > +	if (!runtime) {
> > 		intel_display_driver_suspend_access(xe);
> > 		intel_encoder_suspend_all(&xe->display);
> > 	}
> > @@ -358,7 +363,8 @@ void xe_display_pm_suspend(struct xe_device *xe, bool runtime)
> > void xe_display_pm_suspend_late(struct xe_device *xe)
> > {
> > 	bool s2idle = suspend_to_idle();
> > -	if (!xe->info.probe_display)
> > +
> > +	if (!has_display(xe))
> > 		return;
> > 
> > 	intel_power_domains_suspend(xe, s2idle);
> > @@ -368,7 +374,7 @@ void xe_display_pm_suspend_late(struct xe_device *xe)
> > 
> > void xe_display_pm_runtime_resume(struct xe_device *xe)
> > {
> > -	if (!xe->info.probe_display)
> > +	if (!has_display(xe))
> > 		return;
> > 
> > 	intel_hpd_poll_disable(xe);
> > @@ -379,7 +385,7 @@ void xe_display_pm_runtime_resume(struct xe_device *xe)
> > 
> > void xe_display_pm_resume_early(struct xe_device *xe)
> > {
> > -	if (!xe->info.probe_display)
> > +	if (!has_display(xe))
> > 		return;
> > 
> > 	intel_display_power_resume_early(xe);
> > @@ -391,23 +397,22 @@ void xe_display_pm_resume(struct xe_device *xe, bool runtime)
> > {
> > 	struct intel_display *display = &xe->display;
> > 
> > -	if (!xe->info.probe_display)
> > +	if (!has_display(xe))
> > 		return;
> > 
> > 	intel_dmc_resume(xe);
> > 
> > -	if (has_display(xe))
> > -		drm_mode_config_reset(&xe->drm);
> > +	drm_mode_config_reset(&xe->drm);
> > 
> > 	intel_display_driver_init_hw(xe);
> > 	intel_hpd_init(xe);
> > 
> > -	if (!runtime && has_display(xe))
> > +	if (!runtime)
> > 		intel_display_driver_resume_access(xe);
> > 
> > 	/* MST sideband requires HPD interrupts enabled */
> > 	intel_dp_mst_resume(xe);
> > -	if (!runtime && has_display(xe)) {
> > +	if (!runtime) {
> > 		intel_display_driver_resume(xe);
> > 		drm_kms_helper_poll_enable(&xe->drm);
> > 		intel_display_driver_enable_user_access(xe);
> > @@ -441,7 +446,7 @@ int xe_display_probe(struct xe_device *xe)
> > 	if (err)
> > 		return err;
> > 
> > -	if (has_display(xe))
> > +	if (HAS_DISPLAY(&xe->display))
> > 		return 0;
> > 
> > no_display:
> > -- 
> > 2.46.0
> > 

  reply	other threads:[~2024-09-05 13:44 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-09-04 17:37 [RFC] drm/xe/display: Merge xe's display info probe and i915 HAS_DISPLAY checks Rodrigo Vivi
2024-09-04 18:29 ` ✓ Fi.CI.BAT: success for " Patchwork
2024-09-04 19:34 ` [RFC] " Cavitt, Jonathan
2024-09-05 13:29 ` Lucas De Marchi
2024-09-05 13:44   ` Rodrigo Vivi [this message]
2024-09-05 18:16     ` Jani Nikula
2024-09-05 15:38 ` ✗ Fi.CI.IGT: failure for " 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=Ztm1yy9JxTUgQMyd@intel.com \
    --to=rodrigo.vivi@intel.com \
    --cc=intel-gfx@lists.freedesktop.org \
    --cc=intel-xe@lists.freedesktop.org \
    --cc=jani.nikula@intel.com \
    --cc=lucas.demarchi@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