Skip to contentSkip to conversation

drm/i915: Don't check power domains state in intel_power_domains_init_hw()

PATCH Superseded Imre Deak 3 participants 11 messages Aug 28, 2018 – Aug 29, 2018

Subsystems (2)
lore
Share & tools
lore

Imre Deak

author
#0 lore raw

PATCH 1/1 drm/i915: Don't check power domains state in intel_power_domains_init_hw()

During power domains initialization we acquire power well references for power wells in the INIT power domain. The rest of power wells - which BIOS could have left enabled - we can only acquire references as needed during display HW readout. Thus during initialization these latter power wells can have a refcount of 0 while still being enabled. To avoid the false-positive state mismatch error this causes remove the check from intel_power_domains_init_hw() and rely on the state check in intel_power_domains_enable() which follows the HW readout.

Fixes 6dfc4a8f134f ("drm/i915: Verify power domains after enabling them")
Cc Chris Wilson <[email protected]>

References: https://bugs.freedesktop.org/show_bug.cgi?id=107411

---
 drivers/gpu/drm/i915/intel_runtime_pm.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)
drivers/gpu/drm/i915/intel_runtime_pm.c +1 -2
index 2852395125cd..73d3a6ed3446 100644
@@ -3767,9 +3767,8 @@ void intel_power_domains_init_hw(struct drm_i915_private *dev_priv, bool resume)
 	if (!i915_modparams.disable_power_well)
 		intel_display_power_get(dev_priv, POWER_DOMAIN_INIT);
 	intel_power_domains_sync_hw(dev_priv);
+
 	power_domains->initializing = false;
-
-	intel_power_domains_verify_state(dev_priv);
 }
 
 /**
Signature · 6 lines
2.13.2

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
#1 lore raw

Quoting Imre Deak (2018-08-28 12:40:43)

Quoted 8 lines

During power domains initialization we acquire power well references for power wells in the INIT power domain. The rest of power wells - which BIOS could have left enabled - we can only acquire references as needed during display HW readout. Thus during initialization these latter power wells can have a refcount of 0 while still being enabled. To avoid the false-positive state mismatch error this causes remove the check from intel_power_domains_init_hw() and rely on the state check in intel_power_domains_enable() which follows the HW readout.

Missing from above is a quick explanation of how those extraneous powerwells are sanitizied. If we don't do the HW readout (i915.disable_display?) do we not then leave the powerwell active and so complain in a later verify_state()?
-Chris
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

author
#2 lore raw
Quoted 10 lines · Chris Wilson

Quoting Imre Deak (2018-08-28 12:40:43)

Quoted 8 lines

During power domains initialization we acquire power well references for power wells in the INIT power domain. The rest of power wells - which BIOS could have left enabled - we can only acquire references as needed during display HW readout. Thus during initialization these latter power wells can have a refcount of 0 while still being enabled. To avoid the false-positive state mismatch error this causes remove the check from intel_power_domains_init_hw() and rely on the state check in intel_power_domains_enable() which follows the HW readout.

Quoted 4 lines

Missing from above is a quick explanation of how those extraneous powerwells are sanitizied. If we don't do the HW readout (i915.disable_display?) do we not then leave the powerwell active and so complain in a later verify_state()?

These power wells (AUX and DDI on ICL) can only be enabled/disabled at a specific spot in the modeset sequence, otherwise the power well enable / disable operation will time out. That's the reason they're not part of the INIT domain. For these we will acquire the references during HW readout when noticing that the corresponding display HW block is active and drop them when disabling these HW blocks (normally or as part of display state sanitizatoionin). That way we'll ensure the proper spot mentioned above for power well enabling/disabling.

--Imre
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

author
#3 lore raw
Quoted 16 lines · Imre Deak
Quoted 10 lines · Chris Wilson

Quoting Imre Deak (2018-08-28 12:40:43)

Quoted 8 lines

During power domains initialization we acquire power well references for power wells in the INIT power domain. The rest of power wells - which BIOS could have left enabled - we can only acquire references as needed during display HW readout. Thus during initialization these latter power wells can have a refcount of 0 while still being enabled. To avoid the false-positive state mismatch error this causes remove the check from intel_power_domains_init_hw() and rely on the state check in intel_power_domains_enable() which follows the HW readout.

Quoted 4 lines

Missing from above is a quick explanation of how those extraneous powerwells are sanitizied. If we don't do the HW readout (i915.disable_display?) do we not then leave the powerwell active and so complain in a later verify_state()?

Quoted 8 lines

These power wells (AUX and DDI on ICL) can only be enabled/disabled at a specific spot in the modeset sequence, otherwise the power well enable / disable operation will time out. That's the reason they're not part of the INIT domain. For these we will acquire the references during HW readout when noticing that the corresponding display HW block is active and drop them when disabling these HW blocks (normally or as part of display state sanitizatoionin). That way we'll ensure the proper spot mentioned above for power well enabling/disabling.

And that's also the reason why display HW readout had to be added back for i915.disable_display .

--Imre
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#4 lore raw

Quoting Imre Deak (2018-08-28 12:56:38)

Quoted 26 lines · Imre Deak
Quoted 16 lines · Chris Wilson
Quoted 10 lines

Quoting Imre Deak (2018-08-28 12:40:43)

Quoted 8 lines

During power domains initialization we acquire power well references for power wells in the INIT power domain. The rest of power wells - which BIOS could have left enabled - we can only acquire references as needed during display HW readout. Thus during initialization these latter power wells can have a refcount of 0 while still being enabled. To avoid the false-positive state mismatch error this causes remove the check from intel_power_domains_init_hw() and rely on the state check in intel_power_domains_enable() which follows the HW readout.

Quoted 4 lines

Missing from above is a quick explanation of how those extraneous powerwells are sanitizied. If we don't do the HW readout (i915.disable_display?) do we not then leave the powerwell active and so complain in a later verify_state()?

Quoted 8 lines

These power wells (AUX and DDI on ICL) can only be enabled/disabled at a specific spot in the modeset sequence, otherwise the power well enable / disable operation will time out. That's the reason they're not part of the INIT domain. For these we will acquire the references during HW readout when noticing that the corresponding display HW block is active and drop them when disabling these HW blocks (normally or as part of display state sanitizatoionin). That way we'll ensure the proper spot mentioned above for power well enabling/disabling.

Quoted 2 lines

And that's also the reason why display HW readout had to be added back for i915.disable_display .

Just add the snippet to the changelog: "...as needed during display HW readout and so must defer sanitization until then (also implying that we must always do HW readout to cleanup unused powerwells)."

A reminder in code as well:
/* powerwells are sanitized by i_forget_the_function_name() */ -Chris
_______________________________________________
Intel-gfx mailing list [email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#5 lore raw

✓ Fi.CI.BAT: success for drm/i915: Don't check power domains state in intel_power_domains_init_hw()

In reply to Imre Deak
== Series Details ==

== Series Details ==

Series: drm/i915: Don't check power domains state in intel_power_domains_init_hw()
URL : https://patchwork.freedesktop.org/series/48794/
State : success

== Summary ==

= CI Bug Log - changes from CI_DRM_4714 -> Patchwork_10027 =

== Summary - SUCCESS ==

No regressions found.

External URL: https://patchwork.freedesktop.org/api/1.0/series/48794/revisions/1/mbox/

== Known issues ==

Here are the changes found in Patchwork_10027 that come from known issues:

=== IGT changes ===

    ==== Issues hit ====

    igt@drv_selftest@live_guc:
      fi-skl-guc:         NOTRUN -> DMESG-WARN (fdo#107258, fdo#107175)

    igt@kms_frontbuffer_tracking@basic:
      {fi-byt-clapper}:   PASS -> FAIL (fdo#103167)

    
    ==== Possible fixes ====

    {igt@amdgpu/amd_prime@i915-to-amd}:
      fi-bxt-j4205:       INCOMPLETE (fdo#103927) -> SKIP

    igt@drv_selftest@live_hangcheck:
      fi-kbl-7567u:       DMESG-FAIL (fdo#106947, fdo#106560) -> PASS

    igt@kms_chamelium@dp-edid-read:
      fi-kbl-7500u:       FAIL (fdo#103841) -> PASS

    igt@kms_pipe_crc_basic@suspend-read-crc-pipe-b:
      fi-blb-e6850:       INCOMPLETE -> PASS

    {igt@pm_rpm@module-reload}:
      fi-bxt-j4205:       DMESG-FAIL -> PASS

    igt@prime_vgem@basic-fence-flip:
      fi-ilk-650:         FAIL (fdo#104008) -> PASS

{name}: This element is suppressed. This means it is ignored when computing
the status of the difference (SUCCESS, WARNING, or FAILURE).

fdo#103167 https://bugs.freedesktop.org/show_bug.cgi?id=103167
fdo#103841 https://bugs.freedesktop.org/show_bug.cgi?id=103841
fdo#103927 https://bugs.freedesktop.org/show_bug.cgi?id=103927
fdo#104008 https://bugs.freedesktop.org/show_bug.cgi?id=104008
fdo#106560 https://bugs.freedesktop.org/show_bug.cgi?id=106560
fdo#106947 https://bugs.freedesktop.org/show_bug.cgi?id=106947
fdo#107175 https://bugs.freedesktop.org/show_bug.cgi?id=107175
fdo#107258 https://bugs.freedesktop.org/show_bug.cgi?id=107258

== Participating hosts (53 -> 49) ==

Additional (1): fi-skl-guc
Missing (5): fi-ctg-p8600 fi-ilk-m540 fi-byt-squawks fi-bsw-cyan fi-hsw-4200u

== Build changes ==

* Linux: CI_DRM_4714 -> Patchwork_10027

CI_DRM_4714: 06fa287b780cfea38a1c599cc1fc3611262668cf @ git://anongit.freedesktop.org/gfx-ci/linux
IGT_4610: c40743d3fce5055682d31610519758dd7379c0f8 @ git://anongit.freedesktop.org/xorg/app/intel-gpu-tools
Patchwork_10027: c97f549cc54c33bb30dc825369511ac43b1921b2 @ git://anongit.freedesktop.org/gfx-ci/linux

== Linux commits ==

c97f549cc54c drm/i915: Don't check power domains state in intel_power_domains_init_hw()

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_10027/issues.html
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

author
#6 lore raw

PATCH v2 1/1 drm/i915: Don't check power domains state in intel_power_domains_init_hw()

In reply to Imre Deak

During power domains initialization we acquire power well references for power wells in the INIT power domain. The rest of power wells - which BIOS could have left enabled - we can only acquire references as needed during display HW readout and so must defer sanitization until then (also implying that we must always do HW readout to cleanup unused power wells).

Thus during initialization these latter power wells can have a refcount of 0 while still being enabled. To avoid the false-positive state mismatch error this causes remove the check from intel_power_domains_init_hw() and rely on the state check in intel_power_domains_enable() which follows the HW readout.

v2:
- Add comment to log and code clarifying how unused power wells get disabled. (Chris)

Fixes 6dfc4a8f134f ("drm/i915: Verify power domains after enabling them")
Cc Chris Wilson <[email protected]>

References: https://bugs.freedesktop.org/show_bug.cgi?id=107411

---
 drivers/gpu/drm/i915/intel_runtime_pm.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)
drivers/gpu/drm/i915/intel_runtime_pm.c +5 -5
index 2852395125cd..480dadb1047b 100644
@@ -3724,9 +3724,10 @@ static void intel_power_domains_verify_state(struct drm_i915_private *dev_priv);
  *
  * This function initializes the hardware power domain state and enables all
  * power wells belonging to the INIT power domain. Power wells in other
- * domains (and not in the INIT domain) are referenced or disabled during the
- * modeset state HW readout. After that the reference count of each power well
- * must match its HW enabled state, see intel_power_domains_verify_state().
+ * domains (and not in the INIT domain) are referenced or disabled by
+ * intel_modeset_readout_hw_state(). After that the reference count of each
+ * power well must match its HW enabled state, see
+ * intel_power_domains_verify_state().
  *
  * It will return with power domains disabled (to be enabled later by
  * intel_power_domains_enable()) and must be paired with
@@ -3767,9 +3768,8 @@ void intel_power_domains_init_hw(struct drm_i915_private *dev_priv, bool resume)
 	if (!i915_modparams.disable_power_well)
 		intel_display_power_get(dev_priv, POWER_DOMAIN_INIT);
 	intel_power_domains_sync_hw(dev_priv);
+
 	power_domains->initializing = false;
-
-	intel_power_domains_verify_state(dev_priv);
 }
 
 /**
Signature · 6 lines
2.13.2

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Reviewed-by on v2 1/1
#7 lore raw

Quoting Imre Deak (2018-08-28 13:22:31)

Quoted 13 lines

During power domains initialization we acquire power well references for power wells in the INIT power domain. The rest of power wells - which BIOS could have left enabled - we can only acquire references as needed during display HW readout and so must defer sanitization until then (also implying that we must always do HW readout to cleanup unused power wells).

Thus during initialization these latter power wells can have a refcount of 0 while still being enabled. To avoid the false-positive state mismatch error this causes remove the check from intel_power_domains_init_hw() and rely on the state check in intel_power_domains_enable() which follows the HW readout.

Quoted 8 lines

v2:
- Add comment to log and code clarifying how unused power wells get disabled. (Chris)

Fixes 6dfc4a8f134f ("drm/i915: Verify power domains after enabling them")
Cc Chris Wilson <[email protected]>

References: https://bugs.freedesktop.org/show_bug.cgi?id=107411

That's enough to allow me to keep following along,

-Chris
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#8 lore raw

✓ Fi.CI.BAT: success for drm/i915: Don't check power domains state in intel_power_domains_init_hw() (rev2)

In reply to Imre Deak
== Series Details ==

== Series Details ==

Series: drm/i915: Don't check power domains state in intel_power_domains_init_hw() (rev2)
URL : https://patchwork.freedesktop.org/series/48794/
State : success

== Summary ==

= CI Bug Log - changes from CI_DRM_4714 -> Patchwork_10028 =

== Summary - SUCCESS ==

No regressions found.

External URL: https://patchwork.freedesktop.org/api/1.0/series/48794/revisions/2/mbox/

== Known issues ==

Here are the changes found in Patchwork_10028 that come from known issues:

=== IGT changes ===

    ==== Issues hit ====

    {igt@amdgpu/amd_basic@userptr}:
      {fi-kbl-8809g}:     PASS -> INCOMPLETE (fdo#107402)

    igt@drv_selftest@live_guc:
      fi-skl-guc:         NOTRUN -> DMESG-WARN (fdo#107175, fdo#107258)

    igt@drv_selftest@live_hangcheck:
      fi-kbl-guc:         PASS -> DMESG-FAIL (fdo#106947, fdo#107710)

    igt@kms_pipe_crc_basic@read-crc-pipe-b:
      {fi-byt-clapper}:   PASS -> FAIL (fdo#107362) +1

    
    ==== Possible fixes ====

    {igt@amdgpu/amd_prime@i915-to-amd}:
      fi-bxt-j4205:       INCOMPLETE (fdo#103927) -> SKIP

    igt@drv_selftest@live_hangcheck:
      fi-kbl-7567u:       DMESG-FAIL (fdo#106947, fdo#106560) -> PASS

    igt@kms_chamelium@dp-edid-read:
      fi-kbl-7500u:       FAIL (fdo#103841) -> PASS

    igt@kms_pipe_crc_basic@suspend-read-crc-pipe-b:
      fi-blb-e6850:       INCOMPLETE -> PASS

    {igt@pm_rpm@module-reload}:
      fi-bxt-j4205:       DMESG-FAIL (fdo#107712) -> PASS

{name}: This element is suppressed. This means it is ignored when computing
the status of the difference (SUCCESS, WARNING, or FAILURE).

fdo#103841 https://bugs.freedesktop.org/show_bug.cgi?id=103841
fdo#103927 https://bugs.freedesktop.org/show_bug.cgi?id=103927
fdo#106560 https://bugs.freedesktop.org/show_bug.cgi?id=106560
fdo#106947 https://bugs.freedesktop.org/show_bug.cgi?id=106947
fdo#107175 https://bugs.freedesktop.org/show_bug.cgi?id=107175
fdo#107258 https://bugs.freedesktop.org/show_bug.cgi?id=107258
fdo#107362 https://bugs.freedesktop.org/show_bug.cgi?id=107362
fdo#107402 https://bugs.freedesktop.org/show_bug.cgi?id=107402
fdo#107710 https://bugs.freedesktop.org/show_bug.cgi?id=107710
fdo#107712 https://bugs.freedesktop.org/show_bug.cgi?id=107712

== Participating hosts (53 -> 49) ==

Additional (1): fi-skl-guc
Missing (5): fi-ctg-p8600 fi-ilk-m540 fi-byt-squawks fi-bsw-cyan fi-hsw-4200u

== Build changes ==

* Linux: CI_DRM_4714 -> Patchwork_10028

CI_DRM_4714: 06fa287b780cfea38a1c599cc1fc3611262668cf @ git://anongit.freedesktop.org/gfx-ci/linux
IGT_4610: c40743d3fce5055682d31610519758dd7379c0f8 @ git://anongit.freedesktop.org/xorg/app/intel-gpu-tools
Patchwork_10028: 51662a027a87f68fb7fb910b64c83adc91fa4dcb @ git://anongit.freedesktop.org/gfx-ci/linux

== Linux commits ==

51662a027a87 drm/i915: Don't check power domains state in intel_power_domains_init_hw()

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_10028/issues.html
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#9 lore raw

✓ Fi.CI.IGT: success for drm/i915: Don't check power domains state in intel_power_domains_init_hw() (rev2)

In reply to Imre Deak
== Series Details ==

== Series Details ==

Series: drm/i915: Don't check power domains state in intel_power_domains_init_hw() (rev2)
URL : https://patchwork.freedesktop.org/series/48794/
State : success

== Summary ==

= CI Bug Log - changes from CI_DRM_4714_full -> Patchwork_10028_full =

== Summary - SUCCESS ==

No regressions found.

== Known issues ==

Here are the changes found in Patchwork_10028_full that come from known issues:

=== IGT changes ===

    ==== Issues hit ====

    igt@drv_suspend@shrink:
      shard-kbl:          PASS -> INCOMPLETE (fdo#106886, fdo#103665)

    igt@gem_ppgtt@blt-vs-render-ctx0:
      shard-kbl:          PASS -> INCOMPLETE (fdo#103665, fdo#106023)

    
    ==== Possible fixes ====

    igt@gem_ctx_isolation@vcs0-s3:
      shard-kbl:          INCOMPLETE (fdo#103665, fdo#107556) -> PASS

    igt@kms_setmode@basic:
      shard-apl:          FAIL (fdo#99912) -> PASS

    igt@perf@blocking:
      shard-hsw:          FAIL (fdo#102252) -> PASS

fdo#102252 https://bugs.freedesktop.org/show_bug.cgi?id=102252
fdo#103665 https://bugs.freedesktop.org/show_bug.cgi?id=103665
fdo#106023 https://bugs.freedesktop.org/show_bug.cgi?id=106023
fdo#106886 https://bugs.freedesktop.org/show_bug.cgi?id=106886
fdo#107556 https://bugs.freedesktop.org/show_bug.cgi?id=107556
fdo#99912 https://bugs.freedesktop.org/show_bug.cgi?id=99912

== Participating hosts (5 -> 5) ==

No changes in participating hosts

== Build changes ==

* Linux: CI_DRM_4714 -> Patchwork_10028

CI_DRM_4714: 06fa287b780cfea38a1c599cc1fc3611262668cf @ git://anongit.freedesktop.org/gfx-ci/linux
IGT_4610: c40743d3fce5055682d31610519758dd7379c0f8 @ git://anongit.freedesktop.org/xorg/app/intel-gpu-tools
Patchwork_10028: 51662a027a87f68fb7fb910b64c83adc91fa4dcb @ git://anongit.freedesktop.org/gfx-ci/linux
piglit_4509: fdc5a4ca11124ab8413c7988896eec4c97336694 @ git://anongit.freedesktop.org/piglit

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_10028/shards.html
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

author Applied
#10 lore raw

Re: ✓ Fi.CI.IGT: success for drm/i915: Don't check power domains state in intel_power_domains_init_hw() (rev2)

Quoted 5 lines · Patchwork

== Series Details ==

Series: drm/i915: Don't check power domains state in intel_power_domains_init_hw() (rev2)
URL : https://patchwork.freedesktop.org/series/48794/
State : success

Pushed to -dinq, thanks for the review.

Quoted 55 lines

== Summary ==

= CI Bug Log - changes from CI_DRM_4714_full -> Patchwork_10028_full =

== Summary - SUCCESS ==

No regressions found.

== Known issues ==

Here are the changes found in Patchwork_10028_full that come from known issues:

=== IGT changes ===

    ==== Issues hit ====

    igt@drv_suspend@shrink:
      shard-kbl:          PASS -> INCOMPLETE (fdo#106886, fdo#103665)

    igt@gem_ppgtt@blt-vs-render-ctx0:
      shard-kbl:          PASS -> INCOMPLETE (fdo#103665, fdo#106023)

    
    ==== Possible fixes ====

    igt@gem_ctx_isolation@vcs0-s3:
      shard-kbl:          INCOMPLETE (fdo#103665, fdo#107556) -> PASS

    igt@kms_setmode@basic:
      shard-apl:          FAIL (fdo#99912) -> PASS

    igt@perf@blocking:
      shard-hsw:          FAIL (fdo#102252) -> PASS

fdo#102252 https://bugs.freedesktop.org/show_bug.cgi?id=102252
fdo#103665 https://bugs.freedesktop.org/show_bug.cgi?id=103665
fdo#106023 https://bugs.freedesktop.org/show_bug.cgi?id=106023
fdo#106886 https://bugs.freedesktop.org/show_bug.cgi?id=106886
fdo#107556 https://bugs.freedesktop.org/show_bug.cgi?id=107556
fdo#99912 https://bugs.freedesktop.org/show_bug.cgi?id=99912

== Participating hosts (5 -> 5) ==

No changes in participating hosts

== Build changes ==

* Linux: CI_DRM_4714 -> Patchwork_10028

Quoted 8 lines

CI_DRM_4714: 06fa287b780cfea38a1c599cc1fc3611262668cf @ git://anongit.freedesktop.org/gfx-ci/linux
IGT_4610: c40743d3fce5055682d31610519758dd7379c0f8 @ git://anongit.freedesktop.org/xorg/app/intel-gpu-tools
Patchwork_10028: 51662a027a87f68fb7fb910b64c83adc91fa4dcb @ git://anongit.freedesktop.org/gfx-ci/linux
piglit_4509: fdc5a4ca11124ab8413c7988896eec4c97336694 @ git://anongit.freedesktop.org/piglit

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_10028/shards.html

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx