Skip to contentSkip to conversation

drm/i915: Disable read only ppgtt support for gen11

PATCH In review Mika Kuoppala 4 participants 5 messages Apr 11, 2019

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 drm/i915: Disable read only ppgtt support for gen11

On gen11 writing to read only ppgtt page causes a gpu hang. This behaviour is different than with previous gen where read only ppgtt access is supported. On those, the write is just dropped without visible side effects.

Disable ro ppgtt support on gen11 until a solution can be found to bring it into line with its predecessors.

References: HSDES#1807136187
References: https://bugzilla.freedesktop.org/show_bug.cgi?id=108569

---
 drivers/gpu/drm/i915/i915_gem_gtt.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)
drivers/gpu/drm/i915/i915_gem_gtt.c +7 -2
index 736c845eb77f..caae8cdafc1a 100644
@@ -1548,8 +1548,13 @@ static struct i915_hw_ppgtt *gen8_ppgtt_create(struct drm_i915_private *i915)
 
 	ppgtt_init(i915, ppgtt);
 
-	/* From bdw, there is support for read-only pages in the PPGTT. */
-	ppgtt->vm.has_read_only = true;
+	/*
+	 * From bdw, there is hw support for read-only pages in the PPGTT.
+	 *
+	 * Gen11 has HSDES#:1807136187 unresolved. Disable ro support
+	 * for now.
+	 */
+	ppgtt->vm.has_read_only = INTEL_GEN(i915) != 11;
 
 	/* There are only few exceptions for gen >=6. chv and bxt.
 	 * And we are not sure about the latter so play safe for now.
Signature · 6 lines
2.17.1

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Acked-by on v1 1/1
#1 lore raw

Quoting Mika Kuoppala (2019-04-11 09:30:34)

Quoted 5 lines

On gen11 writing to read only ppgtt page causes a gpu hang. This behaviour is different than with previous gen where read only ppgtt access is supported. On those, the write is just dropped without visible side effects.

Quoted 7 lines

Disable ro ppgtt support on gen11 until a solution can be found to bring it into line with its predecessors.

References: HSDES#1807136187
References: https://bugzilla.freedesktop.org/show_bug.cgi?id=108569

Lets see if I remembered to stick those vm->has_read_only checks everywhere we needed them (and remembered to exercise them -- a bit of a catch 22, if I forgot the check, I probably forgot the test as well).

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

#2 lore raw

✗ Fi.CI.BAT: failure for drm/i915: Disable read only ppgtt support for gen11

In reply to Mika Kuoppala
== Series Details ==

== Series Details ==

Series: drm/i915: Disable read only ppgtt support for gen11
URL : https://patchwork.freedesktop.org/series/59323/
State : failure

== Summary ==

CI Bug Log - changes from CI_DRM_5912 -> Patchwork_12761
====================================================

Summary
-------

**FAILURE**

Serious unknown changes coming with Patchwork_12761 absolutely need to be verified manually.

If you think the reported changes have nothing to do with the changes introduced in Patchwork_12761, please notify your bug team to allow them to document this new failure mode, which will reduce false positives in CI.

External URL: https://patchwork.freedesktop.org/api/1.0/series/59323/revisions/1/mbox/

Possible new issues
-------------------

Here are the unknown changes that may have been introduced in Patchwork_12761:

### IGT changes ###

#### Possible regressions ####

* igt@gem_render_tiled_blits@basic:
- fi-icl-y: NOTRUN -> INCOMPLETE

Known issues
------------

Here are the changes found in Patchwork_12761 that come from known issues:

### IGT changes ###

#### Issues hit ####

* igt@gem_exec_basic@basic-bsd2:
- fi-icl-y: NOTRUN -> SKIP [fdo#109276] +7

* igt@gem_exec_parse@basic-rejected:
- fi-icl-y: NOTRUN -> SKIP [fdo#109289] +1

* igt@i915_selftest@live_hangcheck:
- fi-bxt-dsi: PASS -> INCOMPLETE [fdo#103927]

* igt@kms_busy@basic-flip-c:
- fi-byt-j1900: NOTRUN -> SKIP [fdo#109271] / [fdo#109278]

* igt@kms_chamelium@hdmi-crc-fast:
- fi-byt-j1900: NOTRUN -> SKIP [fdo#109271] +52

* igt@kms_chamelium@hdmi-edid-read:
- fi-hsw-peppy: NOTRUN -> SKIP [fdo#109271] +46

* igt@kms_force_connector_basic@force-edid:
- fi-glk-dsi: NOTRUN -> SKIP [fdo#109271] +26

* igt@kms_frontbuffer_tracking@basic:
- fi-hsw-peppy: NOTRUN -> DMESG-FAIL [fdo#102614] / [fdo#107814]
- fi-glk-dsi: NOTRUN -> FAIL [fdo#103167]

* igt@kms_pipe_crc_basic@read-crc-pipe-a-frame-sequence:
- fi-byt-clapper: PASS -> FAIL [fdo#103191]

#### Possible fixes ####

* igt@gem_ctx_switch@basic-default:
- fi-icl-y: INCOMPLETE [fdo#108569] -> PASS

* igt@i915_selftest@live_hangcheck:
- fi-skl-iommu: INCOMPLETE [fdo#108602] / [fdo#108744] -> PASS

* igt@kms_cursor_legacy@basic-flip-before-cursor-varying-size:
- fi-glk-dsi: INCOMPLETE [fdo#103359] / [k.org#198133] -> PASS

[fdo#102614]: https://bugs.freedesktop.org/show_bug.cgi?id=102614
[fdo#103167]: https://bugs.freedesktop.org/show_bug.cgi?id=103167
[fdo#103191]: https://bugs.freedesktop.org/show_bug.cgi?id=103191
[fdo#103359]: https://bugs.freedesktop.org/show_bug.cgi?id=103359
[fdo#103927]: https://bugs.freedesktop.org/show_bug.cgi?id=103927
[fdo#107814]: https://bugs.freedesktop.org/show_bug.cgi?id=107814
[fdo#108569]: https://bugs.freedesktop.org/show_bug.cgi?id=108569
[fdo#108602]: https://bugs.freedesktop.org/show_bug.cgi?id=108602
[fdo#108744]: https://bugs.freedesktop.org/show_bug.cgi?id=108744
[fdo#109271]: https://bugs.freedesktop.org/show_bug.cgi?id=109271
[fdo#109276]: https://bugs.freedesktop.org/show_bug.cgi?id=109276
[fdo#109278]: https://bugs.freedesktop.org/show_bug.cgi?id=109278
[fdo#109289]: https://bugs.freedesktop.org/show_bug.cgi?id=109289
[k.org#198133]: https://bugzilla.kernel.org/show_bug.cgi?id=198133

Participating hosts (50 -> 43)
------------------------------

Additional (2): fi-byt-j1900 fi-hsw-peppy
Missing (9): fi-kbl-soraka fi-ilk-m540 fi-hsw-4200u fi-byt-squawks fi-icl-u2 fi-bsw-cyan fi-snb-2520m fi-ctg-p8600 fi-bdw-samus

Build changes
-------------

* Linux: CI_DRM_5912 -> Patchwork_12761

CI_DRM_5912: d7c1889b2e971f44d10c69ec204991f3264d2412 @ git://anongit.freedesktop.org/gfx-ci/linux
IGT_4943: 5941f371b0fe25084d4b1c49882faa8d41d44c9f @ git://anongit.freedesktop.org/xorg/app/intel-gpu-tools
Patchwork_12761: 4df3122417ae11d385696ad010b40e2f16a38b14 @ git://anongit.freedesktop.org/gfx-ci/linux

== Linux commits ==

4df3122417ae drm/i915: Disable read only ppgtt support for gen11

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12761/
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

maintainer
#3 lore raw
In reply to Mika Kuoppala
Quoted 23 lines · Mika Kuoppala

On gen11 writing to read only ppgtt page causes a gpu hang. This behaviour is different than with previous gen where read only ppgtt access is supported. On those, the write is just dropped without visible side effects.

Disable ro ppgtt support on gen11 until a solution can be found to bring it into line with its predecessors.

References: HSDES#1807136187
References: https://bugzilla.freedesktop.org/show_bug.cgi?id=108569

---
 drivers/gpu/drm/i915/i915_gem_gtt.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)
drivers/gpu/drm/i915/i915_gem_gtt.c +0 -0
index 736c845eb77f..caae8cdafc1a 100644
@@ -1548,8 +1548,13 @@ static struct i915_hw_ppgtt *gen8_ppgtt_create(struct drm_i915_private *i915)
 
 	ppgtt_init(i915, ppgtt);

drivers/gpu/drm/i915/i915_gem_gtt.c static struct i915_hw_ppgtt *gen8_ppgtt_create(struct drm_i915_private *i915) · line 1557

Quoted 10 lines
drivers/gpu/drm/i915/i915_gem_gtt.c +7 -2
 
-	/* From bdw, there is support for read-only pages in the PPGTT. */
-	ppgtt->vm.has_read_only = true;
+	/*
+	 * From bdw, there is hw support for read-only pages in the PPGTT.
+	 *
+	 * Gen11 has HSDES#:1807136187 unresolved. Disable ro support
+	 * for now.
+	 */
+	ppgtt->vm.has_read_only = INTEL_GEN(i915) != 11;

I believe it is safest to use < 11, no?!

Quoted 10 lines
drivers/gpu/drm/i915/i915_gem_gtt.c +0 -0
 	/* There are only few exceptions for gen >=6. chv and bxt.
 	 * And we are not sure about the latter so play safe for now.
Signature · 6 lines
2.17.1

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

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

#4 lore raw

Quoting Rodrigo Vivi (2019-04-11 20:47:48)

Quoted 25 lines · Mika Kuoppala
Quoted 24 lines

On gen11 writing to read only ppgtt page causes a gpu hang. This behaviour is different than with previous gen where read only ppgtt access is supported. On those, the write is just dropped without visible side effects.

Disable ro ppgtt support on gen11 until a solution can be found to bring it into line with its predecessors.

References: HSDES#1807136187
References: https://bugzilla.freedesktop.org/show_bug.cgi?id=108569

---
 drivers/gpu/drm/i915/i915_gem_gtt.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)
drivers/gpu/drm/i915/i915_gem_gtt.c +0 -0
index 736c845eb77f..caae8cdafc1a 100644
@@ -1548,8 +1548,13 @@ static struct i915_hw_ppgtt *gen8_ppgtt_create(struct drm_i915_private *i915)
 
      ppgtt_init(i915, ppgtt);
 
Quoted 11 lines

drivers/gpu/drm/i915/i915_gem_gtt.c static struct i915_hw_ppgtt *gen8_ppgtt_create(struct drm_i915_private *i915) · line 1557

Quoted 9 lines
drivers/gpu/drm/i915/i915_gem_gtt.c +7 -2
-     /* From bdw, there is support for read-only pages in the PPGTT. */
-     ppgtt->vm.has_read_only = true;
+     /*
+      * From bdw, there is hw support for read-only pages in the PPGTT.
+      *
+      * Gen11 has HSDES#:1807136187 unresolved. Disable ro support
+      * for now.
+      */
+     ppgtt->vm.has_read_only = INTEL_GEN(i915) != 11;

I believe it is safest to use < 11, no?!

No. It's an issue in Icelake.
-Chris
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx