Skip to contentSkip to conversation

Splitting up platform-specific calls

RFC PATCH Superseded Casey Bowman 4 participants 13 messages Jan 20, 2022 – Feb 10, 2022

Subsystems (2)

v1 of 6 Superseded Latest is v6 →

lore
Share & tools
lore
#0 lore raw

cover Splitting up platform-specific calls

In this RFC I would like to ask the community their thoughts on how we can best handle splitting architecture-specific calls.

I would like to address the following:

1. How do we want to split architecture calls? Different object files per platform? Separate function calls within the same object file?

2. How do we address dummy functions? If we have a function call that is used for one or more platforms, but is not used in another, what should we do for this case?

I've given an example of splitting an architecture call in my patch with run_as_guest() being split into different implementations for x86 and arm64 in separate object files, sharing a single header.

Another suggestion from Michael ([email protected]) involved using a single object file, a single header, and splitting various functions calls via ifdefs in the header file.

I would appreciate any input on how we can avoid scaling issues when including multiple architectures and multiple functions (as the number of function calls will inevitably increase with more architectures).

Casey Bowman (1):
i915/drm: Split out x86 and arm64 functionality

 drivers/gpu/drm/i915/Makefile              |  4 +++
 drivers/gpu/drm/i915/i915_drv.h            |  6 +---
 drivers/gpu/drm/i915/i915_platform.h       | 16 +++++++++++
 drivers/gpu/drm/i915/i915_platform_arm64.c | 33 ++++++++++++++++++++++
 drivers/gpu/drm/i915/i915_platform_x86.c   | 33 ++++++++++++++++++++++
 5 files changed, 87 insertions(+), 5 deletions(-)
 create mode 100644 drivers/gpu/drm/i915/i915_platform.h
 create mode 100644 drivers/gpu/drm/i915/i915_platform_arm64.c
 create mode 100644 drivers/gpu/drm/i915/i915_platform_x86.c
Signature · 1 line
2.25.1
#1 lore raw

RFC PATCH 1/1 i915/drm: Split out x86 and arm64 functionality

Some x86 checks are unnecessary on arm64 systems, so they are being split out to avoid being used. There may be further arm64 implementations created in the future for this area, so it's better to split this out now.

---
 drivers/gpu/drm/i915/Makefile              |  4 +++
 drivers/gpu/drm/i915/i915_drv.h            |  6 +---
 drivers/gpu/drm/i915/i915_platform.h       | 16 +++++++++++
 drivers/gpu/drm/i915/i915_platform_arm64.c | 33 ++++++++++++++++++++++
 drivers/gpu/drm/i915/i915_platform_x86.c   | 33 ++++++++++++++++++++++
 5 files changed, 87 insertions(+), 5 deletions(-)
 create mode 100644 drivers/gpu/drm/i915/i915_platform.h
 create mode 100644 drivers/gpu/drm/i915/i915_platform_arm64.c
 create mode 100644 drivers/gpu/drm/i915/i915_platform_x86.c
drivers/gpu/drm/i915/Makefile +4 -0
index 213c5f9fae32..dd66fe57934d 100644
@@ -320,6 +320,10 @@ i915-y += intel_gvt.o
 include $(src)/gvt/Makefile
 endif
 
+# Architecture-specific calls
+i915-$(CONFIG_X86)   += i915_platform_x86.o
+i915-$(CONFIG_ARM64) += i915_platform_arm64.o
+
 obj-$(CONFIG_DRM_I915) += i915.o
 obj-$(CONFIG_DRM_I915_GVT_KVMGT) += gvt/kvmgt.o
 
drivers/gpu/drm/i915/i915_drv.h +1 -5
index 290dfd40c7b3..e688270c8257 100644
@@ -107,6 +107,7 @@
 #include "gt/intel_timeline.h"
 #include "i915_vma.h"
 
+#include "i915_platform.h"
 
 /* General customization:
  */
@@ -1543,11 +1544,6 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915,
 #define INTEL_DISPLAY_ENABLED(dev_priv) \
 	(drm_WARN_ON(&(dev_priv)->drm, !HAS_DISPLAY(dev_priv)), !(dev_priv)->params.disable_display)
 
-static inline bool run_as_guest(void)
-{
-	return !hypervisor_is_type(X86_HYPER_NATIVE);
-}
-
 #define HAS_D12_PLANE_MINIMIZATION(dev_priv) (IS_ROCKETLAKE(dev_priv) || \
 					      IS_ALDERLAKE_S(dev_priv))
 
drivers/gpu/drm/i915/i915_platform.h new file +16 -0
new file mode 100644
index 000000000000..300f93d20f58
@@ -0,0 +1,16 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2022 Intel Corporation
+ */
+
+#ifndef _I915_PLATFORM_
+#define _I915_PLATFORM_
+
+#include <linux/types.h>
+#include <linux/bug.h>
+
+/* Start of i915_drv functionality */
+bool run_as_guest(void);
+/* End of i915_drv functionality */
+
+#endif
drivers/gpu/drm/i915/i915_platform_arm64.c new file +33 -0
new file mode 100644
index 000000000000..95692c4dc75f
@@ -0,0 +1,33 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright © 2022 Intel Corporation
+ */
+
+/*
+ * Read before adding/removing content!
+ *
+ * Ensure that all functions defined here are also defined
+ * in the i915_platform_x86.c file.
+ *
+ * If the function is a dummy function, be sure to add
+ * a DRM_WARN() call to note that the function is a
+ * dummy function to users so that we can better track
+ * any issues that arise due to changes in either file.
+ *
+ * Also be sure to label Start/End of sections where
+ * functions originate from. These files will host
+ * architecture-specific content from a myriad of files,
+ * labeling the sections will help devs keep track of
+ * where the calls interact.
+ */
+
+#include "i915_platform.h"
+
+/* Start of i915_drv functionality */
+/* Intel VT-d is not used on ARM64 systems */
+bool run_as_guest(void)
+{
+	WARN(1, "%s not supported on arm64 platforms.", __func__);
+	return false;
+}
+/* End of i915_drv functionality */
drivers/gpu/drm/i915/i915_platform_x86.c new file +33 -0
new file mode 100644
index 000000000000..9a7174ad2147
@@ -0,0 +1,33 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright © 2022 Intel Corporation
+ */
+
+/*
+ * Read before adding/removing content!
+ *
+ * Ensure that all functions defined here are also defined
+ * in the i915_platform_arm64.c file.
+ *
+ * If the function is a dummy function, be sure to add
+ * a DRM_WARN() call to note that the function is a
+ * dummy function to users so that we can better track
+ * any issues that arise due to changes in either file.
+ *
+ * Also be sure to label Start/End of sections where
+ * functions originate from. These files will host
+ * architecture-specific content from a myriad of files,
+ * labeling the sections will help devs keep track of
+ * where the calls interact.
+ */
+
+#include "i915_platform.h"
+
+#include <asm/hypervisor.h>
+
+/* Start of i915_drv functionality */
+bool run_as_guest(void)
+{
+	return !hypervisor_is_type(X86_HYPER_NATIVE);
+}
+/* End of i915_drv functionality */
Signature · 1 line
2.25.1
#2 lore raw

[Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for Splitting up platform-specific calls

In reply to Casey Bowman
== Series Details ==

== Series Details ==

Series: Splitting up platform-specific calls
URL : https://patchwork.freedesktop.org/series/99126/
State : warning

== Summary ==

$ dim checkpatch origin/drm-tip
cfedaf9b9519 i915/drm: Split out x86 and arm64 functionality
-:53: WARNING:FILE_PATH_CHANGES: added, moved or deleted file(s), does MAINTAINERS need updating?
#53: 
new file mode 100644

-:112: CHECK:LINE_SPACING: Please use a blank line after function/struct/union/enum declarations #112: FILE: drivers/gpu/drm/i915/i915_platform_arm64.c:33:
+}
+/* End of i915_drv functionality */

-:151: CHECK:LINE_SPACING: Please use a blank line after function/struct/union/enum declarations #151: FILE: drivers/gpu/drm/i915/i915_platform_x86.c:33:
+}
+/* End of i915_drv functionality */

total: 0 errors, 1 warnings, 2 checks, 110 lines checked

#4 lore raw

[Intel-gfx] ✓ Fi.CI.BAT: success for Splitting up platform-specific calls

In reply to Casey Bowman
== Series Details ==

== Series Details ==

Series: Splitting up platform-specific calls
URL : https://patchwork.freedesktop.org/series/99126/
State : success

== Summary ==

CI Bug Log - changes from CI_DRM_11115 -> Patchwork_22048
====================================================

Summary
-------

**SUCCESS**

No regressions found.

External URL: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/index.html

Participating hosts (42 -> 42)
------------------------------

Additional (2): bat-jsl-2 fi-pnv-d510
Missing (2): fi-bsw-cyan fi-bdw-samus

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

Here are the changes found in Patchwork_22048 that come from known issues:

### IGT changes ###

#### Issues hit ####

* igt@amdgpu/amd_cs_nop@sync-fork-compute0:
- fi-snb-2600: NOTRUN -> [SKIP][1] ([fdo#109271]) +17 similar issues
[1]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/fi-snb-2600/igt@amdgpu/[email protected]

* igt@gem_flink_basic@bad-flink:
- fi-skl-6600u: [PASS][2] -> [INCOMPLETE][3] ([i915#4547])
[2]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/fi-skl-6600u/igt@[email protected]
[3]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/fi-skl-6600u/igt@[email protected]

* igt@i915_selftest@live@hangcheck:
- fi-hsw-4770: [PASS][4] -> [INCOMPLETE][5] ([i915#4785])
[4]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/fi-hsw-4770/igt@i915_selftest@[email protected]
[5]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/fi-hsw-4770/igt@i915_selftest@[email protected]

* igt@prime_vgem@basic-userptr:
- fi-pnv-d510: NOTRUN -> [SKIP][6] ([fdo#109271]) +57 similar issues
[6]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/fi-pnv-d510/igt@[email protected]

* igt@runner@aborted:
- fi-skl-6600u: NOTRUN -> [FAIL][7] ([i915#2722] / [i915#4312])
[7]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/fi-skl-6600u/igt@[email protected]
- fi-hsw-4770: NOTRUN -> [FAIL][8] ([fdo#109271] / [i915#1436] / [i915#4312])
[8]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/fi-hsw-4770/igt@[email protected]

#### Possible fixes ####

* igt@gem_exec_suspend@basic-s3@smem:
- {bat-rpls-1}: [INCOMPLETE][9] ([i915#4898]) -> [PASS][10]
[9]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/bat-rpls-1/igt@gem_exec_suspend@[email protected]
[10]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/bat-rpls-1/igt@gem_exec_suspend@[email protected]

* igt@i915_selftest@live@gtt:
- fi-bdw-5557u: [DMESG-FAIL][11] -> [PASS][12]
[11]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/fi-bdw-5557u/igt@i915_selftest@[email protected]
[12]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/fi-bdw-5557u/igt@i915_selftest@[email protected]

* igt@i915_selftest@live@hangcheck:
- fi-snb-2600: [INCOMPLETE][13] ([i915#3921]) -> [PASS][14]
[13]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/fi-snb-2600/igt@i915_selftest@[email protected]
[14]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/fi-snb-2600/igt@i915_selftest@[email protected]

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

[fdo#109271]: https://bugs.freedesktop.org/show_bug.cgi?id=109271
[fdo#109285]: https://bugs.freedesktop.org/show_bug.cgi?id=109285
[fdo#109308]: https://bugs.freedesktop.org/show_bug.cgi?id=109308
[fdo#111825]: https://bugs.freedesktop.org/show_bug.cgi?id=111825
[fdo#111827]: https://bugs.freedesktop.org/show_bug.cgi?id=111827
[i915#1072]: https://gitlab.freedesktop.org/drm/intel/issues/1072
[i915#1155]: https://gitlab.freedesktop.org/drm/intel/issues/1155
[i915#1436]: https://gitlab.freedesktop.org/drm/intel/issues/1436
[i915#1845]: https://gitlab.freedesktop.org/drm/intel/issues/1845
[i915#1849]: https://gitlab.freedesktop.org/drm/intel/issues/1849
[i915#2190]: https://gitlab.freedesktop.org/drm/intel/issues/2190
[i915#2722]: https://gitlab.freedesktop.org/drm/intel/issues/2722
[i915#3282]: https://gitlab.freedesktop.org/drm/intel/issues/3282
[i915#3301]: https://gitlab.freedesktop.org/drm/intel/issues/3301
[i915#3637]: https://gitlab.freedesktop.org/drm/intel/issues/3637
[i915#3708]: https://gitlab.freedesktop.org/drm/intel/issues/3708
[i915#3921]: https://gitlab.freedesktop.org/drm/intel/issues/3921
[i915#4103]: https://gitlab.freedesktop.org/drm/intel/issues/4103
[i915#4312]: https://gitlab.freedesktop.org/drm/intel/issues/4312
[i915#4391]: https://gitlab.freedesktop.org/drm/intel/issues/4391
[i915#4547]: https://gitlab.freedesktop.org/drm/intel/issues/4547
[i915#4613]: https://gitlab.freedesktop.org/drm/intel/issues/4613
[i915#4785]: https://gitlab.freedesktop.org/drm/intel/issues/4785
[i915#4873]: https://gitlab.freedesktop.org/drm/intel/issues/4873
[i915#4898]: https://gitlab.freedesktop.org/drm/intel/issues/4898
[i915#533]: https://gitlab.freedesktop.org/drm/intel/issues/533

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

* Linux: CI_DRM_11115 -> Patchwork_22048

CI-20190529: 20190529
CI_DRM_11115: 4e12213687264ffccb45d72fe638f94d3ca666bd @ git://anongit.freedesktop.org/gfx-ci/linux
IGT_6329: 38f656fdd61119105ecfa2c4dac157cd7dcad204 @ https://gitlab.freedesktop.org/drm/igt-gpu-tools.git
Patchwork_22048: cfedaf9b9519ab335e7ccd4bc2e474dcd49f117c @ git://anongit.freedesktop.org/gfx-ci/linux

== Linux commits ==

cfedaf9b9519 i915/drm: Split out x86 and arm64 functionality

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/index.html

#5 lore raw

[Intel-gfx] ✗ Fi.CI.IGT: failure for Splitting up platform-specific calls

In reply to Casey Bowman
== Series Details ==

== Series Details ==

Series: Splitting up platform-specific calls
URL : https://patchwork.freedesktop.org/series/99126/
State : failure

== Summary ==

CI Bug Log - changes from CI_DRM_11115_full -> Patchwork_22048_full
====================================================

Summary
-------

**FAILURE**

Serious unknown changes coming with Patchwork_22048_full absolutely need to be verified manually.

If you think the reported changes have nothing to do with the changes introduced in Patchwork_22048_full, please notify your bug team to allow them to document this new failure mode, which will reduce false positives in CI.

Participating hosts (10 -> 10)
------------------------------

No changes in participating hosts

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

Here are the unknown changes that may have been introduced in Patchwork_22048_full:

### IGT changes ###

#### Possible regressions ####

* igt@gem_exec_schedule@smoketest-all:
- shard-tglb: [PASS][1] -> [INCOMPLETE][2]
[1]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-tglb2/igt@[email protected]
[2]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-tglb6/igt@[email protected]

#### Warnings ####

* igt@runner@aborted:
- shard-apl: ([FAIL][3], [FAIL][4], [FAIL][5], [FAIL][6], [FAIL][7], [FAIL][8]) ([fdo#109271] / [i915#180] / [i915#1814] / [i915#3002] / [i915#4312]) -> ([FAIL][9], [FAIL][10], [FAIL][11], [FAIL][12], [FAIL][13], [FAIL][14]) ([i915#180] / [i915#1814] / [i915#3002] / [i915#4312])
[3]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-apl2/igt@[email protected]
[4]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-apl6/igt@[email protected]
[5]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-apl1/igt@[email protected]
[6]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-apl1/igt@[email protected]
[7]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-apl1/igt@[email protected]
[8]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-apl8/igt@[email protected]
[9]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-apl7/igt@[email protected]
[10]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-apl7/igt@[email protected]
[11]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-apl7/igt@[email protected]
[12]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-apl6/igt@[email protected]
[13]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-apl8/igt@[email protected]
[14]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-apl7/igt@[email protected]

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

Here are the changes found in Patchwork_22048_full that come from known issues:

### IGT changes ###

#### Issues hit ####

* igt@gem_ctx_isolation@preservation-s3@rcs0:
- shard-apl: [PASS][15] -> [DMESG-WARN][16] ([i915#180])
[15]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-apl3/igt@gem_ctx_isolation@[email protected]
[16]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-apl8/igt@gem_ctx_isolation@[email protected]

* igt@gem_ctx_persistence@engines-queued:
- shard-snb: NOTRUN -> [SKIP][17] ([fdo#109271] / [i915#1099])
[17]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-snb4/igt@[email protected]

* igt@gem_eio@unwedge-stress:
- shard-iclb: [PASS][18] -> [TIMEOUT][19] ([i915#2481] / [i915#3070])
[18]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-iclb8/igt@[email protected]
[19]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-iclb2/igt@[email protected]

* igt@gem_exec_balancer@parallel-keep-in-fence:
- shard-iclb: [PASS][20] -> [SKIP][21] ([i915#4525])
[20]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-iclb1/igt@[email protected]
[21]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-iclb3/igt@[email protected]

* igt@gem_exec_fair@basic-deadline:
- shard-glk: [PASS][22] -> [FAIL][23] ([i915#2846])
[22]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-glk8/igt@[email protected]
[23]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-glk8/igt@[email protected]

* igt@gem_exec_fair@basic-flow@rcs0:
- shard-tglb: [PASS][24] -> [FAIL][25] ([i915#2842])
[24]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-tglb3/igt@gem_exec_fair@[email protected]
[25]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-tglb3/igt@gem_exec_fair@[email protected]

* igt@gem_exec_fair@basic-none-share@rcs0:
- shard-iclb: [PASS][26] -> [FAIL][27] ([i915#2842])
[26]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-iclb4/igt@gem_exec_fair@[email protected]
[27]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-iclb1/igt@gem_exec_fair@[email protected]

* igt@gem_exec_fair@basic-none-vip@rcs0:
- shard-glk: [PASS][28] -> [FAIL][29] ([i915#2842])
[28]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-glk5/igt@gem_exec_fair@[email protected]
[29]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-glk2/igt@gem_exec_fair@[email protected]

* igt@gem_exec_fair@basic-pace@rcs0:
- shard-kbl: [PASS][30] -> [FAIL][31] ([i915#2842]) +3 similar issues
[30]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-kbl4/igt@gem_exec_fair@[email protected]
[31]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-kbl7/igt@gem_exec_fair@[email protected]

* igt@gem_exec_fair@basic-pace@vcs1:
- shard-iclb: NOTRUN -> [FAIL][32] ([i915#2842])
[32]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-iclb4/igt@gem_exec_fair@[email protected]

* igt@gem_exec_fair@basic-throttle@rcs0:
- shard-iclb: [PASS][33] -> [FAIL][34] ([i915#2849])
[33]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-iclb3/igt@gem_exec_fair@[email protected]
[34]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-iclb6/igt@gem_exec_fair@[email protected]

* igt@gem_exec_params@no-vebox:
- shard-skl: NOTRUN -> [SKIP][35] ([fdo#109271]) +153 similar issues
[35]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-skl1/igt@[email protected]

* igt@gem_exec_schedule@pi-ringfull@vecs0:
- shard-skl: NOTRUN -> [INCOMPLETE][36] ([i915#4374])
[36]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-skl3/igt@gem_exec_schedule@[email protected]

* igt@gem_exec_whisper@basic-contexts-priority-all:
- shard-glk: [PASS][37] -> [DMESG-WARN][38] ([i915#118])
[37]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-glk6/igt@[email protected]
[38]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-glk4/igt@[email protected]

* igt@gem_lmem_swapping@heavy-random:
- shard-apl: NOTRUN -> [SKIP][39] ([fdo#109271] / [i915#4613]) +1 similar issue
[39]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-apl8/igt@[email protected]

* igt@gem_lmem_swapping@parallel-multi:
- shard-skl: NOTRUN -> [SKIP][40] ([fdo#109271] / [i915#4613]) +1 similar issue
[40]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-skl6/igt@[email protected]

* igt@gem_pread@exhaustion:
- shard-skl: NOTRUN -> [WARN][41] ([i915#2658])
[41]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-skl6/igt@[email protected]

* igt@gem_userptr_blits@dmabuf-sync:
- shard-kbl: NOTRUN -> [SKIP][42] ([fdo#109271] / [i915#3323])
[42]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-kbl1/igt@[email protected]
- shard-skl: NOTRUN -> [SKIP][43] ([fdo#109271] / [i915#3323])
[43]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-skl3/igt@[email protected]

* igt@gem_workarounds@suspend-resume-fd:
- shard-kbl: [PASS][44] -> [DMESG-WARN][45] ([i915#180])
[44]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-kbl3/igt@[email protected]
[45]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-kbl1/igt@[email protected]

* igt@gen9_exec_parse@allowed-single:
- shard-skl: [PASS][46] -> [DMESG-WARN][47] ([i915#1436] / [i915#716])
[46]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-skl9/igt@[email protected]
[47]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-skl8/igt@[email protected]

* igt@i915_pm_dc@dc6-dpms:
- shard-kbl: NOTRUN -> [FAIL][48] ([i915#454])
[48]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-kbl1/igt@[email protected]

* igt@i915_pm_rpm@modeset-lpsp-stress:
- shard-apl: NOTRUN -> [SKIP][49] ([fdo#109271]) +48 similar issues
[49]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-apl4/igt@[email protected]

* igt@i915_pm_rpm@sysfs-read:
- shard-snb: NOTRUN -> [SKIP][50] ([fdo#109271]) +21 similar issues
[50]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-snb4/igt@[email protected]

* igt@kms_async_flips@alternate-sync-async-flip:
- shard-skl: [PASS][51] -> [FAIL][52] ([i915#2521])
[51]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-skl10/igt@[email protected]
[52]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-skl10/igt@[email protected]

* igt@kms_big_fb@x-tiled-max-hw-stride-64bpp-rotate-0-hflip:
- shard-skl: NOTRUN -> [SKIP][53] ([fdo#109271] / [i915#3777])
[53]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-skl1/igt@[email protected]

* igt@kms_big_fb@x-tiled-max-hw-stride-64bpp-rotate-180-hflip:
- shard-apl: NOTRUN -> [SKIP][54] ([fdo#109271] / [i915#3777])
[54]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-apl4/igt@[email protected]

* igt@kms_big_fb@y-tiled-max-hw-stride-32bpp-rotate-180-async-flip:
- shard-skl: NOTRUN -> [FAIL][55] ([i915#3743])
[55]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-skl6/igt@kms_big_fb@y-tiled-max-hw-stride-32bpp-rotate-180-async-flip.html

* igt@kms_ccs@pipe-a-ccs-on-another-bo-y_tiled_gen12_rc_ccs_cc:
- shard-skl: NOTRUN -> [SKIP][56] ([fdo#109271] / [i915#3886]) +6 similar issues
[56]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-skl1/igt@kms_ccs@pipe-a-ccs-on-another-bo-y_tiled_gen12_rc_ccs_cc.html

* igt@kms_ccs@pipe-a-missing-ccs-buffer-y_tiled_gen12_rc_ccs_cc:
- shard-apl: NOTRUN -> [SKIP][57] ([fdo#109271] / [i915#3886]) +2 similar issues
[57]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-apl4/igt@kms_ccs@pipe-a-missing-ccs-buffer-y_tiled_gen12_rc_ccs_cc.html

* igt@kms_ccs@pipe-c-crc-sprite-planes-basic-y_tiled_gen12_mc_ccs:
- shard-tglb: NOTRUN -> [SKIP][58] ([i915#3689] / [i915#3886])
[58]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-tglb5/igt@kms_ccs@pipe-c-crc-sprite-planes-basic-y_tiled_gen12_mc_ccs.html

* igt@kms_ccs@pipe-c-random-ccs-data-y_tiled_gen12_rc_ccs_cc:
- shard-kbl: NOTRUN -> [SKIP][59] ([fdo#109271] / [i915#3886]) +3 similar issues
[59]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-kbl1/igt@kms_ccs@pipe-c-random-ccs-data-y_tiled_gen12_rc_ccs_cc.html

* igt@kms_chamelium@hdmi-mode-timings:
- shard-kbl: NOTRUN -> [SKIP][60] ([fdo#109271] / [fdo#111827]) +5 similar issues
[60]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-kbl7/igt@[email protected]

* igt@kms_color_chamelium@pipe-b-ctm-negative:
- shard-snb: NOTRUN -> [SKIP][61] ([fdo#109271] / [fdo#111827])
[61]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-snb4/igt@[email protected]

* igt@kms_color_chamelium@pipe-c-ctm-0-25:
- shard-apl: NOTRUN -> [SKIP][62] ([fdo#109271] / [fdo#111827]) +5 similar issues
[62]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-apl4/igt@[email protected]

* igt@kms_color_chamelium@pipe-d-ctm-red-to-blue:
- shard-skl: NOTRUN -> [SKIP][63] ([fdo#109271] / [fdo#111827]) +11 similar issues
[63]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-skl1/igt@[email protected]

* igt@kms_cursor_crc@pipe-c-cursor-64x21-onscreen:
- shard-apl: [PASS][64] -> [DMESG-WARN][65] ([i915#62]) +38 similar issues
[64]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-apl3/igt@[email protected]
[65]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-apl7/igt@[email protected]

* igt@kms_cursor_legacy@cursor-vs-flip-toggle:
- shard-apl: [PASS][66] -> [DMESG-WARN][67] ([i915#533] / [i915#62])
[66]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-apl3/igt@[email protected]
[67]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-apl7/igt@[email protected]

* igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions:
- shard-skl: [PASS][68] -> [FAIL][69] ([i915#2346])
[68]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-skl9/igt@[email protected]
[69]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-skl8/igt@[email protected]

* igt@kms_fbcon_fbt@fbc-suspend:
- shard-kbl: [PASS][70] -> [INCOMPLETE][71] ([i915#180] / [i915#636])
[70]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-kbl7/igt@[email protected]
[71]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-kbl1/igt@[email protected]

* igt@kms_flip@flip-vs-expired-vblank@a-edp1:
- shard-skl: [PASS][72] -> [FAIL][73] ([i915#79])
[72]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-skl9/igt@kms_flip@[email protected]
[73]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-skl8/igt@kms_flip@[email protected]

* igt@kms_flip@flip-vs-suspend@b-edp1:
- shard-skl: [PASS][74] -> [INCOMPLETE][75] ([i915#4839])
[74]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-skl4/igt@kms_flip@[email protected]
[75]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-skl4/igt@kms_flip@[email protected]

* igt@kms_flip_scaled_crc@flip-32bpp-ytile-to-32bpp-ytileccs-downscaling:
- shard-iclb: [PASS][76] -> [SKIP][77] ([i915#3701])
[76]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-iclb8/igt@kms_flip_scaled_crc@flip-32bpp-ytile-to-32bpp-ytileccs-downscaling.html
[77]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-iclb2/igt@kms_flip_scaled_crc@flip-32bpp-ytile-to-32bpp-ytileccs-downscaling.html

* igt@kms_flip_scaled_crc@flip-32bpp-ytileccs-to-64bpp-ytile-upscaling:
- shard-glk: [PASS][78] -> [FAIL][79] ([i915#4911]) +1 similar issue
[78]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-glk5/igt@kms_flip_scaled_crc@flip-32bpp-ytileccs-to-64bpp-ytile-upscaling.html
[79]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-glk8/igt@kms_flip_scaled_crc@flip-32bpp-ytileccs-to-64bpp-ytile-upscaling.html

* igt@kms_frontbuffer_tracking@psr-1p-primscrn-indfb-msflip-blt:
- shard-kbl: NOTRUN -> [SKIP][80] ([fdo#109271]) +61 similar issues
[80]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-kbl1/igt@[email protected]

* igt@kms_hdr@bpc-switch-suspend:
- shard-skl: [PASS][81] -> [FAIL][82] ([i915#1188])
[81]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-skl7/igt@[email protected]
[82]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-skl1/igt@[email protected]

* igt@kms_pipe_crc_basic@nonblocking-crc-pipe-d-frame-sequence:
- shard-kbl: NOTRUN -> [SKIP][83] ([fdo#109271] / [i915#533]) +1 similar issue
[83]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-kbl1/igt@[email protected]
- shard-skl: NOTRUN -> [SKIP][84] ([fdo#109271] / [i915#533])
[84]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-skl3/igt@[email protected]

* igt@kms_plane_alpha_blend@pipe-a-constant-alpha-max:
- shard-skl: NOTRUN -> [FAIL][85] ([fdo#108145] / [i915#265]) +1 similar issue
[85]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-skl6/igt@[email protected]

* igt@kms_plane_alpha_blend@pipe-b-alpha-transparent-fb:
- shard-apl: NOTRUN -> [FAIL][86] ([i915#265])
[86]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-apl4/igt@[email protected]

* igt@kms_plane_alpha_blend@pipe-c-alpha-basic:
- shard-kbl: NOTRUN -> [FAIL][87] ([fdo#108145] / [i915#265])
[87]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-kbl1/igt@[email protected]

* igt@kms_psr2_sf@primary-plane-update-sf-dmg-area:
- shard-apl: NOTRUN -> [SKIP][88] ([fdo#109271] / [i915#658])
[88]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-apl8/igt@[email protected]

* igt@kms_psr2_su@page_flip-p010:
- shard-kbl: NOTRUN -> [SKIP][89] ([fdo#109271] / [i915#658])
[89]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-kbl1/igt@kms_psr2_su@page_flip-p010.html

* igt@kms_psr2_su@page_flip-xrgb8888:
- shard-skl: NOTRUN -> [SKIP][90] ([fdo#109271] / [i915#658]) +1 similar issue
[90]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-skl1/igt@kms_psr2_su@page_flip-xrgb8888.html

* igt@kms_psr@psr2_primary_mmap_cpu:
- shard-iclb: [PASS][91] -> [SKIP][92] ([fdo#109441]) +1 similar issue
[91]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-iclb2/igt@kms_psr@psr2_primary_mmap_cpu.html
[92]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-iclb4/igt@kms_psr@psr2_primary_mmap_cpu.html

* igt@kms_setmode@basic:
- shard-glk: [PASS][93] -> [FAIL][94] ([i915#31])
[93]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-glk8/igt@[email protected]
[94]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-glk4/igt@[email protected]

* igt@kms_universal_plane@disable-primary-vs-flip-pipe-b:
- shard-apl: [PASS][95] -> [DMESG-WARN][96] ([i915#1982] / [i915#62])
[95]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-apl4/igt@[email protected]
[96]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-apl7/igt@[email protected]

* igt@sysfs_clients@fair-3:
- shard-kbl: NOTRUN -> [SKIP][97] ([fdo#109271] / [i915#2994]) +1 similar issue
[97]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-kbl1/igt@[email protected]
- shard-skl: NOTRUN -> [SKIP][98] ([fdo#109271] / [i915#2994]) +3 similar issues
[98]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-skl3/igt@[email protected]

#### Possible fixes ####

* igt@gem_eio@in-flight-suspend:
- shard-skl: [INCOMPLETE][99] ([i915#4843]) -> [PASS][100]
[99]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-skl7/igt@[email protected]
[100]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-skl6/igt@[email protected]

* igt@gem_exec_balancer@parallel-out-fence:
- shard-iclb: [SKIP][101] ([i915#4525]) -> [PASS][102]
[101]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-iclb6/igt@[email protected]
[102]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-iclb4/igt@[email protected]

* igt@gem_exec_fair@basic-none-solo@rcs0:
- shard-kbl: [FAIL][103] ([i915#2842]) -> [PASS][104] +5 similar issues
[103]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-kbl6/igt@gem_exec_fair@[email protected]
[104]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-kbl3/igt@gem_exec_fair@[email protected]
- shard-apl: [FAIL][105] ([i915#2842]) -> [PASS][106]
[105]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-apl3/igt@gem_exec_fair@[email protected]
[106]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-apl8/igt@gem_exec_fair@[email protected]

* igt@gem_exec_fair@basic-pace-share@rcs0:
- shard-tglb: [FAIL][107] ([i915#2842]) -> [PASS][108]
[107]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-tglb5/igt@gem_exec_fair@[email protected]
[108]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-tglb5/igt@gem_exec_fair@[email protected]
- shard-glk: [FAIL][109] ([i915#2842]) -> [PASS][110]
[109]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-glk7/igt@gem_exec_fair@[email protected]
[110]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-glk7/igt@gem_exec_fair@[email protected]

* igt@gem_softpin@allocator-evict-all-engines:
- shard-glk: [FAIL][111] ([i915#4171]) -> [PASS][112]
[111]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-glk8/igt@[email protected]
[112]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-glk4/igt@[email protected]

* igt@i915_module_load@reload:
- shard-skl: [DMESG-WARN][113] ([i915#1982]) -> [PASS][114]
[113]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-skl3/igt@[email protected]
[114]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-skl4/igt@[email protected]

* igt@i915_suspend@fence-restore-tiled2untiled:
- shard-snb: [INCOMPLETE][115] -> [PASS][116]
[115]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-snb7/igt@[email protected]
[116]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-snb4/igt@[email protected]

* igt@kms_cursor_crc@pipe-a-cursor-64x64-onscreen:
- shard-glk: [DMESG-WARN][117] ([i915#118]) -> [PASS][118]
[117]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-glk9/igt@[email protected]
[118]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-glk3/igt@[email protected]

* igt@kms_cursor_crc@pipe-c-cursor-suspend:
- shard-kbl: [DMESG-WARN][119] ([i915#180]) -> [PASS][120] +1 similar issue
[119]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-kbl1/igt@[email protected]
[120]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-kbl7/igt@[email protected]

* igt@kms_flip_scaled_crc@flip-32bpp-ytile-to-64bpp-ytile-downscaling:
- shard-iclb: [SKIP][121] ([i915#3701]) -> [PASS][122]
[121]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-iclb2/igt@kms_flip_scaled_crc@flip-32bpp-ytile-to-64bpp-ytile-downscaling.html
[122]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-iclb4/igt@kms_flip_scaled_crc@flip-32bpp-ytile-to-64bpp-ytile-downscaling.html

* igt@kms_frontbuffer_tracking@fbc-suspend:
- shard-apl: [DMESG-WARN][123] ([i915#180]) -> [PASS][124] +3 similar issues
[123]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-apl8/igt@[email protected]
[124]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-apl8/igt@[email protected]

* igt@kms_hdr@bpc-switch:
- shard-skl: [FAIL][125] ([i915#1188]) -> [PASS][126]
[125]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-skl7/igt@[email protected]
[126]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-skl9/igt@[email protected]

* igt@kms_psr@psr2_cursor_plane_move:
- shard-iclb: [SKIP][127] ([fdo#109441]) -> [PASS][128] +1 similar issue
[127]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-iclb8/igt@kms_psr@psr2_cursor_plane_move.html
[128]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-iclb2/igt@kms_psr@psr2_cursor_plane_move.html

* igt@perf@polling:
- shard-skl: [FAIL][129] ([i915#1542]) -> [PASS][130]
[129]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-skl9/igt@[email protected]
[130]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-skl8/igt@[email protected]

* igt@perf@polling-parameterized:
- shard-glk: [FAIL][131] ([i915#1542]) -> [PASS][132]
[131]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-glk6/igt@[email protected]
[132]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-glk7/igt@[email protected]

#### Warnings ####

* igt@i915_pm_dc@dc3co-vpb-simulation:
- shard-iclb: [SKIP][133] ([i915#658]) -> [SKIP][134] ([i915#588])
[133]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-iclb1/igt@[email protected]
[134]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-iclb2/igt@[email protected]

* igt@kms_plane_alpha_blend@pipe-b-alpha-basic:
- shard-apl: [FAIL][135] ([fdo#108145] / [i915#265]) -> [DMESG-FAIL][136] ([fdo#108145] / [i915#265] / [i915#62])
[135]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-apl3/igt@[email protected]
[136]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-apl7/igt@[email protected]

* igt@kms_psr2_sf@overlay-plane-update-continuous-sf:
- shard-iclb: [SKIP][137] ([i915#2920]) -> [SKIP][138] ([fdo#111068] / [i915#658])
[137]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-iclb2/igt@[email protected]
[138]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-iclb5/igt@[email protected]

* igt@kms_psr2_su@page_flip-xrgb8888:
- shard-iclb: [SKIP][139] ([fdo#109642] / [fdo#111068] / [i915#658]) -> [FAIL][140] ([i915#4148])
[139]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-iclb1/igt@kms_psr2_su@page_flip-xrgb8888.html
[140]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-iclb2/igt@kms_psr2_su@page_flip-xrgb8888.html

* igt@kms_sysfs_edid_timing:
- shard-apl: [FAIL][141] ([IGT#2]) -> [DMESG-FAIL][142] ([IGT#2] / [i915#62])
[141]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-apl3/igt@kms_sysfs_edid_timing.html
[142]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-apl7/igt@kms_sysfs_edid_timing.html

* igt@runner@aborted:
- shard-kbl: ([FAIL][143], [FAIL][144], [FAIL][145], [FAIL][146]) ([i915#1436] / [i915#180] / [i915#1814] / [i915#3002] / [i915#4312]) -> ([FAIL][147], [FAIL][148], [FAIL][149], [FAIL][150]) ([i915#180] / [i915#3002] / [i915#4312] / [i915#92])
[143]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-kbl6/igt@[email protected]
[144]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-kbl1/igt@[email protected]
[145]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-kbl1/igt@[email protected]
[146]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-kbl1/igt@[email protected]
[147]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-kbl1/igt@[email protected]
[148]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-kbl1/igt@[email protected]
[149]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-kbl1/igt@[email protected]
[150]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/shard-kbl1/igt@[email protected]
- shard-skl: ([FAIL][151], [FAIL][152], [FAIL][153], [FAIL][154]) ([i915#2029] / [i915#3002] / [i915#4312]) -> ([FAIL][155], [FAIL][156], [FAIL][157], [FAIL][158], [FAIL][159]) ([i915#1436] / [i915#2426] / [i915#3002] / [i915#4312])
[151]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-skl4/igt@[email protected]
[152]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11115/shard-skl8/igt@[email protected]
[153]: https://intel-gfx-ci.01.or

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_22048/index.html

#6 lore raw
In reply to Casey Bowman

CC'ing more reviewers for comments.

Quoted 39 lines · Casey Bowman

In this RFC I would like to ask the community their thoughts on how we can best handle splitting architecture-specific calls.

I would like to address the following:

1. How do we want to split architecture calls? Different object files per platform? Separate function calls within the same object file?

2. How do we address dummy functions? If we have a function call that is used for one or more platforms, but is not used in another, what should we do for this case?

I've given an example of splitting an architecture call in my patch with run_as_guest() being split into different implementations for x86 and arm64 in separate object files, sharing a single header.

Another suggestion from Michael ([email protected]) involved using a single object file, a single header, and splitting various functions calls via ifdefs in the header file.

I would appreciate any input on how we can avoid scaling issues when including multiple architectures and multiple functions (as the number of function calls will inevitably increase with more architectures).

Casey Bowman (1):
i915/drm: Split out x86 and arm64 functionality

  drivers/gpu/drm/i915/Makefile              |  4 +++
  drivers/gpu/drm/i915/i915_drv.h            |  6 +---
  drivers/gpu/drm/i915/i915_platform.h       | 16 +++++++++++
  drivers/gpu/drm/i915/i915_platform_arm64.c | 33 ++++++++++++++++++++++
  drivers/gpu/drm/i915/i915_platform_x86.c   | 33 ++++++++++++++++++++++
  5 files changed, 87 insertions(+), 5 deletions(-)

create mode 100644 drivers/gpu/drm/i915/i915_platform.h create mode 100644 drivers/gpu/drm/i915/i915_platform_arm64.c create mode 100644 drivers/gpu/drm/i915/i915_platform_x86.c

#7 lore raw
In reply to Casey Bowman
Quoted 18 lines · Casey Bowman

In this RFC I would like to ask the community their thoughts on how we can best handle splitting architecture-specific calls.

I would like to address the following:

1. How do we want to split architecture calls? Different object files per platform? Separate function calls within the same object file?

2. How do we address dummy functions? If we have a function call that is used for one or more platforms, but is not used in another, what should we do for this case?

I've given an example of splitting an architecture call in my patch with run_as_guest() being split into different implementations for x86 and arm64 in separate object files, sharing a single header.

Quoted 7 lines

Another suggestion from Michael ([email protected]) involved using a single object file, a single header, and splitting various functions calls via ifdefs in the header file.

I would appreciate any input on how we can avoid scaling issues when including multiple architectures and multiple functions (as the number of function calls will inevitably increase with more architectures).

Personally I think the functionality is best kept organized by, well, functionality, not by platform. Otherwise the platform files will contain all sorts of code with the only common denominator being the platform.

You're also likely to have static platform specific functions, which would needlessly have to be made non-static if the split was by files.

I'd just put the implementations for different platforms next to each other:

#if IS_ENABLED(CONFIG_X86)
...
#elif IS_ENABLED(CONFIG_ARM64)
...
#endif

Usually the declarations would be identical and there'd only be one, without #ifs in the header.

BR,
Jani.

Quoted 13 lines

Casey Bowman (1):
i915/drm: Split out x86 and arm64 functionality

 drivers/gpu/drm/i915/Makefile              |  4 +++
 drivers/gpu/drm/i915/i915_drv.h            |  6 +---
 drivers/gpu/drm/i915/i915_platform.h       | 16 +++++++++++
 drivers/gpu/drm/i915/i915_platform_arm64.c | 33 ++++++++++++++++++++++
 drivers/gpu/drm/i915/i915_platform_x86.c   | 33 ++++++++++++++++++++++
 5 files changed, 87 insertions(+), 5 deletions(-)
 create mode 100644 drivers/gpu/drm/i915/i915_platform.h
 create mode 100644 drivers/gpu/drm/i915/i915_platform_arm64.c
 create mode 100644 drivers/gpu/drm/i915/i915_platform_x86.c
Signature · 1 line
Jani Nikula, Intel Open Source Graphics Center
#9 lore raw
In reply to Casey Bowman
Quoted 8 lines · Casey Bowman

In this RFC I would like to ask the community their thoughts on how we can best handle splitting architecture-specific calls.

I would like to address the following:

1. How do we want to split architecture calls? Different object files per platform? Separate function calls within the same object file?

If we are talking about per-platform divergence of significant functions (not necessarily in size but their height position in the i915 stack) I agree with Jani that top-level per platform organisation is not the best choice.

On the other hand I doubt that there should be many, if any, such functions. In practice I think it should be only low level stuff which diverges.

On a concrete example..

Quoted 3 lines

2. How do we address dummy functions? If we have a function call that is used for one or more platforms, but is not used in another, what should we do for this case?

... depends on the situation. Sometimes a flavour of "warn on once" can be okay I guess, but also why not build bug on? Because..

Quoted 5 lines

I've given an example of splitting an architecture call in my patch with run_as_guest() being split into different implementations for x86 and arm64 in separate object files, sharing a single header.

... run_as_guest may be a very tricky example, given it is called from intel_vtd_active which has a number of callers.

What is correct behaviour on Arm in this example? None of these call sites will run on Arm? Or that you expect the warn on added in this patch to trigger as a demonstration? If so, what is the plan going forward? We can take one example and talk about it hypothetically:

./i915_driver.c: drm_printf(p, "iommu: %s\n", enableddisabled(intel_vtd_active(i915)));

What is the "fix" (refactor) for Arm here? Looks like a new top-level function is needed which does not carry the intel_vtd_ prefix but something more generic. That one could then legitimately "warn on once", while for intel_vtd_active it would be wrong to do so.

And when I say it is needed.. well perhaps it is not strictly needed in this case, but in some other cases I think we go back to the problem I stated some months ago and that is that I suspect use of intel_vtd_active is overloaded. I think it is currently used to answer all these questions: 1. Is the IOMMU active, just for information.; 2. Is the IOMMU active and we want to counteract the performance hit by say using huge pages, adjusting the display bandwidth calculations or whatever. (In which case we also may want to distinguish between pass-through and translation modes.); 3. Is a potentially buggy IOMMU active and we need to work around it. All these under one kind of worked with one iommu implementation but does it with a different IOMMU?

Which I mean leads to end conclusion that this particular function is a tricky example to answer the questions asked. :)

Quoted 4 lines

Another suggestion from Michael ([email protected]) involved using a single object file, a single header, and splitting various functions calls via ifdefs in the header file.

In principle, mostly what you have outlined sounds acceptable to me, with the difference that I would not use i915_platform, but for this particular example something like i915_hypervisor prefix.

Then I would prepare i915 with the same scheme kernel uses, not just for source file divergence, but header file as well. That is:

some_source.c:

#include "i915_hypervisor.h"

i915_hypervisor.h:

#include "platform/i915_hypervisor.h"

Then in i915 root you could have:

platforms/x86/include/platform/i915_hypervisor.h
platforms/arm/include/platform/i915_hypervisor.h

And some kbuild stuff to make that work. Is this doable and does it make sense? Per-platform source files could live in there as well.

Same scheme for i915_clflush would work as well.

Regards,

Tvrtko

Quoted 18 lines

I would appreciate any input on how we can avoid scaling issues when including multiple architectures and multiple functions (as the number of function calls will inevitably increase with more architectures).

Casey Bowman (1):
i915/drm: Split out x86 and arm64 functionality

  drivers/gpu/drm/i915/Makefile              |  4 +++
  drivers/gpu/drm/i915/i915_drv.h            |  6 +---
  drivers/gpu/drm/i915/i915_platform.h       | 16 +++++++++++
  drivers/gpu/drm/i915/i915_platform_arm64.c | 33 ++++++++++++++++++++++
  drivers/gpu/drm/i915/i915_platform_x86.c   | 33 ++++++++++++++++++++++
  5 files changed, 87 insertions(+), 5 deletions(-)

create mode 100644 drivers/gpu/drm/i915/i915_platform.h create mode 100644 drivers/gpu/drm/i915/i915_platform_arm64.c create mode 100644 drivers/gpu/drm/i915/i915_platform_x86.c

#8 lore raw
In reply to Jani Nikula
Quoted 37 lines · Jani Nikula
Quoted 18 lines · Casey Bowman

In this RFC I would like to ask the community their thoughts on how we can best handle splitting architecture-specific calls.

I would like to address the following:

1. How do we want to split architecture calls? Different object files per platform? Separate function calls within the same object file?

2. How do we address dummy functions? If we have a function call that is used for one or more platforms, but is not used in another, what should we do for this case?

I've given an example of splitting an architecture call in my patch with run_as_guest() being split into different implementations for x86 and arm64 in separate object files, sharing a single header.

Quoted 7 lines

Another suggestion from Michael ([email protected]) involved using a single object file, a single header, and splitting various functions calls via ifdefs in the header file.

I would appreciate any input on how we can avoid scaling issues when including multiple architectures and multiple functions (as the number of function calls will inevitably increase with more architectures).

Personally I think the functionality is best kept organized by, well, functionality, not by platform. Otherwise the platform files will contain all sorts of code with the only common denominator being the platform.

You're also likely to have static platform specific functions, which would needlessly have to be made non-static if the split was by files.

I'd just put the implementations for different platforms next to each other:

Quoted 8 lines

#if IS_ENABLED(CONFIG_X86)
...
#elif IS_ENABLED(CONFIG_ARM64)
...
#endif

Usually the declarations would be identical and there'd only be one, without #ifs in the header.

Thanks for the feedback, Jani.

I think this is the proper takeaway for future functions that do have separate implementations for differing architectures.

As for null behavior, as in the example I gave, I think Tvrtko is right about run_as_guest being a tricky example. I think I need to re-evaluate that function and think of another solution altogether for that instance.

I think this will also be the precedent for null cases, where we will need to rethink implementations for cases that don't really have some arch-specific implementation other than returning null (though some exceptions may exist).

Quoted 16 lines

BR,
Jani.

Quoted 12 lines

Casey Bowman (1):
i915/drm: Split out x86 and arm64 functionality

  drivers/gpu/drm/i915/Makefile              |  4 +++
  drivers/gpu/drm/i915/i915_drv.h            |  6 +---
  drivers/gpu/drm/i915/i915_platform.h       | 16 +++++++++++
  drivers/gpu/drm/i915/i915_platform_arm64.c | 33 ++++++++++++++++++++++
  drivers/gpu/drm/i915/i915_platform_x86.c   | 33 ++++++++++++++++++++++
  5 files changed, 87 insertions(+), 5 deletions(-)

create mode 100644 drivers/gpu/drm/i915/i915_platform.h create mode 100644 drivers/gpu/drm/i915/i915_platform_arm64.c create mode 100644 drivers/gpu/drm/i915/i915_platform_x86.c

#10 lore raw
In reply to Tvrtko Ursulin
Quoted 6 lines · Tvrtko Ursulin
Quoted 4 lines · Casey Bowman

In this RFC I would like to ask the community their thoughts on how we can best handle splitting architecture-specific calls.

Quoted 8 lines
Quoted 4 lines

I would like to address the following:

1. How do we want to split architecture calls? Different object files per platform? Separate function calls within the same object file?

If we are talking about per-platform divergence of significant functions (not necessarily in size but their height position in the i915 stack) I agree with Jani that top-level per platform organisation is not the best choice.

On the other hand I doubt that there should be many, if any, such functions. In practice I think it should be only low level stuff which diverges.

I agree, as said with my reply to Jani, I think maybe going forward for arch-specific code, #if IS_ENABLED calls should be used sparingly, only in the cases where we do have that arch-specific implementation (like low level calls), instead of just a 'return null', as in my case.

Quoted 8 lines

On a concrete example..

Quoted 3 lines

2. How do we address dummy functions? If we have a function call that is used for one or more platforms, but is not used in another, what should we do for this case?

... depends on the situation. Sometimes a flavour of "warn on once" can be okay I guess, but also why not build bug on? Because..

True, with Jani's and your comments, I'm thinking that in the case of when we look at a potential arch-specific function where we're just returning null or something similar, we should be re-evaluating the function and seeing if we should make a different change there.

Quoted 7 lines
Quoted 6 lines

I've given an example of splitting an architecture call in my patch with run_as_guest() being split into different implementations for x86 and arm64 in separate object files, sharing a single header.

Quoted 7 lines

... run_as_guest may be a very tricky example, given it is called from intel_vtd_active which has a number of callers.

What is correct behaviour on Arm in this example? None of these call sites will run on Arm? Or that you expect the warn on added in this patch to trigger as a demonstration? If so, what is the plan going forward? We can take one example and talk about it hypothetically:

./i915_driver.c:        drm_printf(p, "iommu: %s\n", enableddisabled(intel_vtd_active(i915)));

What is the "fix" (refactor) for Arm here? Looks like a new top-level function is needed which does not carry the intel_vtd_ prefix but something more generic. That one could then legitimately "warn on once", while for intel_vtd_active it would be wrong to do so.

Good point, run_as_guest might be a bit more challenging of an example.

In my mind, I was thinking of just simply returning null for arm64 here as I don't believe arm64 would be making use of iommu for our cases (at least, in the short term). I think this example function specifically needs to get reworked, as you mentioned, in some way, possibly refactoring intel_vtd_active or something along those lines.

Quoted 22 lines

And when I say it is needed.. well perhaps it is not strictly needed in this case, but in some other cases I think we go back to the problem I stated some months ago and that is that I suspect use of intel_vtd_active is overloaded. I think it is currently used to answer all these questions: 1. Is the IOMMU active, just for information.; 2. Is the IOMMU active and we want to counteract the performance hit by say using huge pages, adjusting the display bandwidth calculations or whatever. (In which case we also may want to distinguish between pass-through and translation modes.); 3. Is a potentially buggy IOMMU active and we need to work around it. All these under one kind of worked with one iommu implementation but does it with a different IOMMU?

Which I mean leads to end conclusion that this particular function is a tricky example to answer the questions asked. :)

Quoted 4 lines

Another suggestion from Michael ([email protected]) involved using a single object file, a single header, and splitting various functions calls via ifdefs in the header file.

In principle, mostly what you have outlined sounds acceptable to me, with the difference that I would not use i915_platform, but for this particular example something like i915_hypervisor prefix.

Then I would prepare i915 with the same scheme kernel uses, not just for source file divergence, but header file as well. That is:

some_source.c:

#include "i915_hypervisor.h"

i915_hypervisor.h:

#include "platform/i915_hypervisor.h"

Quoted 8 lines

Then in i915 root you could have:

platforms/x86/include/platform/i915_hypervisor.h
platforms/arm/include/platform/i915_hypervisor.h

And some kbuild stuff to make that work. Is this doable and does it make sense? Per-platform source files could live in there as well.

Same scheme for i915_clflush would work as well.

I like the idea of getting more specific for the calls here, but I'm somewhat afraid of obfuscating these functions to their own files in addition to scaling issues if we do have more and more arch-specific calls, along with more architectures in the future.

This just seems like it could blow up the driver in what could ultimately be unnecessary organization for a fewer number of calls if we just add a few platforms and different calls. What do you think?

Regards,
Casey

Quoted 23 lines

Regards,

Tvrtko

Quoted 18 lines

I would appreciate any input on how we can avoid scaling issues when including multiple architectures and multiple functions (as the number of function calls will inevitably increase with more architectures).

Casey Bowman (1):
   i915/drm: Split out x86 and arm64 functionality

  drivers/gpu/drm/i915/Makefile              |  4 +++ drivers/gpu/drm/i915/i915_drv.h            |  6 +--- drivers/gpu/drm/i915/i915_platform.h       | 16 +++++++++++ drivers/gpu/drm/i915/i915_platform_arm64.c | 33 ++++++++++++++++++++++ drivers/gpu/drm/i915/i915_platform_x86.c   | 33 ++++++++++++++++++++++

  5 files changed, 87 insertions(+), 5 deletions(-)

  create mode 100644 drivers/gpu/drm/i915/i915_platform.h create mode 100644 drivers/gpu/drm/i915/i915_platform_arm64.c create mode 100644 drivers/gpu/drm/i915/i915_platform_x86.c

#11 lore raw
Quoted 73 lines · Casey Bowman
Quoted 6 lines · Tvrtko Ursulin
Quoted 4 lines · Casey Bowman

In this RFC I would like to ask the community their thoughts on how we can best handle splitting architecture-specific calls.

Quoted 8 lines
Quoted 4 lines

I would like to address the following:

1. How do we want to split architecture calls? Different object files per platform? Separate function calls within the same object file?

If we are talking about per-platform divergence of significant functions (not necessarily in size but their height position in the i915 stack) I agree with Jani that top-level per platform organisation is not the best choice.

On the other hand I doubt that there should be many, if any, such functions. In practice I think it should be only low level stuff which diverges.

I agree, as said with my reply to Jani, I think maybe going forward for arch-specific code, #if IS_ENABLED calls should be used sparingly, only in the cases where we do have that arch-specific implementation (like low level calls), instead of just a 'return null', as in my case.

Quoted 8 lines

On a concrete example..

Quoted 3 lines

2. How do we address dummy functions? If we have a function call that is used for one or more platforms, but is not used in another, what should we do for this case?

... depends on the situation. Sometimes a flavour of "warn on once" can be okay I guess, but also why not build bug on? Because..

True, with Jani's and your comments, I'm thinking that in the case of when we look at a potential arch-specific function where we're just returning null or something similar, we should be re-evaluating the function and seeing if we should make a different change there.

Quoted 7 lines
Quoted 6 lines

I've given an example of splitting an architecture call in my patch with run_as_guest() being split into different implementations for x86 and arm64 in separate object files, sharing a single header.

Quoted 7 lines

... run_as_guest may be a very tricky example, given it is called from intel_vtd_active which has a number of callers.

What is correct behaviour on Arm in this example? None of these call sites will run on Arm? Or that you expect the warn on added in this patch to trigger as a demonstration? If so, what is the plan going forward? We can take one example and talk about it hypothetically:

./i915_driver.c:        drm_printf(p, "iommu: %s\n", enableddisabled(intel_vtd_active(i915)));

What is the "fix" (refactor) for Arm here? Looks like a new top-level function is needed which does not carry the intel_vtd_ prefix but something more generic. That one could then legitimately "warn on once", while for intel_vtd_active it would be wrong to do so.

Good point, run_as_guest might be a bit more challenging of an example.

In my mind, I was thinking of just simply returning null for arm64 here as I don't believe arm64 would be making use of iommu for our cases (at least, in the short term). I think this example function specifically needs to get reworked, as you mentioned, in some way, possibly refactoring intel_vtd_active or something along those lines.

Quoted 27 lines

And when I say it is needed.. well perhaps it is not strictly needed in this case, but in some other cases I think we go back to the problem I stated some months ago and that is that I suspect use of intel_vtd_active is overloaded. I think it is currently used to answer all these questions: 1. Is the IOMMU active, just for information.; 2. Is the IOMMU active and we want to counteract the performance hit by say using huge pages, adjusting the display bandwidth calculations or whatever. (In which case we also may want to distinguish between pass-through and translation modes.); 3. Is a potentially buggy IOMMU active and we need to work around it. All these under one kind of worked with one iommu implementation but does it with a different IOMMU?

Which I mean leads to end conclusion that this particular function is a tricky example to answer the questions asked. :)

Quoted 4 lines

Another suggestion from Michael ([email protected]) involved using a single object file, a single header, and splitting various functions calls via ifdefs in the header file.

In principle, mostly what you have outlined sounds acceptable to me, with the difference that I would not use i915_platform, but for this particular example something like i915_hypervisor prefix.

Then I would prepare i915 with the same scheme kernel uses, not just for source file divergence, but header file as well. That is:

some_source.c:

#include "i915_hypervisor.h"

i915_hypervisor.h:

#include "platform/i915_hypervisor.h"

Then in i915 root you could have:

platforms/x86/include/platform/i915_hypervisor.h
platforms/arm/include/platform/i915_hypervisor.h

Quoted 8 lines
Quoted 3 lines

And some kbuild stuff to make that work. Is this doable and does it make sense? Per-platform source files could live in there as well.

Same scheme for i915_clflush would work as well.

I like the idea of getting more specific for the calls here, but I'm somewhat afraid of obfuscating these functions to their own files in addition to scaling issues if we do have more and more arch-specific calls, along with more architectures in the future.

This just seems like it could blow up the driver in what could ultimately be unnecessary organization for a fewer number of calls if we just add a few platforms and different calls. What do you think?

I don't have a good crystal ball to see how many of these you would end up during the Arm porting effort, just my gut feeling that it shouldn't be a problem to add a few files.

And I don't think it would be obfuscating anything, on the contrary it makes things very clear in respect where the platform dependent boundary is and what are all the bits that diverge per platforms.

Personally I would just do this all as part of your other series which touches clflush and in that way establish a pattern from the start. Plus that way I sleep easy not thinking how we started penalizing old platforms with needless new function calls on hot paths. But it is my opinion only and other people may think differently.

Regards,

Tvrtko

#12 lore raw
Quoted 79 lines · Tvrtko Ursulin
Quoted 77 lines · Casey Bowman
Quoted 6 lines · Tvrtko Ursulin
Quoted 4 lines · Casey Bowman

In this RFC I would like to ask the community their thoughts on how we can best handle splitting architecture-specific calls.

Quoted 8 lines
Quoted 4 lines

I would like to address the following:

1. How do we want to split architecture calls? Different object files per platform? Separate function calls within the same object file?

If we are talking about per-platform divergence of significant functions (not necessarily in size but their height position in the i915 stack) I agree with Jani that top-level per platform organisation is not the best choice.

On the other hand I doubt that there should be many, if any, such functions. In practice I think it should be only low level stuff which diverges.

I agree, as said with my reply to Jani, I think maybe going forward for arch-specific code, #if IS_ENABLED calls should be used sparingly, only in the cases where we do have that arch-specific implementation (like low level calls), instead of just a 'return null', as in my case.

Quoted 8 lines

On a concrete example..

Quoted 3 lines

2. How do we address dummy functions? If we have a function call that is used for one or more platforms, but is not used in another, what should we do for this case?

... depends on the situation. Sometimes a flavour of "warn on once" can be okay I guess, but also why not build bug on? Because..

True, with Jani's and your comments, I'm thinking that in the case of when we look at a potential arch-specific function where we're just returning null or something similar, we should be re-evaluating the function and seeing if we should make a different change there.

Quoted 7 lines
Quoted 6 lines

I've given an example of splitting an architecture call in my patch with run_as_guest() being split into different implementations for x86 and arm64 in separate object files, sharing a single header.

Quoted 7 lines

... run_as_guest may be a very tricky example, given it is called from intel_vtd_active which has a number of callers.

What is correct behaviour on Arm in this example? None of these call sites will run on Arm? Or that you expect the warn on added in this patch to trigger as a demonstration? If so, what is the plan going forward? We can take one example and talk about it hypothetically:

./i915_driver.c:        drm_printf(p, "iommu: %s\n", enableddisabled(intel_vtd_active(i915)));

What is the "fix" (refactor) for Arm here? Looks like a new top-level function is needed which does not carry the intel_vtd_ prefix but something more generic. That one could then legitimately "warn on once", while for intel_vtd_active it would be wrong to do so.

Good point, run_as_guest might be a bit more challenging of an example.

In my mind, I was thinking of just simply returning null for arm64 here as I don't believe arm64 would be making use of iommu for our cases (at least, in the short term). I think this example function specifically needs to get reworked, as you mentioned, in some way, possibly refactoring intel_vtd_active or something along those lines.

Quoted 30 lines

And when I say it is needed.. well perhaps it is not strictly needed in this case, but in some other cases I think we go back to the problem I stated some months ago and that is that I suspect use of intel_vtd_active is overloaded. I think it is currently used to answer all these questions: 1. Is the IOMMU active, just for information.; 2. Is the IOMMU active and we want to counteract the performance hit by say using huge pages, adjusting the display bandwidth calculations or whatever. (In which case we also may want to distinguish between pass-through and translation modes.); 3. Is a potentially buggy IOMMU active and we need to work around it. All these under one kind of worked with one iommu implementation but does it with a different IOMMU?

Which I mean leads to end conclusion that this particular function is a tricky example to answer the questions asked. :)

Quoted 4 lines

Another suggestion from Michael ([email protected]) involved using a single object file, a single header, and splitting various functions calls via ifdefs in the header file.

In principle, mostly what you have outlined sounds acceptable to me, with the difference that I would not use i915_platform, but for this particular example something like i915_hypervisor prefix.

Then I would prepare i915 with the same scheme kernel uses, not just for source file divergence, but header file as well. That is:

some_source.c:

#include "i915_hypervisor.h"

i915_hypervisor.h:

#include "platform/i915_hypervisor.h"

Then in i915 root you could have:

platforms/x86/include/platform/i915_hypervisor.h
platforms/arm/include/platform/i915_hypervisor.h

And some kbuild stuff to make that work. Is this doable and does it make sense? Per-platform source files could live in there as well.

Same scheme for i915_clflush would work as well.

Quoted 8 lines
Quoted 4 lines

I like the idea of getting more specific for the calls here, but I'm somewhat afraid of obfuscating these functions to their own files in addition to scaling issues if we do have more and more arch-specific calls, along with more architectures in the future.

This just seems like it could blow up the driver in what could ultimately be unnecessary organization for a fewer number of calls if we just add a few platforms and different calls. What do you think?

I don't have a good crystal ball to see how many of these you would end up during the Arm porting effort, just my gut feeling that it shouldn't be a problem to add a few files.

And I don't think it would be obfuscating anything, on the contrary it makes things very clear in respect where the platform dependent boundary is and what are all the bits that diverge per platforms.

Fair enough, I'll look into writing up something as an update for further review.

Quoted 2 lines

Personally I would just do this all as part of your other series which touches clflush and in that way establish a pattern from the start. Plus that way I sleep easy not thinking how we started penalizing old platforms with needless new function calls on hot paths. But it is my opinion only and other people may think differently.

I believe that series is authored by Michael, not myself :P
But I do agree with trying to establishing a precedent here for future, similar calls.
It would definitely be helpful to get more opinions for consensus on this, as we know there's many ways to go about this.

In the meantime, to form a basic guideline that others can help mold, I believe the precedent as of the moment for porting i915 functionality for other architectures is:
- If the functionality has low-level code that must be ported to another arch, use #if IS_ENABLED to split out the function calls by arch config (this should be sparingly used).
- If the functionality is specific to x86 architecture (has library calls that only pertain to x86), split out the call to a platform dependent area (so long as there isn't an already-existing generic call available that can be used in a refactor).

Does this sort of guideline make sense/sound correct? I want to make something we can reference when porting various pieces of functionality throughout the driver.

Regards,
Casey