[Intel-gfx] [PULL] gvt-fixes

Hi, Here is one gvt fix for bug in AUX CH register message length get. Please help to pick. Thanks! -- The following changes since commit e354f67733115b4453268f61e6e072e9b1ea7a2f: drm/i915: Fix an error handling path in igt_write_huge() (2023-07-25 08:38:12 +0100) are available in the Git

Re: [Intel-gfx] [PATCH v4 3/7] drm/i915/vdsc: Add function to read any PPS register

Hi Suraj, I agree with overall approach, but I suggest that we use num of vdsc engines per pipe to avoid changes in future. This would perhaps need a small patch first to get num of vdsc engines per pipe though, more details inline for the case: On 7/20/2023 1:37 PM, Suraj Kandpal wrote:

Re: [Intel-gfx] [PATCH v4 2/7] drm/i915/vdsc: Add a check for dsc split cases

On 7/20/2023 1:37 PM, Suraj Kandpal wrote: In intel_vdsc_get_config we only read the primary dsc engine register and not take into account if the other dsc engine is in use and if both registers have the same value or not this patche fixes that by adding a check. --v3 -Remove superfluos new

[Intel-gfx] linux-next: build warning after merge of the drm-misc tree

Hi all, After merging the drm-misc tree, today's linux-next build (htmldocs) produced this warning: include/drm/drm_panel.h:270: warning: Function parameter or member 'follower_lock' not described in 'drm_panel' Introduced by commit de0874165b83 ("drm/panel: Add a way for other devices to

[Intel-gfx] linux-next: build warning after merge of the drm-misc tree

Hi all, After merging the drm-misc tree, today's linux-next build (htmldocs) produced this warning: Documentation/gpu/todo.rst:469: ERROR: Unexpected indentation. Introduced by commit d2aacaf07395 ("drm/panel: Check for already prepared/enabled in drm_panel") -- Cheers, Stephen Rothwell

Re: [Intel-gfx] [PATCH v4.1] drm/i915/gvt: Explicitly check that vGPU is attached before shadowing

Reviewed-by: Yan Zhao Tested-by: Yan Zhao On Tue, Aug 01, 2023 at 04:05:21PM -0700, Sean Christopherson wrote: > Move the check that a vGPU is attached from is_2MB_gtt_possible() all the > way up to shadow_ppgtt_mm() to avoid unnecessary work, and to make it more > obvious that a future cleanup

[Intel-gfx] ✗ Fi.CI.IGT: failure for drm/i915/gvt: KVM: KVMGT fixes and page-track cleanups (rev10)

== Series Details == Series: drm/i915/gvt: KVM: KVMGT fixes and page-track cleanups (rev10) URL : https://patchwork.freedesktop.org/series/112196/ State : failure == Summary == CI Bug Log - changes from CI_DRM_13457_full -> Patchwork_112196v10_full

[Intel-gfx] ✓ Fi.CI.BAT: success for drm/i915/gvt: KVM: KVMGT fixes and page-track cleanups (rev10)

== Series Details == Series: drm/i915/gvt: KVM: KVMGT fixes and page-track cleanups (rev10) URL : https://patchwork.freedesktop.org/series/112196/ State : success == Summary == CI Bug Log - changes from CI_DRM_13457 -> Patchwork_112196v10

[Intel-gfx] ✓ Fi.CI.IGT: success for drm/i915: Fix Wa_22016122933 implementation (rev2)

== Series Details == Series: drm/i915: Fix Wa_22016122933 implementation (rev2) URL : https://patchwork.freedesktop.org/series/121735/ State : success == Summary == CI Bug Log - changes from CI_DRM_13457_full -> Patchwork_121735v2_full

[Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for drm/i915/gvt: KVM: KVMGT fixes and page-track cleanups (rev10)

== Series Details == Series: drm/i915/gvt: KVM: KVMGT fixes and page-track cleanups (rev10) URL : https://patchwork.freedesktop.org/series/112196/ State : warning == Summary == Error: dim checkpatch failed 3f274ee4efc5 drm/i915/gvt: Verify pfn is "valid" before dereferencing "struct page"

Re: [Intel-gfx] [PATCH v4 06/29] drm/i915/gvt: Explicitly check that vGPU is attached before shadowing

On Tue, Aug 01, 2023, Yan Zhao wrote: > On Fri, Jul 28, 2023 at 06:35:12PM -0700, Sean Christopherson wrote: > > Move the check that a vGPU is attacked from is_2MB_gtt_possible() all the > typo: "attacked" --> "attached" > > > way up to shadow_ppgtt_mm() to avoid unnecessary work, and to make it

[Intel-gfx] [PATCH v4.1] drm/i915/gvt: Explicitly check that vGPU is attached before shadowing

Move the check that a vGPU is attached from is_2MB_gtt_possible() all the way up to shadow_ppgtt_mm() to avoid unnecessary work, and to make it more obvious that a future cleanup of is_2MB_gtt_possible() isn't introducing a bug. is_2MB_gtt_possible() has only one caller,

[Intel-gfx] ✓ Fi.CI.IGT: success for Invalidate TLB cache for all GT's

== Series Details == Series: Invalidate TLB cache for all GT's URL : https://patchwork.freedesktop.org/series/121732/ State : success == Summary == CI Bug Log - changes from CI_DRM_13457_full -> Patchwork_121732v1_full Summary ---

[Intel-gfx] ✓ Fi.CI.IGT: success for Replace acronym with full platform name in defines

== Series Details == Series: Replace acronym with full platform name in defines URL : https://patchwork.freedesktop.org/series/121731/ State : success == Summary == CI Bug Log - changes from CI_DRM_13457_full -> Patchwork_121731v1_full

[Intel-gfx] ✓ Fi.CI.BAT: success for drm/i915: Fix Wa_22016122933 implementation (rev2)

== Series Details == Series: drm/i915: Fix Wa_22016122933 implementation (rev2) URL : https://patchwork.freedesktop.org/series/121735/ State : success == Summary == CI Bug Log - changes from CI_DRM_13457 -> Patchwork_121735v2 Summary

[Intel-gfx] ✗ Fi.CI.SPARSE: warning for drm/i915: Fix Wa_22016122933 implementation (rev2)

== Series Details == Series: drm/i915: Fix Wa_22016122933 implementation (rev2) URL : https://patchwork.freedesktop.org/series/121735/ State : warning == Summary == Error: dim sparse failed Sparse version: v0.6.2 Fast mode used, each commit won't be checked separately.

[Intel-gfx] ✗ Fi.CI.SPARSE: warning for fix DRM_USE_DYNAMIC_DEBUG regression (rev3)

== Series Details == Series: fix DRM_USE_DYNAMIC_DEBUG regression (rev3) URL : https://patchwork.freedesktop.org/series/113363/ State : warning == Summary == Error: dim sparse failed Sparse version: v0.6.2 Fast mode used, each commit won't be checked separately.

[Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for fix DRM_USE_DYNAMIC_DEBUG regression (rev3)

== Series Details == Series: fix DRM_USE_DYNAMIC_DEBUG regression (rev3) URL : https://patchwork.freedesktop.org/series/113363/ State : warning == Summary == Error: dim checkpatch failed 7aeb4d232ed1 drm: use correct ccflags-y syntax f2caf4b42667 test-dyndbg: fixup CLASSMAP usage error

[Intel-gfx] ✗ Fi.CI.BAT: failure for drm/i915: Fix Wa_22016122933 implementation

== Series Details == Series: drm/i915: Fix Wa_22016122933 implementation URL : https://patchwork.freedesktop.org/series/121735/ State : failure == Summary == CI Bug Log - changes from CI_DRM_13457 -> Patchwork_121735v1 Summary ---

[Intel-gfx] ✗ Fi.CI.SPARSE: warning for drm/i915: Fix Wa_22016122933 implementation

== Series Details == Series: drm/i915: Fix Wa_22016122933 implementation URL : https://patchwork.freedesktop.org/series/121735/ State : warning == Summary == Error: dim sparse failed Sparse version: v0.6.2 Fast mode used, each commit won't be checked separately.

[Intel-gfx] [PATCH v5 22/22] checkpatch: reword long-line warn about commit-msg

Reword the warning to complain about line length 1st, since thats whats actually tested. Cc: a...@canonical.com Cc: j...@perches.com Signed-off-by: Jim Cromie --- scripts/checkpatch.pl | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/scripts/checkpatch.pl

[Intel-gfx] [PATCH v5 21/22] dyndbg-doc: add classmap info to howto

Add some basic info on classmap usage and api Signed-off-by: Jim Cromie --- .../admin-guide/dynamic-debug-howto.rst | 64 ++- 1 file changed, 63 insertions(+), 1 deletion(-) diff --git a/Documentation/admin-guide/dynamic-debug-howto.rst

[Intel-gfx] [PATCH v5 18/22] dyndbg-test: build it with just CONFIG_DYNAMIC_DEBUG_CORE

Make the test-module buildable with CONFIG_DYNAMIC_DEBUG_CORE; add CFLAGS_$ofile defns to supply -DDYNAMIC_DEBUG_MODULE to cc. Change the Kconfig entry to allow building with just _CORE, and fix the help text. Signed-off-by: Jim Cromie --- lib/Kconfig.debug | 10 +- lib/Makefile |

[Intel-gfx] [PATCH v5 20/22] drm-drivers: DRM_CLASSMAP_USE in 2nd batch of drivers, helpers

Add a DRM_CLASSMAP_USE declaration to 2nd batch of helpers and *_drv.c files. For drivers, add the decl just above the module's PARAMs, since it identifies the "inherited" drm.debug param. Note: with CONFIG_DRM_USE_DYNAMIC_DEBUG=y, a module not also declaring DRM_CLASSMAP_USE will have its

[Intel-gfx] [PATCH v5 19/22] drm: restore CONFIG_DRM_USE_DYNAMIC_DEBUG un-BROKEN

Lots of burn-in testing needed before signing, upstreaming. NOTE: I set default Y to maximize testing by default. Is there a better way to do this ? Signed-off-by: Jim Cromie --- drivers/gpu/drm/Kconfig | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git

[Intel-gfx] [PATCH v5 16/22] dyndbg: refactor ddebug_classparam_clamp_input

Extract input validation code, from param_set_dyndbg_module_classes() (the sys-node >handler) to new: ddebug_classparam_clamp_input(kp), call it from former. It takes kernel-param arg, so it can complain about "foo: bad input". Reuse ddparam_clamp_input(kp) in ddebug_sync_classbits(), to

[Intel-gfx] [PATCH v5 14/22] dyndbg-API: fix CONFIG_DRM_USE_DYNAMIC_DEBUG regression

DECLARE_DYNDBG_CLASSMAP() has a design error; it fails a basic K rule: "define once, refer many times". When DRM_USE_DYNAMIC_DEBUG=y, DECLARE_DYNDBG_CLASSMAP() is used across DRM core & drivers; they all repeat the same classmap-defn args, which must match for the modules to respond together when

[Intel-gfx] [PATCH v5 17/22] dyndbg-API: promote DYNDBG_CLASSMAP_PARAM to API

move macro from test-dynamic-debug.c into header, and refine it. Distinguish the 2 use cases of DYNDBG_CLASSMAP_PARAM* 1.DYNDBG_CLASSMAP_PARAM_REF for DRM, to pass in extern __drm_debug by name. dyndbg keeps bits in it, so drm can still use it as before 2.DYNDBG_CLASSMAP_PARAM new

[Intel-gfx] [PATCH v5 15/22] dyndbg: add for_each_boxed_vector

Add a for_each iterator to walk a counted vector member in a struct (ie the box), and use it to replace 8 open-coded loops. Signed-off-by: Jim Cromie --- lib/dynamic_debug.c | 20 +++- 1 file changed, 11 insertions(+), 9 deletions(-) diff --git a/lib/dynamic_debug.c

[Intel-gfx] [PATCH v5 12/22] dyndbg-API: remove DD_CLASS_TYPE_(DISJOINT|LEVEL)_NAMES and code

Remove the NAMED class types; these 2 classmap types accept class names at the PARAM interface, for example: echo +DRM_UT_CORE,-DRM_UT_KMS > /sys/module/drm/parameters/debug_names The code works, but its only used by test-dynamic-debug, and wasn't asked for by anyone else, so simplify things

[Intel-gfx] [PATCH v5 13/22] checkpatch: special case for file-scoped extern linker-symbol

"externs should be avoided in .c files" needs an exception for linker symbols, like those that mark the start, stop of many kernel sections. Since code already checks REALNAME to avoid linker-scripts entirely, add a new else-if block to look at them instead. As a simple heuristic, treat all

[Intel-gfx] [PATCH v5 13/22] checkpatch: file-scoped extern special case for linker-symbol

"externs should be avoided in .c files" needs an exception for linker symbols, like those that mark the start, stop of many kernel sections. Since code already checks REALNAME to avoid linker-scripts entirely, add a new else-if block to look at them instead. As a simple heuristic, treat all

[Intel-gfx] [PATCH v5 09/22] dyndbg: silence debugs with no-change updates

check for actual changes before announcing them, declutter logs. Signed-off-by: Jim Cromie --- lib/dynamic_debug.c | 12 +++- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/lib/dynamic_debug.c b/lib/dynamic_debug.c index 2a5cbb68d88d..a8973d1a6896 100644 ---

[Intel-gfx] [PATCH v5 11/22] dyndbg: tighten fn-sig of ddebug_apply_class_bitmap

old_bits arg is currently a pointer to the input bits, but this could allow inadvertent changes to the input by the fn. Disallow this. And constify new_bits while here. Signed-off-by: Jim Cromie --- lib/dynamic_debug.c | 21 +++-- 1 file changed, 11 insertions(+), 10

[Intel-gfx] [PATCH v5 08/22] dyndbg: reduce verbose/debug clutter

currently, for verbose=3, these are logged (blank lines for clarity): dyndbg: query 0: "class DRM_UT_CORE +p" mod:* dyndbg: split into words: "class" "DRM_UT_CORE" "+p" dyndbg: op='+' dyndbg: flags=0x1 dyndbg: *flagsp=0x1 *maskp=0x dyndbg: parsed: func="" file="" module=""

[Intel-gfx] [PATCH v5 10/22] dyndbg: tighten ddebug_class_name() 1st arg type

Change function's 1st arg-type, and deref in the caller. The fn doesn't need any other fields in the struct. no functional change. Signed-off-by: Jim Cromie --- lib/dynamic_debug.c | 10 +- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/lib/dynamic_debug.c

[Intel-gfx] [PATCH v5 06/22] dyndbg: split param_set_dyndbg_classes to module/wrapper fns

rename param_set_dyndbg_classes: add _module_ name & arg, old name is wrapper to new. New arg allows caller to specify that only one module is affected by a prdbgs update. Outer fn preserves kernel_param interface, passing NULL to inner fn. This selectivity will be used later to narrow the scope

[Intel-gfx] [PATCH v5 07/22] dyndbg: drop NUM_TYPE_ARRAY

ARRAY_SIZE works here, since array decl is complete. no functional change Signed-off-by: Jim Cromie --- include/linux/dynamic_debug.h | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/include/linux/dynamic_debug.h b/include/linux/dynamic_debug.h index

[Intel-gfx] [PATCH v5 05/22] dyndbg: ddebug_apply_class_bitmap - add module arg, select on it

Add query_module param to ddebug_apply_class_bitmap(). This allows its caller to update just one module, or all (as currently). We'll use this later to propagate drm.debug to each USEr as they're modprobed. No functional change. Signed-off-by: Jim Cromie --- after `modprobe i915`, heres the

[Intel-gfx] [PATCH v5 04/22] dyndbg: replace classmap list with a vector

Classmaps are stored/linked in a section/array, but are each added to the module's ddebug_table.maps list-head. This is unnecessary; even when ddebug_attach_classmap() is handling the builtin section (with classmaps for multiple builtin modules), its contents are ordered, so a module's possibly

[Intel-gfx] [PATCH v5 03/22] dyndbg: make ddebug_class_param union members same size

struct ddebug_class_param keeps a ref to the state-storage of the param, make both flavors use the same unsigned long under-type. ISTM this is simpler and safer. Signed-off-by: Jim Cromie --- include/linux/dynamic_debug.h | 2 +- lib/dynamic_debug.c | 2 +- 2 files changed, 2

[Intel-gfx] [PATCH v5 02/22] test-dyndbg: fixup CLASSMAP usage error

more careful reading of test output reveals: lib/test_dynamic_debug.c:103 [test_dynamic_debug]do_cats =pmf "doing categories\n" lib/test_dynamic_debug.c:105 [test_dynamic_debug]do_cats =p "LOW msg\n" class:MID lib/test_dynamic_debug.c:106 [test_dynamic_debug]do_cats =p "MID msg\n" class:HI

[Intel-gfx] [PATCH v5 01/22] drm: use correct ccflags-y syntax

Incorrect CFLAGS- usage failed to add -DDYNAMIC_DEBUG_MODULE, which broke builds with: CONFIG_DRM_USE_DYNAMIC_DEBUG=y CONFIG_DYNAMIC_DEBUG_CORE=y but without DYNAMIC_DEBUG Nobody noticed because a larger regression emerged. Also add subdir-ccflags so that all drivers pick up the addition.

[Intel-gfx] [PATCH v5 00/22] fix DRM_USE_DYNAMIC_DEBUG regression

Hi Jason, Daniel, etal This is V5, Im hoping to land this one. patchwork will probably call this set v3 113361 fix DRM_USE_DYNAMIC_DEBUG regression - revs 1,2 111652 DRM_USE_DYNAMIC_DEBUG regression - older, also 2 revs It (patch 14 mainly): Fixes: aad0214f3026 ("dyndbg: add

[Intel-gfx] ✓ Fi.CI.BAT: success for Invalidate TLB cache for all GT's

== Series Details == Series: Invalidate TLB cache for all GT's URL : https://patchwork.freedesktop.org/series/121732/ State : success == Summary == CI Bug Log - changes from CI_DRM_13457 -> Patchwork_121732v1 Summary --- **SUCCESS**

[Intel-gfx] ✗ Fi.CI.SPARSE: warning for Invalidate TLB cache for all GT's

== Series Details == Series: Invalidate TLB cache for all GT's URL : https://patchwork.freedesktop.org/series/121732/ State : warning == Summary == Error: dim sparse failed Sparse version: v0.6.2 Fast mode used, each commit won't be checked separately.

[Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for Invalidate TLB cache for all GT's

== Series Details == Series: Invalidate TLB cache for all GT's URL : https://patchwork.freedesktop.org/series/121732/ State : warning == Summary == Error: dim checkpatch failed 0c96cf631102 drm/i915/gt: Move TLB invalidation to its own file Traceback (most recent call last): File

[Intel-gfx] ✓ Fi.CI.BAT: success for Replace acronym with full platform name in defines

== Series Details == Series: Replace acronym with full platform name in defines URL : https://patchwork.freedesktop.org/series/121731/ State : success == Summary == CI Bug Log - changes from CI_DRM_13457 -> Patchwork_121731v1 Summary

[Intel-gfx] ✓ Fi.CI.IGT: success for drm/i915/uncore: unclaimed reg debug race fix

== Series Details == Series: drm/i915/uncore: unclaimed reg debug race fix URL : https://patchwork.freedesktop.org/series/121726/ State : success == Summary == CI Bug Log - changes from CI_DRM_13455_full -> Patchwork_121726v1_full Summary

[Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for Replace acronym with full platform name in defines

== Series Details == Series: Replace acronym with full platform name in defines URL : https://patchwork.freedesktop.org/series/121731/ State : warning == Summary == Error: dim checkpatch failed cc897b4cb6a1 drm/i915/hsw: s/HSW/HASWELL for platform/subplatform defines -:48:

[Intel-gfx] ✗ Fi.CI.SPARSE: warning for Replace acronym with full platform name in defines

== Series Details == Series: Replace acronym with full platform name in defines URL : https://patchwork.freedesktop.org/series/121731/ State : warning == Summary == Error: dim sparse failed Sparse version: v0.6.2 Fast mode used, each commit won't be checked separately.

[Intel-gfx] ✗ Fi.CI.BUILD: warning for Replace acronym with full platform name in defines

== Series Details == Series: Replace acronym with full platform name in defines URL : https://patchwork.freedesktop.org/series/121731/ State : warning == Summary == Error: patch https://patchwork.freedesktop.org/api/1.0/series/121731/revisions/1/mbox/ not found

[Intel-gfx] [PATCH 3/3] drm/i915/gt: Apply workaround 22016122933 correctly

WA_22016122933 was recently applied to all MeteorLake engines, which is simultaneously too broad (should only apply to Media engines) and too specific (should apply to all platforms that use the same media engine as MeteorLake). Correct this in cases where coherency settings are modified. There

[Intel-gfx] [PATCH 0/3] drm/i915: Fix Wa_22016122933 implementation

WA_22016122933 was recently applied to all MeteorLake engines, which is simultaneously too broad (should only apply to Media engines) and too specific (should apply to all platforms that use the same media engine as MeteorLake). Correct this for all current use cases. There are two additional

[Intel-gfx] [PATCH 1/3] drm/i915/gt: Simplify shmem_create_from_object map_type selection

The object pin created for shmem_create_from_object is just a single use mapping with the sole purpose of reading the contents of the whole object in bulk. And the whole source object is also even a throw-away. Ergo, the additional logic required by i915_coherent_map_type can be safely dropped

[Intel-gfx] [PATCH 2/3] drm/i915: Make i915_coherent_map_type GT-centric

Refactor i915_coherent_map_type to be GT-centric rather than device-centric. Each GT may require different coherency handling due to hardware workarounds. Since the function now takes a GT instead of the i915, the function is renamed and moved to the gt folder. Suggested-by: Matt Roper

[Intel-gfx] ✗ Fi.CI.IGT: failure for series starting with [1/3] drm/i915/gt: Simplify shmem_create_from_object map_type selection (rev2)

== Series Details == Series: series starting with [1/3] drm/i915/gt: Simplify shmem_create_from_object map_type selection (rev2) URL : https://patchwork.freedesktop.org/series/121373/ State : failure == Summary == CI Bug Log - changes from CI_DRM_13455_full -> Patchwork_121373v2_full

Re: [Intel-gfx] [PATCH 0/4] Invalidate TLB cache for all GT's

On 8/1/23 16:19, Andi Shyti wrote: Hi, This series revisits a once-trendy topic: TLB invalidation support for multi-gt. It's a theme that's been passed around and reshaped by several of us. I've filtered out most of the original changes from this series, initially sent by Mauro [1]. His

Re: [Intel-gfx] [PATCH 2/3] drm/i915: Make i915_coherent_map_type GT-centric

Hi Jonathan, Apparently checkpatch is complaining... [...] > diff --git a/drivers/gpu/drm/i915/gt/intel_lrc.c > b/drivers/gpu/drm/i915/gt/intel_lrc.c > index 119deb9f938c..1897cb5aa2a2 100644 > --- a/drivers/gpu/drm/i915/gt/intel_lrc.c > +++ b/drivers/gpu/drm/i915/gt/intel_lrc.c > @@ -1191,7

[Intel-gfx] [PATCH 4/4] drm/i915: Remove unnecessary include

The inclusion of intel_gt_defines.h was initially added to i915_drv.h to provide the definition of I915_MAX_GT, where it was originally defined. However, since I915_MAX_GT is now included in i915_gem_object_types.h, it sis no longer required in i915_drv.h. Signed-off-by: Andi Shyti Cc: Chris

[Intel-gfx] [PATCH 3/4] drm/i915: Invalidate the TLBs on each GT

From: Chris Wilson With multi-GT devices, the object may have been bound on each GT. Invalidate the TLBs across all GT before releasing the pages back to the system. Signed-off-by: Chris Wilson Cc: Fei Yang Signed-off-by: Mauro Carvalho Chehab Signed-off-by: Andi Shyti ---

[Intel-gfx] [PATCH 2/4] i915/drm/gt: Move the gt defines in the gt directory

Create a new intel_gt_defines.h inside the gt/ directory as a placeholder for all the generic GT based defines. As of now place only I915_MAX_GT. Co-developed-by: Chris Wilson Signed-off-by: Andi Shyti --- drivers/gpu/drm/i915/gt/intel_gt_defines.h | 11 +++

[Intel-gfx] [PATCH 1/4] drm/i915/gt: Move TLB invalidation to its own file

From: Chris Wilson Prepare for supporting more TLB invalidation scenarios by moving the current MMIO invalidation to its own file. Signed-off-by: Chris Wilson Signed-off-by: Mauro Carvalho Chehab Reviewed-by: Andi Shyti Signed-off-by: Andi Shyti --- drivers/gpu/drm/i915/Makefile

[Intel-gfx] [PATCH 0/4] Invalidate TLB cache for all GT's

Hi, This series revisits a once-trendy topic: TLB invalidation support for multi-gt. It's a theme that's been passed around and reshaped by several of us. I've filtered out most of the original changes from this series, initially sent by Mauro [1]. His ideas were inspired by some changes from

[Intel-gfx] [PATCH v3 14/14] drm/i915/adls: s/ADLS_RPLS/RAPTORLAKE_S in platform and subplatform defines

Driver refers to the platform Alderlake S as ADLS_RPLS in places and RAPTORLAKE_S in some. v2: - Unrolled wrapper IS_ADLS_GRAPHICS_STEP v3: - Replace IS_RAPTORLAKE_S instead of IS_ADLS_RPLS. (Tvrtko/Lucas). - Remove unused macro IS_ADLS_GRAPHICS/DISPLAY_STEP - Change the subject Cc: Tvrtko

[Intel-gfx] [PATCH v1 13/14] drm/i915/adln: s/ADLP/ALDERLAKE_P in ADLN defines

From: Anusha Srivatsa Follow consistent naming convention. Replace ADLP with ALDERLAKE_P Signed-off-by: Anusha Srivatsa Acked-by: Jani Nikula --- drivers/gpu/drm/i915/gt/uc/intel_guc_hwconfig.c | 2 +- drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c| 2 +- drivers/gpu/drm/i915/i915_drv.h

[Intel-gfx] [PATCH v2 11/14] drm/i915/rplp: s/ADLP_RPLP/RAPTORLAKE_P for RPLP defines

Follow consistent naming convention. Replace ADLP with ALDERLAKE_P. v2: - Replace IS_ADLP_RPLP with IS_RAPTORLAKE_P. (Tvrtko/Lucas) - Change the subject Cc: Tvrtko Ursulin Cc: Jani Nikula Cc: Anusha Srivatsa Signed-off-by: Dnyaneshwar Bhadane Reviewed-by: Anusha Srivatsa Acked-by: Jani

[Intel-gfx] [PATCH v2 12/14] drm/i915/rplu: s/ADLP_RPLU/RAPTORLAKE_U in RPLU defines

Follow consistent naming convention. Replace ADLP with ALDERLAKE_P v2: - Replace IS_ADLP_RPLU with IS_RAPTORLAKE_U (Tvrtko/Lucas) - Change the subject Cc: Tvrtko Ursulin Cc: Jani Nikula Cc: Anusha Srivatsa Signed-off-by: Dnyaneshwar Bhadane Reviewed-by: Anusha Srivatsa Acked-by: Jani Nikula

[Intel-gfx] [PATCH v3 10/14] drm/i915/adlp: s/ADLP/ALDERLAKE_P for display and graphics step

Driver refers to the platform Alderlake P as ADLP in places and ALDERLAKE_P in some. Making the consistent change to avoid confusion of the right naming convention for the platform. v2: - Unrolled wrapper IS_ADLP_GRAPHICS_STEP and Replace - Added IS_ALDERLAKE_P() && IS_GRAPHICS_STEP()

[Intel-gfx] [PATCH v4 09/14] drm/i915/tgl: s/TGL/TIGERLAKE for platform/subplatform defines

Follow consistent naming convention. Replace TGL with TIGERLAKE.Replace IS_TGL_DISPLAY_STEP with IS_TIGERLAKE() && IS_DISPLAY_STEP(). v2: - s/TGL/tgl in the subject prefix(Anusha) v3: - Unrolled wrapper IS_TGL_DISPLAY_STEP and Replace - Replace IS_PLATFORM && DISPLAY_STEP (Jani/Tvrtko). v4: -

[Intel-gfx] [PATCH v3 07/14] drm/i915/rkl: s/RKL/ROCKETLAKE for platform/subplatform defines

Follow consistent naming convention. Replace RKL with ROCKETLAKE.Replace IS_RKL_GRAPHICS_STEP with IS_ROCKETLAKE && IS_DISPLAY_STEP. v2: - s/RKL/rkl in the subject prefix(Anusha) v3: - Unrolled wrapper IS_RKL_DISPLAY_STEP. - Replace IS_PLATFORM && IS_DISPLAY_STEP (Jani/Tvrtko) Cc: Tvrtko

[Intel-gfx] [PATCH v5 08/14] drm/i915/jsl: s/JSL/JASPERLAKE for platform/subplatform defines

Follow consistent naming convention. Replace JSL with JASPERLAKE. Unroll IS_JSL_EHL() define with IS_JASPERLAKE() || IS_ELKHARTLAKE() condition. Change in the display step define for Jasperlake. v2: - Change subject prefix skl instead of SKL(Anusha) v3: - Remove the use of define IS_JSL_EHL. -

[Intel-gfx] [PATCH v1 05/14] drm/i915/cfl: s/CFL/COFFEELAKE for platform/subplatform defines

Follow consistent naming convention. Replace CFL with COFFEELAKE. Signed-off-by: Dnyaneshwar Bhadane Reviewed-by: Anusha Srivatsa Acked-by: Jani Nikula --- drivers/gpu/drm/i915/display/intel_ddi_buf_trans.c | 4 ++-- drivers/gpu/drm/i915/i915_drv.h| 8 2 files

[Intel-gfx] [PATCH v1 06/14] drm/i915/cml: s/CML/COMETLAKE for platform/subplatform defines

Follow consistent naming convention. Replace CML with COMETLAKE. Signed-off-by: Dnyaneshwar Bhadane Reviewed-by: Anusha Srivatsa Acked-by: Jani Nikula --- drivers/gpu/drm/i915/display/intel_ddi_buf_trans.c | 4 ++-- drivers/gpu/drm/i915/i915_drv.h| 6 +++--- 2 files

[Intel-gfx] [PATCH v4 04/14] drm/i915/kbl: s/KBL/KABYLAKE for platform/subplatform defines

Follow consistent naming convention. Replace KBL with KABYLAKE.Replace IS_KBL_GRAPHICS_STEP with IS_KABYLAKE () && IS_GRAPHICS_STEP(). v2: - s/KBL/kbl in the subject prefix(Anusha) v3: - Unrolled wrapper IS_KBL_GRAPHICS_STEP. - Replace with IS_PLATFORM && DISPLAY_STEP(tvrtko/jani) v4: - Removed

[Intel-gfx] [PATCH v4 03/14] drm/i915/skl: s/SKL/SKYLAKE for platform/subplatform defines

Follow consistent naming convention. Replace SKL with SKYLAKE and Replace IS_SKL_GRAPHICS_STEP with IS_SKYLAKE() && IS_GRAPHICS_STEP(). v2: - Change subject skl instead of SKL(Anusha) v3: - Unrolled wrapper IS_SKL_GRAPHICS_STEP. - Replace with IS_PLATFORM && DISPLAY_STEP(tvrtko/jani) v4: -

[Intel-gfx] [PATCH v1 01/14] drm/i915/hsw: s/HSW/HASWELL for platform/subplatform defines

Follow consistent naming convention. Replace HSW with HASWELL. Signed-off-by: Dnyaneshwar Bhadane Reviewed-by: Anusha Srivatsa Acked-by: Jani Nikula --- drivers/gpu/drm/i915/display/intel_cdclk.c | 2 +- drivers/gpu/drm/i915/display/intel_display.c| 2 +-

[Intel-gfx] [PATCH v1 02/14] drm/i915/bdw: s/BDW/BROADWELL for platform/subplatform defines

Follow consistent naming convention. Replace BDW with BROADWELL. Signed-off-by: Dnyaneshwar Bhadane Reviewed-by: Anusha Srivatsa Acked-by: Jani Nikula --- drivers/gpu/drm/i915/display/intel_cdclk.c | 4 ++-- drivers/gpu/drm/i915/display/intel_display.c | 2 +-

[Intel-gfx] [PATCH v1 00/14] Replace acronym with full platform name in defines

Replacing the acronym used in platform/sub platform defines. This series covers Haswell, Broadwell, Skylake, Kabylake, Coffeelake, Cometlake, Rocketlake, Jasperlake, Elkhartlake, Tigerlake, Alderlake, platoform define.This way there is a consistent pattern to how platforms are referred.splitting

Re: [Intel-gfx] [PATCH v4 00/14] Replace acronym with full platform name in defines.

On Thu, 27 Jul 2023, Dnyaneshwar Bhadane wrote: > Replacing the acronym used in platform/sub platform defines. > This series covers Haswell, Broadwell, Skylake, Kabylake, Coffeelake, > Cometlake, Rocketlake, Jasperlake, Elkhartlake, Tigerlake, Alderlake, > platform define.This way there is a

[Intel-gfx] ✓ Fi.CI.BAT: success for drm/i915/uncore: unclaimed reg debug race fix

== Series Details == Series: drm/i915/uncore: unclaimed reg debug race fix URL : https://patchwork.freedesktop.org/series/121726/ State : success == Summary == CI Bug Log - changes from CI_DRM_13455 -> Patchwork_121726v1 Summary ---

[Intel-gfx] ✗ Fi.CI.SPARSE: warning for drm/i915/uncore: unclaimed reg debug race fix

== Series Details == Series: drm/i915/uncore: unclaimed reg debug race fix URL : https://patchwork.freedesktop.org/series/121726/ State : warning == Summary == Error: dim sparse failed Sparse version: v0.6.2 Fast mode used, each commit won't be checked separately.

[Intel-gfx] [RESEND 2/2] drm/i915/uncore: fix race around i915->params.mmio_debug

Only check the conditions for unclaimed reg debug once to avoid locking problems when i915->params.mmio_debug changes between header and footer. Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/8749 Cc: Lee Shawn C Reviewed-by: Tvrtko Ursulin Signed-off-by: Jani Nikula ---

[Intel-gfx] [RESEND 1/2] drm/i915/uncore: split unclaimed_reg_debug() to header and footer

Make it easier to have different logic for the two for follow-up. Cc: Lee Shawn C Reviewed-by: Tvrtko Ursulin Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/intel_uncore.c | 37 + 1 file changed, 22 insertions(+), 15 deletions(-) diff --git

[Intel-gfx] [RESEND 0/2] drm/i915/uncore: unclaimed reg debug race fix

Resend of the first two patches of [1], dropping the third. BR, Jani. [1] https://patchwork.freedesktop.org/series/120167/ Jani Nikula (2): drm/i915/uncore: split unclaimed_reg_debug() to header and footer drm/i915/uncore: fix race around i915->params.mmio_debug

[Intel-gfx] ✓ Fi.CI.BAT: success for series starting with [1/3] drm/i915/gt: Simplify shmem_create_from_object map_type selection (rev2)

== Series Details == Series: series starting with [1/3] drm/i915/gt: Simplify shmem_create_from_object map_type selection (rev2) URL : https://patchwork.freedesktop.org/series/121373/ State : success == Summary == CI Bug Log - changes from CI_DRM_13455 -> Patchwork_121373v2

[Intel-gfx] ✗ Fi.CI.SPARSE: warning for series starting with [1/3] drm/i915/gt: Simplify shmem_create_from_object map_type selection (rev2)

== Series Details == Series: series starting with [1/3] drm/i915/gt: Simplify shmem_create_from_object map_type selection (rev2) URL : https://patchwork.freedesktop.org/series/121373/ State : warning == Summary == Error: dim sparse failed Sparse version: v0.6.2 Fast mode used, each commit

[Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for series starting with [1/3] drm/i915/gt: Simplify shmem_create_from_object map_type selection (rev2)

== Series Details == Series: series starting with [1/3] drm/i915/gt: Simplify shmem_create_from_object map_type selection (rev2) URL : https://patchwork.freedesktop.org/series/121373/ State : warning == Summary == Error: dim checkpatch failed 45c69ff3c0cc drm/i915/gt: Simplify

Re: [Intel-gfx] [PATCH v4 29/29] drm/i915/gvt: Drop final dependencies on KVM internal details

On 7/29/2023 4:35 AM, Sean Christopherson wrote: Open code gpa_to_gfn() in kvmgt_page_track_write() and drop KVMGT's dependency on kvm_host.h, i.e. include only kvm_page_track.h. KVMGT assumes "gfn == gpa >> PAGE_SHIFT" all over the place, including a few lines below in the same function with

Re: [Intel-gfx] [PATCH v4 19/29] drm/i915/gvt: switch from ->track_flush_slot() to ->track_remove_region()

On 7/29/2023 4:35 AM, Sean Christopherson wrote: From: Yan Zhao Switch from the poorly named and flawed ->track_flush_slot() to the newly introduced ->track_remove_region(). From KVMGT's perspective, the two hooks are functionally equivalent, the only difference being that

Re: [Intel-gfx] [PATCH v4 17/29] drm/i915/gvt: Don't bother removing write-protection on to-be-deleted slot

On 7/29/2023 4:35 AM, Sean Christopherson wrote: When handling a slot "flush", don't call back into KVM to drop write protection for gfns in the slot. Now that KVM rejects attempts to move memory slots while KVMGT is attached, the only time a slot is "flushed" is when it's being removed, i.e.

Re: [Intel-gfx] [PATCH v4 15/29] KVM: drm/i915/gvt: Drop @vcpu from KVM's ->track_write() hook

On 7/29/2023 4:35 AM, Sean Christopherson wrote: Drop @vcpu from KVM's ->track_write() hook provided for external users of the page-track APIs now that KVM itself doesn't use the page-track mechanism. Reviewed-by: Yan Zhao Tested-by: Yongwei Ma Signed-off-by: Sean Christopherson ---

Re: [Intel-gfx] [PATCH v4 11/29] drm/i915/gvt: Protect gfn hash table with vgpu_lock

On 7/29/2023 4:35 AM, Sean Christopherson wrote: Use vgpu_lock instead of KVM's mmu_lock to protect accesses to the hash table used to track which gfns are write-protected when shadowing the guest's GTT, and hoist the acquisition of vgpu_lock from intel_vgpu_page_track_handler() out to its sole

Re: [Intel-gfx] [PATCH v4 10/29] drm/i915/gvt: Drop unused helper intel_vgpu_reset_gtt()

On 7/29/2023 4:35 AM, Sean Christopherson wrote: Drop intel_vgpu_reset_gtt() as it no longer has any callers. In addition to eliminating dead code, this eliminates the last possible scenario where __kvmgt_protect_table_find() can be reached without holding vgpu_lock. Requiring vgpu_lock to be

Re: [Intel-gfx] [PATCH v4 09/29] drm/i915/gvt: Use an "unsigned long" to iterate over memslot gfns

On 7/29/2023 4:35 AM, Sean Christopherson wrote: Use an "unsigned long" instead of an "int" when iterating over the gfns in a memslot. The number of pages in the memslot is tracked as an "unsigned long", e.g. KVMGT could theoretically break if a KVM memslot larger than 16TiB were deleted (2^32

Re: [Intel-gfx] [PATCH v4 08/29] drm/i915/gvt: Don't rely on KVM's gfn_to_pfn() to query possible 2M GTT

On 7/29/2023 4:35 AM, Sean Christopherson wrote: Now that gvt_pin_guest_page() explicitly verifies the pinned PFN is a transparent hugepage page, don't use KVM's gfn_to_pfn() to pre-check if a 2MiB GTT entry is possible and instead just try to map the GFN with a 2MiB entry. Using KVM to query

Re: [Intel-gfx] [PATCH v4 05/29] drm/i915/gvt: Put the page reference obtained by KVM's gfn_to_pfn()

On 7/29/2023 4:35 AM, Sean Christopherson wrote: Put the struct page reference acquired by gfn_to_pfn(), KVM's API is that the caller is ultimately responsible for dropping any reference. Note, kvm_release_pfn_clean() ensures the pfn is actually a refcounted struct page before trying to put any

Re: [Intel-gfx] [PATCH v4 03/29] drm/i915/gvt: Verify hugepages are contiguous in physical address space

On 8/1/2023 4:47 AM, Yan Zhao wrote: Reviewed-by: Yan Zhao On Fri, Jul 28, 2023 at 06:35:09PM -0700, Sean Christopherson wrote: When shadowing a GTT entry with a 2M page, verify that the pfns are contiguous, not just that the struct page pointers are contiguous. The memory map is virtual

Re: [Intel-gfx] [PATCH v4 01/29] drm/i915/gvt: Verify pfn is "valid" before dereferencing "struct page"

On 7/29/2023 4:35 AM, Sean Christopherson wrote: Check that the pfn found by gfn_to_pfn() is actually backed by "struct page" memory prior to retrieving and dereferencing the page. KVM supports backing guest memory with VM_PFNMAP, VM_IO, etc., and so there is no guarantee the pfn returned by

Re: [Intel-gfx] [PATCH v4 04/29] drm/i915/gvt: Don't try to unpin an empty page range

On 7/29/2023 4:35 AM, Sean Christopherson wrote: From: Yan Zhao Attempt to unpin pages in the error path of gvt_pin_guest_page() if and only if at least one page was successfully pinned. Unpinning doesn't cause functional problems, but vfio_device_container_unpin_pages() rightfully warns

  1   2   >