From: Rob Clark <robdclark@chromium.org>

This series extends io-pgtable-arm with a method to retrieve the page
table entries traversed in the process of address translation, and then
beefs up drm/msm gpu devcore dump to include this (and additional info)
in the devcore dump.

This is a respin of https://patchwork.freedesktop.org/series/94968/
(minus a patch that was already merged)

v2:  Fix an armv7/32b build error in the last patch
v3:  Incorperate Will Deacon's suggestion to make the interface
     callback based.
v4:  Actually wire up the callback
v5:  Drop the callback approach
v6:  Make walk-data struct pgtable specific and rename
     io_pgtable_walk_data to arm_lpae_io_pgtable_walk_data
v7:  Re-use the pgtable walker added for arm_lpae_read_and_clear_dirty()
v8:  Pass pte pointer to callback so it can modify the actual pte
v9:  Fix selftests_running case
v10: Call visit cb for all nodes traversed, leave the decision about
     whether to care about non-leaf nodes to the callback
v11: Adjust logic in 3/4 [smostafa@]

Rob Clark (4):
  iommu/io-pgtable-arm: Make pgtable walker more generic
  iommu/io-pgtable-arm: Re-use the pgtable walk for iova_to_phys
  iommu/io-pgtable-arm: Add way to debug pgtable walk
  drm/msm: Extend gpu devcore dumps with pgtbl info

 drivers/gpu/drm/msm/adreno/adreno_gpu.c |  10 ++
 drivers/gpu/drm/msm/msm_gpu.c           |   9 ++
 drivers/gpu/drm/msm/msm_gpu.h           |   8 ++
 drivers/gpu/drm/msm/msm_iommu.c         |  22 ++++
 drivers/gpu/drm/msm/msm_mmu.h           |   3 +-
 drivers/iommu/io-pgtable-arm.c          | 157 +++++++++++++++---------
 include/linux/io-pgtable.h              |  15 +++
 7 files changed, 167 insertions(+), 57 deletions(-)

-- 
2.47.1

On Tue, 10 Dec 2024 08:51:18 -0800, Rob Clark wrote:
> This series extends io-pgtable-arm with a method to retrieve the page
> table entries traversed in the process of address translation, and then
> beefs up drm/msm gpu devcore dump to include this (and additional info)
> in the devcore dump.
> 
> This is a respin of https://patchwork.freedesktop.org/series/94968/
> (minus a patch that was already merged)
> 
> [...]

Applied io-pgtable changes to iommu (arm/smmu/updates), thanks!

[1/4] iommu/io-pgtable-arm: Make pgtable walker more generic
      https://git.kernel.org/iommu/c/821500d5c597
[2/4] iommu/io-pgtable-arm: Re-use the pgtable walk for iova_to_phys
      https://git.kernel.org/iommu/c/d9e589e6ad73
[3/4] iommu/io-pgtable-arm: Add way to debug pgtable walk
      https://git.kernel.org/iommu/c/aff028a8192d

Cheers,
-- 
Will

https://fixes.arm64.dev
https://next.arm64.dev
https://will.arm64.dev

Hi Rob,

On Tue, Dec 10, 2024 at 08:51:18AM -0800, Rob Clark wrote:
> From: Rob Clark <robdclark@chromium.org>
> 
> This series extends io-pgtable-arm with a method to retrieve the page
> table entries traversed in the process of address translation, and then
> beefs up drm/msm gpu devcore dump to include this (and additional info)
> in the devcore dump.
> 
> This is a respin of https://patchwork.freedesktop.org/series/94968/
> (minus a patch that was already merged)
> 
> v2:  Fix an armv7/32b build error in the last patch
> v3:  Incorperate Will Deacon's suggestion to make the interface
>      callback based.
> v4:  Actually wire up the callback
> v5:  Drop the callback approach
> v6:  Make walk-data struct pgtable specific and rename
>      io_pgtable_walk_data to arm_lpae_io_pgtable_walk_data
> v7:  Re-use the pgtable walker added for arm_lpae_read_and_clear_dirty()
> v8:  Pass pte pointer to callback so it can modify the actual pte
> v9:  Fix selftests_running case
> v10: Call visit cb for all nodes traversed, leave the decision about
>      whether to care about non-leaf nodes to the callback
> v11: Adjust logic in 3/4 [smostafa@]

I see the level initialization was not removed as it was in the diff[1], so it
seems to me that’s redundant as the level is set anyway in the callback, and
actually looking at that I see it’s not used or printed from the driver,
so may it can be removed all together, anyway that’s nit that may be Will can
fix it up while merging.

Otherwise, For the whole series
Reviewed-by: Mostafa Saleh <smostafa@google.com>

diff --git a/drivers/iommu/io-pgtable-arm.c b/drivers/iommu/io-pgtable-arm.c
index ed3ba47df350..a884af8fb101 100644
--- a/drivers/iommu/io-pgtable-arm.c
+++ b/drivers/iommu/io-pgtable-arm.c
@@ -731,7 +731,6 @@ static int visit_pgtable_walk(struct io_pgtable_walk_data *walk_data, int lvl,
 {
 	struct arm_lpae_io_pgtable_walk_data *data = walk_data->data;
 	data->ptes[lvl] = *ptep;
-	data->level = lvl + 1;
 	return 0;
 }
 
@@ -746,8 +745,6 @@ static int arm_lpae_pgtable_walk(struct io_pgtable_ops *ops, unsigned long iova,
 		.end = iova + 1,
 	};
 
-	((struct arm_lpae_io_pgtable_walk_data *)wd)->level = 0;
-
 	return __arm_lpae_iopte_walk(data, &walk_data, data->pgd, data->start_level);
 }
 
diff --git a/include/linux/io-pgtable.h b/include/linux/io-pgtable.h
index 0f17a38a304d..c62e344a4058 100644
--- a/include/linux/io-pgtable.h
+++ b/include/linux/io-pgtable.h
@@ -190,7 +190,6 @@ struct io_pgtable_cfg {
  */
 struct arm_lpae_io_pgtable_walk_data {
 	u64 ptes[4];
-	int level;
 };
 
 /**



[1]https://lore.kernel.org/all/Z1AtDgLD0SDUxDHl@google.com/

Thanks,
Mostafa

> 
> Rob Clark (4):
>   iommu/io-pgtable-arm: Make pgtable walker more generic
>   iommu/io-pgtable-arm: Re-use the pgtable walk for iova_to_phys
>   iommu/io-pgtable-arm: Add way to debug pgtable walk
>   drm/msm: Extend gpu devcore dumps with pgtbl info
> 
>  drivers/gpu/drm/msm/adreno/adreno_gpu.c |  10 ++
>  drivers/gpu/drm/msm/msm_gpu.c           |   9 ++
>  drivers/gpu/drm/msm/msm_gpu.h           |   8 ++
>  drivers/gpu/drm/msm/msm_iommu.c         |  22 ++++
>  drivers/gpu/drm/msm/msm_mmu.h           |   3 +-
>  drivers/iommu/io-pgtable-arm.c          | 157 +++++++++++++++---------
>  include/linux/io-pgtable.h              |  15 +++
>  7 files changed, 167 insertions(+), 57 deletions(-)
> 
> -- 
> 2.47.1
> 

On Tue, Jan 07, 2025 at 12:55:55PM +0000, Mostafa Saleh wrote:
> On Tue, Dec 10, 2024 at 08:51:18AM -0800, Rob Clark wrote:
> > From: Rob Clark <robdclark@chromium.org>
> > 
> > This series extends io-pgtable-arm with a method to retrieve the page
> > table entries traversed in the process of address translation, and then
> > beefs up drm/msm gpu devcore dump to include this (and additional info)
> > in the devcore dump.
> > 
> > This is a respin of https://patchwork.freedesktop.org/series/94968/
> > (minus a patch that was already merged)
> > 
> > v2:  Fix an armv7/32b build error in the last patch
> > v3:  Incorperate Will Deacon's suggestion to make the interface
> >      callback based.
> > v4:  Actually wire up the callback
> > v5:  Drop the callback approach
> > v6:  Make walk-data struct pgtable specific and rename
> >      io_pgtable_walk_data to arm_lpae_io_pgtable_walk_data
> > v7:  Re-use the pgtable walker added for arm_lpae_read_and_clear_dirty()
> > v8:  Pass pte pointer to callback so it can modify the actual pte
> > v9:  Fix selftests_running case
> > v10: Call visit cb for all nodes traversed, leave the decision about
> >      whether to care about non-leaf nodes to the callback
> > v11: Adjust logic in 3/4 [smostafa@]
> 
> I see the level initialization was not removed as it was in the diff[1], so it
> seems to me that’s redundant as the level is set anyway in the callback, and
> actually looking at that I see it’s not used or printed from the driver,
> so may it can be removed all together, anyway that’s nit that may be Will can
> fix it up while merging.
> 
> Otherwise, For the whole series
> Reviewed-by: Mostafa Saleh <smostafa@google.com>

I'm happy to drop the 'level' field if it's not used. We can add it back
if/when it's needed. Rob?

Will

On Tue, Jan 7, 2025 at 4:58 AM Will Deacon <will@kernel.org> wrote:
>
> On Tue, Jan 07, 2025 at 12:55:55PM +0000, Mostafa Saleh wrote:
> > On Tue, Dec 10, 2024 at 08:51:18AM -0800, Rob Clark wrote:
> > > From: Rob Clark <robdclark@chromium.org>
> > >
> > > This series extends io-pgtable-arm with a method to retrieve the page
> > > table entries traversed in the process of address translation, and then
> > > beefs up drm/msm gpu devcore dump to include this (and additional info)
> > > in the devcore dump.
> > >
> > > This is a respin of https://patchwork.freedesktop.org/series/94968/
> > > (minus a patch that was already merged)
> > >
> > > v2:  Fix an armv7/32b build error in the last patch
> > > v3:  Incorperate Will Deacon's suggestion to make the interface
> > >      callback based.
> > > v4:  Actually wire up the callback
> > > v5:  Drop the callback approach
> > > v6:  Make walk-data struct pgtable specific and rename
> > >      io_pgtable_walk_data to arm_lpae_io_pgtable_walk_data
> > > v7:  Re-use the pgtable walker added for arm_lpae_read_and_clear_dirty()
> > > v8:  Pass pte pointer to callback so it can modify the actual pte
> > > v9:  Fix selftests_running case
> > > v10: Call visit cb for all nodes traversed, leave the decision about
> > >      whether to care about non-leaf nodes to the callback
> > > v11: Adjust logic in 3/4 [smostafa@]
> >
> > I see the level initialization was not removed as it was in the diff[1], so it
> > seems to me that’s redundant as the level is set anyway in the callback, and
> > actually looking at that I see it’s not used or printed from the driver,
> > so may it can be removed all together, anyway that’s nit that may be Will can
> > fix it up while merging.
> >
> > Otherwise, For the whole series
> > Reviewed-by: Mostafa Saleh <smostafa@google.com>
>
> I'm happy to drop the 'level' field if it's not used. We can add it back
> if/when it's needed. Rob?

That works for me, thx

BR,
-R