So far our parsing of {iommu,msi}-map properties has always blindly
assumed that the output specifiers will always have exactly 1 cell.
This typically does happen to be the case, but is not actually enforced
(and the PCI msi-map binding even explicitly states support for 0 or 1
cells) - as a result we've now ended up with dodgy DTs out in the field
which depend on this behaviour to map a 1-cell specifier for a 2-cell
provider, despite that being bogus per the bindings themselves.

Since there is some potential use[1] in being able to map at least
single input IDs to multi-cell output specifiers (and properly support
0-cell outputs as well), add support for properly parsing and using the
target nodes' #cells values, albeit with the unfortunate complication of
still having to work around expectations of the old behaviour too.
							-- Robin.

Unlike single #{}-cell, it is complex to establish a linear relation
between input 'id' and output specifier for multi-cell properties, thus
it is always expected that len never going to be > 1. 

These changes have been tested on QEMU for the arm64 architecture.

Since, this would also need update in dt-schema, raised PR[2] for the
same.

[1] https://lore.kernel.org/all/20250627-video_cb-v3-0-51e18c0ffbce@quicinc.com/
[2] PR for iommu-map dtschema: https://github.com/devicetree-org/dt-schema/pull/184

V11:
  - Added explicit filter_np parameter to of_map_id() and of_map_msi_id()
    per Dmitry Baryshkov's review feedback, making the filter explicit
    instead of overloading arg->np as both input filter and output parameter.
  - Removed of_node_put() from inside of_map_id(), making the caller responsible
    for reference management. Updated of_msi_xlate() to properly handle reference counting.
  - Collected ACKed by tags, and fixed minor typos.
  Link to v10:
  https://lore.kernel.org/r/20260309-parse_iommu_cells-v10-0-c62fcaa5a1d8@oss.qualcomm.com

V10:
  - Move of_map_iommu_id()/of_map_msi_id() from include/linux/of.h to
    drivers/of/base.c as out-of-line helpers per feedback from Marc Zyngier
    and Rob Herring.
  - Add kernel-doc to document both helpers for discoverability and
    usage clarity.
  - Fix of_map_msi_id() wrapper and all its callers (cdx_msi.c,
    irq-gic-its-msi-parent.c, drivers/of/irq.c) to correctly use the new
    struct of_phandle_args-based API with proper of_node_put() handling
    as per feeback from Dmitry.
  Link to v9:
  https://lore.kernel.org/r/20260301-parse_iommu_cells-v9-0-4d1bceecc5e1@oss.qualcomm.com

V9:
  - Updated TO/CC list based on feedback to include all relevant
    maintainers.
  - No functional changes to the patches themselves.

  Link to V8:
  https://lore.kernel.org/all/20260226074245.3098486-1-vijayanand.jitta@oss.qualcomm.com/

V8:
  - Removed mentions of of_map_args from commit message to match code.

  Link to V7:
  https://lore.kernel.org/all/20260210101157.2145113-1-vijayanand.jitta@oss.qualcomm.com/

V7:
  - Removed of_map_id_args structure and replaced it with
    of_phandle_args as suggested by Dmitry.

  Link to V6:
  https://lore.kernel.org/all/20260121055400.937856-1-vijayanand.jitta@oss.qualcomm.com/

V6:
  - Fixed build error reported by kernel test bot.

  Link to V5:
  https://lore.kernel.org/all/20260118181125.1436036-1-vijayanand.jitta@oss.qualcomm.com/

V5:
  - Fixed Build Warnings.
  - Raised PR for iommu-map dtschema: https://github.com/devicetree-org/dt-schema/pull/184

  Link to V4:
  https://lore.kernel.org/all/20251231114257.2382820-1-vijayanand.jitta@oss.qualcomm.com/

V4:
  - Added Reviewed-by tag.
  - Resolved warnings reported by kernel test bot, minor code
    reorganization.

  Link to V3:
  https://lore.kernel.org/all/20251221213602.2413124-1-vijayanand.jitta@oss.qualcomm.com/

V3:
  - Added Reviewed-by tag.
  - Updated of_map_id_args struct as a wrapper to of_phandle_args and
    added comment description as suggested by Rob Herring.

  Link to V2:
  https://lore.kernel.org/all/20251204095530.8627-1-vijayanand.jitta@oss.qualcomm.com/

V2:
  - Incorporated the patches from Robin that does the clean implementation.
  - Dropped the patches the were adding multi-map support from this series
    as suggested.

V1:
 https://lore.kernel.org/all/cover.1762235099.git.charan.kalla@oss.qualcomm.com/

RFC:
 https://lore.kernel.org/all/20250928171718.436440-1-charan.kalla@oss.qualcomm.com/#r

Signed-off-by: Vijayanand Jitta <vijayanand.jitta@oss.qualcomm.com>
---

---
Charan Teja Kalla (1):
      of: Factor arguments passed to of_map_id() into a struct

Robin Murphy (2):
      of: Add convenience wrappers for of_map_id()
      of: Respect #{iommu,msi}-cells in maps

 drivers/cdx/cdx_msi.c                    |   8 +-
 drivers/iommu/of_iommu.c                 |   6 +-
 drivers/irqchip/irq-gic-its-msi-parent.c |  11 +-
 drivers/of/base.c                        | 213 ++++++++++++++++++++++++-------
 drivers/of/irq.c                         |  11 +-
 drivers/pci/controller/dwc/pci-imx6.c    |  18 ++-
 drivers/pci/controller/pcie-apple.c      |   6 +-
 drivers/xen/grant-dma-ops.c              |   5 +-
 include/linux/of.h                       |  30 ++++-
 9 files changed, 237 insertions(+), 71 deletions(-)
---
base-commit: 3fa5e5702a82d259897bd7e209469bc06368bf31
change-id: 20260301-parse_iommu_cells-1c33768aebba

Best regards,
-- 
Vijayanand Jitta <vijayanand.jitta@oss.qualcomm.com>

On 3/25/2026 4:38 PM, Vijayanand Jitta wrote:
> So far our parsing of {iommu,msi}-map properties has always blindly
> assumed that the output specifiers will always have exactly 1 cell.
> This typically does happen to be the case, but is not actually enforced
> (and the PCI msi-map binding even explicitly states support for 0 or 1
> cells) - as a result we've now ended up with dodgy DTs out in the field
> which depend on this behaviour to map a 1-cell specifier for a 2-cell
> provider, despite that being bogus per the bindings themselves.
> 
> Since there is some potential use[1] in being able to map at least
> single input IDs to multi-cell output specifiers (and properly support
> 0-cell outputs as well), add support for properly parsing and using the
> target nodes' #cells values, albeit with the unfortunate complication of
> still having to work around expectations of the old behaviour too.
> 							-- Robin.
> 
> Unlike single #{}-cell, it is complex to establish a linear relation
> between input 'id' and output specifier for multi-cell properties, thus
> it is always expected that len never going to be > 1. 
> 
> These changes have been tested on QEMU for the arm64 architecture.
> 
> Since, this would also need update in dt-schema, raised PR[2] for the
> same.
> 
> [1] https://lore.kernel.org/all/20250627-video_cb-v3-0-51e18c0ffbce@quicinc.com/
> [2] PR for iommu-map dtschema: https://github.com/devicetree-org/dt-schema/pull/184
> 
> V11:
>   - Added explicit filter_np parameter to of_map_id() and of_map_msi_id()
>     per Dmitry Baryshkov's review feedback, making the filter explicit
>     instead of overloading arg->np as both input filter and output parameter.
>   - Removed of_node_put() from inside of_map_id(), making the caller responsible
>     for reference management. Updated of_msi_xlate() to properly handle reference counting.
>   - Collected ACKed by tags, and fixed minor typos.
>   Link to v10:
>   https://lore.kernel.org/r/20260309-parse_iommu_cells-v10-0-c62fcaa5a1d8@oss.qualcomm.com
> 
> V10:
>   - Move of_map_iommu_id()/of_map_msi_id() from include/linux/of.h to
>     drivers/of/base.c as out-of-line helpers per feedback from Marc Zyngier
>     and Rob Herring.
>   - Add kernel-doc to document both helpers for discoverability and
>     usage clarity.
>   - Fix of_map_msi_id() wrapper and all its callers (cdx_msi.c,
>     irq-gic-its-msi-parent.c, drivers/of/irq.c) to correctly use the new
>     struct of_phandle_args-based API with proper of_node_put() handling
>     as per feeback from Dmitry.
>   Link to v9:
>   https://lore.kernel.org/r/20260301-parse_iommu_cells-v9-0-4d1bceecc5e1@oss.qualcomm.com
> 
> V9:
>   - Updated TO/CC list based on feedback to include all relevant
>     maintainers.
>   - No functional changes to the patches themselves.
> 
>   Link to V8:
>   https://lore.kernel.org/all/20260226074245.3098486-1-vijayanand.jitta@oss.qualcomm.com/
> 
> V8:
>   - Removed mentions of of_map_args from commit message to match code.
> 
>   Link to V7:
>   https://lore.kernel.org/all/20260210101157.2145113-1-vijayanand.jitta@oss.qualcomm.com/
> 
> V7:
>   - Removed of_map_id_args structure and replaced it with
>     of_phandle_args as suggested by Dmitry.
> 
>   Link to V6:
>   https://lore.kernel.org/all/20260121055400.937856-1-vijayanand.jitta@oss.qualcomm.com/
> 
> V6:
>   - Fixed build error reported by kernel test bot.
> 
>   Link to V5:
>   https://lore.kernel.org/all/20260118181125.1436036-1-vijayanand.jitta@oss.qualcomm.com/
> 
> V5:
>   - Fixed Build Warnings.
>   - Raised PR for iommu-map dtschema: https://github.com/devicetree-org/dt-schema/pull/184
> 
>   Link to V4:
>   https://lore.kernel.org/all/20251231114257.2382820-1-vijayanand.jitta@oss.qualcomm.com/
> 
> V4:
>   - Added Reviewed-by tag.
>   - Resolved warnings reported by kernel test bot, minor code
>     reorganization.
> 
>   Link to V3:
>   https://lore.kernel.org/all/20251221213602.2413124-1-vijayanand.jitta@oss.qualcomm.com/
> 
> V3:
>   - Added Reviewed-by tag.
>   - Updated of_map_id_args struct as a wrapper to of_phandle_args and
>     added comment description as suggested by Rob Herring.
> 
>   Link to V2:
>   https://lore.kernel.org/all/20251204095530.8627-1-vijayanand.jitta@oss.qualcomm.com/
> 
> V2:
>   - Incorporated the patches from Robin that does the clean implementation.
>   - Dropped the patches the were adding multi-map support from this series
>     as suggested.
> 
> V1:
>  https://lore.kernel.org/all/cover.1762235099.git.charan.kalla@oss.qualcomm.com/
> 
> RFC:
>  https://lore.kernel.org/all/20250928171718.436440-1-charan.kalla@oss.qualcomm.com/#r
> 
> Signed-off-by: Vijayanand Jitta <vijayanand.jitta@oss.qualcomm.com>
> ---
> 
> ---
> Charan Teja Kalla (1):
>       of: Factor arguments passed to of_map_id() into a struct
> 
> Robin Murphy (2):
>       of: Add convenience wrappers for of_map_id()
>       of: Respect #{iommu,msi}-cells in maps
> 
>  drivers/cdx/cdx_msi.c                    |   8 +-
>  drivers/iommu/of_iommu.c                 |   6 +-
>  drivers/irqchip/irq-gic-its-msi-parent.c |  11 +-
>  drivers/of/base.c                        | 213 ++++++++++++++++++++++++-------
>  drivers/of/irq.c                         |  11 +-
>  drivers/pci/controller/dwc/pci-imx6.c    |  18 ++-
>  drivers/pci/controller/pcie-apple.c      |   6 +-
>  drivers/xen/grant-dma-ops.c              |   5 +-
>  include/linux/of.h                       |  30 ++++-
>  9 files changed, 237 insertions(+), 71 deletions(-)
> ---
> base-commit: 3fa5e5702a82d259897bd7e209469bc06368bf31
> change-id: 20260301-parse_iommu_cells-1c33768aebba
> 
> Best regards,

Hi Robin,

Could this series be pulled into an immutable branch/tag, if it doesn't
make it into the v7.1 merge window ? There are client changes dependent on it,
So it would help to get them moving forward rather than waiting another cycle.

Thanks,
Vijay