of_platform_depopulate can lead to a kernel error when calling 
release_resource()

of_platform_depopulate()
      of_platform_device_destroy()
        platform_device_unregister(platform_device *pdev)
          platform_device_del(platform_device *pdev)
            for (i = 0; i < pdev->num_resources; i++)
                      release_resource()

The reason is that it is trying to release a resource that was not allocated
via insert_resource

of_platform_populate()
      ...
        of_device_alloc()
          pdev = platform_device_alloc()
          # set pdev->resource, similar to platform_device_add_resources()
          of_device_add(platform_device *pdev)
            # similar to platform_device_add(), but note there's no
            # insert_resource() in this path
            device_add(&pdev->dev)

This set of patches modifies release_resource to check for
resource->parent==NULL and throw a warning if there is an error.

base/platform has been fixed for an hypothetical condition where parent is
set but the platform is neither MEM or IO.

Then platform_device_alloc has been modified so it supports of and amba 
devices.

Finally of_device_add has been modified to use platform_device_add().

v1: https://lkml.org/lkml/2015/4/20/435

v2: https://lkml.org/lkml/2015/4/21/99
    https://lkml.org/lkml/2015/4/21/100

Ricardo Ribalda Delgado (4):
  kernel/resource: Invalid memory access in __release_resource
  base/platform: Only insert MEM and IO resources
  base/platform: Continue on insert_resource() error
  of/platform: Use platform_device interface

 drivers/base/platform.c | 20 ++++++++++++--------
 drivers/of/platform.c   |  3 ++-
 kernel/resource.c       |  3 +++
 3 files changed, 17 insertions(+), 9 deletions(-)

-- 
2.1.4

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Reply via email to