On Tue, Apr 30, 2019 at 9:15 AM Fabien DESSENNE <fabien.dessenne@xxxxxx> wrote:
>
> Hi Rob,
>
>
> On 30/04/2019 2:40 AM, Rob Herring wrote:
> > On Tue, Apr 16, 2019 at 04:58:13PM +0200, Fabien Dessenne wrote:
> >> Add the device tree bindings document for the stm32 remoteproc devices.
> >>
> >> Signed-off-by: Fabien Dessenne <fabien.dessenne@xxxxxx>
> >> ---
> >> .../devicetree/bindings/remoteproc/stm32-rproc.txt | 64 ++++++++++++++++++++++
> >> 1 file changed, 64 insertions(+)
> >> create mode 100644 Documentation/devicetree/bindings/remoteproc/stm32-rproc.txt
> >>
> >> diff --git a/Documentation/devicetree/bindings/remoteproc/stm32-rproc.txt b/Documentation/devicetree/bindings/remoteproc/stm32-rproc.txt
> >> new file mode 100644
> >> index 0000000..430132c
> >> --- /dev/null
> >> +++ b/Documentation/devicetree/bindings/remoteproc/stm32-rproc.txt
> >> @@ -0,0 +1,64 @@
> >> +STMicroelectronics STM32 Remoteproc
> >> +-----------------------------------
> >> +This document defines the binding for the remoteproc component that loads and
> >> +boots firmwares on the ST32MP family chipset.
> >> +
> >> +Required properties:
> >> +- compatible: Must be "st,stm32mp1-m4"
> >> +- reg: Address ranges of the remote processor dedicated memories.
> >> + The parent node should provide an appropriate ranges property
> >> + for properly translating these into bus addresses.
> > dma-ranges, but that's independent of 'reg'.
> >
> > It needs to list how many reg regions and what they are.
>
> The "reg" property needs to be removed since it is not used by the
> driver : the "memory-region" property (defined below) provides with all
> the needed memory information.
I'm not sure that's the right direction. reserved-memory nodes
generally should be in the range of system memory (i.e. what /memory
nodes define).
> Unfortunately, when I remove this "reg" property from the DeviceTree, I
> have this warning when building (W=123) the DTB:
>
> "Warning (avoid_unnecessary_addr_size): /mlahb: unnecessary
> #address-cells/#size-cells without "ranges" or child "reg" property"
>
> IMHO, there is something wrong in the dtc script which seems to ignore
> the "dma-ranges" property that needs to have #address-cells/#size-cells
> defined. Just like "ranges".
Perhaps.
>
> The quick patch below (add check for "dma-ranges" ) in
> scripts/dtc/checks.c solves this issue.
>
> static void check_avoid_unnecessary_addr_size(struct check *c, struct
> dt_info *dti,
> struct node *node)
> {
> ...
> if (get_property(node, "ranges") || get_property(node,
> "dma-ranges") || !node->children)
> return;
> ...
>
> Can you confirm that I can remove the "reg" property and ignore the warning?
That should cause another warning because the 'simple-bus' checks
expect to have nodes with addresses as 'simple-bus' means MMIO bus.
That may have been the check which was broken for a while and I just
fixed not too long ago.
Rob
