Re: [PATCH v2 10/12] remoteproc: stm32: Introduce new parse fw ops for synchronisation

[Bjorn Andersson]

On Fri 24 Apr 13:25 PDT 2020, Mathieu Poirier wrote:

> Introduce new parse firmware rproc_ops functions to be used when
> synchonising with an MCU.
> 
> Mainly based on the work published by Arnaud Pouliquen [1].
> 
> [1]. https://patchwork.kernel.org/project/linux-remoteproc/list/?series=239877
> 
> Signed-off-by: Mathieu Poirier <mathieu.poirier@xxxxxxxxxx>
> Reviewed-by: Loic Pallardy <loic.pallardy@xxxxxx>
> ---
>  drivers/remoteproc/stm32_rproc.c | 51 +++++++++++++++++++++++++++++++-
>  1 file changed, 50 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/remoteproc/stm32_rproc.c b/drivers/remoteproc/stm32_rproc.c
> index 86d23c35d805..b8ae8aed5585 100644
> --- a/drivers/remoteproc/stm32_rproc.c
> +++ b/drivers/remoteproc/stm32_rproc.c
> @@ -215,7 +215,34 @@ static int stm32_rproc_elf_load_rsc_table(struct rproc *rproc,
>  	return 0;
>  }
>  
> -static int stm32_rproc_parse_fw(struct rproc *rproc, const struct firmware *fw)
> +static int stm32_rproc_sync_elf_load_rsc_table(struct rproc *rproc,
> +					       const struct firmware *fw)
> +{
> +	struct resource_table *table = NULL;
> +	struct stm32_rproc *ddata = rproc->priv;
> +
> +	if (ddata->rsc_va) {

Does it really make sense to try to sync with a remote that doesn't have
a resource table?

> +		table = (struct resource_table *)ddata->rsc_va;
> +		/* Assuming that the resource table fits in 1kB is fair */
> +		rproc->cached_table = kmemdup(table, RSC_TBL_SIZE, GFP_KERNEL);

It's unfortunate that we need to create a clone of the resource table
that we found in ram, and then return the original memory when the core
ask for the loaded table...

I wonder if we somehow can avoid this in the core (i.e. skip overwriting
table_ptr with the cached_table during stop)

> +		if (!rproc->cached_table)
> +			return -ENOMEM;
> +
> +		rproc->table_ptr = rproc->cached_table;
> +		rproc->table_sz = RSC_TBL_SIZE;
> +		return 0;
> +	}
> +
> +	rproc->cached_table = NULL;
> +	rproc->table_ptr = NULL;
> +	rproc->table_sz = 0;
> +
> +	dev_warn(&rproc->dev, "no resource table found for this firmware\n");
> +	return 0;
> +}
> +
> +static int stm32_rproc_parse_memory_regions(struct rproc *rproc,
> +					    const struct firmware *fw)
>  {
>  	struct device *dev = rproc->dev.parent;
>  	struct device_node *np = dev->of_node;
> @@ -268,9 +295,30 @@ static int stm32_rproc_parse_fw(struct rproc *rproc, const struct firmware *fw)
>  		index++;
>  	}
>  
> +	return 0;
> +}
> +
> +static int stm32_rproc_parse_fw(struct rproc *rproc, const struct firmware *fw)
> +{
> +	int ret = stm32_rproc_parse_memory_regions(rproc, fw);
> +
> +	if (ret)
> +		return ret;
> +
>  	return stm32_rproc_elf_load_rsc_table(rproc, fw);
>  }
>  
> +static int stm32_rproc_sync_parse_fw(struct rproc *rproc,
> +				     const struct firmware *fw)

Rather than having a function parse_fw that is passed no fw and return
some state that was setup in probe, how about just do this during probe?

Regards,
Bjorn

> +{
> +	int ret = stm32_rproc_parse_memory_regions(rproc, fw);
> +
> +	if (ret)
> +		return ret;
> +
> +	return stm32_rproc_sync_elf_load_rsc_table(rproc, fw);
> +}
> +
>  static irqreturn_t stm32_rproc_wdg(int irq, void *data)
>  {
>  	struct platform_device *pdev = data;
> @@ -544,6 +592,7 @@ static struct rproc_ops st_rproc_ops = {
>  static __maybe_unused struct rproc_ops st_rproc_sync_ops = {
>  	.start		= stm32_rproc_sync_start,
>  	.stop		= stm32_rproc_stop,
> +	.parse_fw       = stm32_rproc_sync_parse_fw,
>  };
>  
>  static const struct of_device_id stm32_rproc_match[] = {
> -- 
> 2.20.1
>