Re: [PATCH v6 4/5] remoteproc: ingenic: Added remoteproc driver

[Paul Cercueil <paul@crapouillou.net>]

Hi Bjorn,

Le dim. 19 avril 2020 à 23:37, Bjorn Andersson 
<bjorn.andersson@linaro.org> a écrit :
> On Fri 17 Apr 10:00 PDT 2020, Paul Cercueil wrote:
> 
>>  This driver is used to boot, communicate with and load firmwares to 
>> the
>>  MIPS co-processor found in the VPU hardware of the JZ47xx SoCs from
>>  Ingenic.
>> 
>>  Signed-off-by: Paul Cercueil <paul@crapouillou.net>
>>  Signed-off-by: kbuild test robot <lkp@intel.com>
>>  Signed-off-by: Julia Lawall <julia.lawall@lip6.fr>
> 
> Please read Documentation/process/submitting-patches.rst about
> "Developer's Certificate of Origin".
> 
> I suspect that you incorporated review feedback on previous revisions
> from kbuild and Julia, this is generally omitted from the actual 
> commit
> message.

Julia / kbuild sent a patch to fix an error in the driver, so my patch 
now has code from Julia / kbuild. That document clearly says that I 
should add their signed-off-by. Or what do you mean?

>>  Acked-by: Mathieu Poirier <mathieu.poirier@linaro.org>
>>  ---
>> 
>>  Notes:
>>      v2: Remove exception for always-mapped memories
>>      v3: - Use clk_bulk API
>>      	- Move device-managed code to its own patch [3/4]
>>      	- Move devicetree table right above ingenic_rproc_driver
>>      	- Removed #ifdef CONFIG_OF around devicetree table
>>      	- Removed .owner = THIS_MODULE in ingenic_rproc_driver
>>      	- Removed useless platform_set_drvdata()
>>      v4: - Add fix reported by Julia
>>      	- Change Kconfig symbol to INGENIC_VPU_RPROC
>>      	- Add documentation to struct vpu
>>      	- disable_irq_nosync() -> disable_irq()
>>      v5: No change
>>      v6: Instead of prepare/unprepare callbacks, use PM runtime 
>> callbacks
>> 
>>   drivers/remoteproc/Kconfig         |   8 +
>>   drivers/remoteproc/Makefile        |   1 +
>>   drivers/remoteproc/ingenic_rproc.c | 282 
>> +++++++++++++++++++++++++++++
>>   3 files changed, 291 insertions(+)
>>   create mode 100644 drivers/remoteproc/ingenic_rproc.c
>> 
>>  diff --git a/drivers/remoteproc/Kconfig b/drivers/remoteproc/Kconfig
>>  index fbaed079b299..31da3e6c6281 100644
>>  --- a/drivers/remoteproc/Kconfig
>>  +++ b/drivers/remoteproc/Kconfig
>>  @@ -240,6 +240,14 @@ config STM32_RPROC
>> 
>>   	  This can be either built-in or a loadable module.
>> 
>>  +config INGENIC_VPU_RPROC
> 
> Please try to keep things alphabetically ordered.
> 
>>  +	tristate "Ingenic JZ47xx VPU remoteproc support"
>>  +	depends on MIPS || COMPILE_TEST
>>  +	help
>>  +	  Say y or m here to support the VPU in the JZ47xx SoCs from 
>> Ingenic.
>>  +	  This can be either built-in or a loadable module.
>>  +	  If unsure say N.
>>  +
>>   endif # REMOTEPROC
>> 
>>   endmenu
> [..]
>>  diff --git a/drivers/remoteproc/ingenic_rproc.c 
>> b/drivers/remoteproc/ingenic_rproc.c
> [..]
>>  +/**
>>  + * struct vpu - Ingenic VPU remoteproc private structure
>>  + * @irq: interrupt number
>>  + * @clks: pointers to the VPU and AUX clocks
> 
> aux_base is missing
> 
>>  + * @mem_info: array of struct vpu_mem_info, which contain the 
>> mapping info of
>>  + *            each of the external memories
>>  + * @dev: private pointer to the device
>>  + */
>>  +struct vpu {
>>  +	int irq;
>>  +	struct clk_bulk_data clks[2];
>>  +	void __iomem *aux_base;
>>  +	struct vpu_mem_info mem_info[ARRAY_SIZE(vpu_mem_map)];
>>  +	struct device *dev;
>>  +};
> [..]
>>  +static void *ingenic_rproc_da_to_va(struct rproc *rproc, u64 da, 
>> size_t len)
>>  +{
>>  +	struct vpu *vpu = rproc->priv;
>>  +	void __iomem *va = NULL;
>>  +	unsigned int i;
>>  +
>>  +	if (len <= 0)
> 
> len can't be negative (also, does it add value to check for and fail 
> len
> == 0?)
> 
>>  +		return NULL;
>>  +
>>  +	for (i = 0; i < ARRAY_SIZE(vpu_mem_map); i++) {
>>  +		const struct vpu_mem_info *info = &vpu->mem_info[i];
>>  +		const struct vpu_mem_map *map = info->map;
>>  +
>>  +		if (da >= map->da && (da + len) < (map->da + info->len)) {
>>  +			va = info->base + (da - map->da);
>>  +			break;
>>  +		}
>>  +	}
>>  +
>>  +	return (__force void *)va;
>>  +}
> [..]
>>  +static struct platform_driver ingenic_rproc_driver = {
>>  +	.probe = ingenic_rproc_probe,
>>  +	.driver = {
>>  +		.name = "ingenic-vpu",
>>  +#ifdef CONFIG_PM
> 
> Please omit the #ifdef here.

If I do, then the PM callbacks will be compiled in even if CONFIG_PM is 
disabled. That means dead code and I see no reason why you would want 
that.

If you don't mind, I'd like to keep the #ifdef CONFIG_PM for now, until 
this patchset is merged: https://lkml.org/lkml/2020/4/13/582

Then it would become a one-liner:
.pm = pm_ptr(&ingenic_rproc_pm),

Cheers,
-Paul

>>  +		.pm = &ingenic_rproc_pm,
>>  +#endif
>>  +		.of_match_table = of_match_ptr(ingenic_rproc_of_matches),
> 
> Please omit the of_match_ptr()
> 
> Regards,
> Bjorn
> 
>>  +	},
>>  +};
>>  +module_platform_driver(ingenic_rproc_driver);
>>  +
>>  +MODULE_LICENSE("GPL");
>>  +MODULE_AUTHOR("Paul Cercueil <paul@crapouillou.net>");
>>  +MODULE_DESCRIPTION("Ingenic JZ47xx Remote Processor control 
>> driver");
>>  --
>>  2.25.1
>>