On 12/8/2017 9:33 AM, Sinan Kaya wrote:

On 12/8/2017 4:11 AM, Lothar Waßmann wrote:

quoted

quoted

While implementing the ACPI piece, I have to convert an unsigned long to
(const void *) in ACPI code so that the APIs are compatible.

Just one more remark: Do you need write access to the data the pointer
returned by device_get_match_data() or of_fwnode_get_match_data()
points to?
If not, the return type of those functions should be 'const void *'.

Yes, the only reason driver is trying to obtain this data pointer is to modify
members of it. I did a quick test with your suggestion.

I guess I should soften my statement here. I look at examples, they seem to be read.
Even for my HIDMA case:

static const struct of_device_id hidma_match[] = {
	{.compatible = "qcom,hidma-1.0",},
	{.compatible = "qcom,hidma-1.1", .data = (void *)(HIDMA_MSI_CAP),},
	{.compatible = "qcom,hidma-1.2",
	 .data = (void *)(HIDMA_MSI_CAP | HIDMA_IDENTITY_CAP),},
	{},
}; 

I can change the return type to (const void*) if Rob, Rafael and Sakari agrees.
I didn't really like converting a long to const void*. That's my personal opinion.

test.c: In function ‘main’:
test.c:15:2: error: assignment of member ‘m’ in read-only object
  t->m = 3;
  ^

struct test
{
        int m;
};

int main(void)
{
        const void *ptr;
        unsigned long l =4;
        const struct test *t;

        ptr = (const void *)l;

        t = ptr;
        t->m = 3;
}

-- 
Sinan Kaya
Qualcomm Datacenter Technologies, Inc. as an affiliate of Qualcomm Technologies, Inc.
Qualcomm Technologies, Inc. is a member of the Code Aurora Forum, a Linux Foundation Collaborative Project.