Re: [RFC PATCH v1 1/4] arm64: Implement infrastructure for stack trace reliability checks

[Madhavan T. Venkataraman]

On 4/1/21 10:27 AM, Mark Brown wrote:
> On Tue, Mar 30, 2021 at 02:09:52PM -0500, madvenka@xxxxxxxxxxxxxxxxxxx wrote:
>> From: "Madhavan T. Venkataraman" <madvenka@xxxxxxxxxxxxxxxxxxx>
>>
>> Implement a check_reliability() function that will contain checks for the
>> presence of various features and conditions that can render the stack trace
>> unreliable.
> 
> This looks good to me with one minor stylistic thing:
> 
>> +/*
>> + * Special functions where the stack trace is unreliable.
>> + */
>> +static struct function_range	special_functions[] = {
>> +	{ 0, 0 }
>> +};
> 
> Might be good to put a comment here saying that this is terminating the
> list rather than detecting a NULL function pointer:
> 
> 	{ /* sentinel */ }
> 
> is a common idiom for that.
> 
> Given that it's a fixed array we could also...
> 
>> +	for (func = special_functions; func->start; func++) {
>> +		if (pc >= func->start && pc < func->end)
> 
> ...do these as
> 
> 	for (i = 0; i < ARRAY_SIZE(special_functions); i++) 
> 
> so you don't need something like that, though that gets awkward when you
> have to write out special_functions[i].field a lot.
> 
> So many different potential colours for the bikeshed!
I will make the above changes.

Thanks!

Madhavan