Re: [PATCH v2 1/2] x86: Check return values from early_memremap calls

[Ross Philipson <ross.philipson@oracle.com>]

On 11/10/22 11:07, Dave Hansen wrote:
> On 11/10/22 07:45, Ross Philipson wrote:
>>   	dt = early_memremap(initial_dtb, map_len);
>> +	if (!dt) {
>> +		pr_warn("failed to memremap initial dtb\n");
>> +		return;
>> +	}
> 
> Are all of these new pr_warn/err()'s really adding much value?  They all
> look pretty generic.  It makes me wonder if we should just spit out a
> generic message in early_memremap() and save all the callers the trouble.

These changes were prompted by some comments on an earlier patch set I 
sent. It was requested that I fix the other missing checks for NULL 
returns from these functions but I thought that was out of scope for 
that patch set. So I agreed to submit this set and add the checks making 
things consistent.

> 
> Oh, and don't we try to refer to functions() with parenthesis?

Yes I can fix that.

Thanks
Ross

>