Re: [PATCH v4 03/16] x86/virt/tdx: Simplify tdmr_get_pamt_sz()

[Binbin Wu <binbin.wu@linux.intel.com>]

On 11/25/2025 3:47 AM, Edgecombe, Rick P wrote:
> On Mon, 2025-11-24 at 17:26 +0800, Binbin Wu wrote:
>> Reviewed-by: Binbin Wu <binbin.wu@linux.intel.com>
> Thanks.
>
>> One nit below.
>>
>> [...]
>>> @@ -535,26 +518,18 @@ static int tdmr_set_up_pamt(struct tdmr_info *tdmr,
>>>     	 * in overlapped TDMRs.
>>>     	 */
>>>     	pamt = alloc_contig_pages(tdmr_pamt_size >> PAGE_SHIFT, GFP_KERNEL,
>>> -			nid, &node_online_map);
>>> -	if (!pamt)
>>> +				  nid, &node_online_map);
>>> +	if (!pamt) {
>>> +		/*
>>> +		 * tdmr->pamt_4k_base is zero so the
>>> +		 * error path will skip freeing.
>>> +		 */
>>>     		return -ENOMEM;
>> Nit:
>> Do you think it's OK to move the comment up so to avoid multiple lines of
>> comments as well as the curly braces?
> Yea, I think that is a good point. But I'm also thinking that this comment is
> not clear enough. There is no error path to speak of in this function, so maybe:
>
> 	/*
> 	 * tdmr->pamt_4k_base is still zero so the error
> 	 * path of the caller will skip freeing the pamt.
> 	 */
>
> If you agree I will keep your RB.
Yes, please.

>>           /* tdmr->pamt_4k_base is zero so the error path will skip freeing. */
>>           if (!pamt)
>>               return -ENOMEM;