Re: [PATCH v2] cxl/acpi: Verify CHBS length for CXL2.0

["Zhijian Li (Fujitsu)" <lizhijian@fujitsu.com>]

On 09/04/2025 09:55, Alison Schofield wrote:
> On Mon, Apr 07, 2025 at 04:02:17PM +0800, Li Zhijian wrote:
>> Per CXL Spec r3.1 Table 9-21, both CXL1.1 and CXL2.0 have defined their
>> own length, verify it to avoid an invalid CHBS.
> 
> Can you comment on removing the comparison to CXL_RCRB_SIZE. At some
> point someone thought that was useful. As you remove it, please
> explain.

Thanks for your feedback.

Okay, I will update it in the commit log in V3

Remove the redundant macro CXL_RCRB_SIZE, as it has the same meaning as
ACPI_CEDT_CHBS_LENGTH_CXL11, which name is more comprehensible.

Thanks
Zhijian

> 
> 
>>
>> Signed-off-by: Li Zhijian <lizhijian@fujitsu.com>
>> ---
>> V2: don't factor out, just validate # Dan
>> ---
>>   drivers/cxl/acpi.c | 8 ++++----
>>   1 file changed, 4 insertions(+), 4 deletions(-)
>>
>> diff --git a/drivers/cxl/acpi.c b/drivers/cxl/acpi.c
>> index cb14829bb9be..2e63e50b2c40 100644
>> --- a/drivers/cxl/acpi.c
>> +++ b/drivers/cxl/acpi.c
>> @@ -11,8 +11,6 @@
>>   #include "cxlpci.h"
>>   #include "cxl.h"
>>   
>> -#define CXL_RCRB_SIZE	SZ_8K
>> -
>>   struct cxl_cxims_data {
>>   	int nr_maps;
>>   	u64 xormaps[] __counted_by(nr_maps);
>> @@ -478,8 +476,10 @@ static int cxl_get_chbs_iter(union acpi_subtable_headers *header, void *arg,
>>   
>>   	chbs = (struct acpi_cedt_chbs *) header;
>>   
>> -	if (chbs->cxl_version == ACPI_CEDT_CHBS_VERSION_CXL11 &&
>> -	    chbs->length != CXL_RCRB_SIZE)
>> +	if ((chbs->cxl_version == ACPI_CEDT_CHBS_VERSION_CXL11 &&
>> +	    chbs->length != ACPI_CEDT_CHBS_LENGTH_CXL11) ||
>> +	   (chbs->cxl_version == ACPI_CEDT_CHBS_VERSION_CXL20 &&
>> +	    chbs->length != ACPI_CEDT_CHBS_LENGTH_CXL20))
>>   		return 0;
>>   
>>   	if (!chbs->base)
>> -- 
>> 2.47.0
>>