Re: [PATCH v4 1/4] media: venus: hfi_parser: add check to avoid out of bound access

[Vikash Garodia]

On 2/20/2025 8:46 PM, Hans Verkuil wrote:
> On 2/7/25 09:24, Vikash Garodia wrote:
>> There is a possibility that init_codecs is invoked multiple times during
>> manipulated payload from video firmware. In such case, if codecs_count
>> can get incremented to value more than MAX_CODEC_NUM, there can be OOB
>> access. Reset the count so that it always starts from beginning.
>>
>> Cc: stable@xxxxxxxxxxxxxxx
>> Fixes: 1a73374a04e5 ("media: venus: hfi_parser: add common capability parser")
>> Reviewed-by: Bryan O'Donoghue <bryan.odonoghue@xxxxxxxxxx>
>> Signed-off-by: Vikash Garodia <quic_vgarodia@xxxxxxxxxxx>
>> ---
>>  drivers/media/platform/qcom/venus/hfi_parser.c | 1 +
>>  1 file changed, 1 insertion(+)
>>
>> diff --git a/drivers/media/platform/qcom/venus/hfi_parser.c b/drivers/media/platform/qcom/venus/hfi_parser.c
>> index 3df241dc3a118bcdeb2c28a6ffdb907b644d5653..1cc17f3dc8948160ea6c3015d2c03e475b8aa29e 100644
>> --- a/drivers/media/platform/qcom/venus/hfi_parser.c
>> +++ b/drivers/media/platform/qcom/venus/hfi_parser.c
>> @@ -17,6 +17,7 @@ typedef void (*func)(struct hfi_plat_caps *cap, const void *data,
>>  static void init_codecs(struct venus_core *core)
>>  {
>>  	struct hfi_plat_caps *caps = core->caps, *cap;
>> +	core->codecs_count = 0;
> 
> This really should be moved down to before the 'if'. There is no reason to mix the assignment
> with variable declarations.
Thats correct, will move it below.
> 
>>  	unsigned long bit;
>>  
>>  	if (hweight_long(core->dec_codecs) + hweight_long(core->enc_codecs) > MAX_CODEC_NUM)
>>
> 
> Regards,
> 
> 	Hans
Regards,
Vikash