Re: [PATCH v1 3/5] misc: fastrpc: Fix initial memory allocation for Audio PD memory pool

[Ekansh Gupta <ekansh.gupta@oss.qualcomm.com>]

On 5/19/2025 5:11 PM, Srinivas Kandagatla wrote:
> On 5/13/25 05:28, Ekansh Gupta wrote:
>> The initially allocated memory is not properly included in the pool,
>> leading to potential issues with memory management. Set the number
>> of pages to one to ensure that the initially allocated memory is
>> correctly added to the Audio PD memory pool.
>>
>> Fixes: 0871561055e66 ("misc: fastrpc: Add support for audiopd")
>> Cc: stable@kernel.org
>> Signed-off-by: Ekansh Gupta <ekansh.gupta@oss.qualcomm.com>
>> ---
>>  drivers/misc/fastrpc.c | 7 ++++---
>>  1 file changed, 4 insertions(+), 3 deletions(-)
>>
>> diff --git a/drivers/misc/fastrpc.c b/drivers/misc/fastrpc.c
>> index d4e38b5e5e6c..b629e24f00bc 100644
>> --- a/drivers/misc/fastrpc.c
>> +++ b/drivers/misc/fastrpc.c
>> @@ -1311,6 +1311,9 @@ static int fastrpc_init_create_static_process(struct fastrpc_user *fl,
>>  		goto err;
>>  	}
>>  
>> +	inbuf.client_id = fl->client_id;
>> +	inbuf.namelen = init.namelen;
> inbuf is not used till the invoke call, why are we moving these two
> lines here?

I just moved it above so that the pagelen is updated properly.

Would you prefer storing pagelen to some local variable and then eventually
assigning it to inbuf.pagelen before making the invoke call?

>
>> +	inbuf.pageslen = 0;
>
>
>>  	if (!fl->cctx->audio_init_mem) {
>>  		err = fastrpc_remote_heap_alloc(fl, fl->sctx->dev, init.memlen,
>>  						&buf);
>> @@ -1335,12 +1338,10 @@ static int fastrpc_init_create_static_process(struct fastrpc_user *fl,
>>  			list_add_tail(&buf->node, &fl->cctx->rhmaps);
>>  			spin_unlock_irqrestore(&fl->cctx->lock, flags);
>>  			fl->cctx->audio_init_mem = true;
>> +			inbuf.pageslen = 1;
>>  		}
>>  	}
>>  
>> -	inbuf.client_id = fl->client_id;
>> -	inbuf.namelen = init.namelen;
>> -	inbuf.pageslen = 0;
>>  	fl->pd = USER_PD;
>>  
>>  	args[0].ptr = (u64)(uintptr_t)&inbuf;