Re: [PATCH v4] mmc: core: don't set limits.discard_granularity as 0

[Coly Li]

On 2020/10/2 20:57, Martin K. Petersen wrote:
> 
> Coly,
> 
>> diff --git a/drivers/mmc/core/queue.c b/drivers/mmc/core/queue.c
>> index 6c022ef0f84d..350d0cc4ee62 100644
>> --- a/drivers/mmc/core/queue.c
>> +++ b/drivers/mmc/core/queue.c
>> @@ -190,7 +190,7 @@ static void mmc_queue_setup_discard(struct request_queue *q,
>>  	q->limits.discard_granularity = card->pref_erase << 9;
>>  	/* granularity must not be greater than max. discard */
>>  	if (card->pref_erase > max_discard)
>> -		q->limits.discard_granularity = 0;
>> +		q->limits.discard_granularity = SECTOR_SIZE;
>>  	if (mmc_can_secure_erase_trim(card))
>>  		blk_queue_flag_set(QUEUE_FLAG_SECERASE, q);
>>  }
> 

Hi Martin,

> The granularity should probably be set to the logical block size instead
> of SECTOR_SIZE. However, looking at mmc_setup_queue() it doesn't appear
> the logical block size is read from the CSD until after discard has been
> configured. So that will require a bit of code shuffling.
> 

Copied, then let me try to modify mmc_setup_queue() too, and set the
discard_granularity from SECTOR_SIZE to logical block size.

Thanks for the hint.

Coly Li