On 2020/10/1 17:27, Vicente Bergas wrote:
> On Thu, Oct 1, 2020 at 11:07 AM Adrian Hunter <adrian.hunter@xxxxxxxxx> wrote:
>>
>> On 1/10/20 11:38 am, Vicente Bergas wrote:
>>> On Thu, Oct 1, 2020 at 9:18 AM Coly Li <colyli@xxxxxxx> wrote:
>>>>
>>>> In mmc_queue_setup_discard() the mmc driver queue's discard_granularity
>>>> might be set as 0 (when card->pref_erase > max_discard) while the mmc
>>>> device still declares to support discard operation. This is buggy and
>>>> triggered the following kernel warning message,
>>>>
>>>> WARNING: CPU: 0 PID: 135 at __blkdev_issue_discard+0x200/0x294
>>>> CPU: 0 PID: 135 Comm: f2fs_discard-17 Not tainted 5.9.0-rc6 #1
>>>> Hardware name: Google Kevin (DT)
>>>> pstate: 00000005 (nzcv daif -PAN -UAO BTYPE=--)
>>>> pc : __blkdev_issue_discard+0x200/0x294
>>>> lr : __blkdev_issue_discard+0x54/0x294
>>>> sp : ffff800011dd3b10
>>>> x29: ffff800011dd3b10 x28: 0000000000000000 x27: ffff800011dd3cc4 x26: ffff800011dd3e18 x25: 000000000004e69b x24: 0000000000000c40 x23: ffff0000f1deaaf0 x22: ffff0000f2849200 x21: 00000000002734d8 x20: 0000000000000008 x19: 0000000000000000 x18: 0000000000000000 x17: 0000000000000000 x16: 0000000000000000 x15: 0000000000000000 x14: 0000000000000394 x13: 0000000000000000 x12: 0000000000000000 x11: 0000000000000000 x10: 00000000000008b0 x9 : ffff800011dd3cb0 x8 : 000000000004e69b x7 : 0000000000000000 x6 : ffff0000f1926400 x5 : ffff0000f1940800 x4 : 0000000000000000 x3 : 0000000000000c40 x2 : 0000000000000008 x1 : 00000000002734d8 x0 : 0000000000000000 Call trace:
>>>> __blkdev_issue_discard+0x200/0x294
>>>> __submit_discard_cmd+0x128/0x374
>>>> __issue_discard_cmd_orderly+0x188/0x244
>>>> __issue_discard_cmd+0x2e8/0x33c
>>>> issue_discard_thread+0xe8/0x2f0
>>>> kthread+0x11c/0x120
>>>> ret_from_fork+0x10/0x1c
>>>> ---[ end trace e4c8023d33dfe77a ]---
>>>>
>>>> This patch fixes the issue by setting discard_granularity as SECTOR_SIZE
>>>> instead of 0 when (card->pref_erase > max_discard) is true. Now no more
>>>> complain from __blkdev_issue_discard() for the improper value of discard
>>>> granularity.
>>>>
>>>> This issue is exposed after commit b35fd7422c2f ("block: check queue's
>>>> limits.discard_granularity in __blkdev_issue_discard()"), a "Fixes:" tag
>>>> is also added for the commit to make sure people won't miss this patch
>>>> after applying the change of __blkdev_issue_discard().
>>>>
>>>> Fixes: e056a1b5b67b ("mmc: queue: let host controllers specify maximum discard timeout")
>>>> Fixes: b35fd7422c2f ("block: check queue's limits.discard_granularity in __blkdev_issue_discard()").
>>>> Reported-by: Vicente Bergas <vicencb@xxxxxxxxx>
>>>
>>> Hi Coly, Adrian,
>>> would you like me to reply with a tested-by?
>>> before testing it: what are the chances of losing data?
>>
>> Very low I would hope, but that is no reason not to have backups :-)
>
> I do have backup, it is for the hassle of reinstalling the OS and
> restoring the backup.
> I'll test it later, maybe today.
>
>>> Regards,
>>> Vicente.
>>>
>>>> Signed-off-by: Coly Li <colyli@xxxxxxx>
>>>> Acked-by: Adrian Hunter <adrian.hunter@xxxxxxxxx>
>>>> Cc: Ulf Hansson <ulf.hansson@xxxxxxxxxx>
>>>> ---
>>>> Changelog,
>>>> v3, add Fixes tag for both commits.
>>>> v2, change commit id of the Fixes tag.
>>>> v1, initial version.
>>>>
>>>> drivers/mmc/core/queue.c | 2 +-
>>>> 1 file changed, 1 insertion(+), 1 deletion(-)
>>>>
>>>> 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 */
>
> Just out of curiosity, wouldn't it be more performant something like:
> q->limits.discard_granularity = minimum(card->pref_erase, max_discard) << 9;
> or is just that SECTOR_SIZE is a number guaranteed to work?
>
The discard_granularity is a hint for the minimal acceptable length of
the discard request. The discard request might be accepted by device
when the discard request covers a discard_granularity range. The
discard_granularity rage is the minimum available discard request
length, it is a LBA range which starts on discard_granularity aligned
LBA, and length is discard_granularity.
Therefore setting discard_granularity to SECTOR_SIZE won't hurt discard
performance, the discard bio split does not depend on it. It is a hint
for a minimal acceptable length of discard request.
Coly Li
