Re: [PATCH v2 3/3] nbd: correct the maximum value for discard sectors

From: Damien Le Moal
Date: Wed Aug 07 2024 - 13:20:29 EST


On 2024/08/07 6:56, Josef Bacik wrote:
> On Tue, Aug 06, 2024 at 03:30:56PM +0200, Wouter Verhelst wrote:
>> The version of the NBD protocol implemented by the kernel driver
>> currently has a 32 bit field for length values. As the NBD protocol uses
>> bytes as a unit of length, length values larger than 2^32 bytes cannot
>> be expressed.
>>
>> Update the max_hw_discard_sectors field to match that.
>>
>> Signed-off-by: Wouter Verhelst <w@xxxxxxx>
>> Fixes: 268283244c0f018dec8bf4a9c69ce50684561f46
>> ---
>> drivers/block/nbd.c | 2 +-
>> 1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c
>> index 20e9f9fdeaae..1457f0c8a4a4 100644
>> --- a/drivers/block/nbd.c
>> +++ b/drivers/block/nbd.c
>> @@ -339,7 +339,7 @@ static int __nbd_set_size(struct nbd_device *nbd, loff_t bytesize,
>>
>> lim = queue_limits_start_update(nbd->disk->queue);
>> if (nbd->config->flags & NBD_FLAG_SEND_TRIM)
>> - lim.max_hw_discard_sectors = UINT_MAX;
>> + lim.max_hw_discard_sectors = UINT_MAX / blksize;
>
> We use 512 as the "sectors" measurement throughout the block layer, so our limit
> is actually
>
> UINT32_MAX >> 9

UINT_MAX >> SECTOR_SHIFT

would be better.

>
> since we can only send at most UINT32_MAX as our length. Fix it to be that for
> both patches and you should be good. Thanks,
>
> Josef
>

--
Damien Le Moal
Western Digital Research