[PATCH 2/5] dm: dm-zoned: pass GFP_KERNEL to blkdev_zone_mgmt

Mikulas Patocka mpatocka at redhat.com
Tue Jan 23 12:18:45 PST 2024



On Tue, 23 Jan 2024, Johannes Thumshirn wrote:

> The call to blkdev_zone_mgmt() in dm-zoned is only used to perform a
> ZONE_RESET operation when freeing a zone.
> 
> This is not done in the IO path, so we can use GFP_KERNEL here, as it will
> never recurse back into the driver on reclaim.

Hi

This doesn't seem safe to me. There's a possible call chain dmz_handle_bio 
-> dmz_put_chunk_mapping -> dmz_free_zone -> dmz_reset_zone -> 
blkdev_zone_mgmt -> recursion (via GFP_KERNEL) back into the dm-zoned 
target.

Mikulas

> Signed-off-by: Johannes Thumshirn <johannes.thumshirn at wdc.com>
> ---
>  drivers/md/dm-zoned-metadata.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/md/dm-zoned-metadata.c b/drivers/md/dm-zoned-metadata.c
> index fdfe30f7b697..a39734f0cb7d 100644
> --- a/drivers/md/dm-zoned-metadata.c
> +++ b/drivers/md/dm-zoned-metadata.c
> @@ -1658,7 +1658,7 @@ static int dmz_reset_zone(struct dmz_metadata *zmd, struct dm_zone *zone)
>  
>  		ret = blkdev_zone_mgmt(dev->bdev, REQ_OP_ZONE_RESET,
>  				       dmz_start_sect(zmd, zone),
> -				       zmd->zone_nr_sectors, GFP_NOIO);
> +				       zmd->zone_nr_sectors, GFP_KERNEL);
>  		if (ret) {
>  			dmz_dev_err(dev, "Reset zone %u failed %d",
>  				    zone->id, ret);
> 
> -- 
> 2.43.0
> 
> 




More information about the Linux-nvme mailing list