Messages in this thread | | | Date | Sat, 9 Oct 2021 11:46:23 +0300 | Subject | Re: [PATCH] block: nbd: fix sanity check for first_minor | From | Pavel Skripkin <> |
| |
On 10/9/21 08:35, Yan, Zheng wrote:
+CC Christoph
> From: "Yan, Zheng" <yanzheng03@kuaishou.com> > > Device's minor is a 20-bits number, max value is 0xfffff. > > Fixes: b1a811633f ("block: nbd: add sanity check for first_minor"). > Signed-off-by: "Yan, Zheng" <yanzheng03@kuaishou.com> > --- > 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 1183f7872b71..53cd038c96e9 100644 > --- a/drivers/block/nbd.c > +++ b/drivers/block/nbd.c > @@ -1753,7 +1753,7 @@ static struct nbd_device *nbd_dev_add(int index, unsigned int refs) > * byte in __device_add_disk(). > */ > disk->first_minor = index << part_shift; > - if (disk->first_minor > 0xff) { > + if (disk->first_minor > MINORMASK) { > err = -EINVAL; > goto out_free_idr; > } >
I've looked into what confused me with 0xff and I found that uapi's kdev_t.h has following MINOR definition:
#define MINOR(dev) ((dev) & 0xff)
But kernel's kdev_t.h is really allows 20 bit minor number :)
Also, can you, please, fix up comment above this check, since it contains info about max minor 0xff and it can confuse readers.
With regards, Pavel Skripkin
| |