[PATCH 1/2] dm-raid: Fix minimal device check

[PATCH 1/2] dm-raid: Fix minimal device check

[Andy Grover]

Check raid_type->minimal_devs instead of parity_devs. minimal_devs is used
nowhere else (and makes sense to use here) so I think this is what was
intended.

Signed-off-by: Andy Grover <agrover@redhat.com>
---
 drivers/md/dm-raid.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/md/dm-raid.c b/drivers/md/dm-raid.c
index a090121..5941cfd 100644
--- a/drivers/md/dm-raid.c
+++ b/drivers/md/dm-raid.c
@@ -151,7 +151,7 @@ static struct raid_set *context_alloc(struct dm_target *ti, struct raid_type *ra
 	unsigned i;
 	struct raid_set *rs;
 
-	if (raid_devs <= raid_type->parity_devs) {
+	if (raid_devs < raid_type->minimal_devs) {
 		ti->error = "Insufficient number of devices";
 		return ERR_PTR(-EINVAL);
 	}
-- 
2.4.3

[PATCH 2/2] dm-raid: Fix minimal_devs for raid6

[Andy Grover]

A raid6 device can be created from 3 devs. This previously didn't cause
an error because minimal_devs was not being checked.

Signed-off-by: Andy Grover <agrover@redhat.com>
---
 drivers/md/dm-raid.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/md/dm-raid.c b/drivers/md/dm-raid.c
index 5941cfd..1ded052 100644
--- a/drivers/md/dm-raid.c
+++ b/drivers/md/dm-raid.c
@@ -90,9 +90,9 @@ static struct raid_type {
 	{"raid5_ra", "RAID5 (right asymmetric)",	1, 2, 5, ALGORITHM_RIGHT_ASYMMETRIC},
 	{"raid5_ls", "RAID5 (left symmetric)",		1, 2, 5, ALGORITHM_LEFT_SYMMETRIC},
 	{"raid5_rs", "RAID5 (right symmetric)",		1, 2, 5, ALGORITHM_RIGHT_SYMMETRIC},
-	{"raid6_zr", "RAID6 (zero restart)",		2, 4, 6, ALGORITHM_ROTATING_ZERO_RESTART},
-	{"raid6_nr", "RAID6 (N restart)",		2, 4, 6, ALGORITHM_ROTATING_N_RESTART},
-	{"raid6_nc", "RAID6 (N continue)",		2, 4, 6, ALGORITHM_ROTATING_N_CONTINUE}
+	{"raid6_zr", "RAID6 (zero restart)",		2, 3, 6, ALGORITHM_ROTATING_ZERO_RESTART},
+	{"raid6_nr", "RAID6 (N restart)",		2, 3, 6, ALGORITHM_ROTATING_N_RESTART},
+	{"raid6_nc", "RAID6 (N continue)",		2, 3, 6, ALGORITHM_ROTATING_N_CONTINUE}
 };
 
 static char *raid10_md_layout_to_format(int layout)
-- 
2.4.3

Re: [PATCH 2/2] dm-raid: Fix minimal_devs for raid6

[Heinz Mauelshagen]

NACK

4 devs is actually the mandatory minimum for raid6;
minimum of 2 data stripes plus 2 devices for parity and q-syndrome.

FWIW: the 2 devices minimum for the raid4/5 layouts are correct too,
            because those personalities are able to drive 2 devices as raid1
            (i.e. allow for takeover from raid1 with 2 legs)
            in order to reshape them to 'real' raid4/5 by adding N>0 devices

Heinz

On 10/28/2015 07:50 PM, Andy Grover wrote:
> A raid6 device can be created from 3 devs. This previously didn't cause
> an error because minimal_devs was not being checked.
>
> Signed-off-by: Andy Grover <agrover@redhat.com>
> ---
>   drivers/md/dm-raid.c | 6 +++---
>   1 file changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/md/dm-raid.c b/drivers/md/dm-raid.c
> index 5941cfd..1ded052 100644
> --- a/drivers/md/dm-raid.c
> +++ b/drivers/md/dm-raid.c
> @@ -90,9 +90,9 @@ static struct raid_type {
>   	{"raid5_ra", "RAID5 (right asymmetric)",	1, 2, 5, ALGORITHM_RIGHT_ASYMMETRIC},
>   	{"raid5_ls", "RAID5 (left symmetric)",		1, 2, 5, ALGORITHM_LEFT_SYMMETRIC},
>   	{"raid5_rs", "RAID5 (right symmetric)",		1, 2, 5, ALGORITHM_RIGHT_SYMMETRIC},
> -	{"raid6_zr", "RAID6 (zero restart)",		2, 4, 6, ALGORITHM_ROTATING_ZERO_RESTART},
> -	{"raid6_nr", "RAID6 (N restart)",		2, 4, 6, ALGORITHM_ROTATING_N_RESTART},
> -	{"raid6_nc", "RAID6 (N continue)",		2, 4, 6, ALGORITHM_ROTATING_N_CONTINUE}
> +	{"raid6_zr", "RAID6 (zero restart)",		2, 3, 6, ALGORITHM_ROTATING_ZERO_RESTART},
> +	{"raid6_nr", "RAID6 (N restart)",		2, 3, 6, ALGORITHM_ROTATING_N_RESTART},
> +	{"raid6_nc", "RAID6 (N continue)",		2, 3, 6, ALGORITHM_ROTATING_N_CONTINUE}
>   };
>   
>   static char *raid10_md_layout_to_format(int layout)

Re: [PATCH 1/2] dm-raid: Fix minimal device check

[Heinz Mauelshagen]

NACK

md can drive a single disk raid5.



On 10/28/2015 07:50 PM, Andy Grover wrote:
> Check raid_type->minimal_devs instead of parity_devs. minimal_devs is used
> nowhere else (and makes sense to use here) so I think this is what was
> intended.
>
> Signed-off-by: Andy Grover <agrover@redhat.com>
> ---
>   drivers/md/dm-raid.c | 2 +-
>   1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/md/dm-raid.c b/drivers/md/dm-raid.c
> index a090121..5941cfd 100644
> --- a/drivers/md/dm-raid.c
> +++ b/drivers/md/dm-raid.c
> @@ -151,7 +151,7 @@ static struct raid_set *context_alloc(struct dm_target *ti, struct raid_type *ra
>   	unsigned i;
>   	struct raid_set *rs;
>   
> -	if (raid_devs <= raid_type->parity_devs) {
> +	if (raid_devs < raid_type->minimal_devs) {
>   		ti->error = "Insufficient number of devices";
>   		return ERR_PTR(-EINVAL);
>   	}

Re: [PATCH 1/2] dm-raid: Fix minimal device check

[Andy Grover]

On 10/28/2015 12:39 PM, Heinz Mauelshagen wrote:
>
> NACK
>
> md can drive a single disk raid5.

Ok thanks. Please withdraw these two patches.

-- Andy