[PATCH dm-zoned-tools 1/2] README: fix small typo host-manmaged --> host-managed

[PATCH dm-zoned-tools 1/2] README: fix small typo host-manmaged --> host-managed

[Luis R. Rodriguez]

Small typo fix.

Signed-off-by: Luis R. Rodriguez <mcgrof@kernel.org>
---
 README | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/README b/README
index 61830dd..7762694 100644
--- a/README
+++ b/README
@@ -23,7 +23,7 @@ http://www.t13.org/Documents/UploadedDocuments/docs2015/
 di537r05-Zoned_Device_ATA_Command_Set_ZAC.pdf
 
 dm-zoned implementation focused on simplicity and on minimizing overhead (CPU,
-memory and storage overhead). For a 10TB host-manmaged disk with 256 MB zones,
+memory and storage overhead). For a 10TB host-managed disk with 256 MB zones,
 dm-zoned memory usage per disk instance is at most 4.5 MB and as little as 5
 zones will be used internally for storing metadata and performaing reclaim
 operations.
-- 
2.17.0

[PATCH dm-zoned-tools 2/2] README: fix example script

[Luis R. Rodriguez]

The target type should be "zoned", not "dm-zoned".

Signed-off-by: Luis R. Rodriguez <mcgrof@kernel.org>
---
 README | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

The README does note explain if one should use dmzadm to format a drive
if using dm zone or only if one is to use a filesystem which supports
zones, such as f2fs. Or if we should *always* use dmzadm format regardless
of the circumstance.

Could this be clarfied? What are the downsides to not using dmzadm
format on a drive and say a user just going straight to use the dm zone
and then mkfs on top of it?

diff --git a/README b/README
index 7762694..9aea28e 100644
--- a/README
+++ b/README
@@ -184,7 +184,7 @@ options="$@"
 
 modprobe dm-zoned
 
-echo "0 `blockdev --getsize ${dev}` dm-zoned ${dev} ${options}" | \
+echo "0 `blockdev --getsize ${dev}` zoned ${dev} ${options}" | \
 dmsetup create zoned-`basename ${dev}`
 ]]
 
-- 
2.17.0

Re: [PATCH dm-zoned-tools 1/2] README: fix small typo host-manmaged --> host-managed

[Damien Le Moal]

On 2018/04/30 12:45, Luis R. Rodriguez wrote:
> Small typo fix.
> 
> Signed-off-by: Luis R. Rodriguez <mcgrof@kernel.org>
> ---
>  README | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/README b/README
> index 61830dd..7762694 100644
> --- a/README
> +++ b/README
> @@ -23,7 +23,7 @@ http://www.t13.org/Documents/UploadedDocuments/docs2015/
>  di537r05-Zoned_Device_ATA_Command_Set_ZAC.pdf
>  
>  dm-zoned implementation focused on simplicity and on minimizing overhead (CPU,
> -memory and storage overhead). For a 10TB host-manmaged disk with 256 MB zones,
> +memory and storage overhead). For a 10TB host-managed disk with 256 MB zones,
>  dm-zoned memory usage per disk instance is at most 4.5 MB and as little as 5
>  zones will be used internally for storing metadata and performaing reclaim
>  operations.

Thanks Luis !

Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>


-- 
Damien Le Moal
Western Digital Research

Re: [PATCH dm-zoned-tools 2/2] README: fix example script

[Damien Le Moal]

On 2018/04/30 12:45, Luis R. Rodriguez wrote:
> The target type should be "zoned", not "dm-zoned".
> 
> Signed-off-by: Luis R. Rodriguez <mcgrof@kernel.org>
> ---
>  README | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> The README does note explain if one should use dmzadm to format a drive
> if using dm zone or only if one is to use a filesystem which supports
> zones, such as f2fs. Or if we should *always* use dmzadm format regardless
> of the circumstance.

dmzadm is only for dm-zoned device mapper. If the file system natively support
zoned block devices, it is the FS problem to format the drive accordingly to
ZBC/ZAC constraints. And for these files systems, dm-zoned is useless.

> Could this be clarfied? What are the downsides to not using dmzadm
> format on a drive and say a user just going straight to use the dm zone
> and then mkfs on top of it?

OK. I will look into it and clarify.

> 
> diff --git a/README b/README
> index 7762694..9aea28e 100644
> --- a/README
> +++ b/README
> @@ -184,7 +184,7 @@ options="$@"
>  
>  modprobe dm-zoned
>  
> -echo "0 `blockdev --getsize ${dev}` dm-zoned ${dev} ${options}" | \
> +echo "0 `blockdev --getsize ${dev}` zoned ${dev} ${options}" | \
>  dmsetup create zoned-`basename ${dev}`
>  ]]

Yes, indeed...

Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>

-- 
Damien Le Moal
Western Digital Research

Re: [PATCH dm-zoned-tools 2/2] README: fix example script

[Luis R. Rodriguez]

On Mon, Apr 30, 2018 at 08:36:09PM +0000, Damien Le Moal wrote:
> On 2018/04/30 12:45, Luis R. Rodriguez wrote:
> > The target type should be "zoned", not "dm-zoned".
> > 
> > Signed-off-by: Luis R. Rodriguez <mcgrof@kernel.org>
> > ---
> >  README | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> > 
> > The README does note explain if one should use dmzadm to format a drive
> > if using dm zone or only if one is to use a filesystem which supports
> > zones, such as f2fs. Or if we should *always* use dmzadm format regardless
> > of the circumstance.
> 
> dmzadm is only for dm-zoned device mapper. If the file system natively support
> zoned block devices, it is the FS problem to format the drive accordingly to
> ZBC/ZAC constraints.

The README could be update to reflect this then, just to be sure. Only now
that I realize it has the dm prefix did I realize this was an obvious
requirement :P

> And for these files systems, dm-zoned is useless.

Sure, make sense.

  Luis