[dm-crypt] Advanced format (4k sector disks) and dm-crypt

[dm-crypt] Advanced format (4k sector disks) and dm-crypt

[Garðar Arnarsson]

Did a quick search but came up empty handed.

I'm wondering if there is anything special that needs to be passed to
be able to use dm-crypt (cryptsetup) with 4096 byte sector drives.

I know md, lvm, partitions, filesystems all need to have special
triggers for 4K drives (even thouh I think it's done automatically for
all of them nowadays) but I see no mention of dm-crypt/cryptsetup.

Re: [dm-crypt] Advanced format (4k sector disks) and dm-crypt

[Arno Wagner]

On Sat, Jan 18, 2014 at 15:14:29 CET, Garðar Arnarsson wrote:
> Did a quick search but came up empty handed.
> 
> I'm wondering if there is anything special that needs to be passed to
> be able to use dm-crypt (cryptsetup) with 4096 byte sector drives.
> 
> I know md, lvm, partitions, filesystems all need to have special
> triggers for 4K drives (even thouh I think it's done automatically for
> all of them nowadays) but I see no mention of dm-crypt/cryptsetup.

Do not worry. Dm-crypt uses 512B sectors, but there is only a 
negligible performance loss. This matters far, far more on the 
filesystem layer. If that layer does it right, dm-crypt will 
allways process 8 x 512B at a time, as it does process exacly 
what the uppeer layer requests. And alignment in LUKS is done
so that 4k sectors will not be broken up for the data area.

Hence this is not a concern and that is the reason it is
not mentioned. I think I will add an FAQ entry for this.

Arno
-- 
Arno Wagner,     Dr. sc. techn., Dipl. Inform.,    Email: arno@wagner.name
GnuPG: ID: CB5D9718  FP: 12D6 C03B 1B30 33BB 13CF  B774 E35C 5FA1 CB5D 9718
----
There are two ways of constructing a software design: One way is to make it
so simple that there are obviously no deficiencies, and the other way is to
make it so complicated that there are no obvious deficiencies. The first
method is far more difficult.  --Tony Hoare

Re: [dm-crypt] Advanced format (4k sector disks) and dm-crypt

[Arno Wagner]

On Sat, Jan 18, 2014 at 20:24:47 CET, Arno Wagner wrote:
> On Sat, Jan 18, 2014 at 15:14:29 CET, Garðar Arnarsson wrote:
> > Did a quick search but came up empty handed.
> > 
> > I'm wondering if there is anything special that needs to be passed to
> > be able to use dm-crypt (cryptsetup) with 4096 byte sector drives.
> > 
> > I know md, lvm, partitions, filesystems all need to have special
> > triggers for 4K drives (even thouh I think it's done automatically for
> > all of them nowadays) but I see no mention of dm-crypt/cryptsetup.
> 
> Do not worry. Dm-crypt uses 512B sectors, but there is only a 
> negligible performance loss. This matters far, far more on the 
> filesystem layer. If that layer does it right, dm-crypt will 
> allways process 8 x 512B at a time, as it does process exacly 
> what the uppeer layer requests. And alignment in LUKS is done
> so that 4k sectors will not be broken up for the data area.
> 
> Hence this is not a concern and that is the reason it is
> not mentioned. I think I will add an FAQ entry for this.
> 
> Arno

Just added Item 2.17 to the FAQ. 

Arno
-- 
Arno Wagner,     Dr. sc. techn., Dipl. Inform.,    Email: arno@wagner.name
GnuPG: ID: CB5D9718  FP: 12D6 C03B 1B30 33BB 13CF  B774 E35C 5FA1 CB5D 9718
----
There are two ways of constructing a software design: One way is to make it
so simple that there are obviously no deficiencies, and the other way is to
make it so complicated that there are no obvious deficiencies. The first
method is far more difficult.  --Tony Hoare

Re: [dm-crypt] Advanced format (4k sector disks) and dm-crypt

[Garðar Arnarsson]

Thank you very much for the reply and explaination.
I suspected that there was not a concern to dm-crypt in regards to 4K
sectors since I did not find any mantion of it when searching the web
and it's good to have that confirmed.

On Sat, Jan 18, 2014 at 7:36 PM, Arno Wagner <arno@wagner.name> wrote:
> On Sat, Jan 18, 2014 at 20:24:47 CET, Arno Wagner wrote:
>> On Sat, Jan 18, 2014 at 15:14:29 CET, Garðar Arnarsson wrote:
>> > Did a quick search but came up empty handed.
>> >
>> > I'm wondering if there is anything special that needs to be passed to
>> > be able to use dm-crypt (cryptsetup) with 4096 byte sector drives.
>> >
>> > I know md, lvm, partitions, filesystems all need to have special
>> > triggers for 4K drives (even thouh I think it's done automatically for
>> > all of them nowadays) but I see no mention of dm-crypt/cryptsetup.
>>
>> Do not worry. Dm-crypt uses 512B sectors, but there is only a
>> negligible performance loss. This matters far, far more on the
>> filesystem layer. If that layer does it right, dm-crypt will
>> allways process 8 x 512B at a time, as it does process exacly
>> what the uppeer layer requests. And alignment in LUKS is done
>> so that 4k sectors will not be broken up for the data area.
>>
>> Hence this is not a concern and that is the reason it is
>> not mentioned. I think I will add an FAQ entry for this.
>>
>> Arno
>
> Just added Item 2.17 to the FAQ.
>
> Arno
> --
> Arno Wagner,     Dr. sc. techn., Dipl. Inform.,    Email: arno@wagner.name
> GnuPG: ID: CB5D9718  FP: 12D6 C03B 1B30 33BB 13CF  B774 E35C 5FA1 CB5D 9718
> ----
> There are two ways of constructing a software design: One way is to make it
> so simple that there are obviously no deficiencies, and the other way is to
> make it so complicated that there are no obvious deficiencies. The first
> method is far more difficult.  --Tony Hoare
> _______________________________________________
> dm-crypt mailing list
> dm-crypt@saout.de
> http://www.saout.de/mailman/listinfo/dm-crypt

Re: [dm-crypt] Advanced format (4k sector disks) and dm-crypt

[Arno Wagner]

On Sun, Jan 19, 2014 at 19:07:39 CET, Garðar Arnarsson wrote:
> Thank you very much for the reply and explaination.
> I suspected that there was not a concern to dm-crypt in regards to 4K
> sectors since I did not find any mantion of it when searching the web
> and it's good to have that confirmed.

And thank you for asking! If you are not very familiar with the
lower-level mechanisms used, it is a valid queation. And the FAQ
and other documentation can only be improved when people ask 
questions.

Arno


> On Sat, Jan 18, 2014 at 7:36 PM, Arno Wagner <arno@wagner.name> wrote:
> > On Sat, Jan 18, 2014 at 20:24:47 CET, Arno Wagner wrote:
> >> On Sat, Jan 18, 2014 at 15:14:29 CET, Garðar Arnarsson wrote:
> >> > Did a quick search but came up empty handed.
> >> >
> >> > I'm wondering if there is anything special that needs to be passed to
> >> > be able to use dm-crypt (cryptsetup) with 4096 byte sector drives.
> >> >
> >> > I know md, lvm, partitions, filesystems all need to have special
> >> > triggers for 4K drives (even thouh I think it's done automatically for
> >> > all of them nowadays) but I see no mention of dm-crypt/cryptsetup.
> >>
> >> Do not worry. Dm-crypt uses 512B sectors, but there is only a
> >> negligible performance loss. This matters far, far more on the
> >> filesystem layer. If that layer does it right, dm-crypt will
> >> allways process 8 x 512B at a time, as it does process exacly
> >> what the uppeer layer requests. And alignment in LUKS is done
> >> so that 4k sectors will not be broken up for the data area.
> >>
> >> Hence this is not a concern and that is the reason it is
> >> not mentioned. I think I will add an FAQ entry for this.
> >>
> >> Arno
> >
> > Just added Item 2.17 to the FAQ.
> >
> > Arno
> > --
> > Arno Wagner,     Dr. sc. techn., Dipl. Inform.,    Email: arno@wagner.name
> > GnuPG: ID: CB5D9718  FP: 12D6 C03B 1B30 33BB 13CF  B774 E35C 5FA1 CB5D 9718
> > ----
> > There are two ways of constructing a software design: One way is to make it
> > so simple that there are obviously no deficiencies, and the other way is to
> > make it so complicated that there are no obvious deficiencies. The first
> > method is far more difficult.  --Tony Hoare
> > _______________________________________________
> > dm-crypt mailing list
> > dm-crypt@saout.de
> > http://www.saout.de/mailman/listinfo/dm-crypt
> _______________________________________________
> dm-crypt mailing list
> dm-crypt@saout.de
> http://www.saout.de/mailman/listinfo/dm-crypt

-- 
Arno Wagner,     Dr. sc. techn., Dipl. Inform.,    Email: arno@wagner.name
GnuPG: ID: CB5D9718  FP: 12D6 C03B 1B30 33BB 13CF  B774 E35C 5FA1 CB5D 9718
----
There are two ways of constructing a software design: One way is to make it
so simple that there are obviously no deficiencies, and the other way is to
make it so complicated that there are no obvious deficiencies. The first
method is far more difficult.  --Tony Hoare