mbox series

[API-NEXT,0/4] Add sha-1 and sha-512

Message ID 1487776105-32713-1-git-send-email-petri.savolainen@linaro.org
Headers show
Series Add sha-1 and sha-512 | expand

Message

Petri Savolainen Feb. 22, 2017, 3:08 p.m. UTC 
Add new algorithm enumerations so that vendor IPSEC implementations and IPSEC 
test applications can proceed. Odp-linux crypto implementation and validation 
tests follow later.

Petri Savolainen (4):
  abi: event: add ODP_EVENT_IPSEC_RESULT
  api: crypto: add sha-1 and sha-512 enumerations
  linux-gen: crypto: sha-1 and sha-512 not implemented yet
  validation: crypto: add stubs for sha-1 and sha-512 tests

 include/odp/api/spec/crypto.h                      | 18 ++++++++++++
 include/odp/arch/default/api/abi/event.h           |  3 +-
 platform/linux-generic/odp_crypto.c                |  2 ++
 test/common_plat/validation/api/crypto/crypto.h    |  2 ++
 .../validation/api/crypto/odp_crypto_test_inp.c    | 32 ++++++++++++++++++++++
 5 files changed, 56 insertions(+), 1 deletion(-)

-- 
2.8.1

Comments

Savolainen, Petri (Nokia - FI/Espoo) March 1, 2017, 11:42 a.m. UTC | #1 
> -----Original Message-----

> From: Dmitry Eremin-Solenikov [mailto:dmitry.ereminsolenikov@linaro.org]

> Sent: Tuesday, February 28, 2017 8:21 PM

> To: Maxim Uvarov <maxim.uvarov@linaro.org>

> Cc: Petri Savolainen <petri.savolainen@linaro.org>

> Subject: Re: [lng-odp] [API-NEXT PATCH 0/4] Add sha-1 and sha-512

> 

> On 28 February 2017 at 20:04, Maxim Uvarov <maxim.uvarov@linaro.org>

> wrote:

> > On 02/28/17 19:35, Dmitry Eremin-Solenikov wrote:

> >> Hello,

> >>

> >> On 23.02.2017 16:06, Dmitry Eremin-Solenikov wrote:

> >>> On 22.02.2017 18:08, Petri Savolainen wrote:

> >>>> Add new algorithm enumerations so that vendor IPSEC implementations

> and IPSEC

> >>>> test applications can proceed. Odp-linux crypto implementation and

> validation

> >>>> tests follow later.

> >>>>

> >>>> Petri Savolainen (4):

> >>>>   abi: event: add ODP_EVENT_IPSEC_RESULT

> >>>>   api: crypto: add sha-1 and sha-512 enumerations

> >>>>   linux-gen: crypto: sha-1 and sha-512 not implemented yet

> >>>>   validation: crypto: add stubs for sha-1 and sha-512 tests

> >>>

> >>> Reviewed-by: Dmitry Eremin-Solenikov

> <dmitry.ereminsolenikov@linaro.org>

> >>

> >> Just wondering, is there any issue with this patchset, that stops it

> >> from being merged?

> >>

> >

> > Dmitri, you said

> > """

> > If nobody steps up, I will take a look onto implementing funcionality

> > and tests. I have several issues with the current code anyway.

> > """

> >

> > so I'm waiting for feedback.

> 

> Ah. But is there any reason for holding back the API without

> implementation?

> Sorry for maybe dump question, I'm still diving in development process.

> 

> --

> With best wishes

> Dmitry


There's no reason to wait for implementation. API spec can go into api-next branch before the implementation and tests.

Nikhil asked/suggested if SHA-2 should be used instead of SHA-256, SHA-512, etc. I still think that it's better to use the same names for the algorithms, as the standards and industry (SDKs, HW manuals) use. E.g. FIPS standard for SHA (http://nvlpubs.nist.gov/nistpubs/FIPS/NIST.FIPS.180-4.pdf and http://nvlpubs.nist.gov/nistpubs/FIPS/NIST.FIPS.202.pdf ) and RFCs use these names for the algorithms: SHA-1, SHA-256, SHA-512, SHA3-256, SHA3-512, SHAKE128. SHA-2 and SHA-3 are family names, but not individual algorithm names. SHA-1 is single algorithm (not a family).

So, Maxim please merge this as is. Dmitry then adds implementation changes on top.

-Petri
Maxim Uvarov March 2, 2017, 2:07 p.m. UTC | #2 
Merged,
Will be in the repo soon.

Maxim.

On 1 March 2017 at 14:42, Savolainen, Petri (Nokia - FI/Espoo) <
petri.savolainen@nokia-bell-labs.com> wrote:

>

>

> > -----Original Message-----

> > From: Dmitry Eremin-Solenikov [mailto:dmitry.ereminsolenikov@linaro.org]

> > Sent: Tuesday, February 28, 2017 8:21 PM

> > To: Maxim Uvarov <maxim.uvarov@linaro.org>

> > Cc: Petri Savolainen <petri.savolainen@linaro.org>

> > Subject: Re: [lng-odp] [API-NEXT PATCH 0/4] Add sha-1 and sha-512

> >

> > On 28 February 2017 at 20:04, Maxim Uvarov <maxim.uvarov@linaro.org>

> > wrote:

> > > On 02/28/17 19:35, Dmitry Eremin-Solenikov wrote:

> > >> Hello,

> > >>

> > >> On 23.02.2017 16:06, Dmitry Eremin-Solenikov wrote:

> > >>> On 22.02.2017 18:08, Petri Savolainen wrote:

> > >>>> Add new algorithm enumerations so that vendor IPSEC implementations

> > and IPSEC

> > >>>> test applications can proceed. Odp-linux crypto implementation and

> > validation

> > >>>> tests follow later.

> > >>>>

> > >>>> Petri Savolainen (4):

> > >>>>   abi: event: add ODP_EVENT_IPSEC_RESULT

> > >>>>   api: crypto: add sha-1 and sha-512 enumerations

> > >>>>   linux-gen: crypto: sha-1 and sha-512 not implemented yet

> > >>>>   validation: crypto: add stubs for sha-1 and sha-512 tests

> > >>>

> > >>> Reviewed-by: Dmitry Eremin-Solenikov

> > <dmitry.ereminsolenikov@linaro.org>

> > >>

> > >> Just wondering, is there any issue with this patchset, that stops it

> > >> from being merged?

> > >>

> > >

> > > Dmitri, you said

> > > """

> > > If nobody steps up, I will take a look onto implementing funcionality

> > > and tests. I have several issues with the current code anyway.

> > > """

> > >

> > > so I'm waiting for feedback.

> >

> > Ah. But is there any reason for holding back the API without

> > implementation?

> > Sorry for maybe dump question, I'm still diving in development process.

> >

> > --

> > With best wishes

> > Dmitry

>

> There's no reason to wait for implementation. API spec can go into

> api-next branch before the implementation and tests.

>

> Nikhil asked/suggested if SHA-2 should be used instead of SHA-256,

> SHA-512, etc. I still think that it's better to use the same names for the

> algorithms, as the standards and industry (SDKs, HW manuals) use. E.g. FIPS

> standard for SHA (http://nvlpubs.nist.gov/nistpubs/FIPS/NIST.FIPS.180-4.

> pdf and http://nvlpubs.nist.gov/nistpubs/FIPS/NIST.FIPS.202.pdf ) and

> RFCs use these names for the algorithms: SHA-1, SHA-256, SHA-512, SHA3-256,

> SHA3-512, SHAKE128. SHA-2 and SHA-3 are family names, but not individual

> algorithm names. SHA-1 is single algorithm (not a family).

>

> So, Maxim please merge this as is. Dmitry then adds implementation changes

> on top.

>

> -Petri

>

>