Hi all,
I'd like to better understand the future plans for shash.
Currently I can see that (please correct me if I'm wrong):
- all the hash algorithms are migrated to shash
- most of the kernel code uses the ahash API
- still some pieces are using the hash API (e.g. IMA)
I want to build a module on top of hash tfm API.
What should I use? I guess ahash.
In fact I don't need async, while I'd like export/import functions.
So, can I already use shash?
What about the (near) future? Is it better (for sync) to use hash, or to
wait for shash [*], or to use ahash in any case?
BTW, should IMA be converted to ahash?
This could be a good exercise to start...
Best,
[*] I made some preliminary tests and shash gave me some problems.
Further, looking at the code, it seems it is not yet completely
implemented w.r.t. the tfm API.
--
Emanuele Cesena <[email protected]>
Il corpo non ha ideali
Emanuele Cesena <[email protected]> wrote:
>
> In fact I don't need async, while I'd like export/import functions.
If you are processing a large (i.e., > 1000 bytes) amount of data,
then you should use the async interface. Otherwise the sync
interface should be good enough.
Cheers,
--
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt