Maamoun TK writes:
> Thank you for merging the patch. There is a very tiny change that keeps
> itching me since I submitted the patch, the following PR
> https://git.lysator.liu.se/nettle/nettle/-/merge_requests/59 would scratch
> that itch.
Merged.
/Niels
--
Niels Möller. PGP key
On Mon, Apr 3, 2023 at 7:37 AM Niels Möller wrote:
> Maamoun TK writes:
>
> > I updated the GHASH implementation for x86_64 architecture to use 2-way
> > block processing, it seems to me from Agner Fog's instruction tables that
> > this change makes sense starting from Broadwell arch for intel
Maamoun TK writes:
> I updated the GHASH implementation for x86_64 architecture to use 2-way
> block processing, it seems to me from Agner Fog's instruction tables that
> this change makes sense starting from Broadwell arch for intel processors
> while it has zero improvement on zen3/zen4
re options and fat setup still missing.
>
> Regards,
> /Niels
>
> C x86_64/gcm-hash.asm
>
> ifelse(`
>Copyright (C) 2022 Niels Möller
>
>This file is part of GNU Nettle.
>
>GNU Nettle is free software: you can redistribute it and/or
>modify i
Maamoun TK writes:
>> I've done the needed changes for the C, the x86_64, arm64 and powerpc64
>> implementations. s390x code also needs update, I hope to get to that in
>> a few days (unless someone else wants to do that).
>
> I handled the s390x part and pushed a MR for changes.
I've now
On Tue, Feb 22, 2022 at 9:55 PM Niels Möller wrote:
> ni...@lysator.liu.se (Niels Möller) writes:
>
> > I'm considering reorganizing the internal gcm functions. I think I'd
> > like to have
> >
> > void
> > _nettle_ghash_set_key (struct gcm_key *gcm, const union nettle_block16
> *key);
> >
>
ni...@lysator.liu.se (Niels Möller) writes:
> I'm considering reorganizing the internal gcm functions. I think I'd
> like to have
>
> void
> _nettle_ghash_set_key (struct gcm_key *gcm, const union nettle_block16
> *key);
>
> which sets the key (typically, the key block is zero encrypte using
ni...@lysator.liu.se (Niels Möller) writes:
> I've written a first version of a gcm_hash for x86_64, using the
> pclmulqdq (carryless mul) instructions. With only a single block at a time,
> no interleaving, this gives to 4.3 GByte/s,
I've added proper config and fat setup and merged this. It
per cycle, by interleaving, we could perhaps increase
performance by another factor of two.
See below. Configure options and fat setup still missing.
Regards,
/Niels
C x86_64/gcm-hash.asm
ifelse(`
Copyright (C) 2022 Niels Möller
This file is part of GNU Nettle.
GNU Nettle is free