Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

chacha20: Process 4 blocks at a time in AVX2 backend #267

Merged
merged 7 commits into from
Aug 29, 2021
Merged

chacha20: Process 4 blocks at a time in AVX2 backend #267

merged 7 commits into from
Aug 29, 2021

Commits on Aug 28, 2021

  1. chacha20: Refactor usage of blocks variant of avx2::StateWords 

    It is now agnostic of the number of blocks processed, which is now a
constant.
    @str4d
    str4d committed Aug 28, 2021
    Configuration menu
    Copy the full SHA
    ebbb642 View commit details
    Browse the repository at this point in the history

  2. chacha20: Pass around &mut avx2::StateWord instead of &mut __m256i

    @str4d
    str4d committed Aug 28, 2021
    Configuration menu
    Copy the full SHA
    687f953 View commit details
    Browse the repository at this point in the history

Commits on Aug 29, 2021

  1. chacha20: Add avx2::StateWord methods for required ops; MSRV 1.51+

    @str4d
    str4d committed Aug 29, 2021
    Configuration menu
    Copy the full SHA
    d0c4f97 View commit details
    Browse the repository at this point in the history

  2. chacha20: Change buffer_pos to a u16 

    For a 4-block buffer, we need to be able to represent the past-the-end
buffer position of 256, which is too large for a `u8`.
    @str4d
    str4d committed Aug 29, 2021
    Configuration menu
    Copy the full SHA
    f089676 View commit details
    Browse the repository at this point in the history

  3. chacha20: Switch to 4-block buffer for SSE2 / AVX2 backend

    @str4d
    str4d committed Aug 29, 2021
    Configuration menu
    Copy the full SHA
    85d81c0 View commit details
    Browse the repository at this point in the history

  4. chacha20: Add a BlockRngResults wrapper type 

    When the non-soft backend is being used, its 4-block buffer size results
in a `BlockRngCore::Results` type of `[u32; 64]` which doesn't implement
`Default`. We replace it with a wrapper type on which we implement the
necessary traits.
    @str4d
    str4d committed Aug 29, 2021
    Configuration menu
    Copy the full SHA
    05e6b48 View commit details
    Browse the repository at this point in the history

  5. ctr: Allow upper-case acronyms (clippy lint)

    @str4d
    str4d committed Aug 29, 2021
    Configuration menu
    Copy the full SHA
    e8419a4 View commit details
    Browse the repository at this point in the history