This series improves crc32c() performance on lengths >= 512 bytes on
newer x86_64 CPUs by enabling the VPCLMULQDQ (vector carryless
multiplication) optimized CRC code.

This series targets crc-next.

Eric Biggers (2):
  lib/crc: x86: Reorganize crc-pclmul static_call initialization
  lib/crc: x86/crc32c: Enable VPCLMULQDQ optimization where beneficial

 lib/crc/x86/crc-pclmul-consts.h   | 47 +++++++++++++++++++++++++++++-
 lib/crc/x86/crc-pclmul-template.h | 31 +++++++++-----------
 lib/crc/x86/crc-t10dif.h          |  9 +++++-
 lib/crc/x86/crc32.h               | 48 +++++++++++++++++++++++++++++--
 lib/crc/x86/crc64.h               | 15 ++++++++--
 5 files changed, 127 insertions(+), 23 deletions(-)


base-commit: 9b0236f4efb889869f7d4f3f084f508cc0433ec9
-- 
2.50.1

On Sun, 20 Jul 2025 at 08:50, Eric Biggers <ebiggers@kernel.org> wrote:
>
> This series improves crc32c() performance on lengths >= 512 bytes on
> newer x86_64 CPUs by enabling the VPCLMULQDQ (vector carryless
> multiplication) optimized CRC code.
>
> This series targets crc-next.
>
> Eric Biggers (2):
>   lib/crc: x86: Reorganize crc-pclmul static_call initialization
>   lib/crc: x86/crc32c: Enable VPCLMULQDQ optimization where beneficial
>

Acked-by: Ard Biesheuvel <ardb@kernel.org>