Abseil Team 99275763ac Use even faster reduction algorithm in FinalizePclmulStream() ...

My previous CL optimized the Barrett reduction.  But since this is CRC32C and
scalar instructions for it are available, there is actually no need for Barrett
reduction at all.  Just use two 64-bit CRC32C instructions to reduce fullCRC.

This improves CRC32C performance on 2048-byte messages on Skylake by another 2%
or so.

PiperOrigin-RevId: 739977426
Change-Id: I4611af88cd32ed7a995e772a13c30e3bdcec8de9

2025-03-24 09:59:22 -07:00

..

internal

Use even faster reduction algorithm in FinalizePclmulStream()

2025-03-24 09:59:22 -07:00

BUILD.bazel

Crc: Detect support for pmull and crc instructions on Apple AArch64

2025-01-13 16:45:10 -08:00

CMakeLists.txt

Crc: Detect support for pmull and crc instructions on Apple AArch64

2025-01-13 16:45:10 -08:00

crc32c_benchmark.cc

Use more efficient reduction algorithm in FinalizePclmulStream()

2025-03-20 18:06:56 -07:00

crc32c_test.cc

Implement AbslStringify for crc32c_t in order to support absl::StrFormat natively

2023-08-01 15:24:56 -07:00

crc32c.cc

Move the implementation of absl::ComputeCrc32c to the header file, to

2025-02-19 09:16:21 -08:00

crc32c.h

Move the implementation of absl::ComputeCrc32c to the header file, to

2025-02-19 09:16:21 -08:00