CRC64 library rework and x86 CRC optimization
Get series with b4
b4 am [email protected]Versions newest first · linked by author and title
- v2 Latest posting 2/211 patches 9 repliesIn review
- v2 posting 1/212 patches 7 repliesSuperseded
- v111 patches 6 repliesSuperseded
Patches in v2 posting 2/2 No applies recorded yet
| # | Subject and files | Size | Replies | Applied |
|---|---|---|---|---|
| 1/11 | lib/crc64-rocksoft: stop wrapping the crypto APIblock/ include/ lib/ +3 more | +12 | 1 reply | |
| 2/11 | crypto: crc64-rocksoft - remove from crypto APIcrypto/ crypto/ crypto/ +3 more | +0 | 1 reply | |
| 3/11 | lib/crc64: rename CRC64-Rocksoft to CRC64-NVMEblock/ include/ lib/ +1 more | +18 | 2 replies | |
| 4/11 | lib/crc_kunit.c: add test and benchmark for CRC64-NVMElib/ | +29 | 1 reply | |
| 5/11 | lib/crc64: add support for arch-optimized implementationsinclude/ lib/ lib/ | +37 | 1 reply | |
| 6/11 | x86: move ZMM exclusion list into CPU feature flagarch/ arch/ arch/ | +24 | 0 replies | |
| 7/11 | scripts/gen-crc-consts: add gen-crc-consts.pyMAINTAINERS scripts/ | +215 | 0 replies | |
| 8/11 | x86/crc: add "template" for [V]PCLMULQDQ based CRC functionsarch/ arch/ | +659 | 0 replies | |
| 9/11 | x86/crc32: implement crc32_le using new templatearch/ arch/ arch/ | +65 | 0 replies | |
| 10/11 | x86/crc-t10dif: implement crc_t10dif using new templatearch/ arch/ arch/ +3 more | +64 | 0 replies | |
| 11/11 | x86/crc64: implement crc64_be and crc64_nvme using new templatearch/ arch/ arch/ +2 more | +158 | 0 replies |
Reviews trailers per patch
10 of 11 patches have a Reviewed-by or Acked-by of their own; the series has one for all of them.
| Patch | Ard Biesheuvel | Ingo Molnar | Keith Busch | Martin K. Petersen |
|---|---|---|---|---|
| Series | A-b | R-b | ||
| 1/11 | R-b | |||
| 2/11 | R-b | |||
| 3/11 | R-b | |||
| 4/11 | R-b | |||
| 5/11 | R-b | |||
| 6/11 | A-b | A-b | ||
| 7/11 | A-b | |||
| 8/11 | A-b | |||
| 9/11 | A-b | |||
| 10/11 | A-b | |||
| 11/11 |
Trailers given in replies are filled; those carried in the patches from earlier versions are outlined with dashes. Scroll the table sideways for all 4 reviewers.
Cover letter [PATCH v2 00/11] CRC64 library rework and x86 CRC optimization
This patchset applies to commit 72deda0abee6e7 and is also available at:
git fetch https://git.kernel.org/pub/scm/linux/kernel/git/ebiggers/linux.git crc-x86-v2
This is the next major set of CRC library improvements, targeting 6.15.
Patches 1-5 rework the CRC64 library along the lines of what I did for CRC32 and CRC-T10DIF in 6.14. They add direct support for architecture-specific optimizations, fix the naming of the NVME CRC64 variant, and eliminate a pointless use of the crypto API.
Patches 6-10 replace the existing x86 PCLMULQDQ optimized CRC code with new code that is shared among the different CRC variants and also adds VPCLMULQDQ support, greatly improving performance on recent CPUs. Patch 11 wires up the same optimization to crc64_be() and crc64_nvme() (a.k.a. the old "crc64_rocksoft") which previously were unoptimized, improving the performance of those CRC functions by as much as 100x. crc64_be is used by bcachefs, and crc64_nvme is used by blk-integrity.
Eric Biggers (11):
lib/crc64-rocksoft: stop wrapping the crypto API
crypto: crc64-rocksoft - remove from crypto API lib/crc64: rename CRC64-Rocksoft to CRC64-NVME lib/crc_kunit.c: add test and benchmark for CRC64-NVME lib/crc64: add support for arch-optimized implementations
x86: move ZMM exclusion list into CPU feature flag scripts/gen-crc-consts: add gen-crc-consts.py x86/crc: add "template" for [V]PCLMULQDQ based CRC functions x86/crc32: implement crc32_le using new template x86/crc-t10dif: implement crc_t10dif using new template x86/crc64: implement crc64_be and crc64_nvme using new template
MAINTAINERS | 1 + arch/x86/Kconfig | 3 +- arch/x86/crypto/aesni-intel_glue.c | 22 +- arch/x86/include/asm/cpufeatures.h | 1 + arch/x86/kernel/cpu/intel.c | 22 ++ arch/x86/lib/Makefile | 5 +- arch/x86/lib/crc-pclmul-consts.h | 195 ++++++++++ arch/x86/lib/crc-pclmul-template.S | 578 ++++++++++++++++++++++++++++ arch/x86/lib/crc-pclmul-template.h | 81 ++++ arch/x86/lib/crc-t10dif-glue.c | 23 +- arch/x86/lib/crc16-msb-pclmul.S | 6 + arch/x86/lib/crc32-glue.c | 37 +- arch/x86/lib/crc32-pclmul.S | 219 +---------- arch/x86/lib/crc64-glue.c | 50 +++ arch/x86/lib/crc64-pclmul.S | 7 + arch/x86/lib/crct10dif-pcl-asm_64.S | 332 ---------------- block/Kconfig | 2 +- block/t10-pi.c | 2 +- crypto/Kconfig | 11 - crypto/Makefile | 1 - crypto/crc64_rocksoft_generic.c | 89 ----- crypto/testmgr.c | 7 - crypto/testmgr.h | 12 - include/linux/crc64.h | 38 +- lib/Kconfig | 16 +- lib/Makefile | 1 - lib/crc64-rocksoft.c | 126 ------ lib/crc64.c | 49 +-- lib/crc_kunit.c | 30 +- lib/gen_crc64table.c | 10 +- scripts/gen-crc-consts.py | 214 ++++++++++ 31 files changed, 1270 insertions(+), 920 deletions(-) create mode 100644 arch/x86/lib/crc-pclmul-consts.h create mode 100644 arch/x86/lib/crc-pclmul-template.S create mode 100644 arch/x86/lib/crc-pclmul-template.h create mode 100644 arch/x86/lib/crc16-msb-pclmul.S create mode 100644 arch/x86/lib/crc64-glue.c create mode 100644 arch/x86/lib/crc64-pclmul.S delete mode 100644 arch/x86/lib/crct10dif-pcl-asm_64.S delete mode 100644 crypto/crc64_rocksoft_generic.c delete mode 100644 lib/crc64-rocksoft.c create mode 100755 scripts/gen-crc-consts.py
base-commit: 72deda0abee6e705ae71a93f69f55e33be5bca5c
Signature · 1 line
2.48.1