Skip to content
  1. Sep 11, 2017
  2. Sep 09, 2017
    • Andy Polyakov's avatar
      sha/asm/keccak1600-armv8.pl: fix return value buglet and ... · 236dd463
      Andy Polyakov authored
      
      
      ... script data load.
      
      On related note an attempt was made to merge rotations with logical
      operations. I mean as we know, ARM ISA has merged rotate-n-logical
      instructions which can be used here. And they were used to improve
      keccak1600-armv4 performance. But not here. Even though this approach
      resulted in improvement on Cortex-A53 proportional to reduction of
      amount of instructions, ~8%, it didn't exactly worked out on
      non-Cortex cores. Presumably because they break merged instructions
      to separate μ-ops, which results in higher *operations* count. X-Gene
      and Denver went ~20% slower and Apple A7 - 40%. The optimization was
      therefore dismissed.
      
      Reviewed-by: default avatarRich Salz <rsalz@openssl.org>
      236dd463
  3. Sep 08, 2017
  4. Sep 07, 2017
  5. Sep 06, 2017
  6. Sep 05, 2017
  7. Sep 04, 2017
  8. Sep 03, 2017
  9. Sep 02, 2017
  10. Sep 01, 2017
  11. Aug 31, 2017