Home | History | Annotate | only in /src/external/lgpl3/gmp/dist/mpn/x86_64/fastsse
Up to higher level directory
NameDateSize
com-palignr.asm27-Sep-20206.7K
com.asm27-Sep-20204.2K
copyd-palignr.asm22-Aug-20175.6K
copyd.asm27-Sep-20204K
copyi-palignr.asm27-Sep-20206.6K
copyi.asm27-Sep-20204.5K
lshift-movdqu2.asm22-Aug-20174.2K
lshift.asm27-Sep-20204.3K
lshiftc-movdqu2.asm22-Aug-20174.4K
lshiftc.asm27-Sep-20204.6K
README22-Aug-2017603
rshift-movdqu2.asm22-Aug-20174.5K
sec_tabselect.asm27-Sep-20204.7K

README

      1 This directory contains code for x86-64 processors with fast
      2 implementations of SSE operations, hence the name "fastsse".
      3 
      4 Current processors that might benefit from this code are:
      5 
      6   AMD K10
      7   AMD Bulldozer/Piledriver/Steamroller/Excavator
      8   Intel Nocona
      9   Intel Nehalem/Westmere
     10   Intel Sandybridge/Ivybridge
     11   Intel Haswell/Broadwell
     12   VIA Nano
     13 
     14 Current processors that do not benefit from this code are:
     15 
     16   AMD K8
     17   AMD Bobcat
     18   Intel Atom
     19 
     20 Intel Conroe/Penryn is a border case; its handling of non-aligned
     21 128-bit memory operands is poor.  VIA Nano also have poor handling of
     22 non-aligned operands.
     23