[commit: ghc] master's head updated: Merge branch 'wip/simd' (680441d)

[git at git.haskell.org]

Repository : ssh://git@git.haskell.org/ghc

Branch 'master' now includes:

     bdcf210 Ensure that globalRegMaybe returns accurate information for XMM registers.
     e02c506 Do not assume that XMM registers are used to pass floating point arguments.
     0f89b9e Enable passing vector arguments in xmm registers on x86-32.
     da5a647 Do not expose LLVM-only primops in GHC.PrimopWrappers.
     16b350a SIMD primops are now generated using schemas that are polymorphic in width and element type.
     638cd12 Flesh out 128-bit wide SIMD primops.
     0c6cf2a Add support for -mavx and -mavx2 flags.
     9d47e58 Add Cmm support for 256-bit-wide values.
     e074c1c Add support for 256-bit-wide vectors.
     afdb2fc Set LLVM option -stack-alignment=32 when compiling AVX instructions.
     f8c5167 Fixup stack spills when generating AVX instructions.
     0b561f1 Pass 256-bit-wide vectors in registers.
     7624815 Add 256-bit-wide SIMD primitives.
     03e33c9 Add support for -mavx512* flags.
     49f4c12 Add Cmm support for 512-bit-wide values.
     26a960c Add support for 512-bit-wide vectors.
     c5add86 Pass 512-bit-wide vectors in registers.
     7dda67b Add 512-bit-wide SIMD primitives.
     d2b9526 By default, only pass 128-bit SIMD vectors in registers on X86-64.
     1ed36c5 Enable -msse to be specified by itself.
     25eeb67 Check that SIMD vector instructions are compatible with current set of dynamic flags.
     680441d Merge branch 'wip/simd'