Code

Simd enablement#2945

Code

Closed

shawnl wants to merge 18 commits intoziglang:masterziglang/zig:masterfrom shawnl:simd5shawnl/zig:simd5Copy head branch name to clipboard

+3,147-275

Commits

Commits on Nov 21, 2019

• @floatCast for vectors, and allow safe casts from non-comptime integers.

  Show description for 6f80beb
  shawnl

  committed
  6f80beb

  Copy full SHA for 6f80beb

• Vector type field access (len, and all those of the child element)

  Show description for f005607
  shawnl

  committed
  f005607

  Copy full SHA for f005607

• @boolToInt for vectors, fix codegen of @boolToInt

  Show description for 9e27863
  shawnl

  committed
  9e27863

  Copy full SHA for 9e27863

• std: branching on vectors, all(), any(), and none(), and select()

  shawnl

  committed
  44b140f

  Copy full SHA for 44b140f

• @bitcast

  Show description for 203408d
  shawnl

  committed
  203408d

  Copy full SHA for 203408d

• fixup comptime @byteswap to have the types

  shawnl

  committed
  79e306d

  Copy full SHA for 79e306d

• @gather and @scatter

  Show description for 46d10d3
  shawnl

  committed
  46d10d3

  Copy full SHA for 46d10d3

• allow ^, &, |, and ~ on bools and vectors of bools. (also ^= &= and |=)

  Show description for f91b0fe
  shawnl

  committed
  f91b0fe

  Copy full SHA for f91b0fe

• Widening and shortening of vectors (in LLVM, sext, zext, fext, and trunc), with safety checks.

  Show description for 9666ba2
  shawnl

  committed
  9666ba2

  Copy full SHA for 9666ba2

• stage1/codegen: vector_all and vector_any helpers

  shawnl

  committed
  be5a24e

  Copy full SHA for be5a24e

• get_c_type for vectors

  shawnl

  committed
  174a6d4

  Copy full SHA for 174a6d4

• docs: @inttofloat @floatCast

  shawnl

  committed
  25c5ad1

  Copy full SHA for 25c5ad1

• emit error for reachable compile error (vector length mix-match)

  shawnl

  committed
  df35065

  Copy full SHA for df35065

• work around bizarre bug (attempt 6)

  shawnl

  committed
  f1360b3

  Copy full SHA for f1360b3

• stage1: accept vectors for all the float operations we have

  Show description for 727c420
  shawnl

  committed
  727c420

  Copy full SHA for 727c420

• rem_zero fixup

  shawnl

  committed
  60ae741

  Copy full SHA for 60ae741

• update glibc to include libmvec

  Show description for bf57996
  shawnl

  committed
  bf57996

  Copy full SHA for bf57996

• initial C ABI testing for vectors

  Show description for 10c4c4a
  shawnl

  committed
  10c4c4a

  Copy full SHA for 10c4c4a