[Numpy-svn] [numpy/numpy] 46a3fc: ENH: unroll vector minmax loop

[GitHub]

  Branch: refs/heads/master
  Home:   https://github.com/numpy/numpy
  Commit: 46a3fc10ef27be3637d72b5a6befb440a012dc87
      https://github.com/numpy/numpy/commit/46a3fc10ef27be3637d72b5a6befb440a012dc87
  Author: Julian Taylor <jtaylor.debian at googlemail.com>
  Date:   2013-10-19 (Sat, 19 Oct 2013)

  Changed paths:
    M numpy/core/src/umath/simd.inc.src

  Log Message:
  -----------
  ENH: unroll vector minmax loop

Improves speed by 10% on intel cpus.
Simplify code by moving the fenv support check into the dispatcher. fenv works
on all common platforms (including windows), the fallback is not worth it for
the exotic platforms where it might not be available.


  Commit: b0dd6901d54e687571e44c4fa1d9c7db0d00e299
      https://github.com/numpy/numpy/commit/b0dd6901d54e687571e44c4fa1d9c7db0d00e299
  Author: Julian Taylor <jtaylor.debian at googlemail.com>
  Date:   2013-10-19 (Sat, 19 Oct 2013)

  Changed paths:
    M numpy/core/tests/test_umath.py
    M numpy/testing/utils.py

  Log Message:
  -----------
  ENH: accept callable as message in assert_

Allows deferring evaluation until failure.
Used for blocked minmax test which evaluate array representations for
the error message many thousand times accumulating to a full second
useless runtime.


  Commit: 43df2ee1a0e8dd27f91063a4f5e78afe06f1970b
      https://github.com/numpy/numpy/commit/43df2ee1a0e8dd27f91063a4f5e78afe06f1970b
  Author: Charles Harris <charlesr.harris at gmail.com>
  Date:   2013-10-19 (Sat, 19 Oct 2013)

  Changed paths:
    M numpy/core/src/umath/simd.inc.src
    M numpy/core/tests/test_umath.py
    M numpy/testing/utils.py

  Log Message:
  -----------
  Merge pull request #3945 from juliantaylor/minmax-improve

ENH: unroll vector minmax loop


Compare: https://github.com/numpy/numpy/compare/4533821ca2d7...43df2ee1a0e8