Field G. Van Zee 7574c9947d Added basic flop-counting mechanism (level-3 only). ...

Details:
- Added optional flop counting to all level-3 front-ends, which is
  enabled via BLIS_ENABLE_FLOP_COUNT. The flop count can be
  reset at any time via bli_flop_count_reset() and queried via
  bli_flop_count(). Caveats:
  - flop counts are approximate for her[2]k, syr[2]k, trmm, and
    trsm operations;
  - flop counts ignore extra flops due to non-unit alpha;
  - flop counts do not account for situations where beta is zero.

2015-02-04 12:11:55 -06:00

..

level0

Added high-level implementations of 4m, 3m.

2014-09-16 18:19:32 -05:00

bli_auxinfo_macro_defs.h

Re-implemented micro-panel alignment.

2014-10-20 19:23:06 -05:00

bli_blas_macro_defs.h

Updated copyright headers (continued).

2014-07-14 16:14:33 -05:00

bli_complex_macro_defs.h

Updated copyright headers (continued).

2014-07-14 16:14:33 -05:00

bli_error_macro_defs.h

Updated copyright headers (continued).

2014-07-14 16:14:33 -05:00

bli_extern_defs.h

Updated copyright headers (continued).

2014-07-14 16:14:33 -05:00

bli_f2c.h

Added support for C99 complex types/arithmetic.

2013-07-18 17:53:31 -05:00

bli_genarray_macro_defs.h

Updated copyright headers (continued).

2014-07-14 16:14:33 -05:00

bli_gentfunc_macro_defs.h

Updated copyright headers (continued).

2014-07-14 16:14:33 -05:00

bli_gentprot_macro_defs.h

Updated copyright headers (continued).

2014-07-14 16:14:33 -05:00

bli_kernel_3m_macro_defs.h

Added 30xk support for packm ukernels.

2014-09-22 16:06:15 -05:00

bli_kernel_4m_macro_defs.h

Added 4m_1b implementation for gemm.

2014-12-12 14:31:57 -06:00

bli_kernel_macro_defs.h

Added new piledriver micro-kernels.

2014-10-29 10:52:23 -05:00

bli_kernel_post_macro_defs.h

Retired portions of bli_kernel_3m/4m_macro_defs.h.

2014-09-01 16:23:17 -05:00

bli_kernel_pre_macro_defs.h

Added 4m_1b implementation for gemm.

2014-12-12 14:31:57 -06:00

bli_kernel_prototypes.h

Updated copyright headers (continued).

2014-07-14 16:14:33 -05:00

bli_kernel_rih_macro_defs.h

Added 30xk support for packm ukernels.

2014-09-22 16:06:15 -05:00

bli_kernel_type_defs.h

Pass pack schema into packm_struc_cxk*().

2014-08-31 11:58:50 -05:00

bli_macro_defs.h

Updated copyright headers (continued).

2014-07-14 16:14:33 -05:00

bli_mem_macro_defs.h

Updated copyright headers (continued).

2014-07-14 16:14:33 -05:00

bli_mem_pool_macro_defs.h

Minor 4m/3m consolidation to mem_pool_macro_defs.h.

2015-01-06 12:15:21 -06:00

bli_obj_macro_defs.h

Extended newly relaxed KC to hemm, symm.

2014-10-23 10:50:59 -05:00

bli_param_macro_defs.h

Use correct definition of bli_is_last_iter().

2014-10-17 11:25:59 -05:00

bli_pool_macro_defs.h

Updated copyright headers (continued).

2014-07-14 16:14:33 -05:00

bli_scalar_macro_defs.h

Added high-level implementations of 4m, 3m.

2014-09-16 18:19:32 -05:00

bli_system.h

Updated copyright headers (continued).

2014-07-14 16:14:33 -05:00

bli_type_defs.h

Re-implemented micro-panel alignment.

2014-10-20 19:23:06 -05:00

blis.h

Added basic flop-counting mechanism (level-3 only).

2015-02-04 12:11:55 -06:00