mirror of
https://github.com/amd/blis.git
synced 2026-05-05 15:01:13 +00:00
f808d829c5
Details:
- Updated the API and semantics of packm kernels such that they must now
handle edge cases, meaning that a c-by-k packm kernel must be able to
pack edge cases that are fewer than c rows/columns and be able to
zero-fill the remaining elements. They must also be able to zero-fill
the equivalent region when copying fewer than k columns/rows (which is
needed by trsm). The new packm kernel API is generally:
void packm_kernel
(
conj_t conja,
dim_t cdim,
dim_t n,
dim_t n_max,
ctype* restrict kappa,
ctype* restrict a, inc_t inca, inc_t lda,
ctype* restrict p, inc_t ldp,
cntx_t* restrict cntx
);
where cdim and n are the dimensions (short and long, respectively) of
the submatrix being copied from the source matrix A, and n_max is the
"full" long dimension (corresponding to the k dimension in gemm) of
the micropanel. The "full" short dimension (corresponding to the
register blocksize MR or NR) is not part of the API because it is
known intrinsically by the packm kernel implementation. Thanks to
Devin Matthews for prompting us to make this change (#282).
- Updated all reference packm kernels in ref_kernels/1m according to
above changes, as well as all optimized packm kernels (which only
consisted of those for knl).
- Bumped the major soname version number in 'so_version' to 2. At first
I was considering leaving it unchanged, but I couldn't escape the
reality that the packm kernel API is much closer to an expert API
than it is some obscure helper function interface within the framework
that nobody would ever notice.
- Removed reference packm kernels for mr/nr = 30. The only sub-config
that would have been using those kernels is knc, which is likely no
longer being used by very many people (if any). (This also mostly
offset the larger object code footprint incurred by moving the edge-
case handling into the individual packm kernels.)
- Fixed an obscure race condition for 3mh and 4mh induced methods in
which those implementations were modifying the contexts stored in the
gks rather than a local copy.
- Fixed a minor bug in the testsuite that prevented non-1m-based induced
method implementations of trsm from executing.
195 lines
6.3 KiB
C
195 lines
6.3 KiB
C
/*
|
|
|
|
BLIS
|
|
An object-based framework for developing high-performance BLAS-like
|
|
libraries.
|
|
|
|
Copyright (C) 2014, The University of Texas at Austin
|
|
|
|
Redistribution and use in source and binary forms, with or without
|
|
modification, are permitted provided that the following conditions are
|
|
met:
|
|
- Redistributions of source code must retain the above copyright
|
|
notice, this list of conditions and the following disclaimer.
|
|
- Redistributions in binary form must reproduce the above copyright
|
|
notice, this list of conditions and the following disclaimer in the
|
|
documentation and/or other materials provided with the distribution.
|
|
- Neither the name(s) of the copyright holder(s) nor the names of its
|
|
contributors may be used to endorse or promote products derived
|
|
from this software without specific prior written permission.
|
|
|
|
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
|
|
"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
|
|
LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
|
|
A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
|
|
HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
|
|
SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
|
|
LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
|
|
DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
|
|
THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
|
|
(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
|
|
OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
|
|
|
*/
|
|
|
|
#ifndef BLIS_SET1MS_MXN_H
|
|
#define BLIS_SET1MS_MXN_H
|
|
|
|
// set1ms_mxn
|
|
|
|
#define bli_sset1ms_mxn( schema, offm, offn, m, n, a, y, rs_y, cs_y, ld_y ) \
|
|
{ \
|
|
/* Include real domain version to facilitate macro-izing mixed-datatype
|
|
components of packm. */ \
|
|
}
|
|
|
|
#define bli_dset1ms_mxn( schema, offm, offn, m, n, a, y, rs_y, cs_y, ld_y ) \
|
|
{ \
|
|
/* Include real domain version to facilitate macro-izing mixed-datatype
|
|
components of packm. */ \
|
|
}
|
|
|
|
static void bli_cset1ms_mxn
|
|
(
|
|
const pack_t schema,
|
|
const dim_t offm,
|
|
const dim_t offn,
|
|
const dim_t m,
|
|
const dim_t n,
|
|
scomplex* restrict alpha,
|
|
scomplex* restrict y, const inc_t rs_y, const inc_t cs_y, const inc_t ld_y
|
|
)
|
|
{
|
|
inc_t offm_local = offm;
|
|
inc_t offn_local = offn;
|
|
dim_t m_local = m;
|
|
dim_t n_local = n;
|
|
inc_t rs_y1 = rs_y;
|
|
inc_t cs_y1 = cs_y;
|
|
inc_t rs_y2 = rs_y;
|
|
inc_t cs_y2 = cs_y;
|
|
dim_t i, j;
|
|
|
|
/* Optimization: The loops walk through y with unit stride if y is
|
|
column-stored. If y is row-stored, swap the dimensions and strides
|
|
to preserve unit stride movement. */
|
|
if ( cs_y == 1 )
|
|
{
|
|
bli_swap_incs( &offm_local, &offn_local );
|
|
bli_swap_dims( &m_local, &n_local );
|
|
bli_swap_incs( &rs_y1, &cs_y1 );
|
|
bli_swap_incs( &rs_y2, &cs_y2 );
|
|
}
|
|
|
|
/* Handle 1e and 1r separately. */
|
|
if ( bli_is_1e_packed( schema ) )
|
|
{
|
|
scomplex* restrict y_off_ri = y + (offm_local )*rs_y1
|
|
+ (offn_local )*cs_y1;
|
|
scomplex* restrict y_off_ir = y + (offm_local )*rs_y1
|
|
+ (offn_local )*cs_y1 + ld_y/2;
|
|
|
|
for ( j = 0; j < n_local; ++j )
|
|
for ( i = 0; i < m_local; ++i )
|
|
{
|
|
bli_ccopy1es( *(alpha),
|
|
*(y_off_ri + i*rs_y1 + j*cs_y1),
|
|
*(y_off_ir + i*rs_y1 + j*cs_y1) );
|
|
}
|
|
}
|
|
else /* if ( bli_is_1r_packed( schema ) ) */
|
|
{
|
|
/* Scale the non-unit stride by two for the 1r loop, which steps
|
|
in units of real (not complex) values. */
|
|
if ( rs_y2 == 1 ) { cs_y2 *= 2; }
|
|
else /* if ( cs_y2 == 1 ) */ { rs_y2 *= 2; }
|
|
|
|
float* restrict y_cast = ( float* )y;
|
|
float* restrict y_off_r = y_cast + (offm_local )*rs_y2
|
|
+ (offn_local )*cs_y2;
|
|
float* restrict y_off_i = y_cast + (offm_local )*rs_y2
|
|
+ (offn_local )*cs_y2 + ld_y;
|
|
|
|
for ( j = 0; j < n_local; ++j )
|
|
for ( i = 0; i < m_local; ++i )
|
|
{
|
|
bli_ccopy1rs( *(alpha),
|
|
*(y_off_r + i*rs_y2 + j*cs_y2),
|
|
*(y_off_i + i*rs_y2 + j*cs_y2) );
|
|
}
|
|
}
|
|
}
|
|
|
|
static void bli_zset1ms_mxn
|
|
(
|
|
const pack_t schema,
|
|
const dim_t offm,
|
|
const dim_t offn,
|
|
const dim_t m,
|
|
const dim_t n,
|
|
dcomplex* restrict alpha,
|
|
dcomplex* restrict y, const inc_t rs_y, const inc_t cs_y, const inc_t ld_y
|
|
)
|
|
{
|
|
inc_t offm_local = offm;
|
|
inc_t offn_local = offn;
|
|
dim_t m_local = m;
|
|
dim_t n_local = n;
|
|
inc_t rs_y1 = rs_y;
|
|
inc_t cs_y1 = cs_y;
|
|
inc_t rs_y2 = rs_y;
|
|
inc_t cs_y2 = cs_y;
|
|
dim_t i, j;
|
|
|
|
/* Optimization: The loops walk through y with unit stride if y is
|
|
column-stored. If y is row-stored, swap the dimensions and strides
|
|
to preserve unit stride movement. */
|
|
if ( cs_y == 1 )
|
|
{
|
|
bli_swap_incs( &offm_local, &offn_local );
|
|
bli_swap_dims( &m_local, &n_local );
|
|
bli_swap_incs( &rs_y1, &cs_y1 );
|
|
bli_swap_incs( &rs_y2, &cs_y2 );
|
|
}
|
|
|
|
/* Handle 1e and 1r separately. */
|
|
if ( bli_is_1e_packed( schema ) )
|
|
{
|
|
dcomplex* restrict y_off_ri = y + (offm_local )*rs_y1
|
|
+ (offn_local )*cs_y1;
|
|
dcomplex* restrict y_off_ir = y + (offm_local )*rs_y1
|
|
+ (offn_local )*cs_y1 + ld_y/2;
|
|
|
|
for ( j = 0; j < n_local; ++j )
|
|
for ( i = 0; i < m_local; ++i )
|
|
{
|
|
bli_zcopy1es( *(alpha),
|
|
*(y_off_ri + i*rs_y1 + j*cs_y1),
|
|
*(y_off_ir + i*rs_y1 + j*cs_y1) );
|
|
}
|
|
}
|
|
else /* if ( bli_is_1r_packed( schema ) ) */
|
|
{
|
|
/* Scale the non-unit stride by two for the 1r loop, which steps
|
|
in units of real (not complex) values. */
|
|
if ( rs_y2 == 1 ) { cs_y2 *= 2; }
|
|
else /* if ( cs_y2 == 1 ) */ { rs_y2 *= 2; }
|
|
|
|
double* restrict y_cast = ( double* )y;
|
|
double* restrict y_off_r = y_cast + (offm_local )*rs_y2
|
|
+ (offn_local )*cs_y2;
|
|
double* restrict y_off_i = y_cast + (offm_local )*rs_y2
|
|
+ (offn_local )*cs_y2 + ld_y;
|
|
|
|
for ( j = 0; j < n_local; ++j )
|
|
for ( i = 0; i < m_local; ++i )
|
|
{
|
|
bli_zcopy1rs( *(alpha),
|
|
*(y_off_r + i*rs_y2 + j*cs_y2),
|
|
*(y_off_i + i*rs_y2 + j*cs_y2) );
|
|
}
|
|
}
|
|
}
|
|
|
|
#endif
|