/* BLIS An object-based framework for developing high-performance BLAS-like libraries. Copyright (C) 2014, The University of Texas at Austin Copyright (C) 2018 - 2019, Advanced Micro Devices, Inc. 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. */ #include "blis.h" #define FUNCPTR_T gemm_fp typedef void (*FUNCPTR_T) ( doff_t diagoffb, pack_t schema_a, pack_t schema_b, dim_t m, dim_t n, dim_t k, void* alpha1, void* a, inc_t cs_a, dim_t pd_a, inc_t ps_a, void* b, inc_t rs_b, dim_t pd_b, inc_t ps_b, void* alpha2, void* c, inc_t rs_c, inc_t cs_c, cntx_t* cntx, rntm_t* rntm, thrinfo_t* thread ); static FUNCPTR_T GENARRAY(ftypes,trsm_ru_ker_var2); void bli_trsm_ru_ker_var2 ( obj_t* a, obj_t* b, obj_t* c, cntx_t* cntx, rntm_t* rntm, cntl_t* cntl, thrinfo_t* thread ) { num_t dt_exec = bli_obj_exec_dt( c ); doff_t diagoffb = bli_obj_diag_offset( b ); pack_t schema_a = bli_obj_pack_schema( a ); pack_t schema_b = bli_obj_pack_schema( b ); dim_t m = bli_obj_length( c ); dim_t n = bli_obj_width( c ); dim_t k = bli_obj_width( a ); void* buf_a = bli_obj_buffer_at_off( a ); inc_t cs_a = bli_obj_col_stride( a ); dim_t pd_a = bli_obj_panel_dim( a ); inc_t ps_a = bli_obj_panel_stride( a ); void* buf_b = bli_obj_buffer_at_off( b ); inc_t rs_b = bli_obj_row_stride( b ); dim_t pd_b = bli_obj_panel_dim( b ); inc_t ps_b = bli_obj_panel_stride( b ); void* buf_c = bli_obj_buffer_at_off( c ); inc_t rs_c = bli_obj_row_stride( c ); inc_t cs_c = bli_obj_col_stride( c ); void* buf_alpha1; void* buf_alpha2; FUNCPTR_T f; // Grab the address of the internal scalar buffer for the scalar // attached to A (the non-triangular matrix). This will be the alpha // scalar used in the gemmtrsm subproblems (ie: the scalar that would // be applied to the packed copy of A prior to it being updated by // the trsm subproblem). This scalar may be unit, if for example it // was applied during packing. buf_alpha1 = bli_obj_internal_scalar_buffer( a ); // Grab the address of the internal scalar buffer for the scalar // attached to C. This will be the "beta" scalar used in the gemm-only // subproblems that correspond to micro-panels that do not intersect // the diagonal. We need this separate scalar because it's possible // that the alpha attached to B was reset, if it was applied during // packing. buf_alpha2 = bli_obj_internal_scalar_buffer( c ); // Index into the type combination array to extract the correct // function pointer. f = ftypes[dt_exec]; // Invoke the function. f( diagoffb, schema_a, schema_b, m, n, k, buf_alpha1, buf_a, cs_a, pd_a, ps_a, buf_b, rs_b, pd_b, ps_b, buf_alpha2, buf_c, rs_c, cs_c, cntx, rntm, thread ); } #undef GENTFUNC #define GENTFUNC( ctype, ch, varname ) \ \ void PASTEMAC(ch,varname) \ ( \ doff_t diagoffb, \ pack_t schema_a, \ pack_t schema_b, \ dim_t m, \ dim_t n, \ dim_t k, \ void* alpha1, \ void* a, inc_t cs_a, dim_t pd_a, inc_t ps_a, \ void* b, inc_t rs_b, dim_t pd_b, inc_t ps_b, \ void* alpha2, \ void* c, inc_t rs_c, inc_t cs_c, \ cntx_t* cntx, \ rntm_t* rntm, \ thrinfo_t* thread \ ) \ { \ const num_t dt = PASTEMAC(ch,type); \ \ /* Alias some constants to simpler names. */ \ const dim_t MR = pd_a; \ const dim_t NR = pd_b; \ const dim_t PACKMR = cs_a; \ const dim_t PACKNR = rs_b; \ \ /* Cast the micro-kernel address to its function pointer type. */ \ /* NOTE: We use the lower-triangular gemmtrsm ukernel because, while the current macro-kernel targets the "ru" case (right-side/upper- triangular), it becomes lower-triangular after the kernel operation is transposed so that all kernel instances are of the "left" variety (since those are the only trsm ukernels that exist). */ \ PASTECH(ch,gemmtrsm_ukr_ft) \ gemmtrsm_ukr = bli_cntx_get_l3_vir_ukr_dt( dt, BLIS_GEMMTRSM_L_UKR, cntx ); \ PASTECH(ch,gemm_ukr_ft) \ gemm_ukr = bli_cntx_get_l3_vir_ukr_dt( dt, BLIS_GEMM_UKR, cntx ); \ \ /* Temporary C buffer for edge cases. Note that the strides of this temporary buffer are set so that they match the storage of the original C matrix. For example, if C is column-stored, ct will be column-stored as well. */ \ ctype ct[ BLIS_STACK_BUF_MAX_SIZE \ / sizeof( ctype ) ] \ __attribute__((aligned(BLIS_STACK_BUF_ALIGN_SIZE))); \ const bool col_pref = bli_cntx_ukr_prefers_cols_dt( dt, BLIS_GEMM_VIR_UKR, cntx ); \ const inc_t rs_ct = ( col_pref ? 1 : NR ); \ const inc_t cs_ct = ( col_pref ? MR : 1 ); \ \ ctype* restrict zero = PASTEMAC(ch,0); \ ctype* restrict minus_one = PASTEMAC(ch,m1); \ ctype* restrict a_cast = a; \ ctype* restrict b_cast = b; \ ctype* restrict c_cast = c; \ ctype* restrict alpha1_cast = alpha1; \ ctype* restrict alpha2_cast = alpha2; \ ctype* restrict b1; \ ctype* restrict c1; \ \ doff_t diagoffb_j; \ dim_t k_full; \ dim_t m_iter, m_left; \ dim_t n_iter, n_left; \ dim_t m_cur; \ dim_t n_cur; \ dim_t k_b0111; \ dim_t k_b01; \ dim_t off_b01; \ dim_t off_b11; \ dim_t i, j; \ inc_t rstep_a; \ inc_t cstep_b; \ inc_t rstep_c, cstep_c; \ inc_t istep_a; \ inc_t istep_b; \ inc_t off_scl; \ inc_t ss_b_num; \ inc_t ss_b_den; \ inc_t ps_b_cur; \ inc_t is_b_cur; \ auxinfo_t aux; \ \ /* Assumptions/assertions: rs_a == 1 cs_a == PACKNR pd_a == NR ps_a == stride to next micro-panel of A rs_b == PACKMR cs_b == 1 pd_b == MR ps_b == stride to next micro-panel of B rs_c == (no assumptions) cs_c == (no assumptions) Note that MR/NR and PACKMR/PACKNR have been swapped to reflect the swapping of values in the control tree (ie: those values used when packing). This swapping is needed since we cast right-hand trsm in terms of transposed left-hand trsm. So, if we're going to be transposing the operation, then A needs to be packed with NR and B needs to be packed with MR (remember: B is the triangular matrix in the right-hand side parameter case). */ \ \ /* Safety trap: Certain indexing within this macro-kernel does not work as intended if both MR and NR are odd. */ \ if ( ( bli_is_odd( PACKMR ) && bli_is_odd( NR ) ) || \ ( bli_is_odd( PACKNR ) && bli_is_odd( MR ) ) ) bli_abort(); \ \ /* If any dimension is zero, return immediately. */ \ if ( bli_zero_dim3( m, n, k ) ) return; \ \ /* Safeguard: If the current panel of B is entirely below its diagonal, it is implicitly zero. So we do nothing. */ \ if ( bli_is_strictly_below_diag_n( diagoffb, k, n ) ) return; \ \ /* Compute k_full as k inflated up to a multiple of NR. This is needed because some parameter combinations of trsm reduce k to advance past zero regions in the triangular matrix, and when computing the imaginary stride of B (the non-triangular matrix), which is used by 4m1/3m1 implementations, we need this unreduced value of k. */ \ k_full = ( k % NR != 0 ? k + NR - ( k % NR ) : k ); \ \ /* Compute indexing scaling factor for for 4m or 3m. This is needed because one of the packing register blocksizes (PACKMR or PACKNR) is used to index into the micro-panels of the non- triangular matrix when computing with a diagonal-intersecting micro-panel of the triangular matrix. In the case of 4m or 3m, real values are stored in both sub-panels, and so the indexing needs to occur in units of real values. The value computed here is divided into the complex pointer offset to cause the pointer to be advanced by the correct value. */ \ if ( bli_is_4mi_packed( schema_b ) || \ bli_is_3mi_packed( schema_b ) || \ bli_is_rih_packed( schema_b ) ) off_scl = 2; \ else off_scl = 1; \ \ /* Compute the storage stride scaling. Usually this is just 1. However, in the case of interleaved 3m, we need to scale the offset by 3/2. Note that real-only, imag-only, and summed-only packing formats are not applicable here since trsm is a two- operand operation only (unlike trmm, which is capable of three- operand). */ \ if ( bli_is_3mi_packed( schema_b ) ) { ss_b_num = 3; ss_b_den = 2; } \ else { ss_b_num = 1; ss_b_den = 1; } \ \ /* If there is a zero region to the left of where the diagonal of B intersects the top edge of the panel, adjust the pointer to C and treat this case as if the diagonal offset were zero. This skips over the region that was not packed. (Note we assume the diagonal offset is a multiple of MR; this assumption will hold as long as the cache blocksizes are each a multiple of MR and NR.) */ \ if ( diagoffb > 0 ) \ { \ j = diagoffb; \ n = n - j; \ diagoffb = 0; \ c_cast = c_cast + (j )*cs_c; \ } \ \ /* If there is a zero region below where the diagonal of B intersects the right side of the block, shrink it to prevent "no-op" iterations from executing. */ \ if ( -diagoffb + n < k ) \ { \ k = -diagoffb + n; \ } \ \ /* Check the k dimension, which needs to be a multiple of NR. If k isn't a multiple of NR, we adjust it higher to satisfy the micro- kernel, which is expecting to perform an NR x NR triangular solve. This adjustment of k is consistent with what happened when B was packed: all of its bottom/right edges were zero-padded, and furthermore, the panel that stores the bottom-right corner of the matrix has its diagonal extended into the zero-padded region (as identity). This allows the trsm of that bottom-right panel to proceed without producing any infs or NaNs that would infect the "good" values of the corresponding block of A. */ \ if ( k % NR != 0 ) k += NR - ( k % NR ); \ \ /* NOTE: We don't need to check that n is a multiple of PACKNR since we know that the underlying buffer was already allocated to have an n dimension that is a multiple of PACKNR, with the region between the last column and the next multiple of NR zero-padded accordingly. */ \ \ /* Clear the temporary C buffer in case it has any infs or NaNs. */ \ PASTEMAC(ch,set0s_mxn)( MR, NR, \ ct, rs_ct, cs_ct ); \ \ /* Compute number of primary and leftover components of the m and n dimensions. */ \ n_iter = n / NR; \ n_left = n % NR; \ \ m_iter = m / MR; \ m_left = m % MR; \ \ if ( n_left ) ++n_iter; \ if ( m_left ) ++m_iter; \ \ /* Determine some increments used to step through A, B, and C. */ \ rstep_a = ps_a; \ \ cstep_b = ps_b; \ \ rstep_c = rs_c * MR; \ cstep_c = cs_c * NR; \ \ istep_a = PACKMR * k_full; \ istep_b = PACKNR * k; \ \ if ( bli_is_odd( istep_a ) ) istep_a += 1; \ if ( bli_is_odd( istep_b ) ) istep_b += 1; \ \ /* Save the pack schemas of A and B to the auxinfo_t object. NOTE: We swap the values for A and B since the triangular "A" matrix is actually contained within B. */ \ bli_auxinfo_set_schema_a( schema_b, &aux ); \ bli_auxinfo_set_schema_b( schema_a, &aux ); \ \ /* Save the imaginary stride of A to the auxinfo_t object. NOTE: We swap the values for A and B since the triangular "A" matrix is actually contained within B. */ \ bli_auxinfo_set_is_b( istep_a, &aux ); \ \ b1 = b_cast; \ c1 = c_cast; \ \ /* Loop over the n dimension (NR columns at a time). */ \ for ( j = 0; j < n_iter; ++j ) \ { \ ctype* restrict a1; \ ctype* restrict c11; \ ctype* restrict b01; \ ctype* restrict b11; \ ctype* restrict b2; \ \ diagoffb_j = diagoffb - ( doff_t )j*NR; \ a1 = a_cast; \ c11 = c1; \ \ n_cur = ( bli_is_not_edge_f( j, n_iter, n_left ) ? NR : n_left ); \ \ /* Initialize our next panel of B to be the current panel of B. */ \ b2 = b1; \ \ /* If the current panel of B intersects the diagonal, use a special micro-kernel that performs a fused gemm and trsm. If the current panel of B resides above the diagonal, use a a regular gemm micro-kernel. Otherwise, if it is below the diagonal, it was not packed (because it is implicitly zero) and so we do nothing. */ \ if ( bli_intersects_diag_n( diagoffb_j, k, NR ) ) \ { \ /* Determine the offset to and length of the panel that was packed so we can index into the corresponding location in A. */ \ off_b01 = 0; \ k_b0111 = bli_min( k, -diagoffb_j + NR ); \ k_b01 = k_b0111 - NR; \ off_b11 = k_b01; \ \ /* Compute the addresses of the panel B10 and the triangular block B11. */ \ b01 = b1; \ /* b11 = b1 + ( k_b01 * PACKNR ) / off_scl; */ \ b11 = bli_ptr_inc_by_frac( b1, sizeof( ctype ), k_b01 * PACKNR, off_scl ); \ \ /* Compute the panel stride for the current micro-panel. */ \ is_b_cur = k_b0111 * PACKNR; \ is_b_cur += ( bli_is_odd( is_b_cur ) ? 1 : 0 ); \ ps_b_cur = ( is_b_cur * ss_b_num ) / ss_b_den; \ \ /* Save the 4m1/3m1 imaginary stride of B to the auxinfo_t object. NOTE: We swap the values for A and B since the triangular "A" matrix is actually contained within B. */ \ bli_auxinfo_set_is_a( is_b_cur, &aux ); \ \ /* Loop over the m dimension (MR rows at a time). */ \ for ( i = 0; i < m_iter; ++i ) \ { \ if( bli_trsm_my_iter( i, thread ) ){ \ \ ctype* restrict a10; \ ctype* restrict a11; \ ctype* restrict a2; \ \ m_cur = ( bli_is_not_edge_f( i, m_iter, m_left ) ? MR : m_left ); \ \ /* Compute the addresses of the A10 panel and A11 block. */ \ a10 = a1 + ( off_b01 * PACKMR ) / off_scl; \ a11 = a1 + ( off_b11 * PACKMR ) / off_scl; \ \ /* Compute the addresses of the next panels of A and B. */ \ a2 = a1; \ /*if ( bli_is_last_iter( i, m_iter, 0, 1 ) ) */\ if ( i + bli_thrinfo_num_threads(thread) >= m_iter ) \ { \ a2 = a_cast; \ b2 = b1 + ps_b_cur; \ if ( bli_is_last_iter( j, n_iter, 0, 1 ) ) \ b2 = b_cast; \ } \ \ /* Save addresses of next panels of A and B to the auxinfo_t object. NOTE: We swap the values for A and B since the triangular "A" matrix is actually contained within B. */ \ bli_auxinfo_set_next_a( b2, &aux ); \ bli_auxinfo_set_next_b( a2, &aux ); \ \ /* Handle interior and edge cases separately. */ \ if ( m_cur == MR && n_cur == NR ) \ { \ /* Invoke the fused gemm/trsm micro-kernel. */ \ gemmtrsm_ukr \ ( \ k_b01, \ alpha1_cast, \ b01, \ b11, \ a10, \ a11, \ c11, cs_c, rs_c, \ &aux, \ cntx \ ); \ } \ else \ { \ /* Invoke the fused gemm/trsm micro-kernel. */ \ gemmtrsm_ukr \ ( \ k_b01, \ alpha1_cast, \ b01, \ b11, \ a10, \ a11, \ ct, cs_ct, rs_ct, \ &aux, \ cntx \ ); \ \ /* Copy the result to the bottom edge of C. */ \ PASTEMAC(ch,copys_mxn)( m_cur, n_cur, \ ct, rs_ct, cs_ct, \ c11, rs_c, cs_c ); \ } \ } \ \ a1 += rstep_a; \ c11 += rstep_c; \ } \ \ b1 += ps_b_cur; \ } \ else if ( bli_is_strictly_above_diag_n( diagoffb_j, k, NR ) ) \ { \ /* Save the 4m1/3m1 imaginary stride of B to the auxinfo_t object. NOTE: We swap the values for A and B since the triangular "A" matrix is actually contained within B. */ \ bli_auxinfo_set_is_a( istep_b, &aux ); \ \ /* Loop over the m dimension (MR rows at a time). */ \ for ( i = 0; i < m_iter; ++i ) \ { \ if( bli_trsm_my_iter( i, thread ) ){ \ \ ctype* restrict a2; \ \ m_cur = ( bli_is_not_edge_f( i, m_iter, m_left ) ? MR : m_left ); \ \ /* Compute the addresses of the next panels of A and B. */ \ a2 = a1; \ /*if ( bli_is_last_iter( i, m_iter, 0, 1 ) ) */\ if ( i + bli_thrinfo_num_threads(thread) >= m_iter ) \ { \ a2 = a_cast; \ b2 = b1 + cstep_b; \ if ( bli_is_last_iter( j, n_iter, 0, 1 ) ) \ b2 = b_cast; \ } \ \ /* Save addresses of next panels of A and B to the auxinfo_t object. NOTE: We swap the values for A and B since the triangular "A" matrix is actually contained within B. */ \ bli_auxinfo_set_next_a( b2, &aux ); \ bli_auxinfo_set_next_b( a2, &aux ); \ \ /* Handle interior and edge cases separately. */ \ if ( m_cur == MR && n_cur == NR ) \ { \ /* Invoke the gemm micro-kernel. */ \ gemm_ukr \ ( \ k, \ minus_one, \ b1, \ a1, \ alpha2_cast, \ c11, cs_c, rs_c, \ &aux, \ cntx \ ); \ } \ else \ { \ /* Invoke the gemm micro-kernel. */ \ gemm_ukr \ ( \ k, \ minus_one, \ b1, \ a1, \ zero, \ ct, cs_ct, rs_ct, \ &aux, \ cntx \ ); \ \ /* Add the result to the edge of C. */ \ PASTEMAC(ch,xpbys_mxn)( m_cur, n_cur, \ ct, rs_ct, cs_ct, \ alpha2_cast, \ c11, rs_c, cs_c ); \ } \ } \ \ a1 += rstep_a; \ c11 += rstep_c; \ } \ \ b1 += cstep_b; \ } \ \ c1 += cstep_c; \ } \ } INSERT_GENTFUNC_BASIC0( trsm_ru_ker_var2 )