blis/addon/aocl_gemm/aocl_gemm_s8s8s32os32_utils.c

/*

   BLIS
   An object-based framework for developing high-performance BLAS-like
   libraries.

   Copyright (C) 2023 - 2025, Advanced Micro Devices, Inc. All rights reserved.

   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"
#include "aocl_gemm_interface_apis.h"
#include "lpgemm_types.h"
#include "lpgemm_config.h"
#include "lpgemm_utils_s8.h"
#include "lpgemm_reorder_s8.h"

AOCL_GEMM_GET_REORDER_BUF_SIZE(s8s8s32os32)
{
	if ( ( k <= 0 ) || ( n <= 0 ) )
	{
		return 0; // Error.
	}

	// Check if avx512_vnni ISA is supported, lpgemm matmul only works with it.
	if ( bli_cpuid_is_avx512vnni_supported() == FALSE )
	{
		bli_print_msg(" AVX512_VNNI ISA not supported by processor, "
				"cannot perform s8s8s32 gemm.", __FILE__, __LINE__ );
		return 0; // Error.
	}

	/* Initialize BLIS. */
	bli_init_auto();

	// Set MC, NC, KC, NR, MR.
	aocl_lpgemm_init_global_cntx();

	AOCL_MATRIX_TYPE input_mat_type;
	bli_param_map_char_to_lpmat_type( mat_type, &input_mat_type );

	if ( input_mat_type == A_MATRIX )
	{
		return 0; // A reorder not supported.
	}

	// Extra space since packing does width in multiples of 16. The vnni
	// instruction can be used as long as atleast one zmm register can be fully
	// loaded; and since k_dim needs to be atleast 4, having n_dim atleast 16
	// should give 4x16=64 elements, enough for 1 zmm register.The padding is
	// not rounded to NR (=64), since that would result in memory wastage.
#ifdef BLIS_KERNELS_ZEN4
	dim_t n_reorder;
	if( n == 1 )
	{
		n_reorder = 1;
	}
	else
	{
		n_reorder = make_multiple_of_n( n, 16 );

	}

	// Extra space since packing does length in multiples of 4.
	dim_t k_reorder;
	if( n == 1 )
	{
		k_reorder = k;
	}
	else
	{
		k_reorder = make_multiple_of_n( k, 4 );
	}
#else
	dim_t n_reorder = make_multiple_of_n( n, 16 );
	dim_t k_reorder = make_multiple_of_n( k, 4 );
#endif
	//extra memory of n_reorder * sizeof(int32_t) to store sum of every column of B matrix buffer
	siz_t size_req = sizeof( int8_t ) * k_reorder * n_reorder + ( n_reorder * sizeof( int32_t ) );

	return size_req;
}


AOCL_GEMM_GET_REORDER_BUF_SIZE_SYM_QUANT(s8s8s32os32_sym_quant)
{
	if ( ( k <= 0 ) || ( n <= 0 ) )
	{
		return 0; // Error.
	}

	// Check if avx512_vnni ISA is supported, lpgemm matmul only works with it.
	if ( bli_cpuid_is_avx512vnni_supported() == FALSE )
	{
		bli_print_msg(" AVX512_VNNI ISA not supported by processor, "
				"cannot perform s8s8s32 gemm.", __FILE__, __LINE__ );
		return 0; // Error.
	}

	/* Initialize BLIS. */
	bli_init_auto();

	// Set MC, NC, KC, NR, MR.
	aocl_lpgemm_init_global_cntx();

	AOCL_MATRIX_TYPE input_mat_type;
	bli_param_map_char_to_lpmat_type( mat_type, &input_mat_type );

	if ( input_mat_type == A_MATRIX )
	{
		return 0; // A reorder not supported.
	}

	// Extra space since packing does width in multiples of 16. The vnni
	// instruction can be used as long as atleast one zmm register can be fully
	// loaded; and since k_dim needs to be atleast 4, having n_dim atleast 16
	// should give 4x16=64 elements, enough for 1 zmm register.The padding is
	// not rounded to NR (=64), since that would result in memory wastage.
#ifdef BLIS_KERNELS_ZEN4
	dim_t n_reorder;
	if( n == 1 )
	{
		n_reorder = 1;
	}
	else
	{
		n_reorder = make_multiple_of_n( n, 16 );
	}

	// Extra space since packing does length in multiples of 4.
	dim_t k_reorder;
	if( n == 1 )
	{
		k_reorder = k;
	}
	else
	{
		k_reorder = make_multiple_of_n( k, 4 );
	}
#else
	dim_t n_reorder = make_multiple_of_n( n, 16 );
	dim_t k_reorder = make_multiple_of_n( k, 4 );
#endif
	dim_t group_size = meta_data->group_size;

	if( group_size & 3 )
	{
		bli_print_msg(" Group size should be multiple of 4 for s8s8s32os32_sym_quant", __FILE__, __LINE__ );
		return 0; // Error.
	}

	dim_t num_groups = ( k + group_size - 1 ) / group_size;

	// extra memory to store sum of every column per group of B matrix buffer
	size_t extra_mem_req = num_groups * n_reorder * sizeof( int32_t );

	//extra memory of n_reorder * sizeof(int32_t) to store sum of every column of B matrix buffer
	siz_t size_req = sizeof( int8_t ) * k_reorder * n_reorder + extra_mem_req;

	return size_req;
}

AOCL_GEMM_REORDER(int8_t,s8s8s32os32)
{
	trans_t blis_trans;
	/* Map BLAS chars to their corresponding BLIS enumerated type value. */
	bli_param_map_netlib_to_blis_trans(trans, &blis_trans);

	if ((input_buf_addr == NULL) || (reorder_buf_addr == NULL) ||
		(k <= 0) || (n <= 0) )
	{
		return; // Error.
	}

	inc_t rs_b, cs_b;
	if( ( order == 'r') || ( order == 'R' ) )
	{
		if( ( bli_is_notrans( blis_trans ) && ( ldb < n ) ) ||
		    ( bli_is_trans( blis_trans ) && ( ldb < k ) ) )
		{
			return; // Error.
		}
		else
		{
			rs_b = bli_is_notrans( blis_trans ) ? ldb : 1;
			cs_b = bli_is_notrans( blis_trans ) ? 1 : ldb;
		}
	}
	else if ( ( order == 'c' ) || ( order == 'C' ) )
	{
		if( ( bli_is_notrans( blis_trans ) && ( ldb < k ) ) ||
		    ( bli_is_trans( blis_trans ) && ( ldb < n ) ) )
		{
			return; // Error.
		}
		else
		{
			rs_b = bli_is_notrans( blis_trans ) ? 1 : ldb;
			cs_b = bli_is_notrans( blis_trans ) ? ldb : 1;
		}
	}
	else
	{
		return; // Error
	}

	// Check if avx512_vnni ISA is supported, lpgemm matmul only works with it.
	if ( bli_cpuid_is_avx512vnni_supported() == FALSE )
	{
		bli_print_msg(" AVX512_VNNI ISA not supported by processor, "
				"cannot perform s8s8s32 gemm.", __FILE__, __LINE__ );
		return; // Error.
	}

	/* Initialize BLIS. */
	bli_init_auto();

	// Set MC, NC, KC, NR, MR.
	aocl_lpgemm_init_global_cntx();

	AOCL_MATRIX_TYPE input_mat_type;
	bli_param_map_char_to_lpmat_type( mat_type, &input_mat_type );

	if ( input_mat_type == A_MATRIX )
	{
		return; // A reorder not supported.
	}
#ifdef BLIS_KERNELS_ZEN4
	if( n == 1 )
	{
		int32_t* pack_b_column_sum = ( int32_t* ) ( reorder_buf_addr +
		                               ( sizeof( int8_t ) * n * k ));

		*pack_b_column_sum =  0;

		for( dim_t k0 = 0; k0 < k; k0++ )
		{
			reorder_buf_addr[k0] = input_buf_addr[ k0 * rs_b ];
			*pack_b_column_sum += reorder_buf_addr[k0];
		}
		*pack_b_column_sum *= 128;
		return;
	}
#endif
	// Initialize a local runtime with global settings if necessary. Note
	// that in the case that a runtime is passed in, we make a local copy.
	rntm_t rntm_g;
	bli_rntm_init_from_global( &rntm_g );
	bli_pba_rntm_set_pba( &rntm_g );

	lpgemm_cntx_t* lcntx_g = lpgemm_get_global_cntx_obj( S8S8S32OS32 );

	// Create dummy b_reorder obj.
	lpgemm_obj_t b_reorder;
	b_reorder.storage.aligned_buffer = reorder_buf_addr;

	// Create dummy original b obj;
	lpgemm_obj_t b;
	b.storage.aligned_buffer = ( void* )input_buf_addr;
	b.rs = rs_b;
	b.cs = cs_b;
	b.width = n;
	b.length = k;

	reorderb_nr64_s8s8s32o32( &b, &b_reorder, &rntm_g, lcntx_g );
}


AOCL_GEMM_REORDER_SYM_QUANT(int8_t,s8s8s32os32_sym_quant)
{
	trans_t blis_trans;
	/* Map BLAS chars to their corresponding BLIS enumerated type value. */
	bli_param_map_netlib_to_blis_trans(trans, &blis_trans);

	if ((input_buf_addr == NULL) || (reorder_buf_addr == NULL) ||
		(k <= 0) || (n <= 0) )
	{
		return; // Error.
	}

	inc_t rs_b, cs_b;
	if( ( order == 'r') || ( order == 'R' ) )
	{
		if( ( bli_is_notrans( blis_trans ) && ( ldb < n ) ) ||
		    ( bli_is_trans( blis_trans ) && ( ldb < k ) ) )
		{
			return; // Error.
		}
		else
		{
			rs_b = bli_is_notrans( blis_trans ) ? ldb : 1;
			cs_b = bli_is_notrans( blis_trans ) ? 1 : ldb;
		}
	}
	else if ( ( order == 'c' ) || ( order == 'C' ) )
	{
		if( ( bli_is_notrans( blis_trans ) && ( ldb < k ) ) ||
		    ( bli_is_trans( blis_trans ) && ( ldb < n ) ) )
		{
			return; // Error.
		}
		else
		{
			rs_b = bli_is_notrans( blis_trans ) ? 1 : ldb;
			cs_b = bli_is_notrans( blis_trans ) ? ldb : 1;
		}
	}
	else
	{
		return; // Error
	}

	// Check if avx512_vnni ISA is supported, lpgemm matmul only works with it.
	if ( bli_cpuid_is_avx512vnni_supported() == FALSE )
	{
		bli_print_msg(" AVX512_VNNI ISA not supported by processor, "
				"cannot perform s8s8s32 gemm.", __FILE__, __LINE__ );
		return; // Error.
	}

	dim_t group_size = meta_data->group_size;
	if( group_size & 3 )
	{
		bli_print_msg(" Group size should be multiple of 4 for s8s8s32os32_sym_quant", __FILE__, __LINE__ );
		return; // Error.
	}

	/* Initialize BLIS. */
	bli_init_auto();

	// Set MC, NC, KC, NR, MR.
	aocl_lpgemm_init_global_cntx();

	AOCL_MATRIX_TYPE input_mat_type;
	bli_param_map_char_to_lpmat_type( mat_type, &input_mat_type );

	if ( input_mat_type == A_MATRIX )
	{
		return; // A reorder not supported.
	}

#ifdef BLIS_KERNELS_ZEN4
	if( n == 1 )
	{
        // Calculate the address of the beginning of the column sum buffer that
        // is allocated after the reorder buffer.
        int32_t* pack_b_column_sum = ( int32_t* ) ( reorder_buf_addr + 
                                                    ( k * sizeof( int8_t ) ) );

        // NOTE We're working under the assumption that group_size is a factor
        // of k.
        for ( dim_t k0 = 0; k0 < k; k0 += group_size )
        {
            // Initialize the current column sum to 0.
            *pack_b_column_sum = 0;
            for ( dim_t group = 0; group < group_size; group++ )
            {
                reorder_buf_addr[ k0 + group ] = input_buf_addr[ ( k0 + group ) * rs_b ];
                *pack_b_column_sum += reorder_buf_addr[ k0 + group ];
            }

            *pack_b_column_sum *= 128;
            // Move the pack_b_column_sum pointer one step to the next group.
            pack_b_column_sum += 1;
        }

        return;
	}
#endif

	// Initialize a local runtime with global settings if necessary. Note
	// that in the case that a runtime is passed in, we make a local copy.
	rntm_t rntm_g;
	bli_rntm_init_from_global( &rntm_g );
	bli_pba_rntm_set_pba( &rntm_g );

	lpgemm_cntx_t* lcntx_g = lpgemm_get_global_cntx_obj( S8S8S32OS32 );

	// Create dummy b_reorder obj.
	lpgemm_obj_t b_reorder;
	b_reorder.storage.aligned_buffer = reorder_buf_addr;

	// Create dummy original b obj;
	lpgemm_obj_t b;
	b.storage.aligned_buffer = ( void* )input_buf_addr;
	b.rs = rs_b;
	b.cs = cs_b;
	b.width = n;
	b.length = k;

	reorderb_nr64_s8s8s32o32_sym_quant( &b, &b_reorder, &rntm_g, lcntx_g, group_size );
}

AOCL_GEMM_UNREORDER(int8_t, s8s8s32os32_reference)
{
	if ( ( output_buf_addr == NULL ) || ( reorder_buf_addr == NULL ) ||
	     ( k <= 0 ) || ( n <= 0 ) )
	{
		return; // Error.
	}

	inc_t rs_b, cs_b;

	// Check for the validity of strides.
	if( ( order == 'r' ) || ( order == 'R' ) )
	{
		if( ldb < n ) return; // Error
		else
		{
			rs_b = ldb;
			cs_b = 1;
		}
	}
	else if( ( order == 'c' ) || ( order == 'C' ) )
	{
		if( ldb < k ) return; // Error.
		else
		{
			rs_b = 1;
			cs_b = ldb;
		}
	}
	else
	{
		return; // Error.
	}

	/* Initialize BLIS. */
	bli_init_auto();

	// Set MC, NC, KC, NR, MR.
	aocl_lpgemm_init_global_cntx();

	AOCL_MATRIX_TYPE input_mat_type;
	bli_param_map_char_to_lpmat_type( mat_type, &input_mat_type );

	if ( input_mat_type == A_MATRIX )
	{
		return; // A reorder not supported.
	}
#ifdef BLIS_KERNELS_ZEN4
	if( n == 1 )
	{
		if( rs_b == 1 )
		{
			memcpy( output_buf_addr, reorder_buf_addr, ( k * sizeof( int8_t ) ) );
		}
		else
		{
			for( dim_t k0 = 0; k0 < k; k0++ )
			{
				output_buf_addr[k0*rs_b] = reorder_buf_addr[k0];
			}
		}
		return;
	}
#endif

	// Initialize a local runtime with global settings if necessary. Note
	// that in the case that a runtime is passed in, we make a local copy.
	rntm_t rntm_g;
	bli_rntm_init_from_global( &rntm_g );
	bli_pba_rntm_set_pba( &rntm_g );

	lpgemm_cntx_t* lcntx_g = lpgemm_get_global_cntx_obj( S8S8S32OS32 );

	// Create dummy b_reorder obj.
	lpgemm_obj_t b_reorder;
	b_reorder.storage.aligned_buffer = ( void* ) reorder_buf_addr;

	// Create dummy original b obj;
	lpgemm_obj_t b;
	b.storage.aligned_buffer = ( void* )output_buf_addr;
	b.rs = rs_b;
	b.cs = cs_b;
	b.width = n;
	b.length = k;

	unreorderb_nr64_s8s8s32os32_reference( &b, &b_reorder, &rntm_g, lcntx_g );
}