amxintrin.h

Go to the documentation of this file.

/*===--------------- amxintrin.h - AMX intrinsics -*- C/C++ -*---------------===
 *
 * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
 * See https://llvm.org/LICENSE.txt for license information.
 * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
 *
 *===------------------------------------------------------------------------===
 */
 
#ifndef __IMMINTRIN_H
#error "Never use <amxintrin.h> directly; include <immintrin.h> instead."
#endif /* __IMMINTRIN_H */
 
#ifndef __AMXINTRIN_H
#define __AMXINTRIN_H
#ifdef __x86_64__
 
/* Define the default attributes for the functions in this file. */
#define __DEFAULT_FN_ATTRS_TILE                                                \
  __attribute__((__always_inline__, __nodebug__, __target__("amx-tile")))
#define __DEFAULT_FN_ATTRS_INT8                                                \
  __attribute__((__always_inline__, __nodebug__, __target__("amx-int8")))
#define __DEFAULT_FN_ATTRS_BF16                                                \
  __attribute__((__always_inline__, __nodebug__, __target__("amx-bf16")))
#define __DEFAULT_FN_ATTRS_FP16                                                \
  __attribute__((__always_inline__, __nodebug__, __target__("amx-fp16")))
 
/// Load tile configuration from a 64-byte memory location specified by
/// "mem_addr". The tile configuration includes the tile type palette, the
/// number of bytes per row, and the number of rows. If the specified
/// palette_id is zero, that signifies the init state for both the tile
/// config and the tile data, and the tiles are zeroed. Any invalid
/// configurations will result in #GP fault.
///
/// \headerfile <immintrin.h>
///
/// This intrinsic corresponds to the <c> LDTILECFG </c> instruction.
///
/// \param __config
///    A pointer to 512-bits configuration
static __inline__ void __DEFAULT_FN_ATTRS_TILE
_tile_loadconfig(const void *__config) {
  __builtin_ia32_tile_loadconfig(__config);
}
 
/// Stores the current tile configuration to a 64-byte memory location
/// specified by "mem_addr". The tile configuration includes the tile type
/// palette, the number of bytes per row, and the number of rows. If tiles
/// are not configured, all zeroes will be stored to memory.
///
/// \headerfile <immintrin.h>
///
/// This intrinsic corresponds to the <c> STTILECFG </c> instruction.
///
/// \param __config
///    A pointer to 512-bits configuration
static __inline__ void __DEFAULT_FN_ATTRS_TILE
_tile_storeconfig(void *__config) {
  __builtin_ia32_tile_storeconfig(__config);
}
 
/// Release the tile configuration to return to the init state, which
/// releases all storage it currently holds.
///
/// \headerfile <immintrin.h>
///
/// This intrinsic corresponds to the <c> TILERELEASE </c> instruction.
static __inline__ void __DEFAULT_FN_ATTRS_TILE _tile_release(void) {
  __builtin_ia32_tilerelease();
}
 
/// Load tile rows from memory specifieid by "base" address and "stride" into
/// destination tile "dst" using the tile configuration previously configured
/// via "_tile_loadconfig".
///
/// \headerfile <immintrin.h>
///
/// This intrinsic corresponds to the <c> TILELOADD </c> instruction.
///
/// \param dst
///    A destination tile. Max size is 1024 Bytes.
/// \param base
///    A pointer to base address.
/// \param stride
///    The stride between the rows' data to be loaded in memory.
#define _tile_loadd(dst, base, stride)                                         \
  __builtin_ia32_tileloadd64((dst), ((const void *)(base)),                    \
                             (__SIZE_TYPE__)(stride))
 
/// Load tile rows from memory specifieid by "base" address and "stride" into
/// destination tile "dst" using the tile configuration previously configured
/// via "_tile_loadconfig". This intrinsic provides a hint to the implementation
/// that the data will likely not be reused in the near future and the data
/// caching can be optimized accordingly.
///
/// \headerfile <immintrin.h>
///
/// This intrinsic corresponds to the <c> TILELOADDT1 </c> instruction.
///
/// \param dst
///    A destination tile. Max size is 1024 Bytes.
/// \param base
///    A pointer to base address.
/// \param stride
///    The stride between the rows' data to be loaded in memory.
#define _tile_stream_loadd(dst, base, stride)                                  \
  __builtin_ia32_tileloaddt164((dst), ((const void *)(base)),                  \
                               (__SIZE_TYPE__)(stride))
 
/// Store the tile specified by "src" to memory specifieid by "base" address and
/// "stride" using the tile configuration previously configured via
/// "_tile_loadconfig".
///
/// \headerfile <immintrin.h>
///
/// This intrinsic corresponds to the <c> TILESTORED </c> instruction.
///
/// \param dst
///    A destination tile. Max size is 1024 Bytes.
/// \param base
///    A pointer to base address.
/// \param stride
///    The stride between the rows' data to be stored in memory.
#define _tile_stored(dst, base, stride)                                        \
  __builtin_ia32_tilestored64((dst), ((void *)(base)), (__SIZE_TYPE__)(stride))
 
/// Zero the tile specified by "tdest".
///
/// \headerfile <immintrin.h>
///
/// This intrinsic corresponds to the <c> TILEZERO </c> instruction.
///
/// \param tile
///    The destination tile to be zero. Max size is 1024 Bytes.
#define _tile_zero(tile) __builtin_ia32_tilezero((tile))
 
/// Compute dot-product of bytes in tiles with a source/destination accumulator.
/// Multiply groups of 4 adjacent pairs of signed 8-bit integers in src0 with
/// corresponding signed 8-bit integers in src1, producing 4 intermediate 32-bit
/// results. Sum these 4 results with the corresponding 32-bit integer in "dst",
/// and store the 32-bit result back to tile "dst".
///
/// \headerfile <immintrin.h>
///
/// This intrinsic corresponds to the <c> TDPBSSD </c> instruction.
///
/// \param dst
///    The destination tile. Max size is 1024 Bytes.
/// \param src0
///    The 1st source tile. Max size is 1024 Bytes.
/// \param src1
///    The 2nd source tile. Max size is 1024 Bytes.
#define _tile_dpbssd(dst, src0, src1)                                          \
  __builtin_ia32_tdpbssd((dst), (src0), (src1))
 
/// Compute dot-product of bytes in tiles with a source/destination accumulator.
/// Multiply groups of 4 adjacent pairs of signed 8-bit integers in src0 with
/// corresponding unsigned 8-bit integers in src1, producing 4 intermediate
/// 32-bit results. Sum these 4 results with the corresponding 32-bit integer
/// in "dst", and store the 32-bit result back to tile "dst".
///
/// \headerfile <immintrin.h>
///
/// This intrinsic corresponds to the <c> TDPBSUD </c> instruction.
///
/// \param dst
///    The destination tile. Max size is 1024 Bytes.
/// \param src0
///    The 1st source tile. Max size is 1024 Bytes.
/// \param src1
///    The 2nd source tile. Max size is 1024 Bytes.
#define _tile_dpbsud(dst, src0, src1)                                          \
  __builtin_ia32_tdpbsud((dst), (src0), (src1))
 
/// Compute dot-product of bytes in tiles with a source/destination accumulator.
/// Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in src0 with
/// corresponding signed 8-bit integers in src1, producing 4 intermediate 32-bit
/// results. Sum these 4 results with the corresponding 32-bit integer in "dst",
/// and store the 32-bit result back to tile "dst".
///
/// \headerfile <immintrin.h>
///
/// This intrinsic corresponds to the <c> TDPBUSD </c> instruction.
///
/// \param dst
///    The destination tile. Max size is 1024 Bytes.
/// \param src0
///    The 1st source tile. Max size is 1024 Bytes.
/// \param src1
///    The 2nd source tile. Max size is 1024 Bytes.
#define _tile_dpbusd(dst, src0, src1)                                          \
  __builtin_ia32_tdpbusd((dst), (src0), (src1))
 
/// Compute dot-product of bytes in tiles with a source/destination accumulator.
/// Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in src0 with
/// corresponding unsigned 8-bit integers in src1, producing 4 intermediate
/// 32-bit results. Sum these 4 results with the corresponding 32-bit integer in
/// "dst", and store the 32-bit result back to tile "dst".
///
/// \headerfile <immintrin.h>
///
/// This intrinsic corresponds to the <c> TDPBUUD </c> instruction.
///
/// \param dst
///    The destination tile. Max size is 1024 Bytes.
/// \param src0
///    The 1st source tile. Max size is 1024 Bytes.
/// \param src1
///    The 2nd source tile. Max size is 1024 Bytes.
#define _tile_dpbuud(dst, src0, src1)                                          \
  __builtin_ia32_tdpbuud((dst), (src0), (src1))
 
/// Compute dot-product of BF16 (16-bit) floating-point pairs in tiles src0 and
/// src1, accumulating the intermediate single-precision (32-bit) floating-point
/// elements with elements in "dst", and store the 32-bit result back to tile
/// "dst".
///
/// \headerfile <immintrin.h>
///
/// This intrinsic corresponds to the <c> TDPBF16PS </c> instruction.
///
/// \param dst
///    The destination tile. Max size is 1024 Bytes.
/// \param src0
///    The 1st source tile. Max size is 1024 Bytes.
/// \param src1
///    The 2nd source tile. Max size is 1024 Bytes.
#define _tile_dpbf16ps(dst, src0, src1)                                        \
  __builtin_ia32_tdpbf16ps((dst), (src0), (src1))
 
/// AMX tile register size can be configured, the maximum size is 16x64=1024
/// bytes. Since there is no 2D type in llvm IR, we use vector type to
/// represent 2D tile and the fixed size is maximum amx tile register size.
typedef int _tile1024i __attribute__((__vector_size__(1024), __aligned__(64)));
 
/// This is internal intrinsic. C/C++ user should avoid calling it directly.
static __inline__ _tile1024i __DEFAULT_FN_ATTRS_INT8
_tile_loadd_internal(unsigned short m, unsigned short n, const void *base,
                     __SIZE_TYPE__ stride) {
  return __builtin_ia32_tileloadd64_internal(m, n, base,
                                             (__SIZE_TYPE__)(stride));
}
 
/// This is internal intrinsic. C/C++ user should avoid calling it directly.
static __inline__ _tile1024i __DEFAULT_FN_ATTRS_INT8
_tile_loaddt1_internal(unsigned short m, unsigned short n, const void *base,
                       __SIZE_TYPE__ stride) {
  return __builtin_ia32_tileloaddt164_internal(m, n, base,
                                               (__SIZE_TYPE__)(stride));
}
 
/// This is internal intrinsic. C/C++ user should avoid calling it directly.
static __inline__ _tile1024i __DEFAULT_FN_ATTRS_INT8
_tile_dpbssd_internal(unsigned short m, unsigned short n, unsigned short k,
                      _tile1024i dst, _tile1024i src1, _tile1024i src2) {
  return __builtin_ia32_tdpbssd_internal(m, n, k, dst, src1, src2);
}
 
/// This is internal intrinsic. C/C++ user should avoid calling it directly.
static __inline__ _tile1024i __DEFAULT_FN_ATTRS_INT8
_tile_dpbsud_internal(unsigned short m, unsigned short n, unsigned short k,
                      _tile1024i dst, _tile1024i src1, _tile1024i src2) {
  return __builtin_ia32_tdpbsud_internal(m, n, k, dst, src1, src2);
}
 
/// This is internal intrinsic. C/C++ user should avoid calling it directly.
static __inline__ _tile1024i __DEFAULT_FN_ATTRS_INT8
_tile_dpbusd_internal(unsigned short m, unsigned short n, unsigned short k,
                      _tile1024i dst, _tile1024i src1, _tile1024i src2) {
  return __builtin_ia32_tdpbusd_internal(m, n, k, dst, src1, src2);
}
 
/// This is internal intrinsic. C/C++ user should avoid calling it directly.
static __inline__ _tile1024i __DEFAULT_FN_ATTRS_INT8
_tile_dpbuud_internal(unsigned short m, unsigned short n, unsigned short k,
                      _tile1024i dst, _tile1024i src1, _tile1024i src2) {
  return __builtin_ia32_tdpbuud_internal(m, n, k, dst, src1, src2);
}
 
/// This is internal intrinsic. C/C++ user should avoid calling it directly.
static __inline__ void __DEFAULT_FN_ATTRS_INT8
_tile_stored_internal(unsigned short m, unsigned short n, void *base,
                      __SIZE_TYPE__ stride, _tile1024i tile) {
  return __builtin_ia32_tilestored64_internal(m, n, base,
                                              (__SIZE_TYPE__)(stride), tile);
}
 
/// This is internal intrinsic. C/C++ user should avoid calling it directly.
static __inline__ _tile1024i __DEFAULT_FN_ATTRS_BF16
_tile_dpbf16ps_internal(unsigned short m, unsigned short n, unsigned short k,
                        _tile1024i dst, _tile1024i src1, _tile1024i src2) {
  return __builtin_ia32_tdpbf16ps_internal(m, n, k, dst, src1, src2);
}
 
/// This is internal intrinsic. C/C++ user should avoid calling it directly.
static __inline__ _tile1024i __DEFAULT_FN_ATTRS_FP16
_tile_dpfp16ps_internal(unsigned short m, unsigned short n, unsigned short k,
                        _tile1024i dst, _tile1024i src1, _tile1024i src2) {
  return __builtin_ia32_tdpfp16ps_internal(m, n, k, dst, src1, src2);
}
 
/// This struct pack the shape and tile data together for user. We suggest
/// initializing the struct as early as possible, because compiler depends
/// on the shape information to do configure. The constant value is preferred
/// for optimization by compiler.
typedef struct __tile1024i_str {
  const unsigned short row;
  const unsigned short col;
  _tile1024i tile;
} __tile1024i;
 
/// Load tile rows from memory specifieid by "base" address and "stride" into
/// destination tile "dst".
///
/// \headerfile <immintrin.h>
///
/// This intrinsic corresponds to the <c> TILELOADD </c> instruction.
///
/// \param dst
///    A destination tile. Max size is 1024 Bytes.
/// \param base
///    A pointer to base address.
/// \param stride
///    The stride between the rows' data to be loaded in memory.
__DEFAULT_FN_ATTRS_TILE
static __inline__ void __tile_loadd(__tile1024i *dst, const void *base,
                                    __SIZE_TYPE__ stride) {
  dst->tile = _tile_loadd_internal(dst->row, dst->col, base, stride);
}
 
/// Load tile rows from memory specifieid by "base" address and "stride" into
/// destination tile "dst". This intrinsic provides a hint to the implementation
/// that the data will likely not be reused in the near future and the data
/// caching can be optimized accordingly.
///
/// \headerfile <immintrin.h>
///
/// This intrinsic corresponds to the <c> TILELOADDT1 </c> instruction.
///
/// \param dst
///    A destination tile. Max size is 1024 Bytes.
/// \param base
///    A pointer to base address.
/// \param stride
///    The stride between the rows' data to be loaded in memory.
__DEFAULT_FN_ATTRS_TILE
static __inline__ void __tile_stream_loadd(__tile1024i *dst, const void *base,
                                           __SIZE_TYPE__ stride) {
  dst->tile = _tile_loaddt1_internal(dst->row, dst->col, base, stride);
}
 
/// Compute dot-product of bytes in tiles with a source/destination accumulator.
/// Multiply groups of 4 adjacent pairs of signed 8-bit integers in src0 with
/// corresponding signed 8-bit integers in src1, producing 4 intermediate 32-bit
/// results. Sum these 4 results with the corresponding 32-bit integer in "dst",
/// and store the 32-bit result back to tile "dst".
///
/// \headerfile <immintrin.h>
///
/// This intrinsic corresponds to the <c> TDPBSSD </c> instruction.
///
/// \param dst
///    The destination tile. Max size is 1024 Bytes.
/// \param src0
///    The 1st source tile. Max size is 1024 Bytes.
/// \param src1
///    The 2nd source tile. Max size is 1024 Bytes.
__DEFAULT_FN_ATTRS_INT8
static __inline__ void __tile_dpbssd(__tile1024i *dst, __tile1024i src0,
                                     __tile1024i src1) {
  dst->tile = _tile_dpbssd_internal(src0.row, src1.col, src0.col, dst->tile,
                                    src0.tile, src1.tile);
}
 
/// Compute dot-product of bytes in tiles with a source/destination accumulator.
/// Multiply groups of 4 adjacent pairs of signed 8-bit integers in src0 with
/// corresponding unsigned 8-bit integers in src1, producing 4 intermediate
/// 32-bit results. Sum these 4 results with the corresponding 32-bit integer
/// in "dst", and store the 32-bit result back to tile "dst".
///
/// \headerfile <immintrin.h>
///
/// This intrinsic corresponds to the <c> TDPBSUD </c> instruction.
///
/// \param dst
///    The destination tile. Max size is 1024 Bytes.
/// \param src0
///    The 1st source tile. Max size is 1024 Bytes.
/// \param src1
///    The 2nd source tile. Max size is 1024 Bytes.
__DEFAULT_FN_ATTRS_INT8
static __inline__ void __tile_dpbsud(__tile1024i *dst, __tile1024i src0,
                                     __tile1024i src1) {
  dst->tile = _tile_dpbsud_internal(src0.row, src1.col, src0.col, dst->tile,
                                    src0.tile, src1.tile);
}
 
/// Compute dot-product of bytes in tiles with a source/destination accumulator.
/// Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in src0 with
/// corresponding signed 8-bit integers in src1, producing 4 intermediate 32-bit
/// results. Sum these 4 results with the corresponding 32-bit integer in "dst",
/// and store the 32-bit result back to tile "dst".
///
/// \headerfile <immintrin.h>
///
/// This intrinsic corresponds to the <c> TDPBUSD </c> instruction.
///
/// \param dst
///    The destination tile. Max size is 1024 Bytes.
/// \param src0
///    The 1st source tile. Max size is 1024 Bytes.
/// \param src1
///    The 2nd source tile. Max size is 1024 Bytes.
__DEFAULT_FN_ATTRS_INT8
static __inline__ void __tile_dpbusd(__tile1024i *dst, __tile1024i src0,
                                     __tile1024i src1) {
  dst->tile = _tile_dpbusd_internal(src0.row, src1.col, src0.col, dst->tile,
                                    src0.tile, src1.tile);
}
 
/// Compute dot-product of bytes in tiles with a source/destination accumulator.
/// Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in src0 with
/// corresponding unsigned 8-bit integers in src1, producing 4 intermediate
/// 32-bit results. Sum these 4 results with the corresponding 32-bit integer in
/// "dst", and store the 32-bit result back to tile "dst".
///
/// \headerfile <immintrin.h>
///
/// This intrinsic corresponds to the <c> TDPBUUD </c> instruction.
///
/// \param dst
///    The destination tile. Max size is 1024 Bytes.
/// \param src0
///    The 1st source tile. Max size is 1024 Bytes.
/// \param src1
///    The 2nd source tile. Max size is 1024 Bytes.
__DEFAULT_FN_ATTRS_INT8
static __inline__ void __tile_dpbuud(__tile1024i *dst, __tile1024i src0,
                                     __tile1024i src1) {
  dst->tile = _tile_dpbuud_internal(src0.row, src1.col, src0.col, dst->tile,
                                    src0.tile, src1.tile);
}
 
/// Store the tile specified by "src" to memory specifieid by "base" address and
/// "stride".
///
/// \headerfile <immintrin.h>
///
/// This intrinsic corresponds to the <c> TILESTORED </c> instruction.
///
/// \param base
///    A pointer to base address.
/// \param stride
///    The stride between the rows' data to be stored in memory.
__DEFAULT_FN_ATTRS_TILE
static __inline__ void __tile_stored(void *base, __SIZE_TYPE__ stride,
                                     __tile1024i src) {
  _tile_stored_internal(src.row, src.col, base, stride, src.tile);
}
 
/// Zero the tile specified by "dst".
///
/// \headerfile <immintrin.h>
///
/// This intrinsic corresponds to the <c> TILEZERO </c> instruction.
///
/// \param dst
///    The destination tile to be zero. Max size is 1024 Bytes.
__DEFAULT_FN_ATTRS_TILE
static __inline__ void __tile_zero(__tile1024i *dst) {
  dst->tile = __builtin_ia32_tilezero_internal(dst->row, dst->col);
}
 
/// Compute dot-product of BF16 (16-bit) floating-point pairs in tiles src0 and
/// src1, accumulating the intermediate single-precision (32-bit) floating-point
/// elements with elements in "dst", and store the 32-bit result back to tile
/// "dst".
///
/// \headerfile <immintrin.h>
///
/// This intrinsic corresponds to the <c> TDPBF16PS </c> instruction.
///
/// \param dst
///    The destination tile. Max size is 1024 Bytes.
/// \param src0
///    The 1st source tile. Max size is 1024 Bytes.
/// \param src1
///    The 2nd source tile. Max size is 1024 Bytes.
__DEFAULT_FN_ATTRS_BF16
static __inline__ void __tile_dpbf16ps(__tile1024i *dst, __tile1024i src0,
                                       __tile1024i src1) {
  dst->tile = _tile_dpbf16ps_internal(src0.row, src1.col, src0.col, dst->tile,
                                      src0.tile, src1.tile);
}
 
/// Compute dot-product of FP16 (16-bit) floating-point pairs in tiles src0 and
/// src1, accumulating the intermediate single-precision (32-bit) floating-point
/// elements with elements in "dst", and store the 32-bit result back to tile
/// "dst".
///
/// \headerfile <immintrin.h>
///
/// This intrinsic corresponds to the <c> TDPFP16PS </c> instruction.
///
/// \param dst
///    The destination tile. Max size is 1024 Bytes.
/// \param src0
///    The 1st source tile. Max size is 1024 Bytes.
/// \param src1
///    The 2nd source tile. Max size is 1024 Bytes.
__DEFAULT_FN_ATTRS_FP16
static __inline__ void __tile_dpfp16ps(__tile1024i *dst, __tile1024i src0,
                                       __tile1024i src1) {
  dst->tile = _tile_dpfp16ps_internal(src0.row, src1.col, src0.col, dst->tile,
                                      src0.tile, src1.tile);
}
 
#undef __DEFAULT_FN_ATTRS_TILE
#undef __DEFAULT_FN_ATTRS_INT8
#undef __DEFAULT_FN_ATTRS_BF16
#undef __DEFAULT_FN_ATTRS_FP16
 
#endif /* __x86_64__ */
#endif /* __AMXINTRIN_H */