[clang] [llvm] [X86][AMX] Support AMX-TRANSPOSE, part 2 (PR #115660)

[Feng Zou via cfe-commits]

================
@@ -0,0 +1,94 @@
+/*===----- amxbf16transposeintrin.h - AMX-BF16 and AMX-TRANSPOSE ------------===
+ *
+ * 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 <amxbf16transposeintrin.h> directly; use <immintrin.h> instead."
+#endif /* __IMMINTRIN_H */
+
+#ifndef __AMX_BF16TRANSPOSEINTRIN_H
+#define __AMX_BF16TRANSPOSEINTRIN_H
+#ifdef __x86_64__
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS                                                     \
+  __attribute__((__always_inline__, __nodebug__,                               \
+                 __target__("amx-bf16,amx-transpose")))
+
+/// Compute transpose and dot-product of BF16 (16-bit) floating-point pairs in
+///    tiles \a a and \a b, accumulating the intermediate single-precision
+///    (32-bit) floating-point elements with elements in \a dst, and store the
+///    32-bit result back to tile \a dst.
+///
+/// \headerfile <immintrin.h>
+///
+/// \code
+/// void _tile_tdpbf16ps (__tile dst, __tile a, __tile b)
+/// \endcode
+///
+/// \code{.operation}
+/// FOR m := 0 TO dst.rows - 1
+///	tmp := dst.row[m]
+///	FOR k := 0 TO (a.colsb / 4) - 1
+///		FOR n := 0 TO (dst.colsb / 4) - 1
+///			tmp.bf32[n] += FP32(a.row[m].bf16[2*k+0]) *
+///					FP32(b.row[k].bf16[2*n+0])
+///			tmp.bf32[n] += FP32(a.row[m].bf16[2*k+1]) *
+///					FP32(b.row[k].bf16[2*n+1])
+///		ENDFOR
+///	ENDFOR
+///	write_row_and_zero(dst, m, tmp, dst.colsb)
+/// ENDFOR
+/// zero_upper_rows(dst, dst.rows)
+/// zero_tileconfig_start()
+/// \endcode
+///
+/// This intrinsic corresponds to the \c TTDPBF16PS instruction.
+///
+/// \param dst
+///    The destination tile. Max size is 1024 Bytes.
+/// \param a
+///    The 1st source tile. Max size is 1024 Bytes.
+/// \param b
+///    The 2nd source tile. Max size is 1024 Bytes.
+#define _tile_tdpbf16ps(dst, a, b) __builtin_ia32_ttdpbf16ps(dst, a, b)
----------------
fzou1 wrote:

__builtin_ia32_ttdpbf16ps(dst, a, b) -> __builtin_ia32_ttdpbf16ps((dst), (a), (b))

https://github.com/llvm/llvm-project/pull/115660