pq_crypto 0.6.2 → 0.6.4

data/ext/pqcrypto/vendor/mlkem-native/mlkem/src/native/ppc64le/src/poly_tomont_ppc_asm.S

@@ -0,0 +1,294 @@
+/*
+ * Copyright (c) The mlkem-native project authors
+ * Copyright (c) IBM Corp. 2025, 2026
+ * SPDX-License-Identifier: Apache-2.0 OR ISC OR MIT
+ *
+ * Written by Danny Tsen <dtsen@us.ibm.com>
+ */
+
+/*
+ * Poly_tomont: Inplace conversion of all coefficients of a polynomial
+ *              from normal domain to Montgomery domain
+ *
+ * Arguments:*r: pointer to input/output polynomial
+ */
+
+#include "../../../common.h"
+#if defined(MLK_ARITH_BACKEND_PPC64LE_DEFAULT) && \
+    !defined(MLK_CONFIG_MULTILEVEL_NO_SHARED) && defined(__POWER8_VECTOR__)
+
+/*
+ * WARNING: This file is auto-derived from the mlkem-native source file
+ *   dev/ppc64le/src/poly_tomont_ppc_asm.S using scripts/simpasm. Do not modify it directly.
+ */
+
+.text
+.balign 4
+.global MLK_ASM_NAMESPACE(poly_tomont_ppc_asm)
+MLK_ASM_FN_SYMBOL(poly_tomont_ppc_asm)
+
+        stdu 1, -320(1)
+        mflr 0
+        li 6, 128
+        li 7, 144
+        li 8, 160
+        li 9, 176
+        li 10, 192
+        li 11, 208
+        li 12, 224
+        stxvd2x 52, 6, 1
+        stxvd2x 53, 7, 1
+        stxvd2x 54, 8, 1
+        stxvd2x 55, 9, 1
+        stxvd2x 56, 10, 1
+        stxvd2x 57, 11, 1
+        stxvd2x 58, 12, 1
+        li 6, 240
+        li 7, 256
+        li 8, 272
+        li 9, 288
+        stxvd2x 59, 6, 1
+        stxvd2x 60, 7, 1
+        stxvd2x 61, 8, 1
+        stxvd2x 62, 9, 1
+        li 6, 0
+        li 7, 80
+        li 8, 96
+        lxvd2x 37, 6, 4
+        lxvd2x 32, 7, 4
+        lxvd2x 34, 8, 4
+        vxor 3, 3, 3
+        li 4, -128
+        li 5, -112
+        li 6, -96
+        li 7, -80
+        li 8, -64
+        li 9, -48
+        li 10, -32
+        li 11, -16
+        lxvd2x 45, 0, 3
+        addi 3, 3, 16
+        lxvd2x 50, 0, 3
+        addi 3, 3, 16
+        lxvd2x 55, 0, 3
+        addi 3, 3, 16
+        lxvd2x 39, 0, 3
+        addi 3, 3, 16
+        vmhraddshs 14, 13, 2, 3
+        vmhraddshs 19, 18, 2, 3
+        vmhraddshs 24, 23, 2, 3
+        vmhraddshs 8, 7, 2, 3
+        vmladduhm 13, 13, 0, 3
+        vmladduhm 18, 18, 0, 3
+        vmladduhm 23, 23, 0, 3
+        vmladduhm 7, 7, 0, 3
+        vmladduhm 27, 14, 5, 13
+        vmladduhm 28, 19, 5, 18
+        vmladduhm 29, 24, 5, 23
+        vmladduhm 30, 8, 5, 7
+        lxvd2x 45, 0, 3
+        addi 3, 3, 16
+        lxvd2x 50, 0, 3
+        addi 3, 3, 16
+        lxvd2x 55, 0, 3
+        addi 3, 3, 16
+        lxvd2x 39, 0, 3
+        addi 3, 3, 16
+        vmhraddshs 14, 13, 2, 3
+        vmhraddshs 19, 18, 2, 3
+        vmhraddshs 24, 23, 2, 3
+        vmhraddshs 8, 7, 2, 3
+        vmladduhm 13, 13, 0, 3
+        vmladduhm 18, 18, 0, 3
+        vmladduhm 23, 23, 0, 3
+        vmladduhm 7, 7, 0, 3
+        vmladduhm 13, 14, 5, 13
+        vmladduhm 18, 19, 5, 18
+        vmladduhm 23, 24, 5, 23
+        vmladduhm 7, 8, 5, 7
+        stxvd2x 59, 4, 3
+        stxvd2x 60, 5, 3
+        stxvd2x 61, 6, 3
+        stxvd2x 62, 7, 3
+        stxvd2x 45, 8, 3
+        stxvd2x 50, 9, 3
+        stxvd2x 55, 10, 3
+        stxvd2x 39, 11, 3
+        lxvd2x 45, 0, 3
+        addi 3, 3, 16
+        lxvd2x 50, 0, 3
+        addi 3, 3, 16
+        lxvd2x 55, 0, 3
+        addi 3, 3, 16
+        lxvd2x 39, 0, 3
+        addi 3, 3, 16
+        vmhraddshs 14, 13, 2, 3
+        vmhraddshs 19, 18, 2, 3
+        vmhraddshs 24, 23, 2, 3
+        vmhraddshs 8, 7, 2, 3
+        vmladduhm 13, 13, 0, 3
+        vmladduhm 18, 18, 0, 3
+        vmladduhm 23, 23, 0, 3
+        vmladduhm 7, 7, 0, 3
+        vmladduhm 27, 14, 5, 13
+        vmladduhm 28, 19, 5, 18
+        vmladduhm 29, 24, 5, 23
+        vmladduhm 30, 8, 5, 7
+        lxvd2x 45, 0, 3
+        addi 3, 3, 16
+        lxvd2x 50, 0, 3
+        addi 3, 3, 16
+        lxvd2x 55, 0, 3
+        addi 3, 3, 16
+        lxvd2x 39, 0, 3
+        addi 3, 3, 16
+        vmhraddshs 14, 13, 2, 3
+        vmhraddshs 19, 18, 2, 3
+        vmhraddshs 24, 23, 2, 3
+        vmhraddshs 8, 7, 2, 3
+        vmladduhm 13, 13, 0, 3
+        vmladduhm 18, 18, 0, 3
+        vmladduhm 23, 23, 0, 3
+        vmladduhm 7, 7, 0, 3
+        vmladduhm 13, 14, 5, 13
+        vmladduhm 18, 19, 5, 18
+        vmladduhm 23, 24, 5, 23
+        vmladduhm 7, 8, 5, 7
+        stxvd2x 59, 4, 3
+        stxvd2x 60, 5, 3
+        stxvd2x 61, 6, 3
+        stxvd2x 62, 7, 3
+        stxvd2x 45, 8, 3
+        stxvd2x 50, 9, 3
+        stxvd2x 55, 10, 3
+        stxvd2x 39, 11, 3
+        lxvd2x 45, 0, 3
+        addi 3, 3, 16
+        lxvd2x 50, 0, 3
+        addi 3, 3, 16
+        lxvd2x 55, 0, 3
+        addi 3, 3, 16
+        lxvd2x 39, 0, 3
+        addi 3, 3, 16
+        vmhraddshs 14, 13, 2, 3
+        vmhraddshs 19, 18, 2, 3
+        vmhraddshs 24, 23, 2, 3
+        vmhraddshs 8, 7, 2, 3
+        vmladduhm 13, 13, 0, 3
+        vmladduhm 18, 18, 0, 3
+        vmladduhm 23, 23, 0, 3
+        vmladduhm 7, 7, 0, 3
+        vmladduhm 27, 14, 5, 13
+        vmladduhm 28, 19, 5, 18
+        vmladduhm 29, 24, 5, 23
+        vmladduhm 30, 8, 5, 7
+        lxvd2x 45, 0, 3
+        addi 3, 3, 16
+        lxvd2x 50, 0, 3
+        addi 3, 3, 16
+        lxvd2x 55, 0, 3
+        addi 3, 3, 16
+        lxvd2x 39, 0, 3
+        addi 3, 3, 16
+        vmhraddshs 14, 13, 2, 3
+        vmhraddshs 19, 18, 2, 3
+        vmhraddshs 24, 23, 2, 3
+        vmhraddshs 8, 7, 2, 3
+        vmladduhm 13, 13, 0, 3
+        vmladduhm 18, 18, 0, 3
+        vmladduhm 23, 23, 0, 3
+        vmladduhm 7, 7, 0, 3
+        vmladduhm 13, 14, 5, 13
+        vmladduhm 18, 19, 5, 18
+        vmladduhm 23, 24, 5, 23
+        vmladduhm 7, 8, 5, 7
+        stxvd2x 59, 4, 3
+        stxvd2x 60, 5, 3
+        stxvd2x 61, 6, 3
+        stxvd2x 62, 7, 3
+        stxvd2x 45, 8, 3
+        stxvd2x 50, 9, 3
+        stxvd2x 55, 10, 3
+        stxvd2x 39, 11, 3
+        lxvd2x 45, 0, 3
+        addi 3, 3, 16
+        lxvd2x 50, 0, 3
+        addi 3, 3, 16
+        lxvd2x 55, 0, 3
+        addi 3, 3, 16
+        lxvd2x 39, 0, 3
+        addi 3, 3, 16
+        vmhraddshs 14, 13, 2, 3
+        vmhraddshs 19, 18, 2, 3
+        vmhraddshs 24, 23, 2, 3
+        vmhraddshs 8, 7, 2, 3
+        vmladduhm 13, 13, 0, 3
+        vmladduhm 18, 18, 0, 3
+        vmladduhm 23, 23, 0, 3
+        vmladduhm 7, 7, 0, 3
+        vmladduhm 27, 14, 5, 13
+        vmladduhm 28, 19, 5, 18
+        vmladduhm 29, 24, 5, 23
+        vmladduhm 30, 8, 5, 7
+        lxvd2x 45, 0, 3
+        addi 3, 3, 16
+        lxvd2x 50, 0, 3
+        addi 3, 3, 16
+        lxvd2x 55, 0, 3
+        addi 3, 3, 16
+        lxvd2x 39, 0, 3
+        addi 3, 3, 16
+        vmhraddshs 14, 13, 2, 3
+        vmhraddshs 19, 18, 2, 3
+        vmhraddshs 24, 23, 2, 3
+        vmhraddshs 8, 7, 2, 3
+        vmladduhm 13, 13, 0, 3
+        vmladduhm 18, 18, 0, 3
+        vmladduhm 23, 23, 0, 3
+        vmladduhm 7, 7, 0, 3
+        vmladduhm 13, 14, 5, 13
+        vmladduhm 18, 19, 5, 18
+        vmladduhm 23, 24, 5, 23
+        vmladduhm 7, 8, 5, 7
+        stxvd2x 59, 4, 3
+        stxvd2x 60, 5, 3
+        stxvd2x 61, 6, 3
+        stxvd2x 62, 7, 3
+        stxvd2x 45, 8, 3
+        stxvd2x 50, 9, 3
+        stxvd2x 55, 10, 3
+        stxvd2x 39, 11, 3
+        li 6, 128
+        li 7, 144
+        li 8, 160
+        li 9, 176
+        li 10, 192
+        li 11, 208
+        li 12, 224
+        lxvd2x 52, 6, 1
+        lxvd2x 53, 7, 1
+        lxvd2x 54, 8, 1
+        lxvd2x 55, 9, 1
+        lxvd2x 56, 10, 1
+        lxvd2x 57, 11, 1
+        lxvd2x 58, 12, 1
+        li 6, 240
+        li 7, 256
+        li 8, 272
+        li 9, 288
+        lxvd2x 59, 6, 1
+        lxvd2x 60, 7, 1
+        lxvd2x 61, 8, 1
+        lxvd2x 62, 9, 1
+        mtlr 0
+        addi 1, 1, 320
+        blr
+
+MLK_ASM_FN_SIZE(poly_tomont_ppc_asm)
+
+#endif /* MLK_ARITH_BACKEND_PPC64LE_DEFAULT && !MLK_CONFIG_MULTILEVEL_NO_SHARED \
+          && __POWER8_VECTOR__ */
+
+#if defined(__ELF__)
+.section .note.GNU-stack,"",%progbits
+#endif