/* { dg-final { check-function-bodies "**" "" "-DCHECK_ASM" } } */ #include "test_sme2_acle.h" /* ** mla_0_z0_z0: ** mov (w8|w9|w10|w11), #?0 ** bfmlal za\.s\[\1, 0:1, vgx2\], {z0\.h - z1\.h}, {z0\.h - z1\.h} ** ret */ TEST_ZA_XN (mla_0_z0_z0, svbfloat16x2_t, svmla_za32_bf16_vg2x2 (0, z0, z0), svmla_za32_vg2x2 (0, z0, z0)) /* ** mla_w0_z0_z0: ** mov (w8|w9|w10|w11), w0 ** bfmlal za\.s\[\1, 0:1, vgx2\], {z0\.h - z1\.h}, {z0\.h - z1\.h} ** ret */ TEST_ZA_XN (mla_w0_z0_z0, svbfloat16x2_t, svmla_za32_bf16_vg2x2 (w0, z0, z0), svmla_za32_vg2x2 (w0, z0, z0)) /* ** mla_w8_z0_z4: ** bfmlal za\.s\[w8, 0:1, vgx2\], {z0\.h - z1\.h}, {z4\.h - z5\.h} ** ret */ TEST_ZA_XN (mla_w8_z0_z4, svbfloat16x2_t, svmla_za32_bf16_vg2x2 (w8, z0, z4), svmla_za32_vg2x2 (w8, z0, z4)) /* ** mla_w8_z4_z18: ** bfmlal za\.s\[w8, 0:1, vgx2\], {z4\.h - z5\.h}, {z18\.h - z19\.h} ** ret */ TEST_ZA_XN (mla_w8_z4_z18, svbfloat16x2_t, svmla_za32_bf16_vg2x2 (w8, z4, z18), svmla_za32_vg2x2 (w8, z4, z18)) /* Leave the assembler to check for correctness for misaligned registers. */ /* ** mla_w8_z0_z23: ** ... ** bfmlal za\.s\[w8, 0:1, vgx2\], {z0\.h - z1\.h}, [^\n]+ ** ret */ TEST_ZA_XN (mla_w8_z0_z23, svbfloat16x2_t, svmla_za32_bf16_vg2x2 (w8, z0, z23), svmla_za32_vg2x2 (w8, z0, z23)) /* ** mla_w8_z23_z0: ** ... ** bfmlal za\.s\[w8, 0:1, vgx2\], [^\n]+, {z0\.h - z1\.h} ** ret */ TEST_ZA_XN (mla_w8_z23_z0, svbfloat16x2_t, svmla_za32_bf16_vg2x2 (w8, z23, z0), svmla_za32_vg2x2 (w8, z23, z0)) /* ** mla_w8_z18_z28: ** bfmlal za\.s\[w8, 0:1, vgx2\], {z18\.h - z19\.h}, {z28\.h - z29\.h} ** ret */ TEST_ZA_XN (mla_w8_z18_z28, svbfloat16x2_t, svmla_za32_bf16_vg2x2 (w8, z18, z28), svmla_za32_vg2x2 (w8, z18, z28)) /* ** mla_w8_z28_z4: ** bfmlal za\.s\[w8, 0:1, vgx2\], {z28\.h - z29\.h}, {z4\.h - z5\.h} ** ret */ TEST_ZA_XN (mla_w8_z28_z4, svbfloat16x2_t, svmla_za32_bf16_vg2x2 (w8, z28, z4), svmla_za32_vg2x2 (w8, z28, z4)) /* ** mla_w8p1_z4_z0: ** add (w8|w9|w10|w11), w8, #?1 ** bfmlal za\.s\[\1, 0:1, vgx2\], {z4\.h - z5\.h}, {z0\.h - z1\.h} ** ret */ TEST_ZA_XN (mla_w8p1_z4_z0, svbfloat16x2_t, svmla_za32_bf16_vg2x2 (w8 + 1, z4, z0), svmla_za32_vg2x2 (w8 + 1, z4, z0)) /* ** mla_w8p2_z4_z0: ** bfmlal za\.s\[w8, 2:3, vgx2\], {z4\.h - z5\.h}, {z0\.h - z1\.h} ** ret */ TEST_ZA_XN (mla_w8p2_z4_z0, svbfloat16x2_t, svmla_za32_bf16_vg2x2 (w8 + 2, z4, z0), svmla_za32_vg2x2 (w8 + 2, z4, z0)) /* ** mla_w8p6_z4_z0: ** bfmlal za\.s\[w8, 6:7, vgx2\], {z4\.h - z5\.h}, {z0\.h - z1\.h} ** ret */ TEST_ZA_XN (mla_w8p6_z4_z0, svbfloat16x2_t, svmla_za32_bf16_vg2x2 (w8 + 6, z4, z0), svmla_za32_vg2x2 (w8 + 6, z4, z0)) /* ** mla_w8p7_z4_z0: ** add (w8|w9|w10|w11), w8, #?7 ** bfmlal za\.s\[\1, 0:1, vgx2\], {z4\.h - z5\.h}, {z0\.h - z1\.h} ** ret */ TEST_ZA_XN (mla_w8p7_z4_z0, svbfloat16x2_t, svmla_za32_bf16_vg2x2 (w8 + 7, z4, z0), svmla_za32_vg2x2 (w8 + 7, z4, z0)) /* ** mla_w8p8_z4_z4: ** add (w8|w9|w10|w11), w8, #?8 ** bfmlal za\.s\[\1, 0:1, vgx2\], {z4\.h - z5\.h}, {z4\.h - z5\.h} ** ret */ TEST_ZA_XN (mla_w8p8_z4_z4, svbfloat16x2_t, svmla_za32_bf16_vg2x2 (w8 + 8, z4, z4), svmla_za32_vg2x2 (w8 + 8, z4, z4)) /* ** mla_w8m1_z4_z0: ** sub (w8|w9|w10|w11), w8, #?1 ** bfmlal za\.s\[\1, 0:1, vgx2\], {z4\.h - z5\.h}, {z0\.h - z1\.h} ** ret */ TEST_ZA_XN (mla_w8m1_z4_z0, svbfloat16x2_t, svmla_za32_bf16_vg2x2 (w8 - 1, z4, z0), svmla_za32_vg2x2 (w8 - 1, z4, z0)) /* ** mla_single_0_z1_z0: ** mov (w8|w9|w10|w11), #?0 ** bfmlal za\.s\[\1, 0:1, vgx2\], {z1\.h - z2\.h}, z0\.h ** ret */ TEST_ZA_SINGLE (mla_single_0_z1_z0, svbfloat16x2_t, svbfloat16_t, svmla_single_za32_bf16_vg2x2 (0, z1, z0), svmla_za32_vg2x2 (0, z1, z0)) /* ** mla_single_w0_z1_z0: ** mov (w8|w9|w10|w11), w0 ** bfmlal za\.s\[\1, 0:1, vgx2\], {z1\.h - z2\.h}, z0\.h ** ret */ TEST_ZA_SINGLE (mla_single_w0_z1_z0, svbfloat16x2_t, svbfloat16_t, svmla_single_za32_bf16_vg2x2 (w0, z1, z0), svmla_za32_vg2x2 (w0, z1, z0)) /* ** mla_single_w8_z1_z0: ** bfmlal za\.s\[w8, 0:1, vgx2\], {z1\.h - z2\.h}, z0\.h ** ret */ TEST_ZA_SINGLE (mla_single_w8_z1_z0, svbfloat16x2_t, svbfloat16_t, svmla_single_za32_bf16_vg2x2 (w8, z1, z0), svmla_za32_vg2x2 (w8, z1, z0)) /* ** mla_single_w8p1_z1_z0: ** add (w8|w9|w10|w11), w8, #?1 ** bfmlal za\.s\[\1, 0:1, vgx2\], {z1\.h - z2\.h}, z0\.h ** ret */ TEST_ZA_SINGLE (mla_single_w8p1_z1_z0, svbfloat16x2_t, svbfloat16_t, svmla_single_za32_bf16_vg2x2 (w8 + 1, z1, z0), svmla_za32_vg2x2 (w8 + 1, z1, z0)) /* ** mla_single_w8p4_z20_z0: ** bfmlal za\.s\[w8, 4:5, vgx2\], {z20\.h - z21\.h}, z0\.h ** ret */ TEST_ZA_SINGLE (mla_single_w8p4_z20_z0, svbfloat16x2_t, svbfloat16_t, svmla_single_za32_bf16_vg2x2 (w8 + 4, z20, z0), svmla_za32_vg2x2 (w8 + 4, z20, z0)) /* ** mla_single_w8p6_z27_z0: ** bfmlal za\.s\[w8, 6:7, vgx2\], {z27\.h - z28\.h}, z0\.h ** ret */ TEST_ZA_SINGLE (mla_single_w8p6_z27_z0, svbfloat16x2_t, svbfloat16_t, svmla_single_za32_bf16_vg2x2 (w8 + 6, z27, z0), svmla_za32_vg2x2 (w8 + 6, z27, z0)) /* ** mla_single_w8p7_z1_z0: ** add (w8|w9|w10|w11), w8, #?7 ** bfmlal za\.s\[\1, 0:1, vgx2\], {z1\.h - z2\.h}, z0\.h ** ret */ TEST_ZA_SINGLE (mla_single_w8p7_z1_z0, svbfloat16x2_t, svbfloat16_t, svmla_single_za32_bf16_vg2x2 (w8 + 7, z1, z0), svmla_za32_vg2x2 (w8 + 7, z1, z0)) /* ** mla_single_w8p8_z1_z0: ** add (w8|w9|w10|w11), w8, #?8 ** bfmlal za\.s\[\1, 0:1, vgx2\], {z1\.h - z2\.h}, z0\.h ** ret */ TEST_ZA_SINGLE (mla_single_w8p8_z1_z0, svbfloat16x2_t, svbfloat16_t, svmla_single_za32_bf16_vg2x2 (w8 + 8, z1, z0), svmla_za32_vg2x2 (w8 + 8, z1, z0)) /* ** mla_single_w0m1_z1_z0: ** sub (w8|w9|w10|w11), w0, #?1 ** bfmlal za\.s\[\1, 0:1, vgx2\], {z1\.h - z2\.h}, z0\.h ** ret */ TEST_ZA_SINGLE (mla_single_w0m1_z1_z0, svbfloat16x2_t, svbfloat16_t, svmla_single_za32_bf16_vg2x2 (w0 - 1, z1, z0), svmla_za32_vg2x2 (w0 - 1, z1, z0)) /* ** mla_single_w8_z0_z15: ** str d15, \[sp, #?-16\]! ** bfmlal za\.s\[w8, 0:1, vgx2\], {z0\.h - z1\.h}, z15\.h ** ldr d15, \[sp\], #?16 ** ret */ TEST_ZA_SINGLE_Z15 (mla_single_w8_z0_z15, svbfloat16x2_t, svbfloat16_t, svmla_single_za32_bf16_vg2x2 (w8, z0, z15), svmla_za32_vg2x2 (w8, z0, z15)) /* ** mla_single_w8_z20_z16: ** mov (z[0-7]).d, z16.d ** bfmlal za\.s\[w8, 0:1, vgx2\], {z20\.h - z21\.h}, \1\.h ** ret */ TEST_ZA_SINGLE (mla_single_w8_z20_z16, svbfloat16x2_t, svbfloat16_t, svmla_single_za32_bf16_vg2x2 (w8, z20, z16), svmla_za32_vg2x2 (w8, z20, z16))