/* { dg-do compile } */ /* { dg-final { check-function-bodies "**" "" "-DCHECK_ASM" } } */ #include "test_sve_acle.h" #pragma GCC target "+sve2+faminmax" #if STREAMING_COMPATIBLE #pragma GCC target "+sme2" #endif /* ** amin_f16_m_tied1: ** famin z0\.h, p0/m, z0\.h, z1\.h ** ret */ TEST_UNIFORM_Z (amin_f16_m_tied1, svfloat16_t, z0 = svamin_f16_m (p0, z0, z1), z0 = svamin_m (p0, z0, z1)) /* ** amin_f16_m_tied2: ** mov (z[0-9]+)\.d, z0\.d ** ... ** famin z0\.h, p0/m, z0\.h, \1\.h ** ret */ TEST_UNIFORM_Z (amin_f16_m_tied2, svfloat16_t, z0 = svamin_f16_m (p0, z1, z0), z0 = svamin_m (p0, z1, z0)) /* ** amin_f16_m_untied: ** ... ** famin z0\.h, p0/m, z0\.h, z2\.h ** ret */ TEST_UNIFORM_Z (amin_f16_m_untied, svfloat16_t, z0 = svamin_f16_m (p0, z1, z2), z0 = svamin_m (p0, z1, z2)) /* ** amin_h4_f16_m_tied1: ** mov (z[0-9]+\.h), h4 ** famin z0\.h, p0/m, z0\.h, \1 ** ret */ TEST_UNIFORM_ZD (amin_h4_f16_m_tied1, svfloat16_t, __fp16, z0 = svamin_n_f16_m (p0, z0, d4), z0 = svamin_m (p0, z0, d4)) /* ** amin_h4_f16_m_untied: ** mov (z[0-9]+\.h), h4 ** ... ** famin z0\.h, p0/m, z0\.h, \1 ** ret */ TEST_UNIFORM_ZD (amin_h4_f16_m_untied, svfloat16_t, __fp16, z0 = svamin_n_f16_m (p0, z1, d4), z0 = svamin_m (p0, z1, d4)) /* ** amin_0_f16_m_tied1: ** ... ** famin z0\.h, p0/m, z0\.h, z[0-9]+\.h ** ret */ TEST_UNIFORM_Z (amin_0_f16_m_tied1, svfloat16_t, z0 = svamin_n_f16_m (p0, z0, 0), z0 = svamin_m (p0, z0, 0)) /* ** amin_0_f16_m_untied: ** ... ** famin z0\.h, p0/m, z0\.h, z[0-9]+\.h ** ret */ TEST_UNIFORM_Z (amin_0_f16_m_untied, svfloat16_t, z0 = svamin_n_f16_m (p0, z1, 0), z0 = svamin_m (p0, z1, 0)) /* ** amin_1_f16_m_tied1: ** ... ** famin z0\.h, p0/m, z0\.h, z[0-9]+\.h ** ret */ TEST_UNIFORM_Z (amin_1_f16_m_tied1, svfloat16_t, z0 = svamin_n_f16_m (p0, z0, 1), z0 = svamin_m (p0, z0, 1)) /* ** amin_1_f16_m_untied: ** ... ** famin z0\.h, p0/m, z0\.h, z[0-9]+\.h ** ret */ TEST_UNIFORM_Z (amin_1_f16_m_untied, svfloat16_t, z0 = svamin_n_f16_m (p0, z1, 1), z0 = svamin_m (p0, z1, 1)) /* ** amin_2_f16_m: ** ... ** famin z0\.h, p0/m, z0\.h, z[0-9]+\.h ** ret */ TEST_UNIFORM_Z (amin_2_f16_m, svfloat16_t, z0 = svamin_n_f16_m (p0, z0, 2), z0 = svamin_m (p0, z0, 2)) /* ** amin_f16_z_tied1: ** ... ** famin z0\.h, p0/m, z0\.h, z1\.h ** ret */ TEST_UNIFORM_Z (amin_f16_z_tied1, svfloat16_t, z0 = svamin_f16_z (p0, z0, z1), z0 = svamin_z (p0, z0, z1)) /* ** amin_f16_z_tied2: ** ... ** famin z0\.h, p0/m, z0\.h, z1\.h ** ret */ TEST_UNIFORM_Z (amin_f16_z_tied2, svfloat16_t, z0 = svamin_f16_z (p0, z1, z0), z0 = svamin_z (p0, z1, z0)) /* ** amin_f16_z_untied: ** ( ** ... ** famin z0\.h, p0/m, z0\.h, z2\.h ** | ** ... ** famin z0\.h, p0/m, z0\.h, z1\.h ** ) ** ret */ TEST_UNIFORM_Z (amin_f16_z_untied, svfloat16_t, z0 = svamin_f16_z (p0, z1, z2), z0 = svamin_z (p0, z1, z2)) /* ** amin_h4_f16_z_tied1: ** mov (z[0-9]+\.h), h4 ** ... ** famin z0\.h, p0/m, z0\.h, \1 ** ret */ TEST_UNIFORM_ZD (amin_h4_f16_z_tied1, svfloat16_t, __fp16, z0 = svamin_n_f16_z (p0, z0, d4), z0 = svamin_z (p0, z0, d4)) /* ** amin_h4_f16_z_untied: ** mov (z[0-9]+\.h), h4 ** ( ** ... ** famin z0\.h, p0/m, z0\.h, \1 ** | ** ... ** famin z0\.h, p0/m, z0\.h, z1\.h ** ) ** ret */ TEST_UNIFORM_ZD (amin_h4_f16_z_untied, svfloat16_t, __fp16, z0 = svamin_n_f16_z (p0, z1, d4), z0 = svamin_z (p0, z1, d4)) /* ** amin_0_f16_z_tied1: ** ... ** famin z0\.h, p0/m, z0\.h, z[0-9]+\.h ** ret */ TEST_UNIFORM_Z (amin_0_f16_z_tied1, svfloat16_t, z0 = svamin_n_f16_z (p0, z0, 0), z0 = svamin_z (p0, z0, 0)) /* ** amin_0_f16_z_untied: ** ... ** famin z0\.h, p0/m, z0\.h, z[0-9]+\.h ** ret */ TEST_UNIFORM_Z (amin_0_f16_z_untied, svfloat16_t, z0 = svamin_n_f16_z (p0, z1, 0), z0 = svamin_z (p0, z1, 0)) /* ** amin_1_f16_z_tied1: ** ... ** famin z0\.h, p0/m, z0\.h, z[0-9]+\.h ** ret */ TEST_UNIFORM_Z (amin_1_f16_z_tied1, svfloat16_t, z0 = svamin_n_f16_z (p0, z0, 1), z0 = svamin_z (p0, z0, 1)) /* ** amin_1_f16_z_untied: ** ... ** famin z0\.h, p0/m, z0\.h, z[0-9]+\.h ** ret */ TEST_UNIFORM_Z (amin_1_f16_z_untied, svfloat16_t, z0 = svamin_n_f16_z (p0, z1, 1), z0 = svamin_z (p0, z1, 1)) /* ** amin_2_f16_z: ** ... ** famin z0\.h, p0/m, z0\.h, z[0-9]+\.h ** ret */ TEST_UNIFORM_Z (amin_2_f16_z, svfloat16_t, z0 = svamin_n_f16_z (p0, z0, 2), z0 = svamin_z (p0, z0, 2)) /* ** amin_f16_x_tied1: ** famin z0\.h, p0/m, z0\.h, z1\.h ** ret */ TEST_UNIFORM_Z (amin_f16_x_tied1, svfloat16_t, z0 = svamin_f16_x (p0, z0, z1), z0 = svamin_x (p0, z0, z1)) /* ** amin_f16_x_tied2: ** famin z0\.h, p0/m, z0\.h, z1\.h ** ret */ TEST_UNIFORM_Z (amin_f16_x_tied2, svfloat16_t, z0 = svamin_f16_x (p0, z1, z0), z0 = svamin_x (p0, z1, z0)) /* ** amin_f16_x_untied: ** ( ** ... ** famin z0\.h, p0/m, z0\.h, z2\.h ** | ** ... ** famin z0\.h, p0/m, z0\.h, z1\.h ** ) ** ret */ TEST_UNIFORM_Z (amin_f16_x_untied, svfloat16_t, z0 = svamin_f16_x (p0, z1, z2), z0 = svamin_x (p0, z1, z2)) /* ** amin_h4_f16_x_tied1: ** mov (z[0-9]+\.h), h4 ** famin z0\.h, p0/m, z0\.h, \1 ** ret */ TEST_UNIFORM_ZD (amin_h4_f16_x_tied1, svfloat16_t, __fp16, z0 = svamin_n_f16_x (p0, z0, d4), z0 = svamin_x (p0, z0, d4)) /* ** amin_h4_f16_x_untied: ** mov z0\.h, h4 ** famin z0\.h, p0/m, z0\.h, z1\.h ** ret */ TEST_UNIFORM_ZD (amin_h4_f16_x_untied, svfloat16_t, __fp16, z0 = svamin_n_f16_x (p0, z1, d4), z0 = svamin_x (p0, z1, d4)) /* ** amin_0_f16_x_tied1: ** ... ** famin z0\.h, p0/m, z0\.h, z[0-9]+\.h ** ret */ TEST_UNIFORM_Z (amin_0_f16_x_tied1, svfloat16_t, z0 = svamin_n_f16_x (p0, z0, 0), z0 = svamin_x (p0, z0, 0)) /* ** amin_0_f16_x_untied: ** ... ** famin z0\.h, p0/m, z0\.h, z[0-9]+\.h ** ret */ TEST_UNIFORM_Z (amin_0_f16_x_untied, svfloat16_t, z0 = svamin_n_f16_x (p0, z1, 0), z0 = svamin_x (p0, z1, 0)) /* ** amin_1_f16_x_tied1: ** ... ** famin z0\.h, p0/m, z0\.h, z[0-9]+\.h ** ret */ TEST_UNIFORM_Z (amin_1_f16_x_tied1, svfloat16_t, z0 = svamin_n_f16_x (p0, z0, 1), z0 = svamin_x (p0, z0, 1)) /* ** amin_1_f16_x_untied: ** ... ** famin z0\.h, p0/m, z0\.h, z[0-9]+\.h ** ret */ TEST_UNIFORM_Z (amin_1_f16_x_untied, svfloat16_t, z0 = svamin_n_f16_x (p0, z1, 1), z0 = svamin_x (p0, z1, 1)) /* ** amin_2_f16_x_tied1: ** ... ** famin z0\.h, p0/m, z0\.h, z[0-9]+\.h ** ret */ TEST_UNIFORM_Z (amin_2_f16_x_tied1, svfloat16_t, z0 = svamin_n_f16_x (p0, z0, 2), z0 = svamin_x (p0, z0, 2)) /* ** amin_2_f16_x_untied: ** ... ** famin z0\.h, p0/m, z0\.h, z[0-9]+\.h ** ret */ TEST_UNIFORM_Z (amin_2_f16_x_untied, svfloat16_t, z0 = svamin_n_f16_x (p0, z1, 2), z0 = svamin_x (p0, z1, 2)) /* ** ptrue_amin_f16_x_tied1: ** ... ** ptrue p[0-9]+\.b[^\n]* ** ... ** ret */ TEST_UNIFORM_Z (ptrue_amin_f16_x_tied1, svfloat16_t, z0 = svamin_f16_x (svptrue_b16 (), z0, z1), z0 = svamin_x (svptrue_b16 (), z0, z1)) /* ** ptrue_amin_f16_x_tied2: ** ... ** ptrue p[0-9]+\.b[^\n]* ** ... ** ret */ TEST_UNIFORM_Z (ptrue_amin_f16_x_tied2, svfloat16_t, z0 = svamin_f16_x (svptrue_b16 (), z1, z0), z0 = svamin_x (svptrue_b16 (), z1, z0)) /* ** ptrue_amin_f16_x_untied: ** ... ** ptrue p[0-9]+\.b[^\n]* ** ... ** ret */ TEST_UNIFORM_Z (ptrue_amin_f16_x_untied, svfloat16_t, z0 = svamin_f16_x (svptrue_b16 (), z1, z2), z0 = svamin_x (svptrue_b16 (), z1, z2)) /* ** ptrue_amin_0_f16_x_tied1: ** ... ** ptrue p[0-9]+\.b[^\n]* ** ... ** ret */ TEST_UNIFORM_Z (ptrue_amin_0_f16_x_tied1, svfloat16_t, z0 = svamin_n_f16_x (svptrue_b16 (), z0, 0), z0 = svamin_x (svptrue_b16 (), z0, 0)) /* ** ptrue_amin_0_f16_x_untied: ** ... ** ptrue p[0-9]+\.b[^\n]* ** ... ** ret */ TEST_UNIFORM_Z (ptrue_amin_0_f16_x_untied, svfloat16_t, z0 = svamin_n_f16_x (svptrue_b16 (), z1, 0), z0 = svamin_x (svptrue_b16 (), z1, 0)) /* ** ptrue_amin_1_f16_x_tied1: ** ... ** ptrue p[0-9]+\.b[^\n]* ** ... ** ret */ TEST_UNIFORM_Z (ptrue_amin_1_f16_x_tied1, svfloat16_t, z0 = svamin_n_f16_x (svptrue_b16 (), z0, 1), z0 = svamin_x (svptrue_b16 (), z0, 1)) /* ** ptrue_amin_1_f16_x_untied: ** ... ** ptrue p[0-9]+\.b[^\n]* ** ... ** ret */ TEST_UNIFORM_Z (ptrue_amin_1_f16_x_untied, svfloat16_t, z0 = svamin_n_f16_x (svptrue_b16 (), z1, 1), z0 = svamin_x (svptrue_b16 (), z1, 1)) /* ** ptrue_amin_2_f16_x_tied1: ** ... ** ptrue p[0-9]+\.b[^\n]* ** ... ** ret */ TEST_UNIFORM_Z (ptrue_amin_2_f16_x_tied1, svfloat16_t, z0 = svamin_n_f16_x (svptrue_b16 (), z0, 2), z0 = svamin_x (svptrue_b16 (), z0, 2)) /* ** ptrue_amin_2_f16_x_untied: ** ... ** ptrue p[0-9]+\.b[^\n]* ** ... ** ret */ TEST_UNIFORM_Z (ptrue_amin_2_f16_x_untied, svfloat16_t, z0 = svamin_n_f16_x (svptrue_b16 (), z1, 2), z0 = svamin_x (svptrue_b16 (), z1, 2))