x86: Replace all sse instructions with vex equivilent in avx+ files

Most of these don't really matter as there was no dirty upper state
but we should generally avoid stray sse when its not needed.

The one case that really matters is in svml_d_tanh4_core_avx2.S:

blendvps %xmm0, %xmm8, %xmm7

When there was a dirty upper state.

Tested on x86_64-linux

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_acos4_core_avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_d_acos4_core_avx2.S
index e19bddd2e2c2ff6212149947174f819a317aa44e..73025e8b0f1e1e3db35c5afb268ea63c161c14db 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_d_acos4_core_avx2.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_d_acos4_core_avx2.S
@@ -210,11 +210,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movsd   32(%rsp, %r14, 8), %xmm0
+       vmovsd  32(%rsp, %r14, 8), %xmm0
        call    acos@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movsd   %xmm0, 64(%rsp, %r14, 8)
+       vmovsd  %xmm0, 64(%rsp, %r14, 8)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_acos8_core_avx512.S b/sysdeps/x86_64/fpu/multiarch/svml_d_acos8_core_avx512.S
index f4c72c36182629dcb70aa3581e6a00c1e36ded14..b8cc6dd7764f95ed83869cafdfa228d1b54d1d4d 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_d_acos8_core_avx512.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_d_acos8_core_avx512.S
@@ -232,11 +232,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movsd   64(%rsp, %r14, 8), %xmm0
+       vmovsd  64(%rsp, %r14, 8), %xmm0
        call    acos@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movsd   %xmm0, 128(%rsp, %r14, 8)
+       vmovsd  %xmm0, 128(%rsp, %r14, 8)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_acosh4_core_avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_d_acosh4_core_avx2.S
index 5d0b23b72c3b285316f50795b3d515c61c15227f..126110cf17bbbf7813a6d661756c1ef37be043ea 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_d_acosh4_core_avx2.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_d_acosh4_core_avx2.S
@@ -372,11 +372,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movsd   32(%rsp, %r14, 8), %xmm0
+       vmovsd  32(%rsp, %r14, 8), %xmm0
        call    acosh@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movsd   %xmm0, 64(%rsp, %r14, 8)
+       vmovsd  %xmm0, 64(%rsp, %r14, 8)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_acosh8_core_avx512.S b/sysdeps/x86_64/fpu/multiarch/svml_d_acosh8_core_avx512.S
index b9a1131664f98cd7538b83ea3be1f0e526d752bf..db0ef3b9dd9d1de999734e8c5368fb0e7f28b764 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_d_acosh8_core_avx512.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_d_acosh8_core_avx512.S
@@ -317,11 +317,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movsd   64(%rsp, %r14, 8), %xmm0
+       vmovsd  64(%rsp, %r14, 8), %xmm0
        call    acosh@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movsd   %xmm0, 128(%rsp, %r14, 8)
+       vmovsd  %xmm0, 128(%rsp, %r14, 8)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_asin4_core_avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_d_asin4_core_avx2.S
index ba96089504ecbabc6b5a42c725d25d6b0bb50e5c..612a45da3096f3bbba679706fab0fd640d3017d3 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_d_asin4_core_avx2.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_d_asin4_core_avx2.S
@@ -202,11 +202,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movsd   32(%rsp, %r14, 8), %xmm0
+       vmovsd  32(%rsp, %r14, 8), %xmm0
        call    asin@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movsd   %xmm0, 64(%rsp, %r14, 8)
+       vmovsd  %xmm0, 64(%rsp, %r14, 8)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_asin8_core_avx512.S b/sysdeps/x86_64/fpu/multiarch/svml_d_asin8_core_avx512.S
index 0f5b773b045c92e5daccd426bf879090509a102b..e7b41ab232f2cf3a67ca8e69d2e74963c49ef719 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_d_asin8_core_avx512.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_d_asin8_core_avx512.S
@@ -224,11 +224,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movsd   64(%rsp, %r14, 8), %xmm0
+       vmovsd  64(%rsp, %r14, 8), %xmm0
        call    asin@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movsd   %xmm0, 128(%rsp, %r14, 8)
+       vmovsd  %xmm0, 128(%rsp, %r14, 8)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_asinh4_core_avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_d_asinh4_core_avx2.S
index 131b716c957aeee52547376eb1a2955d6ed80092..1fcbb245b715c05c7f55c346ade14ea4ae346600 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_d_asinh4_core_avx2.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_d_asinh4_core_avx2.S
@@ -429,11 +429,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movsd   32(%rsp, %r14, 8), %xmm0
+       vmovsd  32(%rsp, %r14, 8), %xmm0
        call    asinh@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movsd   %xmm0, 64(%rsp, %r14, 8)
+       vmovsd  %xmm0, 64(%rsp, %r14, 8)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_asinh8_core_avx512.S b/sysdeps/x86_64/fpu/multiarch/svml_d_asinh8_core_avx512.S
index 5bdc6859f0d47d0082236ba9ac8b0ef1ad7ced6d..8445fc8ba408ac9eecebcabdf7860f46b07fcdb4 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_d_asinh8_core_avx512.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_d_asinh8_core_avx512.S
@@ -343,11 +343,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movsd   64(%rsp, %r14, 8), %xmm0
+       vmovsd  64(%rsp, %r14, 8), %xmm0
        call    asinh@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movsd   %xmm0, 128(%rsp, %r14, 8)
+       vmovsd  %xmm0, 128(%rsp, %r14, 8)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_atan24_core_avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_d_atan24_core_avx2.S
index 1b601576cc034e398301170bfba01e5b5f9ea498..a45cae79a1ff7c649cf9062a622c1f9b9d1912d1 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_d_atan24_core_avx2.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_d_atan24_core_avx2.S
@@ -277,12 +277,12 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movsd   32(%rsp, %r14, 8), %xmm0
-       movsd   64(%rsp, %r14, 8), %xmm1
+       vmovsd  32(%rsp, %r14, 8), %xmm0
+       vmovsd  64(%rsp, %r14, 8), %xmm1
        call    atan2@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movsd   %xmm0, 96(%rsp, %r14, 8)
+       vmovsd  %xmm0, 96(%rsp, %r14, 8)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_atan28_core_avx512.S b/sysdeps/x86_64/fpu/multiarch/svml_d_atan28_core_avx512.S
index ef9581075d99a9052615980e94bd6aca30b53d84..c3b0f7940cb0e24d97662e4b74e5071f2a4334e6 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_d_atan28_core_avx512.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_d_atan28_core_avx512.S
@@ -295,12 +295,12 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movsd   64(%rsp, %r14, 8), %xmm0
-       movsd   128(%rsp, %r14, 8), %xmm1
+       vmovsd  64(%rsp, %r14, 8), %xmm0
+       vmovsd  128(%rsp, %r14, 8), %xmm1
        call    atan2@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movsd   %xmm0, 192(%rsp, %r14, 8)
+       vmovsd  %xmm0, 192(%rsp, %r14, 8)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_atanh4_core_avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_d_atanh4_core_avx2.S
index b5cbfd224c31299657c5e785931d0bff26c1a00c..c9c41ef9f405b36592e0ed94497bec49a162353b 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_d_atanh4_core_avx2.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_d_atanh4_core_avx2.S
@@ -339,11 +339,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movsd   32(%rsp, %r14, 8), %xmm0
+       vmovsd  32(%rsp, %r14, 8), %xmm0
        call    atanh@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movsd   %xmm0, 64(%rsp, %r14, 8)
+       vmovsd  %xmm0, 64(%rsp, %r14, 8)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_atanh8_core_avx512.S b/sysdeps/x86_64/fpu/multiarch/svml_d_atanh8_core_avx512.S
index 3193c026dd24a81f5fc9e6c0de84a484c53d0bd3..de4edb3cc006d71881df0f86dc8a91e82c75bdac 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_d_atanh8_core_avx512.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_d_atanh8_core_avx512.S
@@ -274,11 +274,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movsd   64(%rsp, %r14, 8), %xmm0
+       vmovsd  64(%rsp, %r14, 8), %xmm0
        call    atanh@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movsd   %xmm0, 128(%rsp, %r14, 8)
+       vmovsd  %xmm0, 128(%rsp, %r14, 8)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_cbrt4_core_avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_d_cbrt4_core_avx2.S
index 96ecbe05c13f2a17fd3c74903f84cbec88a0f530..71a25f3db85e3f12df0350b6534881874bf61594 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_d_cbrt4_core_avx2.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_d_cbrt4_core_avx2.S
@@ -262,11 +262,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movsd   32(%rsp, %r14, 8), %xmm0
+       vmovsd  32(%rsp, %r14, 8), %xmm0
        call    cbrt@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movsd   %xmm0, 64(%rsp, %r14, 8)
+       vmovsd  %xmm0, 64(%rsp, %r14, 8)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_cosh4_core_avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_d_cosh4_core_avx2.S
index 25df252108de15ba9fb5b3d22417da14abc23347..a3d9104f5eeb4a7ac2d26a56c70a9f21d87772e3 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_d_cosh4_core_avx2.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_d_cosh4_core_avx2.S
@@ -282,11 +282,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movsd   32(%rsp, %r14, 8), %xmm0
+       vmovsd  32(%rsp, %r14, 8), %xmm0
        call    cosh@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movsd   %xmm0, 64(%rsp, %r14, 8)
+       vmovsd  %xmm0, 64(%rsp, %r14, 8)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_cosh8_core_avx512.S b/sysdeps/x86_64/fpu/multiarch/svml_d_cosh8_core_avx512.S
index 066bbc7de6bb2eaba1ba1d1618df312ecb20037b..4ff0e038a36906bbc9902a6ddf363f14e26fd488 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_d_cosh8_core_avx512.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_d_cosh8_core_avx512.S
@@ -231,11 +231,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movsd   64(%rsp, %r14, 8), %xmm0
+       vmovsd  64(%rsp, %r14, 8), %xmm0
        call    cosh@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movsd   %xmm0, 128(%rsp, %r14, 8)
+       vmovsd  %xmm0, 128(%rsp, %r14, 8)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_erfc4_core_avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_d_erfc4_core_avx2.S
index c832b65e3e1d5b3291642d074f7710d3f3bdcedb..6efd2e95ba081555848d4a2c891f606f94d4b485 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_d_erfc4_core_avx2.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_d_erfc4_core_avx2.S
@@ -258,11 +258,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movsd   32(%rsp, %r14, 8), %xmm0
+       vmovsd  32(%rsp, %r14, 8), %xmm0
        call    erfc@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movsd   %xmm0, 64(%rsp, %r14, 8)
+       vmovsd  %xmm0, 64(%rsp, %r14, 8)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_erfc8_core_avx512.S b/sysdeps/x86_64/fpu/multiarch/svml_d_erfc8_core_avx512.S
index 77228814d330d5bb8aa76e22a8e5d4854c0c38a6..42bdfe6f188913d620d8dca54ab15f249807051f 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_d_erfc8_core_avx512.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_d_erfc8_core_avx512.S
@@ -261,11 +261,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movsd   64(%rsp, %r14, 8), %xmm0
+       vmovsd  64(%rsp, %r14, 8), %xmm0
        call    erfc@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movsd   %xmm0, 128(%rsp, %r14, 8)
+       vmovsd  %xmm0, 128(%rsp, %r14, 8)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_exp104_core_avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_d_exp104_core_avx2.S
index 7271bcc1d9bf6c6854400c57ea7bda2d00d473fd..f519bcce45cc72d22328cfb382994885485ddee5 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_d_exp104_core_avx2.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_d_exp104_core_avx2.S
@@ -231,11 +231,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movsd   32(%rsp, %r14, 8), %xmm0
+       vmovsd  32(%rsp, %r14, 8), %xmm0
        call    exp10@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movsd   %xmm0, 64(%rsp, %r14, 8)
+       vmovsd  %xmm0, 64(%rsp, %r14, 8)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_exp108_core_avx512.S b/sysdeps/x86_64/fpu/multiarch/svml_d_exp108_core_avx512.S
index 40b01c3cd08d158b27ff9e1d80b0de5f858a411c..3f0c6701998b8b9732f6217ff1e23e2df2704d64 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_d_exp108_core_avx512.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_d_exp108_core_avx512.S
@@ -191,11 +191,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movsd   64(%rsp, %r14, 8), %xmm0
+       vmovsd  64(%rsp, %r14, 8), %xmm0
        call    exp10@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movsd   %xmm0, 128(%rsp, %r14, 8)
+       vmovsd  %xmm0, 128(%rsp, %r14, 8)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_exp24_core_avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_d_exp24_core_avx2.S
index ced774e89c294f21866b1c3dd4dee951eece9a15..afa00a38bb7dd5c1e29879b0a819d503508551de 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_d_exp24_core_avx2.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_d_exp24_core_avx2.S
@@ -223,11 +223,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movsd   32(%rsp, %r14, 8), %xmm0
+       vmovsd  32(%rsp, %r14, 8), %xmm0
        call    exp2@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movsd   %xmm0, 64(%rsp, %r14, 8)
+       vmovsd  %xmm0, 64(%rsp, %r14, 8)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_exp28_core_avx512.S b/sysdeps/x86_64/fpu/multiarch/svml_d_exp28_core_avx512.S
index 7a85fd8b18df5407d25a0a1feb27ad05ccc19bcd..eee785dbf5ea62bd3e8997a7ef62b66039075eb3 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_d_exp28_core_avx512.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_d_exp28_core_avx512.S
@@ -227,11 +227,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movsd   64(%rsp, %r14, 8), %xmm0
+       vmovsd  64(%rsp, %r14, 8), %xmm0
        call    exp2@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movsd   %xmm0, 128(%rsp, %r14, 8)
+       vmovsd  %xmm0, 128(%rsp, %r14, 8)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_expm14_core_avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_d_expm14_core_avx2.S
index 590341c2434044c027545283ede00f081f52f033..4a3202750f032d262f546367fae5b54914a3cb35 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_d_expm14_core_avx2.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_d_expm14_core_avx2.S
@@ -205,11 +205,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movsd   32(%rsp, %r14, 8), %xmm0
+       vmovsd  32(%rsp, %r14, 8), %xmm0
        call    expm1@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movsd   %xmm0, 64(%rsp, %r14, 8)
+       vmovsd  %xmm0, 64(%rsp, %r14, 8)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_expm18_core_avx512.S b/sysdeps/x86_64/fpu/multiarch/svml_d_expm18_core_avx512.S
index efae1f8b663a23e3c36c631af91ceb4bc31e86d8..0fa17f3a73347de4c57268d986f700f628a380a8 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_d_expm18_core_avx512.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_d_expm18_core_avx512.S
@@ -211,11 +211,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movsd   64(%rsp, %r14, 8), %xmm0
+       vmovsd  64(%rsp, %r14, 8), %xmm0
        call    expm1@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movsd   %xmm0, 128(%rsp, %r14, 8)
+       vmovsd  %xmm0, 128(%rsp, %r14, 8)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_hypot4_core_avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_d_hypot4_core_avx2.S
index ae5738c1b7aede02fa62242131f1a67e9cd42980..5c693d132ebcb454f5e2aa3759e0f2fde8188920 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_d_hypot4_core_avx2.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_d_hypot4_core_avx2.S
@@ -231,12 +231,12 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movsd   32(%rsp, %r14, 8), %xmm0
-       movsd   64(%rsp, %r14, 8), %xmm1
+       vmovsd  32(%rsp, %r14, 8), %xmm0
+       vmovsd  64(%rsp, %r14, 8), %xmm1
        call    hypot@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movsd   %xmm0, 96(%rsp, %r14, 8)
+       vmovsd  %xmm0, 96(%rsp, %r14, 8)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_hypot8_core_avx512.S b/sysdeps/x86_64/fpu/multiarch/svml_d_hypot8_core_avx512.S
index 0c404fd5eec4a8f26b982cb3c77f174ac620c550..a392252c8b1f9f18ca5fc57ee9df37a53385c421 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_d_hypot8_core_avx512.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_d_hypot8_core_avx512.S
@@ -194,12 +194,12 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movsd   64(%rsp, %r14, 8), %xmm0
-       movsd   128(%rsp, %r14, 8), %xmm1
+       vmovsd  64(%rsp, %r14, 8), %xmm0
+       vmovsd  128(%rsp, %r14, 8), %xmm1
        call    hypot@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movsd   %xmm0, 192(%rsp, %r14, 8)
+       vmovsd  %xmm0, 192(%rsp, %r14, 8)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_log104_core_avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_d_log104_core_avx2.S
index 2461c6ad565c53763d93cd07e50c87568ef14a93..9bf45a6dc2b5b2433cd0f8e93cb92eaf93877ddc 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_d_log104_core_avx2.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_d_log104_core_avx2.S
@@ -225,11 +225,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movsd   32(%rsp, %r14, 8), %xmm0
+       vmovsd  32(%rsp, %r14, 8), %xmm0
        call    log10@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movsd   %xmm0, 64(%rsp, %r14, 8)
+       vmovsd  %xmm0, 64(%rsp, %r14, 8)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_log108_core_avx512.S b/sysdeps/x86_64/fpu/multiarch/svml_d_log108_core_avx512.S
index 5d129ef4e513ecd009c9335a70a44b2fd42aca7c..101618cce929688ec67a19feb6b946e93b4ee1ff 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_d_log108_core_avx512.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_d_log108_core_avx512.S
@@ -207,11 +207,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movsd   64(%rsp, %r14, 8), %xmm0
+       vmovsd  64(%rsp, %r14, 8), %xmm0
        call    log10@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movsd   %xmm0, 128(%rsp, %r14, 8)
+       vmovsd  %xmm0, 128(%rsp, %r14, 8)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_log1p4_core_avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_d_log1p4_core_avx2.S
index 13235793e851b98aa6e1460acae3af81f42168ec..39ec0024cfed3a8d6f76da3055d735402d6ee85a 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_d_log1p4_core_avx2.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_d_log1p4_core_avx2.S
@@ -263,11 +263,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movsd   32(%rsp, %r14, 8), %xmm0
+       vmovsd  32(%rsp, %r14, 8), %xmm0
        call    log1p@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movsd   %xmm0, 64(%rsp, %r14, 8)
+       vmovsd  %xmm0, 64(%rsp, %r14, 8)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_log1p8_core_avx512.S b/sysdeps/x86_64/fpu/multiarch/svml_d_log1p8_core_avx512.S
index dd55b5dd18d73c1efb6b4f1da9d27c88e6d03cf5..3033fcb5b34d3a3c3150937ba16776cffd8cf5ee 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_d_log1p8_core_avx512.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_d_log1p8_core_avx512.S
@@ -225,11 +225,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movsd   64(%rsp, %r14, 8), %xmm0
+       vmovsd  64(%rsp, %r14, 8), %xmm0
        call    log1p@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movsd   %xmm0, 128(%rsp, %r14, 8)
+       vmovsd  %xmm0, 128(%rsp, %r14, 8)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_log24_core_avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_d_log24_core_avx2.S
index 25d2edaae5d16f411f950c4d0c7b95541556e10e..84bdb2090dfb0d849565a33d03a3e95453076114 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_d_log24_core_avx2.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_d_log24_core_avx2.S
@@ -223,11 +223,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movsd   32(%rsp, %r14, 8), %xmm0
+       vmovsd  32(%rsp, %r14, 8), %xmm0
        call    log2@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movsd   %xmm0, 64(%rsp, %r14, 8)
+       vmovsd  %xmm0, 64(%rsp, %r14, 8)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_log28_core_avx512.S b/sysdeps/x86_64/fpu/multiarch/svml_d_log28_core_avx512.S
index bcb6736dec337f60d2c5d0c3702bc7044bbb69c1..b3e9bb3ca4c503a3e93096acb75f8afb2f8bb5ac 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_d_log28_core_avx512.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_d_log28_core_avx512.S
@@ -205,11 +205,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movsd   64(%rsp, %r14, 8), %xmm0
+       vmovsd  64(%rsp, %r14, 8), %xmm0
        call    log2@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movsd   %xmm0, 128(%rsp, %r14, 8)
+       vmovsd  %xmm0, 128(%rsp, %r14, 8)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_sinh4_core_avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_d_sinh4_core_avx2.S
index ae166005798cd1d8f344dbab3244e22e21b1eecf..ad2a06ad37d9e20ff8ff5b510292c1203a6f2252 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_d_sinh4_core_avx2.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_d_sinh4_core_avx2.S
@@ -280,11 +280,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movsd   32(%rsp, %r14, 8), %xmm0
+       vmovsd  32(%rsp, %r14, 8), %xmm0
        call    sinh@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movsd   %xmm0, 64(%rsp, %r14, 8)
+       vmovsd  %xmm0, 64(%rsp, %r14, 8)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_sinh8_core_avx512.S b/sysdeps/x86_64/fpu/multiarch/svml_d_sinh8_core_avx512.S
index 075665d57da8e26902566853d6ac25c0dec33a87..7ca915e30fcc5ac3e872cef0edcd0a7d352b2247 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_d_sinh8_core_avx512.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_d_sinh8_core_avx512.S
@@ -271,11 +271,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movsd   64(%rsp, %r14, 8), %xmm0
+       vmovsd  64(%rsp, %r14, 8), %xmm0
        call    sinh@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movsd   %xmm0, 128(%rsp, %r14, 8)
+       vmovsd  %xmm0, 128(%rsp, %r14, 8)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_tan4_core_avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_d_tan4_core_avx2.S
index 01c86736e79daf715da4fd5e2d54c94f559ae594..f26daf316bc9ca15cd8afbadb6f716ec5b379a5e 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_d_tan4_core_avx2.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_d_tan4_core_avx2.S
@@ -267,11 +267,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movsd   32(%rsp, %r14, 8), %xmm0
+       vmovsd  32(%rsp, %r14, 8), %xmm0
        call    tan@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movsd   %xmm0, 64(%rsp, %r14, 8)
+       vmovsd  %xmm0, 64(%rsp, %r14, 8)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_tan8_core_avx512.S b/sysdeps/x86_64/fpu/multiarch/svml_d_tan8_core_avx512.S
index 376479035e66af51ebc5b96bce1186b9a7564e8b..0c90328b0a81047665fccb61bec289077360282e 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_d_tan8_core_avx512.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_d_tan8_core_avx512.S
@@ -239,11 +239,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movsd   64(%rsp, %r14, 8), %xmm0
+       vmovsd  64(%rsp, %r14, 8), %xmm0
        call    tan@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movsd   %xmm0, 128(%rsp, %r14, 8)
+       vmovsd  %xmm0, 128(%rsp, %r14, 8)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_tanh4_core_avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_d_tanh4_core_avx2.S
index 7ddf145b25a01d080c159b0545ec468a00253589..ea41d326ebf2fed4b25f57d0d3a16ff6de1cd9fe 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_d_tanh4_core_avx2.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_d_tanh4_core_avx2.S
@@ -110,7 +110,7 @@ ENTRY(_ZGVdN4v_tanh_avx2)
        vpcmpgtd %xmm11, %xmm9, %xmm10
        vpcmpgtd %xmm8, %xmm9, %xmm0
        vpand   %xmm10, %xmm9, %xmm7
-       blendvps %xmm0, %xmm8, %xmm7
+       vblendvps %xmm0, %xmm8, %xmm7, %xmm7
 
        /*
         * VSHRIMM( I, iIndex, = iIndex, (17 - 4) );
@@ -272,11 +272,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movsd   32(%rsp, %r14, 8), %xmm0
+       vmovsd  32(%rsp, %r14, 8), %xmm0
        call    tanh@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movsd   %xmm0, 64(%rsp, %r14, 8)
+       vmovsd  %xmm0, 64(%rsp, %r14, 8)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_tanh8_core_avx512.S b/sysdeps/x86_64/fpu/multiarch/svml_d_tanh8_core_avx512.S
index 82c011950038d90082d215641c34201f242ca1c5..c995401a248aecd2b992301da4c78ef9f89bc639 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_d_tanh8_core_avx512.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_d_tanh8_core_avx512.S
@@ -286,11 +286,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movsd   64(%rsp, %r14, 8), %xmm0
+       vmovsd  64(%rsp, %r14, 8), %xmm0
        call    tanh@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movsd   %xmm0, 128(%rsp, %r14, 8)
+       vmovsd  %xmm0, 128(%rsp, %r14, 8)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_acosf16_core_avx512.S b/sysdeps/x86_64/fpu/multiarch/svml_s_acosf16_core_avx512.S
index 26fef1f26800dee6d40549b813ebba9f05408bd4..fd84977e952b7a7ce047a316a75c78941b75752e 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_s_acosf16_core_avx512.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_s_acosf16_core_avx512.S
@@ -205,11 +205,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movss   64(%rsp, %r14, 4), %xmm0
+       vmovss  64(%rsp, %r14, 4), %xmm0
        call    acosf@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movss   %xmm0, 128(%rsp, %r14, 4)
+       vmovss  %xmm0, 128(%rsp, %r14, 4)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_acosf8_core_avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_s_acosf8_core_avx2.S
index bf28a5dd005155994c0aa7a35f341599d385dd5e..078fe5a898921d22fb1b0bc92e4ecb4857f2f6f5 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_s_acosf8_core_avx2.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_s_acosf8_core_avx2.S
@@ -198,11 +198,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movss   32(%rsp, %r14, 4), %xmm0
+       vmovss  32(%rsp, %r14, 4), %xmm0
        call    acosf@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movss   %xmm0, 64(%rsp, %r14, 4)
+       vmovss  %xmm0, 64(%rsp, %r14, 4)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_acoshf16_core_avx512.S b/sysdeps/x86_64/fpu/multiarch/svml_s_acoshf16_core_avx512.S
index 3f44e75248d107629258ec392e77eb3ec9698978..65026e647de1f5da062a64dba7df7eafa8c0b1b0 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_s_acoshf16_core_avx512.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_s_acoshf16_core_avx512.S
@@ -290,11 +290,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movss   64(%rsp, %r14, 4), %xmm0
+       vmovss  64(%rsp, %r14, 4), %xmm0
        call    acoshf@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movss   %xmm0, 128(%rsp, %r14, 4)
+       vmovss  %xmm0, 128(%rsp, %r14, 4)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_acoshf8_core_avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_s_acoshf8_core_avx2.S
index 3a70fc144892fe850dab664b2d2619ca8db8e145..489dac033ca77608c78eab7227745e936eab49d0 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_s_acoshf8_core_avx2.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_s_acoshf8_core_avx2.S
@@ -286,11 +286,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movss   32(%rsp, %r14, 4), %xmm0
+       vmovss  32(%rsp, %r14, 4), %xmm0
        call    acoshf@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movss   %xmm0, 64(%rsp, %r14, 4)
+       vmovss  %xmm0, 64(%rsp, %r14, 4)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_asinf16_core_avx512.S b/sysdeps/x86_64/fpu/multiarch/svml_s_asinf16_core_avx512.S
index 4e9984d8709cf4412abd0135f6013f59741015e9..2accef703e12f884c61f066758a832d904f94f8a 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_s_asinf16_core_avx512.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_s_asinf16_core_avx512.S
@@ -198,11 +198,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movss   64(%rsp, %r14, 4), %xmm0
+       vmovss  64(%rsp, %r14, 4), %xmm0
        call    asinf@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movss   %xmm0, 128(%rsp, %r14, 4)
+       vmovss  %xmm0, 128(%rsp, %r14, 4)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_asinf8_core_avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_s_asinf8_core_avx2.S
index 59bea9dc424ec52104422d0e99849d4ee893f6e6..257c8da2f7b020fc5eeaf84b1f18a8c81e4b9345 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_s_asinf8_core_avx2.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_s_asinf8_core_avx2.S
@@ -187,11 +187,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movss   32(%rsp, %r14, 4), %xmm0
+       vmovss  32(%rsp, %r14, 4), %xmm0
        call    asinf@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movss   %xmm0, 64(%rsp, %r14, 4)
+       vmovss  %xmm0, 64(%rsp, %r14, 4)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_asinhf16_core_avx512.S b/sysdeps/x86_64/fpu/multiarch/svml_s_asinhf16_core_avx512.S
index 6b569ecf41904ae98cdb401a154165303090479f..a0c27922e4f270bc850f881f8d35dc07d515c3aa 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_s_asinhf16_core_avx512.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_s_asinhf16_core_avx512.S
@@ -313,11 +313,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movss   64(%rsp, %r14, 4), %xmm0
+       vmovss  64(%rsp, %r14, 4), %xmm0
        call    asinhf@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movss   %xmm0, 128(%rsp, %r14, 4)
+       vmovss  %xmm0, 128(%rsp, %r14, 4)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_asinhf8_core_avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_s_asinhf8_core_avx2.S
index 794030a481189b9675040b5ed10ec2ef395a0061..d6f6c3d5aa33b706b4e2955aff4a4b6e3bd10950 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_s_asinhf8_core_avx2.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_s_asinhf8_core_avx2.S
@@ -361,11 +361,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movss   32(%rsp, %r14, 4), %xmm0
+       vmovss  32(%rsp, %r14, 4), %xmm0
        call    asinhf@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movss   %xmm0, 64(%rsp, %r14, 4)
+       vmovss  %xmm0, 64(%rsp, %r14, 4)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_atan2f16_core_avx512.S b/sysdeps/x86_64/fpu/multiarch/svml_s_atan2f16_core_avx512.S
index 56aa5bb917dc11aa6bbdb8949b27073acfd79901..15ffa4b6c9b7aaccf595e694da631fa9b9f8331a 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_s_atan2f16_core_avx512.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_s_atan2f16_core_avx512.S
@@ -257,12 +257,12 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movss   64(%rsp, %r14, 4), %xmm0
-       movss   128(%rsp, %r14, 4), %xmm1
+       vmovss  64(%rsp, %r14, 4), %xmm0
+       vmovss  128(%rsp, %r14, 4), %xmm1
        call    atan2f@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movss   %xmm0, 192(%rsp, %r14, 4)
+       vmovss  %xmm0, 192(%rsp, %r14, 4)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_atan2f8_core_avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_s_atan2f8_core_avx2.S
index 29ebbb6db2266330dd40799417ac1bb6bd2bcddd..08b18c3e3f1b6d2da6b4ae25003f02b61267c2c9 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_s_atan2f8_core_avx2.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_s_atan2f8_core_avx2.S
@@ -238,12 +238,12 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movss   32(%rsp, %r14, 4), %xmm0
-       movss   64(%rsp, %r14, 4), %xmm1
+       vmovss  32(%rsp, %r14, 4), %xmm0
+       vmovss  64(%rsp, %r14, 4), %xmm1
        call    atan2f@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movss   %xmm0, 96(%rsp, %r14, 4)
+       vmovss  %xmm0, 96(%rsp, %r14, 4)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_atanhf16_core_avx512.S b/sysdeps/x86_64/fpu/multiarch/svml_s_atanhf16_core_avx512.S
index f42462c58180809221623e8e5658cc769779d0e6..94186a14cbd83a41691cbe57d613abfdd596bfa5 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_s_atanhf16_core_avx512.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_s_atanhf16_core_avx512.S
@@ -222,13 +222,13 @@ L(SPECIAL_VALUES_LOOP):
        tzcntl  %ebx, %ebp
 
        /* Scalar math fucntion call to process special input.  */
-       movss   64(%rsp, %rbp, 4), %xmm0
+       vmovss  64(%rsp, %rbp, 4), %xmm0
        call    atanhf@PLT
 
        /* No good way to avoid the store-forwarding fault this will cause on
           return. `lfence` avoids the SF fault but at greater cost as it
           serialized stack/callee save restoration.  */
-       movss   %xmm0, (%rsp, %rbp, 4)
+       vmovss  %xmm0, (%rsp, %rbp, 4)
 
        blsrl   %ebx, %ebx
        jnz     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_atanhf8_core_avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_s_atanhf8_core_avx2.S
index 43eb4238317762b480b58872112d23d0f82de1a7..49ffd7a9b2ff4fd285446ded90ae876505ba1fc1 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_s_atanhf8_core_avx2.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_s_atanhf8_core_avx2.S
@@ -231,13 +231,13 @@ L(SPECIAL_VALUES_LOOP):
        tzcntl  %ebx, %ebp
 
        /* Scalar math fucntion call to process special input.  */
-       movss   32(%rsp, %rbp, 4), %xmm0
+       vmovss  32(%rsp, %rbp, 4), %xmm0
        call    atanhf@PLT
 
        /* No good way to avoid the store-forwarding fault this will cause on
           return. `lfence` avoids the SF fault but at greater cost as it
           serialized stack/callee save restoration.  */
-       movss   %xmm0, (%rsp, %rbp, 4)
+       vmovss  %xmm0, (%rsp, %rbp, 4)
 
        blsrl   %ebx, %ebx
        jnz     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_cbrtf8_core_avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_s_cbrtf8_core_avx2.S
index d24d36163dcd586d14a305ebf29af5aac4834af1..14b58c171a106c6c700e4038f23753aae7403cb3 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_s_cbrtf8_core_avx2.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_s_cbrtf8_core_avx2.S
@@ -304,11 +304,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movss   32(%rsp, %r14, 4), %xmm0
+       vmovss  32(%rsp, %r14, 4), %xmm0
        call    cbrtf@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movss   %xmm0, 64(%rsp, %r14, 4)
+       vmovss  %xmm0, 64(%rsp, %r14, 4)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_coshf16_core_avx512.S b/sysdeps/x86_64/fpu/multiarch/svml_s_coshf16_core_avx512.S
index 6b740bf866a9952fab89bd9f06c1e523180d0139..d1a5ddf5b45ccc9504a6ec9440264085c4267052 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_s_coshf16_core_avx512.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_s_coshf16_core_avx512.S
@@ -228,11 +228,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movss   64(%rsp, %r14, 4), %xmm0
+       vmovss  64(%rsp, %r14, 4), %xmm0
        call    coshf@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movss   %xmm0, 128(%rsp, %r14, 4)
+       vmovss  %xmm0, 128(%rsp, %r14, 4)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_coshf8_core_avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_s_coshf8_core_avx2.S
index 6f29218af19fa5705181ccb840a89be5f4274959..a00650ccd6865544dcbcdb4cd62fdd4e253bd382 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_s_coshf8_core_avx2.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_s_coshf8_core_avx2.S
@@ -242,11 +242,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movss   32(%rsp, %r14, 4), %xmm0
+       vmovss  32(%rsp, %r14, 4), %xmm0
        call    coshf@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movss   %xmm0, 64(%rsp, %r14, 4)
+       vmovss  %xmm0, 64(%rsp, %r14, 4)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_erfcf16_core_avx512.S b/sysdeps/x86_64/fpu/multiarch/svml_s_erfcf16_core_avx512.S
index 9daaa0c06da1a9468289a2cbc08d00f0a8ac8aef..5fb5b2f0f769bcac235b658ace8471c08aab367d 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_s_erfcf16_core_avx512.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_s_erfcf16_core_avx512.S
@@ -218,11 +218,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movss   64(%rsp, %r14, 4), %xmm0
+       vmovss  64(%rsp, %r14, 4), %xmm0
        call    erfcf@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movss   %xmm0, 128(%rsp, %r14, 4)
+       vmovss  %xmm0, 128(%rsp, %r14, 4)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_erfcf8_core_avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_s_erfcf8_core_avx2.S
index 4cafc1bcd5805a6e100a96702c89eaa47fd26d9f..60b9fab000cec8d7a8189db575dcf6d92a1a75e8 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_s_erfcf8_core_avx2.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_s_erfcf8_core_avx2.S
@@ -243,11 +243,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movss   32(%rsp, %r14, 4), %xmm0
+       vmovss  32(%rsp, %r14, 4), %xmm0
        call    erfcf@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movss   %xmm0, 64(%rsp, %r14, 4)
+       vmovss  %xmm0, 64(%rsp, %r14, 4)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_exp10f16_core_avx512.S b/sysdeps/x86_64/fpu/multiarch/svml_s_exp10f16_core_avx512.S
index eb9f3f8d8b858dc52e77c91df6f54c3054d72228..10f0b2cb379da2bd6b93d08f560295da1455d5b8 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_s_exp10f16_core_avx512.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_s_exp10f16_core_avx512.S
@@ -186,11 +186,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movss   64(%rsp, %r14, 4), %xmm0
+       vmovss  64(%rsp, %r14, 4), %xmm0
        call    exp10f@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movss   %xmm0, 128(%rsp, %r14, 4)
+       vmovss  %xmm0, 128(%rsp, %r14, 4)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_exp10f8_core_avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_s_exp10f8_core_avx2.S
index 11244d5a5f2fde25b27bba5af9fd1687c3992a35..275ab42529ce03ab763eaa5e68e44cd0575c6bde 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_s_exp10f8_core_avx2.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_s_exp10f8_core_avx2.S
@@ -238,11 +238,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movss   32(%rsp, %r14, 4), %xmm0
+       vmovss  32(%rsp, %r14, 4), %xmm0
        call    exp10f@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movss   %xmm0, 64(%rsp, %r14, 4)
+       vmovss  %xmm0, 64(%rsp, %r14, 4)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_exp2f16_core_avx512.S b/sysdeps/x86_64/fpu/multiarch/svml_s_exp2f16_core_avx512.S
index 5b406c6e32ec797d07428b91092e5847c27dbb24..8a5f1e3985880f681e7edec617376192d68136d4 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_s_exp2f16_core_avx512.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_s_exp2f16_core_avx512.S
@@ -209,11 +209,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movss   64(%rsp, %r14, 4), %xmm0
+       vmovss  64(%rsp, %r14, 4), %xmm0
        call    exp2f@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movss   %xmm0, 128(%rsp, %r14, 4)
+       vmovss  %xmm0, 128(%rsp, %r14, 4)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_exp2f8_core_avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_s_exp2f8_core_avx2.S
index f7a80a4d64ead89e959d2b08638f0c24f299807b..cc87e66425b522174a1d870bdeab61fca9de6ec7 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_s_exp2f8_core_avx2.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_s_exp2f8_core_avx2.S
@@ -188,11 +188,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movss   32(%rsp, %r14, 4), %xmm0
+       vmovss  32(%rsp, %r14, 4), %xmm0
        call    exp2f@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movss   %xmm0, 64(%rsp, %r14, 4)
+       vmovss  %xmm0, 64(%rsp, %r14, 4)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_expm1f16_core_avx512.S b/sysdeps/x86_64/fpu/multiarch/svml_s_expm1f16_core_avx512.S
index 71d23e632ca551bbabcfc07062bc8f2a6bcd564c..7fe830daa4b2916626802811d7991daca5b4de16 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_s_expm1f16_core_avx512.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_s_expm1f16_core_avx512.S
@@ -194,11 +194,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movss   64(%rsp, %r14, 4), %xmm0
+       vmovss  64(%rsp, %r14, 4), %xmm0
        call    expm1f@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movss   %xmm0, 128(%rsp, %r14, 4)
+       vmovss  %xmm0, 128(%rsp, %r14, 4)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_expm1f8_core_avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_s_expm1f8_core_avx2.S
index 73f862528a0342787574cd1df6cf22bb23390ffc..d5d7fa2791f423dae09c741a80f52c3afa77debc 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_s_expm1f8_core_avx2.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_s_expm1f8_core_avx2.S
@@ -212,11 +212,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movss   32(%rsp, %r14, 4), %xmm0
+       vmovss  32(%rsp, %r14, 4), %xmm0
        call    expm1f@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movss   %xmm0, 64(%rsp, %r14, 4)
+       vmovss  %xmm0, 64(%rsp, %r14, 4)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_hypotf16_core_avx512.S b/sysdeps/x86_64/fpu/multiarch/svml_s_hypotf16_core_avx512.S
index 548936fe6161cd55f8282f9338fe3daa09f87237..c92e3ab065576966fbf7102eb053ae9cfd7c1e21 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_s_hypotf16_core_avx512.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_s_hypotf16_core_avx512.S
@@ -202,12 +202,12 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movss   64(%rsp, %r14, 4), %xmm0
-       movss   128(%rsp, %r14, 4), %xmm1
+       vmovss  64(%rsp, %r14, 4), %xmm0
+       vmovss  128(%rsp, %r14, 4), %xmm1
        call    hypotf@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movss   %xmm0, 192(%rsp, %r14, 4)
+       vmovss  %xmm0, 192(%rsp, %r14, 4)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_hypotf8_core_avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_s_hypotf8_core_avx2.S
index fc97828008af6e8533b8e28beca3d89609a4dfb3..7a26c5accc033c1a68854dc6b69d80c97a574217 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_s_hypotf8_core_avx2.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_s_hypotf8_core_avx2.S
@@ -226,12 +226,12 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movss   32(%rsp, %r14, 4), %xmm0
-       movss   64(%rsp, %r14, 4), %xmm1
+       vmovss  32(%rsp, %r14, 4), %xmm0
+       vmovss  64(%rsp, %r14, 4), %xmm1
        call    hypotf@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movss   %xmm0, 96(%rsp, %r14, 4)
+       vmovss  %xmm0, 96(%rsp, %r14, 4)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_log10f16_core_avx512.S b/sysdeps/x86_64/fpu/multiarch/svml_s_log10f16_core_avx512.S
index b192dfe464b100ef1880e065e94396e3aff1a943..0eb9b23c4e9a80651964a487225a884f909a1204 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_s_log10f16_core_avx512.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_s_log10f16_core_avx512.S
@@ -161,11 +161,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movss   64(%rsp, %r14, 4), %xmm0
+       vmovss  64(%rsp, %r14, 4), %xmm0
        call    log10f@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movss   %xmm0, 128(%rsp, %r14, 4)
+       vmovss  %xmm0, 128(%rsp, %r14, 4)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_log10f8_core_avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_s_log10f8_core_avx2.S
index ea51c28f81247533cce734d033879d336d6cae83..4bdc62e90e026d101959b469131d0e9a092f461d 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_s_log10f8_core_avx2.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_s_log10f8_core_avx2.S
@@ -174,11 +174,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movss   32(%rsp, %r14, 4), %xmm0
+       vmovss  32(%rsp, %r14, 4), %xmm0
        call    log10f@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movss   %xmm0, 64(%rsp, %r14, 4)
+       vmovss  %xmm0, 64(%rsp, %r14, 4)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_log1pf16_core_avx512.S b/sysdeps/x86_64/fpu/multiarch/svml_s_log1pf16_core_avx512.S
index 8fa5068595ff351b11f475cae3894a4798ce9afe..2c864f0c0e7850cf5ea673b7b0531b794913f5cf 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_s_log1pf16_core_avx512.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_s_log1pf16_core_avx512.S
@@ -207,11 +207,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movss   64(%rsp, %r14, 4), %xmm0
+       vmovss  64(%rsp, %r14, 4), %xmm0
        call    log1pf@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movss   %xmm0, 128(%rsp, %r14, 4)
+       vmovss  %xmm0, 128(%rsp, %r14, 4)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_log1pf8_core_avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_s_log1pf8_core_avx2.S
index 54d6a9a685cf478b3107124e8055a59a27ddfaca..7326a2b5ad869221e3e8cdce984a9f06557dc291 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_s_log1pf8_core_avx2.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_s_log1pf8_core_avx2.S
@@ -190,11 +190,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movss   32(%rsp, %r14, 4), %xmm0
+       vmovss  32(%rsp, %r14, 4), %xmm0
        call    log1pf@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movss   %xmm0, 64(%rsp, %r14, 4)
+       vmovss  %xmm0, 64(%rsp, %r14, 4)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_log2f16_core_avx512.S b/sysdeps/x86_64/fpu/multiarch/svml_s_log2f16_core_avx512.S
index 3b0a28fee05fb25fa3c11425a8b9687c36277f8b..02b255dde818f95b22b6134311b072a7e9d7e88f 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_s_log2f16_core_avx512.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_s_log2f16_core_avx512.S
@@ -158,11 +158,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movss   64(%rsp, %r14, 4), %xmm0
+       vmovss  64(%rsp, %r14, 4), %xmm0
        call    log2f@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movss   %xmm0, 128(%rsp, %r14, 4)
+       vmovss  %xmm0, 128(%rsp, %r14, 4)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_log2f8_core_avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_s_log2f8_core_avx2.S
index eaa51121784bbcae85691ab4ae9eb64415212d1d..2245d40f84b6d16f820a8a45fc32d588cb4c0acf 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_s_log2f8_core_avx2.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_s_log2f8_core_avx2.S
@@ -169,11 +169,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movss   32(%rsp, %r14, 4), %xmm0
+       vmovss  32(%rsp, %r14, 4), %xmm0
        call    log2f@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movss   %xmm0, 64(%rsp, %r14, 4)
+       vmovss  %xmm0, 64(%rsp, %r14, 4)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_sinhf16_core_avx512.S b/sysdeps/x86_64/fpu/multiarch/svml_s_sinhf16_core_avx512.S
index fad4847f284a9859495e8f1c438d58f9503146d4..89be733eb262ce355a737af8668ee029baab0156 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_s_sinhf16_core_avx512.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_s_sinhf16_core_avx512.S
@@ -252,11 +252,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movss   64(%rsp, %r14, 4), %xmm0
+       vmovss  64(%rsp, %r14, 4), %xmm0
        call    sinhf@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movss   %xmm0, 128(%rsp, %r14, 4)
+       vmovss  %xmm0, 128(%rsp, %r14, 4)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_sinhf8_core_avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_s_sinhf8_core_avx2.S
index 8c4b46cee29f9cd0e48e8b512159218b6df7ee32..e358e2efee8b0171eea4f2e3c21d17f159d926d1 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_s_sinhf8_core_avx2.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_s_sinhf8_core_avx2.S
@@ -243,11 +243,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movss   32(%rsp, %r14, 4), %xmm0
+       vmovss  32(%rsp, %r14, 4), %xmm0
        call    sinhf@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movss   %xmm0, 64(%rsp, %r14, 4)
+       vmovss  %xmm0, 64(%rsp, %r14, 4)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_tanf16_core_avx512.S b/sysdeps/x86_64/fpu/multiarch/svml_s_tanf16_core_avx512.S
index f2a18f0b2c2a26969958ca94701fd2c23f33825e..4e18cdc0ce84b3f8f86600cde8d13d4185ca21cf 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_s_tanf16_core_avx512.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_s_tanf16_core_avx512.S
@@ -235,11 +235,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %r12d, %r14d
-       movss   64(%rsp, %r14, 4), %xmm0
+       vmovss  64(%rsp, %r14, 4), %xmm0
        call    tanf@PLT
        # LOE rbx r14 r15 r12d r13d xmm0
 
-       movss   %xmm0, 128(%rsp, %r14, 4)
+       vmovss  %xmm0, 128(%rsp, %r14, 4)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_tanf8_core_avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_s_tanf8_core_avx2.S
index cd33fac6435405054ff5f75728384e7f09358d7c..d34e61ac4141dc98cc7e7e51f4a79e79d38d2210 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_s_tanf8_core_avx2.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_s_tanf8_core_avx2.S
@@ -261,11 +261,11 @@ L(SPECIAL_VALUES_LOOP):
 
 L(SCALAR_MATH_CALL):
        movl    %ebx, %r13d
-       movss   32(%rsp, %r13, 4), %xmm0
+       vmovss  32(%rsp, %r13, 4), %xmm0
        call    tanf@PLT
        # LOE r13 r14 r15 ebx r12d xmm0
 
-       movss   %xmm0, 64(%rsp, %r13, 4)
+       vmovss  %xmm0, 64(%rsp, %r13, 4)
 
        /* Process special inputs in loop */
        jmp     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_tanhf16_core_avx512.S b/sysdeps/x86_64/fpu/multiarch/svml_s_tanhf16_core_avx512.S
index 7edc74a116669db1ff5845c4be795ae0dd90ae5c..84f73fdaf90be55b3eaa7828bfbc5cc387dea70a 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_s_tanhf16_core_avx512.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_s_tanhf16_core_avx512.S
@@ -221,13 +221,13 @@ L(SPECIAL_VALUES_LOOP):
        tzcntl  %ebx, %ebp
 
        /* Scalar math fucntion call to process special input.  */
-       movss   64(%rsp, %rbp, 4), %xmm0
+       vmovss  64(%rsp, %rbp, 4), %xmm0
        call    tanhf@PLT
 
        /* No good way to avoid the store-forwarding fault this will cause on
           return. `lfence` avoids the SF fault but at greater cost as it
           serialized stack/callee save restoration.  */
-       movss   %xmm0, (%rsp, %rbp, 4)
+       vmovss  %xmm0, (%rsp, %rbp, 4)
 
        blsrl   %ebx, %ebx
        jnz     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_tanhf8_core_avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_s_tanhf8_core_avx2.S
index 55df346a00c5e8d3a914c5b5abce57e2eb934272..ea3e9f4210d7169eaa5f23aa298698b973f964d6 100644 (file)
--- a/sysdeps/x86_64/fpu/multiarch/svml_s_tanhf8_core_avx2.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_s_tanhf8_core_avx2.S
@@ -240,13 +240,13 @@ L(SPECIAL_VALUES_LOOP):
        tzcntl  %ebx, %ebp
 
        /* Scalar math function call to process special input.  */
-       movss   32(%rsp, %rbp, 4), %xmm0
+       vmovss  32(%rsp, %rbp, 4), %xmm0
        call    tanhf@PLT
 
        /* No good way to avoid the store-forwarding fault this will cause on
           return. `lfence` avoids the SF fault but at greater cost as it
           serialized stack/callee save restoration.  */
-       movss   %xmm0, (%rsp, %rbp, 4)
+       vmovss  %xmm0, (%rsp, %rbp, 4)
 
        blsrl   %ebx, %ebx
        jnz     L(SPECIAL_VALUES_LOOP)

diff --git a/sysdeps/x86_64/multiarch/strrchr-avx2.S b/sysdeps/x86_64/multiarch/strrchr-avx2.S
index bd26ba80d557c9f8cff82009219004b876150da0..eb128a2ae33960cdbcb74ac37ccd5a168915dd6f 100644 (file)
--- a/sysdeps/x86_64/multiarch/strrchr-avx2.S
+++ b/sysdeps/x86_64/multiarch/strrchr-avx2.S
@@ -49,7 +49,7 @@
 
        .section SECTION(.text), "ax", @progbits
 ENTRY(STRRCHR)
-       movd    %esi, %xmm7
+       vmovd   %esi, %xmm7
        movl    %edi, %eax
        /* Broadcast CHAR to YMM4.  */
        VPBROADCAST %xmm7, %ymm7