public inbox for glibc-cvs@sourceware.org
help / color / mirror / Atom feed
* [glibc] x86_64: Fix svml_s_exp10f8_core_avx2.S code formatting
@ 2022-03-08 5:49 Sunil Pandey
0 siblings, 0 replies; only message in thread
From: Sunil Pandey @ 2022-03-08 5:49 UTC (permalink / raw)
To: glibc-cvs
https://sourceware.org/git/gitweb.cgi?p=glibc.git;h=008fb2c8b98c03e35cba49fe45c240a21a5eb205
commit 008fb2c8b98c03e35cba49fe45c240a21a5eb205
Author: Sunil K Pandey <skpgkp2@gmail.com>
Date: Mon Mar 7 10:47:12 2022 -0800
x86_64: Fix svml_s_exp10f8_core_avx2.S code formatting
This commit contains following formatting changes
1. Instructions proceeded by a tab.
2. Instruction less than 8 characters in length have a tab
between it and the first operand.
3. Instruction greater than 7 characters in length have a
space between it and the first operand.
4. Tabs after `#define`d names and their value.
5. 8 space at the beginning of line replaced by tab.
6. Indent comments with code.
7. Remove redundent .text section.
8. 1 space between line content and line comment.
9. Space after all commas.
Reviewed-by: Noah Goldstein <goldstein.w.n@gmail.com>
Diff:
---
.../fpu/multiarch/svml_s_exp10f8_core_avx2.S | 520 ++++++++++-----------
1 file changed, 259 insertions(+), 261 deletions(-)
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 a363ff8539..11244d5a5f 100644
--- a/sysdeps/x86_64/fpu/multiarch/svml_s_exp10f8_core_avx2.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_s_exp10f8_core_avx2.S
@@ -22,7 +22,7 @@
* exp10(x) = 2^x/log10(2) = 2^n * (1 + T[j]) * (1 + P(y))
* where
* x = m*log10(2)/K + y, y in [-log10(2)/K..log10(2)/K]
- * m = n*K + j, m,n,j - signed integer, j in [-K/2..K/2]
+ * m = n*K + j, m, n,j - signed integer, j in [-K/2..K/2]
*
* values of 2^j/K are tabulated
*
@@ -43,289 +43,287 @@
/* Offsets for data table __svml_sexp10_data_internal
*/
-#define _sT 0
-#define _sLg2_10 128
-#define _sShifter 160
-#define _sInvLg2_10hi 192
-#define _sInvLg2_10lo 224
-#define _sPC0 256
-#define _sPC1 288
-#define _sPC2 320
-#define _iIndexMask 352
-#define _iAbsMask 384
-#define _iDomainRange 416
+#define _sT 0
+#define _sLg2_10 128
+#define _sShifter 160
+#define _sInvLg2_10hi 192
+#define _sInvLg2_10lo 224
+#define _sPC0 256
+#define _sPC1 288
+#define _sPC2 320
+#define _iIndexMask 352
+#define _iAbsMask 384
+#define _iDomainRange 416
#include <sysdep.h>
- .text
- .section .text.avx2,"ax",@progbits
+ .section .text.avx2, "ax", @progbits
ENTRY(_ZGVdN8v_exp10f_avx2)
- pushq %rbp
- cfi_def_cfa_offset(16)
- movq %rsp, %rbp
- cfi_def_cfa(6, 16)
- cfi_offset(6, -16)
- andq $-32, %rsp
- subq $96, %rsp
- lea __svml_sexp10_data_internal(%rip), %rax
- vmovups _sShifter+__svml_sexp10_data_internal(%rip), %ymm4
-
-/* Load arument */
- vmovups _sLg2_10+__svml_sexp10_data_internal(%rip), %ymm1
- vmovups _iIndexMask+__svml_sexp10_data_internal(%rip), %ymm2
- vmovaps %ymm0, %ymm3
- vfmadd213ps %ymm4, %ymm3, %ymm1
-
-/* Index and lookup */
- vandps %ymm2, %ymm1, %ymm7
-
-/* iIndex *= sizeof(S); */
- vpslld $2, %ymm7, %ymm10
- vsubps %ymm4, %ymm1, %ymm0
-
-/* Check for overflow\underflow */
- vandps _iAbsMask+__svml_sexp10_data_internal(%rip), %ymm3, %ymm5
- vpcmpgtd _iDomainRange+__svml_sexp10_data_internal(%rip), %ymm5, %ymm6
- vmovmskps %ymm6, %edx
- vmovd %xmm10, %ecx
- vextractf128 $1, %ymm10, %xmm6
- vpextrd $1, %xmm10, %esi
- vpextrd $2, %xmm10, %edi
- vpextrd $3, %xmm10, %r8d
- movslq %ecx, %rcx
- movslq %esi, %rsi
- movslq %edi, %rdi
- movslq %r8d, %r8
- vmovd (%rax,%rcx), %xmm8
- vmovd (%rax,%rsi), %xmm9
- vmovd (%rax,%rdi), %xmm11
- vmovd (%rax,%r8), %xmm12
- vpunpckldq %xmm9, %xmm8, %xmm13
- vpunpckldq %xmm12, %xmm11, %xmm14
- vpunpcklqdq %xmm14, %xmm13, %xmm15
-
-/* R */
- vmovups _sInvLg2_10hi+__svml_sexp10_data_internal(%rip), %ymm13
- vmovd %xmm6, %r9d
- vfnmadd213ps %ymm3, %ymm0, %ymm13
- vpextrd $1, %xmm6, %r10d
- movslq %r9d, %r9
- movslq %r10d, %r10
- vfnmadd132ps _sInvLg2_10lo+__svml_sexp10_data_internal(%rip), %ymm13, %ymm0
- vmovd (%rax,%r9), %xmm4
- vmovd (%rax,%r10), %xmm5
- vpunpckldq %xmm5, %xmm4, %xmm9
-
-/*
- * Polynomial
- * exp10 = 2^N*(Tj+Tj*poly)
- * poly(sN) = {1+later} a0+a1*sR
- */
- vmovups _sPC2+__svml_sexp10_data_internal(%rip), %ymm4
- vfmadd213ps _sPC1+__svml_sexp10_data_internal(%rip), %ymm0, %ymm4
- vpextrd $2, %xmm6, %r11d
- vpextrd $3, %xmm6, %ecx
- movslq %r11d, %r11
- movslq %ecx, %rcx
- vfmadd213ps _sPC0+__svml_sexp10_data_internal(%rip), %ymm0, %ymm4
- vmovd (%rax,%r11), %xmm7
- vmovd (%rax,%rcx), %xmm8
- vpunpckldq %xmm8, %xmm7, %xmm11
-
-/* remove index bits */
- vpandn %ymm1, %ymm2, %ymm0
- vpunpcklqdq %xmm11, %xmm9, %xmm12
-
-/* 2^N */
- vpslld $18, %ymm0, %ymm1
- vinsertf128 $1, %xmm12, %ymm15, %ymm14
-
-/* Tj_l+Tj_h*poly */
- vfmadd213ps %ymm14, %ymm14, %ymm4
-
-/* quick mul 2^N */
- vpaddd %ymm1, %ymm4, %ymm0
-
-/* Finish */
- testl %edx, %edx
-
-/* Go to special inputs processing branch */
- jne L(SPECIAL_VALUES_BRANCH)
- # LOE rbx r12 r13 r14 r15 edx ymm0 ymm3
-
-/* Restore registers
- * and exit the function
- */
+ pushq %rbp
+ cfi_def_cfa_offset(16)
+ movq %rsp, %rbp
+ cfi_def_cfa(6, 16)
+ cfi_offset(6, -16)
+ andq $-32, %rsp
+ subq $96, %rsp
+ lea __svml_sexp10_data_internal(%rip), %rax
+ vmovups _sShifter+__svml_sexp10_data_internal(%rip), %ymm4
+
+ /* Load arument */
+ vmovups _sLg2_10+__svml_sexp10_data_internal(%rip), %ymm1
+ vmovups _iIndexMask+__svml_sexp10_data_internal(%rip), %ymm2
+ vmovaps %ymm0, %ymm3
+ vfmadd213ps %ymm4, %ymm3, %ymm1
+
+ /* Index and lookup */
+ vandps %ymm2, %ymm1, %ymm7
+
+ /* iIndex *= sizeof(S); */
+ vpslld $2, %ymm7, %ymm10
+ vsubps %ymm4, %ymm1, %ymm0
+
+ /* Check for overflow\underflow */
+ vandps _iAbsMask+__svml_sexp10_data_internal(%rip), %ymm3, %ymm5
+ vpcmpgtd _iDomainRange+__svml_sexp10_data_internal(%rip), %ymm5, %ymm6
+ vmovmskps %ymm6, %edx
+ vmovd %xmm10, %ecx
+ vextractf128 $1, %ymm10, %xmm6
+ vpextrd $1, %xmm10, %esi
+ vpextrd $2, %xmm10, %edi
+ vpextrd $3, %xmm10, %r8d
+ movslq %ecx, %rcx
+ movslq %esi, %rsi
+ movslq %edi, %rdi
+ movslq %r8d, %r8
+ vmovd (%rax, %rcx), %xmm8
+ vmovd (%rax, %rsi), %xmm9
+ vmovd (%rax, %rdi), %xmm11
+ vmovd (%rax, %r8), %xmm12
+ vpunpckldq %xmm9, %xmm8, %xmm13
+ vpunpckldq %xmm12, %xmm11, %xmm14
+ vpunpcklqdq %xmm14, %xmm13, %xmm15
+
+ /* R */
+ vmovups _sInvLg2_10hi+__svml_sexp10_data_internal(%rip), %ymm13
+ vmovd %xmm6, %r9d
+ vfnmadd213ps %ymm3, %ymm0, %ymm13
+ vpextrd $1, %xmm6, %r10d
+ movslq %r9d, %r9
+ movslq %r10d, %r10
+ vfnmadd132ps _sInvLg2_10lo+__svml_sexp10_data_internal(%rip), %ymm13, %ymm0
+ vmovd (%rax, %r9), %xmm4
+ vmovd (%rax, %r10), %xmm5
+ vpunpckldq %xmm5, %xmm4, %xmm9
+
+ /*
+ * Polynomial
+ * exp10 = 2^N*(Tj+Tj*poly)
+ * poly(sN) = {1+later} a0+a1*sR
+ */
+ vmovups _sPC2+__svml_sexp10_data_internal(%rip), %ymm4
+ vfmadd213ps _sPC1+__svml_sexp10_data_internal(%rip), %ymm0, %ymm4
+ vpextrd $2, %xmm6, %r11d
+ vpextrd $3, %xmm6, %ecx
+ movslq %r11d, %r11
+ movslq %ecx, %rcx
+ vfmadd213ps _sPC0+__svml_sexp10_data_internal(%rip), %ymm0, %ymm4
+ vmovd (%rax, %r11), %xmm7
+ vmovd (%rax, %rcx), %xmm8
+ vpunpckldq %xmm8, %xmm7, %xmm11
+
+ /* remove index bits */
+ vpandn %ymm1, %ymm2, %ymm0
+ vpunpcklqdq %xmm11, %xmm9, %xmm12
+
+ /* 2^N */
+ vpslld $18, %ymm0, %ymm1
+ vinsertf128 $1, %xmm12, %ymm15, %ymm14
+
+ /* Tj_l+Tj_h*poly */
+ vfmadd213ps %ymm14, %ymm14, %ymm4
+
+ /* quick mul 2^N */
+ vpaddd %ymm1, %ymm4, %ymm0
+
+ /* Finish */
+ testl %edx, %edx
+
+ /* Go to special inputs processing branch */
+ jne L(SPECIAL_VALUES_BRANCH)
+ # LOE rbx r12 r13 r14 r15 edx ymm0 ymm3
+
+ /* Restore registers
+ * and exit the function
+ */
L(EXIT):
- movq %rbp, %rsp
- popq %rbp
- cfi_def_cfa(7, 8)
- cfi_restore(6)
- ret
- cfi_def_cfa(6, 16)
- cfi_offset(6, -16)
-
-/* Branch to process
- * special inputs
- */
+ movq %rbp, %rsp
+ popq %rbp
+ cfi_def_cfa(7, 8)
+ cfi_restore(6)
+ ret
+ cfi_def_cfa(6, 16)
+ cfi_offset(6, -16)
+
+ /* Branch to process
+ * special inputs
+ */
L(SPECIAL_VALUES_BRANCH):
- vmovups %ymm3, 32(%rsp)
- vmovups %ymm0, 64(%rsp)
- # LOE rbx r12 r13 r14 r15 edx ymm0
-
- xorl %eax, %eax
- # LOE rbx r12 r13 r14 r15 eax edx
-
- vzeroupper
- movq %r12, 16(%rsp)
- /* DW_CFA_expression: r12 (r12) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -32; DW_OP_and; DW_OP_const4s: -80; DW_OP_plus) */
- .cfi_escape 0x10, 0x0c, 0x0e, 0x38, 0x1c, 0x0d, 0xe0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0xb0, 0xff, 0xff, 0xff, 0x22
- movl %eax, %r12d
- movq %r13, 8(%rsp)
- /* DW_CFA_expression: r13 (r13) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -32; DW_OP_and; DW_OP_const4s: -88; DW_OP_plus) */
- .cfi_escape 0x10, 0x0d, 0x0e, 0x38, 0x1c, 0x0d, 0xe0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0xa8, 0xff, 0xff, 0xff, 0x22
- movl %edx, %r13d
- movq %r14, (%rsp)
- /* DW_CFA_expression: r14 (r14) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -32; DW_OP_and; DW_OP_const4s: -96; DW_OP_plus) */
- .cfi_escape 0x10, 0x0e, 0x0e, 0x38, 0x1c, 0x0d, 0xe0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0xa0, 0xff, 0xff, 0xff, 0x22
- # LOE rbx r15 r12d r13d
-
-/* Range mask
- * bits check
- */
+ vmovups %ymm3, 32(%rsp)
+ vmovups %ymm0, 64(%rsp)
+ # LOE rbx r12 r13 r14 r15 edx ymm0
+
+ xorl %eax, %eax
+ # LOE rbx r12 r13 r14 r15 eax edx
+
+ vzeroupper
+ movq %r12, 16(%rsp)
+ /* DW_CFA_expression: r12 (r12) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -32; DW_OP_and; DW_OP_const4s: -80; DW_OP_plus) */
+ .cfi_escape 0x10, 0x0c, 0x0e, 0x38, 0x1c, 0x0d, 0xe0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0xb0, 0xff, 0xff, 0xff, 0x22
+ movl %eax, %r12d
+ movq %r13, 8(%rsp)
+ /* DW_CFA_expression: r13 (r13) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -32; DW_OP_and; DW_OP_const4s: -88; DW_OP_plus) */
+ .cfi_escape 0x10, 0x0d, 0x0e, 0x38, 0x1c, 0x0d, 0xe0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0xa8, 0xff, 0xff, 0xff, 0x22
+ movl %edx, %r13d
+ movq %r14, (%rsp)
+ /* DW_CFA_expression: r14 (r14) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -32; DW_OP_and; DW_OP_const4s: -96; DW_OP_plus) */
+ .cfi_escape 0x10, 0x0e, 0x0e, 0x38, 0x1c, 0x0d, 0xe0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0xa0, 0xff, 0xff, 0xff, 0x22
+ # LOE rbx r15 r12d r13d
+
+ /* Range mask
+ * bits check
+ */
L(RANGEMASK_CHECK):
- btl %r12d, %r13d
+ btl %r12d, %r13d
-/* Call scalar math function */
- jc L(SCALAR_MATH_CALL)
- # LOE rbx r15 r12d r13d
+ /* Call scalar math function */
+ jc L(SCALAR_MATH_CALL)
+ # LOE rbx r15 r12d r13d
-/* Special inputs
- * processing loop
- */
+ /* Special inputs
+ * processing loop
+ */
L(SPECIAL_VALUES_LOOP):
- incl %r12d
- cmpl $8, %r12d
-
-/* Check bits in range mask */
- jl L(RANGEMASK_CHECK)
- # LOE rbx r15 r12d r13d
-
- movq 16(%rsp), %r12
- cfi_restore(12)
- movq 8(%rsp), %r13
- cfi_restore(13)
- movq (%rsp), %r14
- cfi_restore(14)
- vmovups 64(%rsp), %ymm0
-
-/* Go to exit */
- jmp L(EXIT)
- /* DW_CFA_expression: r12 (r12) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -32; DW_OP_and; DW_OP_const4s: -80; DW_OP_plus) */
- .cfi_escape 0x10, 0x0c, 0x0e, 0x38, 0x1c, 0x0d, 0xe0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0xb0, 0xff, 0xff, 0xff, 0x22
- /* DW_CFA_expression: r13 (r13) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -32; DW_OP_and; DW_OP_const4s: -88; DW_OP_plus) */
- .cfi_escape 0x10, 0x0d, 0x0e, 0x38, 0x1c, 0x0d, 0xe0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0xa8, 0xff, 0xff, 0xff, 0x22
- /* DW_CFA_expression: r14 (r14) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -32; DW_OP_and; DW_OP_const4s: -96; DW_OP_plus) */
- .cfi_escape 0x10, 0x0e, 0x0e, 0x38, 0x1c, 0x0d, 0xe0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0xa0, 0xff, 0xff, 0xff, 0x22
- # LOE rbx r12 r13 r14 r15 ymm0
-
-/* Scalar math fucntion call
- * to process special input
- */
+ incl %r12d
+ cmpl $8, %r12d
+
+ /* Check bits in range mask */
+ jl L(RANGEMASK_CHECK)
+ # LOE rbx r15 r12d r13d
+
+ movq 16(%rsp), %r12
+ cfi_restore(12)
+ movq 8(%rsp), %r13
+ cfi_restore(13)
+ movq (%rsp), %r14
+ cfi_restore(14)
+ vmovups 64(%rsp), %ymm0
+
+ /* Go to exit */
+ jmp L(EXIT)
+ /* DW_CFA_expression: r12 (r12) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -32; DW_OP_and; DW_OP_const4s: -80; DW_OP_plus) */
+ .cfi_escape 0x10, 0x0c, 0x0e, 0x38, 0x1c, 0x0d, 0xe0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0xb0, 0xff, 0xff, 0xff, 0x22
+ /* DW_CFA_expression: r13 (r13) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -32; DW_OP_and; DW_OP_const4s: -88; DW_OP_plus) */
+ .cfi_escape 0x10, 0x0d, 0x0e, 0x38, 0x1c, 0x0d, 0xe0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0xa8, 0xff, 0xff, 0xff, 0x22
+ /* DW_CFA_expression: r14 (r14) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -32; DW_OP_and; DW_OP_const4s: -96; DW_OP_plus) */
+ .cfi_escape 0x10, 0x0e, 0x0e, 0x38, 0x1c, 0x0d, 0xe0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0xa0, 0xff, 0xff, 0xff, 0x22
+ # LOE rbx r12 r13 r14 r15 ymm0
+
+ /* Scalar math fucntion call
+ * to process special input
+ */
L(SCALAR_MATH_CALL):
- movl %r12d, %r14d
- movss 32(%rsp,%r14,4), %xmm0
- call exp10f@PLT
- # LOE rbx r14 r15 r12d r13d xmm0
+ movl %r12d, %r14d
+ movss 32(%rsp, %r14, 4), %xmm0
+ call exp10f@PLT
+ # LOE rbx r14 r15 r12d r13d xmm0
- movss %xmm0, 64(%rsp,%r14,4)
+ movss %xmm0, 64(%rsp, %r14, 4)
-/* Process special inputs in loop */
- jmp L(SPECIAL_VALUES_LOOP)
- # LOE rbx r15 r12d r13d
+ /* Process special inputs in loop */
+ jmp L(SPECIAL_VALUES_LOOP)
+ # LOE rbx r15 r12d r13d
END(_ZGVdN8v_exp10f_avx2)
- .section .rodata, "a"
- .align 32
+ .section .rodata, "a"
+ .align 32
#ifdef __svml_sexp10_data_internal_typedef
typedef unsigned int VUINT32;
-typedef struct
-{
- __declspec(align(32)) VUINT32 _sT[(1<<5)][1];
- __declspec(align(32)) VUINT32 _sLg2_10[8][1];
- __declspec(align(32)) VUINT32 _sShifter[8][1];
- __declspec(align(32)) VUINT32 _sInvLg2_10hi[8][1];
- __declspec(align(32)) VUINT32 _sInvLg2_10lo[8][1];
- __declspec(align(32)) VUINT32 _sPC0[8][1];
- __declspec(align(32)) VUINT32 _sPC1[8][1];
- __declspec(align(32)) VUINT32 _sPC2[8][1];
- __declspec(align(32)) VUINT32 _iIndexMask[8][1];
- __declspec(align(32)) VUINT32 _iAbsMask[8][1];
- __declspec(align(32)) VUINT32 _iDomainRange[8][1];
+typedef struct {
+ __declspec(align(32)) VUINT32 _sT[(1<<5)][1];
+ __declspec(align(32)) VUINT32 _sLg2_10[8][1];
+ __declspec(align(32)) VUINT32 _sShifter[8][1];
+ __declspec(align(32)) VUINT32 _sInvLg2_10hi[8][1];
+ __declspec(align(32)) VUINT32 _sInvLg2_10lo[8][1];
+ __declspec(align(32)) VUINT32 _sPC0[8][1];
+ __declspec(align(32)) VUINT32 _sPC1[8][1];
+ __declspec(align(32)) VUINT32 _sPC2[8][1];
+ __declspec(align(32)) VUINT32 _iIndexMask[8][1];
+ __declspec(align(32)) VUINT32 _iAbsMask[8][1];
+ __declspec(align(32)) VUINT32 _iDomainRange[8][1];
} __svml_sexp10_data_internal;
#endif
__svml_sexp10_data_internal:
- /*== _sT ==*/
- .long 0x3f800000 // 2^( 0 /32 )
- .long 0x3f82cd87 // 2^( 1 /32 )
- .long 0x3f85aac3 // 2^( 2 /32 )
- .long 0x3f88980f // 2^( 3 /32 )
- .long 0x3f8b95c2 // 2^( 4 /32 )
- .long 0x3f8ea43a // 2^( 5 /32 )
- .long 0x3f91c3d3 // 2^( 6 /32 )
- .long 0x3f94f4f0 // 2^( 7 /32 )
- .long 0x3f9837f0 // 2^( 8 /32 )
- .long 0x3f9b8d3a // 2^( 9 /32 )
- .long 0x3f9ef532 // 2^( 10/32 )
- .long 0x3fa27043 // 2^( 11/32 )
- .long 0x3fa5fed7 // 2^( 12/32 )
- .long 0x3fa9a15b // 2^( 13/32 )
- .long 0x3fad583f // 2^( 14/32 )
- .long 0x3fb123f6 // 2^( 15/32 )
- .long 0x3fb504f3 // 2^( 16/32 )
- .long 0x3fb8fbaf // 2^( 17/32 )
- .long 0x3fbd08a4 // 2^( 18/32 )
- .long 0x3fc12c4d // 2^( 19/32 )
- .long 0x3fc5672a // 2^( 20/32 )
- .long 0x3fc9b9be // 2^( 21/32 )
- .long 0x3fce248c // 2^( 22/32 )
- .long 0x3fd2a81e // 2^( 23/32 )
- .long 0x3fd744fd // 2^( 24/32 )
- .long 0x3fdbfbb8 // 2^( 25/32 )
- .long 0x3fe0ccdf // 2^( 26/32 )
- .long 0x3fe5b907 // 2^( 27/32 )
- .long 0x3feac0c7 // 2^( 28/32 )
- .long 0x3fefe4ba // 2^( 29/32 )
- .long 0x3ff5257d // 2^( 30/32 )
- .long 0x3ffa83b3 // 2^( 31/32 )
- .align 32
- .long 0x42d49a78, 0x42d49a78, 0x42d49a78, 0x42d49a78, 0x42d49a78, 0x42d49a78, 0x42d49a78, 0x42d49a78 /* _sLg2_10*2^K */
- .align 32
- .long 0x4b400000, 0x4b400000, 0x4b400000, 0x4b400000, 0x4b400000, 0x4b400000, 0x4b400000, 0x4b400000 /* _sShifter) */
- .align 32
- .long 0x3c1a2000, 0x3c1a2000, 0x3c1a2000, 0x3c1a2000, 0x3c1a2000, 0x3c1a2000, 0x3c1a2000, 0x3c1a2000 /* _sInvLg2_10hi/2^K hi (24-K-7) bits*/
- .align 32
- .long 0x341a84fc, 0x341a84fc, 0x341a84fc, 0x341a84fc, 0x341a84fc, 0x341a84fc, 0x341a84fc, 0x341a84fc /* _sInvLg2_10lo/2^K lo bits */
- // otherwise exp10(0) won't produce exact 1.0
- .align 32
- .long 0x2fecc868, 0x2fecc868, 0x2fecc868, 0x2fecc868, 0x2fecc868, 0x2fecc868, 0x2fecc868, 0x2fecc868 /* _sPC0 */
- .align 32
- .long 0x40135e1b, 0x40135e1b, 0x40135e1b, 0x40135e1b, 0x40135e1b, 0x40135e1b, 0x40135e1b, 0x40135e1b /* _sPC1 */
- .align 32
- .long 0x4029a8d2, 0x4029a8d2, 0x4029a8d2, 0x4029a8d2, 0x4029a8d2, 0x4029a8d2, 0x4029a8d2, 0x4029a8d2 /* _sPC2 */
- .align 32
- .long 0x0000001f, 0x0000001f, 0x0000001f, 0x0000001f, 0x0000001f, 0x0000001f, 0x0000001f, 0x0000001f /* _iIndexMask =(2^K-1)*/
- //common
- .align 32
- .long 0x7fffffff, 0x7fffffff, 0x7fffffff, 0x7fffffff, 0x7fffffff, 0x7fffffff, 0x7fffffff, 0x7fffffff /* _iAbsMask */
- .align 32
- .long 0x4217b818, 0x4217b818, 0x4217b818, 0x4217b818, 0x4217b818, 0x4217b818, 0x4217b818, 0x4217b818 /* _iDomainRange=-log10(max_denormal=0x007fffff) RZ */
- .align 32
- .type __svml_sexp10_data_internal,@object
- .size __svml_sexp10_data_internal,.-__svml_sexp10_data_internal
+ /* _sT */
+ .long 0x3f800000 // 2^( 0 /32 )
+ .long 0x3f82cd87 // 2^( 1 /32 )
+ .long 0x3f85aac3 // 2^( 2 /32 )
+ .long 0x3f88980f // 2^( 3 /32 )
+ .long 0x3f8b95c2 // 2^( 4 /32 )
+ .long 0x3f8ea43a // 2^( 5 /32 )
+ .long 0x3f91c3d3 // 2^( 6 /32 )
+ .long 0x3f94f4f0 // 2^( 7 /32 )
+ .long 0x3f9837f0 // 2^( 8 /32 )
+ .long 0x3f9b8d3a // 2^( 9 /32 )
+ .long 0x3f9ef532 // 2^( 10/32 )
+ .long 0x3fa27043 // 2^( 11/32 )
+ .long 0x3fa5fed7 // 2^( 12/32 )
+ .long 0x3fa9a15b // 2^( 13/32 )
+ .long 0x3fad583f // 2^( 14/32 )
+ .long 0x3fb123f6 // 2^( 15/32 )
+ .long 0x3fb504f3 // 2^( 16/32 )
+ .long 0x3fb8fbaf // 2^( 17/32 )
+ .long 0x3fbd08a4 // 2^( 18/32 )
+ .long 0x3fc12c4d // 2^( 19/32 )
+ .long 0x3fc5672a // 2^( 20/32 )
+ .long 0x3fc9b9be // 2^( 21/32 )
+ .long 0x3fce248c // 2^( 22/32 )
+ .long 0x3fd2a81e // 2^( 23/32 )
+ .long 0x3fd744fd // 2^( 24/32 )
+ .long 0x3fdbfbb8 // 2^( 25/32 )
+ .long 0x3fe0ccdf // 2^( 26/32 )
+ .long 0x3fe5b907 // 2^( 27/32 )
+ .long 0x3feac0c7 // 2^( 28/32 )
+ .long 0x3fefe4ba // 2^( 29/32 )
+ .long 0x3ff5257d // 2^( 30/32 )
+ .long 0x3ffa83b3 // 2^( 31/32 )
+ .align 32
+ .long 0x42d49a78, 0x42d49a78, 0x42d49a78, 0x42d49a78, 0x42d49a78, 0x42d49a78, 0x42d49a78, 0x42d49a78 /* _sLg2_10*2^K */
+ .align 32
+ .long 0x4b400000, 0x4b400000, 0x4b400000, 0x4b400000, 0x4b400000, 0x4b400000, 0x4b400000, 0x4b400000 /* _sShifter) */
+ .align 32
+ .long 0x3c1a2000, 0x3c1a2000, 0x3c1a2000, 0x3c1a2000, 0x3c1a2000, 0x3c1a2000, 0x3c1a2000, 0x3c1a2000 /* _sInvLg2_10hi/2^K hi (24-K-7) bits */
+ .align 32
+ .long 0x341a84fc, 0x341a84fc, 0x341a84fc, 0x341a84fc, 0x341a84fc, 0x341a84fc, 0x341a84fc, 0x341a84fc /* _sInvLg2_10lo/2^K lo bits */
+ // otherwise exp10(0) won't produce exact 1.0
+ .align 32
+ .long 0x2fecc868, 0x2fecc868, 0x2fecc868, 0x2fecc868, 0x2fecc868, 0x2fecc868, 0x2fecc868, 0x2fecc868 /* _sPC0 */
+ .align 32
+ .long 0x40135e1b, 0x40135e1b, 0x40135e1b, 0x40135e1b, 0x40135e1b, 0x40135e1b, 0x40135e1b, 0x40135e1b /* _sPC1 */
+ .align 32
+ .long 0x4029a8d2, 0x4029a8d2, 0x4029a8d2, 0x4029a8d2, 0x4029a8d2, 0x4029a8d2, 0x4029a8d2, 0x4029a8d2 /* _sPC2 */
+ .align 32
+ .long 0x0000001f, 0x0000001f, 0x0000001f, 0x0000001f, 0x0000001f, 0x0000001f, 0x0000001f, 0x0000001f /* _iIndexMask =(2^K-1) */
+ //common
+ .align 32
+ .long 0x7fffffff, 0x7fffffff, 0x7fffffff, 0x7fffffff, 0x7fffffff, 0x7fffffff, 0x7fffffff, 0x7fffffff /* _iAbsMask */
+ .align 32
+ .long 0x4217b818, 0x4217b818, 0x4217b818, 0x4217b818, 0x4217b818, 0x4217b818, 0x4217b818, 0x4217b818 /* _iDomainRange=-log10(max_denormal=0x007fffff) RZ */
+ .align 32
+ .type __svml_sexp10_data_internal, @object
+ .size __svml_sexp10_data_internal, .-__svml_sexp10_data_internal
^ permalink raw reply [flat|nested] only message in thread
only message in thread, other threads:[~2022-03-08 5:49 UTC | newest]
Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-03-08 5:49 [glibc] x86_64: Fix svml_s_exp10f8_core_avx2.S code formatting Sunil Pandey
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).