mirror of git://sourceware.org/git/glibc.git
x86_64: Fix svml_s_atan2f4_core_sse4.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>
This commit is contained in:
Sunil K Pandey
parent
00dec7a257
commit
d5f5ecab14
|
|
@ -30,355 +30,354 @@
|
|||
|
||||
/* Offsets for data table __svml_satan2_data_internal
|
||||
*/
|
||||
#define sZERO 0
|
||||
#define sSIGN_MASK 16
|
||||
#define sABS_MASK 32
|
||||
#define sPIO2 48
|
||||
#define sPI 64
|
||||
#define sPC8 80
|
||||
#define sPC7 96
|
||||
#define sPC6 112
|
||||
#define sPC5 128
|
||||
#define sPC4 144
|
||||
#define sPC3 160
|
||||
#define sPC2 176
|
||||
#define sPC1 192
|
||||
#define sPC0 208
|
||||
#define iCHK_WORK_SUB 224
|
||||
#define iCHK_WORK_CMP 240
|
||||
#define sZERO 0
|
||||
#define sSIGN_MASK 16
|
||||
#define sABS_MASK 32
|
||||
#define sPIO2 48
|
||||
#define sPI 64
|
||||
#define sPC8 80
|
||||
#define sPC7 96
|
||||
#define sPC6 112
|
||||
#define sPC5 128
|
||||
#define sPC4 144
|
||||
#define sPC3 160
|
||||
#define sPC2 176
|
||||
#define sPC1 192
|
||||
#define sPC0 208
|
||||
#define iCHK_WORK_SUB 224
|
||||
#define iCHK_WORK_CMP 240
|
||||
|
||||
#include <sysdep.h>
|
||||
|
||||
.text
|
||||
.section .text.sse4,"ax",@progbits
|
||||
.section .text.sse4, "ax", @progbits
|
||||
ENTRY(_ZGVbN4vv_atan2f_sse4)
|
||||
subq $88, %rsp
|
||||
cfi_def_cfa_offset(96)
|
||||
movaps %xmm0, %xmm12
|
||||
subq $88, %rsp
|
||||
cfi_def_cfa_offset(96)
|
||||
movaps %xmm0, %xmm12
|
||||
|
||||
/*
|
||||
* #define NO_VECTOR_ZERO_ATAN2_ARGS
|
||||
* Declarations
|
||||
* Variables
|
||||
* Constants
|
||||
* The end of declarations
|
||||
* Implementation
|
||||
* Arguments signs
|
||||
*/
|
||||
movups sABS_MASK+__svml_satan2_data_internal(%rip), %xmm10
|
||||
movaps %xmm1, %xmm13
|
||||
movaps %xmm10, %xmm11
|
||||
andps %xmm12, %xmm10
|
||||
andps %xmm13, %xmm11
|
||||
movaps %xmm10, %xmm7
|
||||
cmpltps %xmm11, %xmm7
|
||||
/*
|
||||
* #define NO_VECTOR_ZERO_ATAN2_ARGS
|
||||
* Declarations
|
||||
* Variables
|
||||
* Constants
|
||||
* The end of declarations
|
||||
* Implementation
|
||||
* Arguments signs
|
||||
*/
|
||||
movups sABS_MASK+__svml_satan2_data_internal(%rip), %xmm10
|
||||
movaps %xmm1, %xmm13
|
||||
movaps %xmm10, %xmm11
|
||||
andps %xmm12, %xmm10
|
||||
andps %xmm13, %xmm11
|
||||
movaps %xmm10, %xmm7
|
||||
cmpltps %xmm11, %xmm7
|
||||
|
||||
/*
|
||||
* 1) If y<x then a= y, b=x, PIO2=0
|
||||
* 2) If y>x then a=-x, b=y, PIO2=Pi/2
|
||||
*/
|
||||
movups sSIGN_MASK+__svml_satan2_data_internal(%rip), %xmm6
|
||||
movaps %xmm7, %xmm0
|
||||
orps %xmm11, %xmm6
|
||||
movaps %xmm10, %xmm4
|
||||
andnps %xmm6, %xmm0
|
||||
movaps %xmm7, %xmm6
|
||||
movaps %xmm11, %xmm5
|
||||
andps %xmm7, %xmm4
|
||||
andnps %xmm10, %xmm6
|
||||
andps %xmm7, %xmm5
|
||||
orps %xmm4, %xmm0
|
||||
orps %xmm5, %xmm6
|
||||
/*
|
||||
* 1) If y<x then a= y, b=x, PIO2=0
|
||||
* 2) If y>x then a=-x, b=y, PIO2=Pi/2
|
||||
*/
|
||||
movups sSIGN_MASK+__svml_satan2_data_internal(%rip), %xmm6
|
||||
movaps %xmm7, %xmm0
|
||||
orps %xmm11, %xmm6
|
||||
movaps %xmm10, %xmm4
|
||||
andnps %xmm6, %xmm0
|
||||
movaps %xmm7, %xmm6
|
||||
movaps %xmm11, %xmm5
|
||||
andps %xmm7, %xmm4
|
||||
andnps %xmm10, %xmm6
|
||||
andps %xmm7, %xmm5
|
||||
orps %xmm4, %xmm0
|
||||
orps %xmm5, %xmm6
|
||||
|
||||
/* Division a/b. */
|
||||
divps %xmm6, %xmm0
|
||||
/* Division a/b. */
|
||||
divps %xmm6, %xmm0
|
||||
|
||||
/* Testing on working interval. */
|
||||
movdqu iCHK_WORK_SUB+__svml_satan2_data_internal(%rip), %xmm14
|
||||
movaps %xmm11, %xmm15
|
||||
movaps %xmm10, %xmm3
|
||||
psubd %xmm14, %xmm15
|
||||
psubd %xmm14, %xmm3
|
||||
movdqa %xmm15, %xmm1
|
||||
movdqu iCHK_WORK_CMP+__svml_satan2_data_internal(%rip), %xmm2
|
||||
movdqa %xmm3, %xmm14
|
||||
pcmpgtd %xmm2, %xmm1
|
||||
pcmpeqd %xmm2, %xmm15
|
||||
pcmpgtd %xmm2, %xmm14
|
||||
pcmpeqd %xmm2, %xmm3
|
||||
por %xmm15, %xmm1
|
||||
por %xmm3, %xmm14
|
||||
por %xmm14, %xmm1
|
||||
/* Testing on working interval. */
|
||||
movdqu iCHK_WORK_SUB+__svml_satan2_data_internal(%rip), %xmm14
|
||||
movaps %xmm11, %xmm15
|
||||
movaps %xmm10, %xmm3
|
||||
psubd %xmm14, %xmm15
|
||||
psubd %xmm14, %xmm3
|
||||
movdqa %xmm15, %xmm1
|
||||
movdqu iCHK_WORK_CMP+__svml_satan2_data_internal(%rip), %xmm2
|
||||
movdqa %xmm3, %xmm14
|
||||
pcmpgtd %xmm2, %xmm1
|
||||
pcmpeqd %xmm2, %xmm15
|
||||
pcmpgtd %xmm2, %xmm14
|
||||
pcmpeqd %xmm2, %xmm3
|
||||
por %xmm15, %xmm1
|
||||
por %xmm3, %xmm14
|
||||
por %xmm14, %xmm1
|
||||
|
||||
/* Polynomial. */
|
||||
movaps %xmm0, %xmm14
|
||||
mulps %xmm0, %xmm14
|
||||
movaps %xmm13, %xmm4
|
||||
movmskps %xmm1, %ecx
|
||||
movaps %xmm14, %xmm15
|
||||
movaps %xmm11, %xmm9
|
||||
mulps %xmm14, %xmm15
|
||||
pxor %xmm13, %xmm9
|
||||
movups sPC8+__svml_satan2_data_internal(%rip), %xmm2
|
||||
movaps %xmm10, %xmm8
|
||||
mulps %xmm15, %xmm2
|
||||
pxor %xmm12, %xmm8
|
||||
movups sPC7+__svml_satan2_data_internal(%rip), %xmm3
|
||||
xorl %edx, %edx
|
||||
mulps %xmm15, %xmm3
|
||||
addps sPC6+__svml_satan2_data_internal(%rip), %xmm2
|
||||
mulps %xmm15, %xmm2
|
||||
addps sPC5+__svml_satan2_data_internal(%rip), %xmm3
|
||||
mulps %xmm15, %xmm3
|
||||
addps sPC4+__svml_satan2_data_internal(%rip), %xmm2
|
||||
mulps %xmm15, %xmm2
|
||||
addps sPC3+__svml_satan2_data_internal(%rip), %xmm3
|
||||
mulps %xmm15, %xmm3
|
||||
addps sPC2+__svml_satan2_data_internal(%rip), %xmm2
|
||||
mulps %xmm2, %xmm15
|
||||
addps sPC1+__svml_satan2_data_internal(%rip), %xmm3
|
||||
mulps %xmm3, %xmm14
|
||||
addps sPC0+__svml_satan2_data_internal(%rip), %xmm15
|
||||
/* Polynomial. */
|
||||
movaps %xmm0, %xmm14
|
||||
mulps %xmm0, %xmm14
|
||||
movaps %xmm13, %xmm4
|
||||
movmskps %xmm1, %ecx
|
||||
movaps %xmm14, %xmm15
|
||||
movaps %xmm11, %xmm9
|
||||
mulps %xmm14, %xmm15
|
||||
pxor %xmm13, %xmm9
|
||||
movups sPC8+__svml_satan2_data_internal(%rip), %xmm2
|
||||
movaps %xmm10, %xmm8
|
||||
mulps %xmm15, %xmm2
|
||||
pxor %xmm12, %xmm8
|
||||
movups sPC7+__svml_satan2_data_internal(%rip), %xmm3
|
||||
xorl %edx, %edx
|
||||
mulps %xmm15, %xmm3
|
||||
addps sPC6+__svml_satan2_data_internal(%rip), %xmm2
|
||||
mulps %xmm15, %xmm2
|
||||
addps sPC5+__svml_satan2_data_internal(%rip), %xmm3
|
||||
mulps %xmm15, %xmm3
|
||||
addps sPC4+__svml_satan2_data_internal(%rip), %xmm2
|
||||
mulps %xmm15, %xmm2
|
||||
addps sPC3+__svml_satan2_data_internal(%rip), %xmm3
|
||||
mulps %xmm15, %xmm3
|
||||
addps sPC2+__svml_satan2_data_internal(%rip), %xmm2
|
||||
mulps %xmm2, %xmm15
|
||||
addps sPC1+__svml_satan2_data_internal(%rip), %xmm3
|
||||
mulps %xmm3, %xmm14
|
||||
addps sPC0+__svml_satan2_data_internal(%rip), %xmm15
|
||||
|
||||
/* if x<0, sPI = Pi, else sPI =0 */
|
||||
movups __svml_satan2_data_internal(%rip), %xmm5
|
||||
xorl %eax, %eax
|
||||
andnps sPIO2+__svml_satan2_data_internal(%rip), %xmm7
|
||||
addps %xmm14, %xmm15
|
||||
cmpleps %xmm5, %xmm4
|
||||
/* if x<0, sPI = Pi, else sPI =0 */
|
||||
movups __svml_satan2_data_internal(%rip), %xmm5
|
||||
xorl %eax, %eax
|
||||
andnps sPIO2+__svml_satan2_data_internal(%rip), %xmm7
|
||||
addps %xmm14, %xmm15
|
||||
cmpleps %xmm5, %xmm4
|
||||
|
||||
/* Reconstruction. */
|
||||
mulps %xmm15, %xmm0
|
||||
andps sPI+__svml_satan2_data_internal(%rip), %xmm4
|
||||
addps %xmm7, %xmm0
|
||||
orps %xmm9, %xmm0
|
||||
addps %xmm4, %xmm0
|
||||
orps %xmm8, %xmm0
|
||||
/* Reconstruction. */
|
||||
mulps %xmm15, %xmm0
|
||||
andps sPI+__svml_satan2_data_internal(%rip), %xmm4
|
||||
addps %xmm7, %xmm0
|
||||
orps %xmm9, %xmm0
|
||||
addps %xmm4, %xmm0
|
||||
orps %xmm8, %xmm0
|
||||
|
||||
/* Special branch for fast (vector) processing of zero arguments */
|
||||
testl %ecx, %ecx
|
||||
/* Special branch for fast (vector) processing of zero arguments */
|
||||
testl %ecx, %ecx
|
||||
|
||||
/* Go to auxilary branch */
|
||||
jne L(AUX_BRANCH)
|
||||
# LOE rbx rbp r12 r13 r14 r15 eax edx xmm0 xmm1 xmm4 xmm5 xmm6 xmm7 xmm8 xmm9 xmm10 xmm11 xmm12 xmm13
|
||||
/* Go to auxilary branch */
|
||||
jne L(AUX_BRANCH)
|
||||
# LOE rbx rbp r12 r13 r14 r15 eax edx xmm0 xmm1 xmm4 xmm5 xmm6 xmm7 xmm8 xmm9 xmm10 xmm11 xmm12 xmm13
|
||||
|
||||
/* Return from auxilary branch
|
||||
* for out of main path inputs
|
||||
*/
|
||||
/* Return from auxilary branch
|
||||
* for out of main path inputs
|
||||
*/
|
||||
|
||||
L(AUX_BRANCH_RETURN):
|
||||
/*
|
||||
* Special branch for fast (vector) processing of zero arguments
|
||||
* The end of implementation
|
||||
*/
|
||||
testl %edx, %edx
|
||||
/*
|
||||
* Special branch for fast (vector) processing of zero arguments
|
||||
* The end of implementation
|
||||
*/
|
||||
testl %edx, %edx
|
||||
|
||||
/* Go to special inputs processing branch */
|
||||
jne L(SPECIAL_VALUES_BRANCH)
|
||||
# LOE rbx rbp r12 r13 r14 r15 eax edx xmm0 xmm12 xmm13
|
||||
/* Go to special inputs processing branch */
|
||||
jne L(SPECIAL_VALUES_BRANCH)
|
||||
# LOE rbx rbp r12 r13 r14 r15 eax edx xmm0 xmm12 xmm13
|
||||
|
||||
/* Restore registers
|
||||
* and exit the function
|
||||
*/
|
||||
/* Restore registers
|
||||
* and exit the function
|
||||
*/
|
||||
|
||||
L(EXIT):
|
||||
addq $88, %rsp
|
||||
cfi_def_cfa_offset(8)
|
||||
ret
|
||||
cfi_def_cfa_offset(96)
|
||||
addq $88, %rsp
|
||||
cfi_def_cfa_offset(8)
|
||||
ret
|
||||
cfi_def_cfa_offset(96)
|
||||
|
||||
/* Branch to process
|
||||
* special inputs
|
||||
*/
|
||||
/* Branch to process
|
||||
* special inputs
|
||||
*/
|
||||
|
||||
L(SPECIAL_VALUES_BRANCH):
|
||||
movups %xmm12, 32(%rsp)
|
||||
movups %xmm13, 48(%rsp)
|
||||
movups %xmm0, 64(%rsp)
|
||||
# LOE rbx rbp r12 r13 r14 r15 eax edx xmm0
|
||||
movups %xmm12, 32(%rsp)
|
||||
movups %xmm13, 48(%rsp)
|
||||
movups %xmm0, 64(%rsp)
|
||||
# LOE rbx rbp r12 r13 r14 r15 eax edx xmm0
|
||||
|
||||
movq %r12, 16(%rsp)
|
||||
cfi_offset(12, -80)
|
||||
movl %eax, %r12d
|
||||
movq %r13, 8(%rsp)
|
||||
cfi_offset(13, -88)
|
||||
movl %edx, %r13d
|
||||
movq %r14, (%rsp)
|
||||
cfi_offset(14, -96)
|
||||
# LOE rbx rbp r15 r12d r13d
|
||||
movq %r12, 16(%rsp)
|
||||
cfi_offset(12, -80)
|
||||
movl %eax, %r12d
|
||||
movq %r13, 8(%rsp)
|
||||
cfi_offset(13, -88)
|
||||
movl %edx, %r13d
|
||||
movq %r14, (%rsp)
|
||||
cfi_offset(14, -96)
|
||||
# LOE rbx rbp r15 r12d r13d
|
||||
|
||||
/* Range mask
|
||||
* bits check
|
||||
*/
|
||||
/* Range mask
|
||||
* bits check
|
||||
*/
|
||||
|
||||
L(RANGEMASK_CHECK):
|
||||
btl %r12d, %r13d
|
||||
btl %r12d, %r13d
|
||||
|
||||
/* Call scalar math function */
|
||||
jc L(SCALAR_MATH_CALL)
|
||||
# LOE rbx rbp r15 r12d r13d
|
||||
/* Call scalar math function */
|
||||
jc L(SCALAR_MATH_CALL)
|
||||
# LOE rbx rbp r15 r12d r13d
|
||||
|
||||
/* Special inputs
|
||||
* processing loop
|
||||
*/
|
||||
/* Special inputs
|
||||
* processing loop
|
||||
*/
|
||||
|
||||
L(SPECIAL_VALUES_LOOP):
|
||||
incl %r12d
|
||||
cmpl $4, %r12d
|
||||
incl %r12d
|
||||
cmpl $4, %r12d
|
||||
|
||||
/* Check bits in range mask */
|
||||
jl L(RANGEMASK_CHECK)
|
||||
# LOE rbx rbp r15 r12d r13d
|
||||
/* Check bits in range mask */
|
||||
jl L(RANGEMASK_CHECK)
|
||||
# LOE rbx rbp r15 r12d r13d
|
||||
|
||||
movq 16(%rsp), %r12
|
||||
cfi_restore(12)
|
||||
movq 8(%rsp), %r13
|
||||
cfi_restore(13)
|
||||
movq (%rsp), %r14
|
||||
cfi_restore(14)
|
||||
movups 64(%rsp), %xmm0
|
||||
movq 16(%rsp), %r12
|
||||
cfi_restore(12)
|
||||
movq 8(%rsp), %r13
|
||||
cfi_restore(13)
|
||||
movq (%rsp), %r14
|
||||
cfi_restore(14)
|
||||
movups 64(%rsp), %xmm0
|
||||
|
||||
/* Go to exit */
|
||||
jmp L(EXIT)
|
||||
cfi_offset(12, -80)
|
||||
cfi_offset(13, -88)
|
||||
cfi_offset(14, -96)
|
||||
# LOE rbx rbp r12 r13 r14 r15 xmm0
|
||||
/* Go to exit */
|
||||
jmp L(EXIT)
|
||||
cfi_offset(12, -80)
|
||||
cfi_offset(13, -88)
|
||||
cfi_offset(14, -96)
|
||||
# LOE rbx rbp r12 r13 r14 r15 xmm0
|
||||
|
||||
/* Scalar math fucntion call
|
||||
* to process special input
|
||||
*/
|
||||
/* Scalar math fucntion call
|
||||
* to process special input
|
||||
*/
|
||||
|
||||
L(SCALAR_MATH_CALL):
|
||||
movl %r12d, %r14d
|
||||
movss 32(%rsp,%r14,4), %xmm0
|
||||
movss 48(%rsp,%r14,4), %xmm1
|
||||
call atan2f@PLT
|
||||
# LOE rbx rbp r14 r15 r12d r13d xmm0
|
||||
movl %r12d, %r14d
|
||||
movss 32(%rsp, %r14, 4), %xmm0
|
||||
movss 48(%rsp, %r14, 4), %xmm1
|
||||
call atan2f@PLT
|
||||
# LOE rbx rbp 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)
|
||||
cfi_restore(12)
|
||||
cfi_restore(13)
|
||||
cfi_restore(14)
|
||||
# LOE rbx rbp r15 r12d r13d
|
||||
/* Process special inputs in loop */
|
||||
jmp L(SPECIAL_VALUES_LOOP)
|
||||
cfi_restore(12)
|
||||
cfi_restore(13)
|
||||
cfi_restore(14)
|
||||
# LOE rbx rbp r15 r12d r13d
|
||||
|
||||
/* Auxilary branch
|
||||
* for out of main path inputs
|
||||
*/
|
||||
/* Auxilary branch
|
||||
* for out of main path inputs
|
||||
*/
|
||||
|
||||
L(AUX_BRANCH):
|
||||
/* Check if both X & Y are not NaNs: iXYnotNAN */
|
||||
movaps %xmm13, %xmm3
|
||||
movaps %xmm12, %xmm2
|
||||
cmpordps %xmm13, %xmm3
|
||||
cmpordps %xmm12, %xmm2
|
||||
/* Check if both X & Y are not NaNs: iXYnotNAN */
|
||||
movaps %xmm13, %xmm3
|
||||
movaps %xmm12, %xmm2
|
||||
cmpordps %xmm13, %xmm3
|
||||
cmpordps %xmm12, %xmm2
|
||||
|
||||
/*
|
||||
* Path for zero arguments (at least one of both)
|
||||
* Check if both args are zeros (den. is zero)
|
||||
*/
|
||||
cmpeqps %xmm5, %xmm6
|
||||
/*
|
||||
* Path for zero arguments (at least one of both)
|
||||
* Check if both args are zeros (den. is zero)
|
||||
*/
|
||||
cmpeqps %xmm5, %xmm6
|
||||
|
||||
/* Check if at least on of Y or Y is zero: iAXAYZERO */
|
||||
pcmpeqd %xmm5, %xmm11
|
||||
pcmpeqd %xmm5, %xmm10
|
||||
andps %xmm2, %xmm3
|
||||
por %xmm10, %xmm11
|
||||
/* Check if at least on of Y or Y is zero: iAXAYZERO */
|
||||
pcmpeqd %xmm5, %xmm11
|
||||
pcmpeqd %xmm5, %xmm10
|
||||
andps %xmm2, %xmm3
|
||||
por %xmm10, %xmm11
|
||||
|
||||
/* Check if at least on of Y or Y is zero and not NaN: iAXAYZEROnotNAN */
|
||||
andps %xmm3, %xmm11
|
||||
/* Check if at least on of Y or Y is zero and not NaN: iAXAYZEROnotNAN */
|
||||
andps %xmm3, %xmm11
|
||||
|
||||
/* Exclude from previous callout mask zero (and not NaN) arguments */
|
||||
movaps %xmm11, %xmm10
|
||||
pandn %xmm1, %xmm10
|
||||
/* Exclude from previous callout mask zero (and not NaN) arguments */
|
||||
movaps %xmm11, %xmm10
|
||||
pandn %xmm1, %xmm10
|
||||
|
||||
/* Set sPIO2 to zero if den. is zero */
|
||||
movaps %xmm6, %xmm1
|
||||
andnps %xmm7, %xmm1
|
||||
andps %xmm5, %xmm6
|
||||
orps %xmm6, %xmm1
|
||||
/* Set sPIO2 to zero if den. is zero */
|
||||
movaps %xmm6, %xmm1
|
||||
andnps %xmm7, %xmm1
|
||||
andps %xmm5, %xmm6
|
||||
orps %xmm6, %xmm1
|
||||
|
||||
/* Res = sign(Y)*(X<0)?(PIO2+PI):PIO2 */
|
||||
pcmpgtd %xmm13, %xmm5
|
||||
orps %xmm9, %xmm1
|
||||
andps %xmm4, %xmm5
|
||||
/* Res = sign(Y)*(X<0)?(PIO2+PI):PIO2 */
|
||||
pcmpgtd %xmm13, %xmm5
|
||||
orps %xmm9, %xmm1
|
||||
andps %xmm4, %xmm5
|
||||
|
||||
/* Merge results from main and spec path */
|
||||
movaps %xmm11, %xmm4
|
||||
addps %xmm5, %xmm1
|
||||
/* Merge results from main and spec path */
|
||||
movaps %xmm11, %xmm4
|
||||
addps %xmm5, %xmm1
|
||||
|
||||
/* Go to callout */
|
||||
movmskps %xmm10, %edx
|
||||
orps %xmm8, %xmm1
|
||||
andnps %xmm0, %xmm4
|
||||
andps %xmm11, %xmm1
|
||||
movaps %xmm4, %xmm0
|
||||
orps %xmm1, %xmm0
|
||||
/* Go to callout */
|
||||
movmskps %xmm10, %edx
|
||||
orps %xmm8, %xmm1
|
||||
andnps %xmm0, %xmm4
|
||||
andps %xmm11, %xmm1
|
||||
movaps %xmm4, %xmm0
|
||||
orps %xmm1, %xmm0
|
||||
|
||||
/* Return to main vector processing path */
|
||||
jmp L(AUX_BRANCH_RETURN)
|
||||
# LOE rbx rbp r12 r13 r14 r15 eax edx xmm0 xmm12 xmm13
|
||||
/* Return to main vector processing path */
|
||||
jmp L(AUX_BRANCH_RETURN)
|
||||
# LOE rbx rbp r12 r13 r14 r15 eax edx xmm0 xmm12 xmm13
|
||||
END(_ZGVbN4vv_atan2f_sse4)
|
||||
|
||||
.section .rodata, "a"
|
||||
.align 16
|
||||
.section .rodata, "a"
|
||||
.align 16
|
||||
|
||||
#ifdef __svml_satan2_data_internal_typedef
|
||||
typedef unsigned int VUINT32;
|
||||
typedef struct {
|
||||
__declspec(align(16)) VUINT32 sZERO[4][1];
|
||||
__declspec(align(16)) VUINT32 sSIGN_MASK[4][1];
|
||||
__declspec(align(16)) VUINT32 sABS_MASK[4][1];
|
||||
__declspec(align(16)) VUINT32 sPIO2[4][1];
|
||||
__declspec(align(16)) VUINT32 sPI[4][1];
|
||||
__declspec(align(16)) VUINT32 sPC8[4][1];
|
||||
__declspec(align(16)) VUINT32 sPC7[4][1];
|
||||
__declspec(align(16)) VUINT32 sPC6[4][1];
|
||||
__declspec(align(16)) VUINT32 sPC5[4][1];
|
||||
__declspec(align(16)) VUINT32 sPC4[4][1];
|
||||
__declspec(align(16)) VUINT32 sPC3[4][1];
|
||||
__declspec(align(16)) VUINT32 sPC2[4][1];
|
||||
__declspec(align(16)) VUINT32 sPC1[4][1];
|
||||
__declspec(align(16)) VUINT32 sPC0[4][1];
|
||||
__declspec(align(16)) VUINT32 iCHK_WORK_SUB[4][1];
|
||||
__declspec(align(16)) VUINT32 iCHK_WORK_CMP[4][1];
|
||||
__declspec(align(16)) VUINT32 sZERO[4][1];
|
||||
__declspec(align(16)) VUINT32 sSIGN_MASK[4][1];
|
||||
__declspec(align(16)) VUINT32 sABS_MASK[4][1];
|
||||
__declspec(align(16)) VUINT32 sPIO2[4][1];
|
||||
__declspec(align(16)) VUINT32 sPI[4][1];
|
||||
__declspec(align(16)) VUINT32 sPC8[4][1];
|
||||
__declspec(align(16)) VUINT32 sPC7[4][1];
|
||||
__declspec(align(16)) VUINT32 sPC6[4][1];
|
||||
__declspec(align(16)) VUINT32 sPC5[4][1];
|
||||
__declspec(align(16)) VUINT32 sPC4[4][1];
|
||||
__declspec(align(16)) VUINT32 sPC3[4][1];
|
||||
__declspec(align(16)) VUINT32 sPC2[4][1];
|
||||
__declspec(align(16)) VUINT32 sPC1[4][1];
|
||||
__declspec(align(16)) VUINT32 sPC0[4][1];
|
||||
__declspec(align(16)) VUINT32 iCHK_WORK_SUB[4][1];
|
||||
__declspec(align(16)) VUINT32 iCHK_WORK_CMP[4][1];
|
||||
} __svml_satan2_data_internal;
|
||||
#endif
|
||||
__svml_satan2_data_internal:
|
||||
.long 0x00000000, 0x00000000, 0x00000000, 0x00000000 // sZERO
|
||||
.align 16
|
||||
.long 0x80000000, 0x80000000, 0x80000000, 0x80000000 // sSIGN_MASK
|
||||
.align 16
|
||||
.long 0x7FFFFFFF, 0x7FFFFFFF, 0x7FFFFFFF, 0x7FFFFFFF // sABS_MASK
|
||||
.align 16
|
||||
.long 0x3FC90FDB, 0x3FC90FDB, 0x3FC90FDB, 0x3FC90FDB // sPIO2
|
||||
.align 16
|
||||
.long 0x40490FDB, 0x40490FDB, 0x40490FDB, 0x40490FDB // sPI
|
||||
.align 16
|
||||
.long 0x3B322CC0, 0x3B322CC0, 0x3B322CC0, 0x3B322CC0 // sA08
|
||||
.align 16
|
||||
.long 0xBC7F2631, 0xBC7F2631, 0xBC7F2631, 0xBC7F2631 // sA07
|
||||
.align 16
|
||||
.long 0x3D2BC384, 0x3D2BC384, 0x3D2BC384, 0x3D2BC384 // sA06
|
||||
.align 16
|
||||
.long 0xBD987629, 0xBD987629, 0xBD987629, 0xBD987629 // sA05
|
||||
.align 16
|
||||
.long 0x3DD96474, 0x3DD96474, 0x3DD96474, 0x3DD96474 // sA04
|
||||
.align 16
|
||||
.long 0xBE1161F8, 0xBE1161F8, 0xBE1161F8, 0xBE1161F8 // sA03
|
||||
.align 16
|
||||
.long 0x3E4CB79F, 0x3E4CB79F, 0x3E4CB79F, 0x3E4CB79F // sA02
|
||||
.align 16
|
||||
.long 0xBEAAAA49, 0xBEAAAA49, 0xBEAAAA49, 0xBEAAAA49 // sA01
|
||||
.align 16
|
||||
.long 0x3F800000, 0x3F800000, 0x3F800000, 0x3F800000 // sA00
|
||||
.align 16
|
||||
.long 0x81000000, 0x81000000, 0x81000000, 0x81000000 //iCHK_WORK_SUB
|
||||
.align 16
|
||||
.long 0xFC000000, 0xFC000000, 0xFC000000, 0xFC000000 //iCHK_WORK_CMP
|
||||
.align 16
|
||||
.type __svml_satan2_data_internal,@object
|
||||
.size __svml_satan2_data_internal,.-__svml_satan2_data_internal
|
||||
.long 0x00000000, 0x00000000, 0x00000000, 0x00000000 // sZERO
|
||||
.align 16
|
||||
.long 0x80000000, 0x80000000, 0x80000000, 0x80000000 // sSIGN_MASK
|
||||
.align 16
|
||||
.long 0x7FFFFFFF, 0x7FFFFFFF, 0x7FFFFFFF, 0x7FFFFFFF // sABS_MASK
|
||||
.align 16
|
||||
.long 0x3FC90FDB, 0x3FC90FDB, 0x3FC90FDB, 0x3FC90FDB // sPIO2
|
||||
.align 16
|
||||
.long 0x40490FDB, 0x40490FDB, 0x40490FDB, 0x40490FDB // sPI
|
||||
.align 16
|
||||
.long 0x3B322CC0, 0x3B322CC0, 0x3B322CC0, 0x3B322CC0 // sA08
|
||||
.align 16
|
||||
.long 0xBC7F2631, 0xBC7F2631, 0xBC7F2631, 0xBC7F2631 // sA07
|
||||
.align 16
|
||||
.long 0x3D2BC384, 0x3D2BC384, 0x3D2BC384, 0x3D2BC384 // sA06
|
||||
.align 16
|
||||
.long 0xBD987629, 0xBD987629, 0xBD987629, 0xBD987629 // sA05
|
||||
.align 16
|
||||
.long 0x3DD96474, 0x3DD96474, 0x3DD96474, 0x3DD96474 // sA04
|
||||
.align 16
|
||||
.long 0xBE1161F8, 0xBE1161F8, 0xBE1161F8, 0xBE1161F8 // sA03
|
||||
.align 16
|
||||
.long 0x3E4CB79F, 0x3E4CB79F, 0x3E4CB79F, 0x3E4CB79F // sA02
|
||||
.align 16
|
||||
.long 0xBEAAAA49, 0xBEAAAA49, 0xBEAAAA49, 0xBEAAAA49 // sA01
|
||||
.align 16
|
||||
.long 0x3F800000, 0x3F800000, 0x3F800000, 0x3F800000 // sA00
|
||||
.align 16
|
||||
.long 0x81000000, 0x81000000, 0x81000000, 0x81000000 // iCHK_WORK_SUB
|
||||
.align 16
|
||||
.long 0xFC000000, 0xFC000000, 0xFC000000, 0xFC000000 // iCHK_WORK_CMP
|
||||
.align 16
|
||||
.type __svml_satan2_data_internal, @object
|
||||
.size __svml_satan2_data_internal, .-__svml_satan2_data_internal
|
||||
|
|
|
|||
Loading…
Reference in New Issue