mirror of
https://sourceware.org/git/glibc.git
synced 2024-12-26 04:31:03 +00:00
x86_64: Fix svml_s_sinhf4_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
f03bdaf054
commit
edc4a5012b
@ -50,13 +50,12 @@
|
||||
|
||||
#include <sysdep.h>
|
||||
|
||||
.text
|
||||
.section .text.sse4,"ax",@progbits
|
||||
.section .text.sse4, "ax", @progbits
|
||||
ENTRY(_ZGVbN4v_sinhf_sse4)
|
||||
subq $72, %rsp
|
||||
cfi_def_cfa_offset(80)
|
||||
|
||||
/*
|
||||
/*
|
||||
* Implementation
|
||||
* Abs argument
|
||||
*/
|
||||
@ -64,7 +63,7 @@ ENTRY(_ZGVbN4v_sinhf_sse4)
|
||||
andps %xmm0, %xmm14
|
||||
movaps %xmm14, %xmm10
|
||||
|
||||
/*
|
||||
/*
|
||||
* Load argument
|
||||
* dM = x/log(2) + RShifter
|
||||
*/
|
||||
@ -72,48 +71,48 @@ ENTRY(_ZGVbN4v_sinhf_sse4)
|
||||
pxor %xmm0, %xmm10
|
||||
mulps %xmm10, %xmm7
|
||||
|
||||
/*
|
||||
/*
|
||||
* Check for overflow\underflow
|
||||
* MORE faster than GE?
|
||||
*/
|
||||
movaps %xmm10, %xmm1
|
||||
movups _sShifter+__svml_ssinh_data_internal(%rip), %xmm2
|
||||
|
||||
/* sR = sX - sN*Log2_hi */
|
||||
/* sR = sX - sN*Log2_hi */
|
||||
movups _sLn2hi+__svml_ssinh_data_internal(%rip), %xmm3
|
||||
addps %xmm2, %xmm7
|
||||
|
||||
/*
|
||||
/*
|
||||
* R
|
||||
* sN = sM - RShifter
|
||||
*/
|
||||
movaps %xmm7, %xmm4
|
||||
|
||||
/*
|
||||
* G1,G2 2^N,2^(-N)
|
||||
/*
|
||||
* G1, G2 2^N, 2^(-N)
|
||||
* iM now is an EXP(2^N)
|
||||
*/
|
||||
pslld $23, %xmm7
|
||||
|
||||
/* sR = (sX - sN*Log2_hi) - sN*Log2_lo */
|
||||
/* sR = (sX - sN*Log2_hi) - sN*Log2_lo */
|
||||
movups _sLn2lo+__svml_ssinh_data_internal(%rip), %xmm5
|
||||
subps %xmm2, %xmm4
|
||||
mulps %xmm4, %xmm3
|
||||
mulps %xmm4, %xmm5
|
||||
subps %xmm3, %xmm10
|
||||
|
||||
/*
|
||||
/*
|
||||
* sinh(r) = r*((a1=1)+r^2*(a3+r^2*(a5+{v1 r^2*a7})))) = r + r*(r^2*(a3+r^2*(a5+r^2*a7))) ....
|
||||
* sSinh_r = (a3+r^2*a5)
|
||||
*/
|
||||
movups _sPC5+__svml_ssinh_data_internal(%rip), %xmm8
|
||||
subps %xmm5, %xmm10
|
||||
|
||||
/* sR2 = sR^2 */
|
||||
/* sR2 = sR^2 */
|
||||
movaps %xmm10, %xmm12
|
||||
mulps %xmm10, %xmm12
|
||||
|
||||
/*
|
||||
/*
|
||||
* sinh(X) = sG2 + sG1*sinh(dR) + sG2*sR2*(a2+sR2*(a4+a6*sR2)
|
||||
* sOut = (a4 +a6*sR2)
|
||||
*/
|
||||
@ -123,52 +122,52 @@ ENTRY(_ZGVbN4v_sinhf_sse4)
|
||||
addps _sPC3+__svml_ssinh_data_internal(%rip), %xmm8
|
||||
addps _sPC4+__svml_ssinh_data_internal(%rip), %xmm9
|
||||
|
||||
/* sSinh_r = r^2*(a3+r^2*a5) */
|
||||
/* sSinh_r = r^2*(a3+r^2*a5) */
|
||||
mulps %xmm12, %xmm8
|
||||
|
||||
/* sOut = a2+sR2*(a4+a6*sR2) */
|
||||
/* sOut = a2+sR2*(a4+a6*sR2) */
|
||||
mulps %xmm12, %xmm9
|
||||
|
||||
/* sSinh_r = r + r*(r^2*(a3+r^2*a5)) */
|
||||
/* sSinh_r = r + r*(r^2*(a3+r^2*a5)) */
|
||||
mulps %xmm10, %xmm8
|
||||
addps _sPC2+__svml_ssinh_data_internal(%rip), %xmm9
|
||||
addps %xmm8, %xmm10
|
||||
|
||||
/* sOut = sR2*(a2+sR2*(a4+a6*sR2) */
|
||||
/* sOut = sR2*(a2+sR2*(a4+a6*sR2) */
|
||||
mulps %xmm9, %xmm12
|
||||
movdqu _iHalf+__svml_ssinh_data_internal(%rip), %xmm6
|
||||
movdqa %xmm6, %xmm13
|
||||
psubd %xmm7, %xmm6
|
||||
paddd %xmm7, %xmm13
|
||||
|
||||
/* sG1 = 2^(N-1)+2^(-N-1) */
|
||||
/* sG1 = 2^(N-1)+2^(-N-1) */
|
||||
movdqa %xmm13, %xmm11
|
||||
|
||||
/* sG2 = 2^(N-1)-2^(-N-1) */
|
||||
/* sG2 = 2^(N-1)-2^(-N-1) */
|
||||
subps %xmm6, %xmm13
|
||||
addps %xmm6, %xmm11
|
||||
|
||||
/* sOut = sG2*sR2*(a2+sR2*(a4+a6*sR2) */
|
||||
/* sOut = sG2*sR2*(a2+sR2*(a4+a6*sR2) */
|
||||
mulps %xmm13, %xmm12
|
||||
|
||||
/* sOut = sG1*sinh(dR)+sG2*sR2*(a2+sR2*(a4+a6*sR2) */
|
||||
/* sOut = sG1*sinh(dR)+sG2*sR2*(a2+sR2*(a4+a6*sR2) */
|
||||
mulps %xmm10, %xmm11
|
||||
pcmpgtd _iDomainRange+__svml_ssinh_data_internal(%rip), %xmm1
|
||||
addps %xmm11, %xmm12
|
||||
movmskps %xmm1, %edx
|
||||
|
||||
/* sOut = sG2 + sG1*sinh(dR) + sG2*sR2*(a2+sR2*(a4+a6*sR2) */
|
||||
/* sOut = sG2 + sG1*sinh(dR) + sG2*sR2*(a2+sR2*(a4+a6*sR2) */
|
||||
addps %xmm12, %xmm13
|
||||
|
||||
/* Ret H */
|
||||
/* Ret H */
|
||||
orps %xmm13, %xmm14
|
||||
testl %edx, %edx
|
||||
|
||||
/* Go to special inputs processing branch */
|
||||
/* Go to special inputs processing branch */
|
||||
jne L(SPECIAL_VALUES_BRANCH)
|
||||
# LOE rbx rbp r12 r13 r14 r15 edx xmm0 xmm14
|
||||
|
||||
/* Restore registers
|
||||
/* Restore registers
|
||||
* and exit the function
|
||||
*/
|
||||
|
||||
@ -179,7 +178,7 @@ L(EXIT):
|
||||
ret
|
||||
cfi_def_cfa_offset(80)
|
||||
|
||||
/* Branch to process
|
||||
/* Branch to process
|
||||
* special inputs
|
||||
*/
|
||||
|
||||
@ -199,18 +198,18 @@ L(SPECIAL_VALUES_BRANCH):
|
||||
cfi_offset(14, -80)
|
||||
# LOE rbx rbp r15 r12d r13d
|
||||
|
||||
/* Range mask
|
||||
/* Range mask
|
||||
* bits check
|
||||
*/
|
||||
|
||||
L(RANGEMASK_CHECK):
|
||||
btl %r12d, %r13d
|
||||
|
||||
/* Call scalar math function */
|
||||
/* Call scalar math function */
|
||||
jc L(SCALAR_MATH_CALL)
|
||||
# LOE rbx rbp r15 r12d r13d
|
||||
|
||||
/* Special inputs
|
||||
/* Special inputs
|
||||
* processing loop
|
||||
*/
|
||||
|
||||
@ -218,7 +217,7 @@ L(SPECIAL_VALUES_LOOP):
|
||||
incl %r12d
|
||||
cmpl $4, %r12d
|
||||
|
||||
/* Check bits in range mask */
|
||||
/* Check bits in range mask */
|
||||
jl L(RANGEMASK_CHECK)
|
||||
# LOE rbx rbp r15 r12d r13d
|
||||
|
||||
@ -230,26 +229,26 @@ L(SPECIAL_VALUES_LOOP):
|
||||
cfi_restore(14)
|
||||
movups 48(%rsp), %xmm14
|
||||
|
||||
/* Go to exit */
|
||||
/* Go to exit */
|
||||
jmp L(EXIT)
|
||||
cfi_offset(12, -64)
|
||||
cfi_offset(13, -72)
|
||||
cfi_offset(14, -80)
|
||||
# LOE rbx rbp r12 r13 r14 r15 xmm14
|
||||
|
||||
/* Scalar math fucntion call
|
||||
/* Scalar math fucntion call
|
||||
* to process special input
|
||||
*/
|
||||
|
||||
L(SCALAR_MATH_CALL):
|
||||
movl %r12d, %r14d
|
||||
movss 32(%rsp,%r14,4), %xmm0
|
||||
movss 32(%rsp, %r14, 4), %xmm0
|
||||
call sinhf@PLT
|
||||
# LOE rbx rbp r14 r15 r12d r13d xmm0
|
||||
|
||||
movss %xmm0, 48(%rsp,%r14,4)
|
||||
movss %xmm0, 48(%rsp, %r14, 4)
|
||||
|
||||
/* Process special inputs in loop */
|
||||
/* Process special inputs in loop */
|
||||
jmp L(SPECIAL_VALUES_LOOP)
|
||||
# LOE rbx rbp r15 r12d r13d
|
||||
END(_ZGVbN4v_sinhf_sse4)
|
||||
@ -259,8 +258,7 @@ END(_ZGVbN4v_sinhf_sse4)
|
||||
|
||||
#ifdef __svml_ssinh_data_internal_typedef
|
||||
typedef unsigned int VUINT32;
|
||||
typedef struct
|
||||
{
|
||||
typedef struct {
|
||||
__declspec(align(16)) VUINT32 _sInvLn2[4][1];
|
||||
__declspec(align(16)) VUINT32 _sLn2hi[4][1];
|
||||
__declspec(align(16)) VUINT32 _sLn2lo[4][1];
|
||||
@ -277,7 +275,7 @@ typedef struct
|
||||
} __svml_ssinh_data_internal;
|
||||
#endif
|
||||
__svml_ssinh_data_internal:
|
||||
.long 0x3FB8AA3B, 0x3FB8AA3B, 0x3FB8AA3B, 0x3FB8AA3B /* _sInvLn2 */ //k=0
|
||||
.long 0x3FB8AA3B, 0x3FB8AA3B, 0x3FB8AA3B, 0x3FB8AA3B /* _sInvLn2 */ // k=0
|
||||
.align 16
|
||||
.long 0x3F317000, 0x3F317000, 0x3F317000, 0x3F317000 /* _sLn2hi */
|
||||
.align 16
|
||||
@ -302,7 +300,7 @@ __svml_ssinh_data_internal:
|
||||
.long 0x3ab6a8a3, 0x3ab6a8a3, 0x3ab6a8a3, 0x3ab6a8a3 /* _sPC6 */
|
||||
// Integer constants
|
||||
.align 16
|
||||
.long 0x3f000000, 0x3f000000, 0x3f000000, 0x3f000000 /* _iHalf*/
|
||||
.long 0x3f000000, 0x3f000000, 0x3f000000, 0x3f000000 /* _iHalf */
|
||||
.align 16
|
||||
.type __svml_ssinh_data_internal,@object
|
||||
.size __svml_ssinh_data_internal,.-__svml_ssinh_data_internal
|
||||
.type __svml_ssinh_data_internal, @object
|
||||
.size __svml_ssinh_data_internal, .-__svml_ssinh_data_internal
|
||||
|
||||
Loading…
Reference in New Issue
Block a user