mirror of
https://sourceware.org/git/glibc.git
synced 2024-11-22 13:00:06 +00:00
x86_64: Fix svml_d_exp28_core_avx512.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
2b7494c4f8
commit
91a317cc97
@ -66,8 +66,7 @@
|
||||
|
||||
#include <sysdep.h>
|
||||
|
||||
.text
|
||||
.section .text.evex512,"ax",@progbits
|
||||
.section .text.evex512, "ax", @progbits
|
||||
ENTRY(_ZGVeN8v_exp2_skx)
|
||||
pushq %rbp
|
||||
cfi_def_cfa_offset(16)
|
||||
@ -79,7 +78,7 @@ ENTRY(_ZGVeN8v_exp2_skx)
|
||||
vmovups poly_coeff5+__svml_dexp2_data_internal_avx512(%rip), %zmm14
|
||||
vmovups poly_coeff6+__svml_dexp2_data_internal_avx512(%rip), %zmm6
|
||||
|
||||
/*
|
||||
/*
|
||||
* Reduced argument
|
||||
* where VREDUCE is available
|
||||
*/
|
||||
@ -89,12 +88,12 @@ ENTRY(_ZGVeN8v_exp2_skx)
|
||||
vmovups poly_coeff3+__svml_dexp2_data_internal_avx512(%rip), %zmm8
|
||||
vmovups __svml_dexp2_data_internal_avx512(%rip), %zmm13
|
||||
|
||||
/* c6*r + c5 */
|
||||
/* c6*r + c5 */
|
||||
vfmadd231pd {rn-sae}, %zmm10, %zmm6, %zmm14
|
||||
vmovups poly_coeff2+__svml_dexp2_data_internal_avx512(%rip), %zmm9
|
||||
vmovups Threshold+__svml_dexp2_data_internal_avx512(%rip), %zmm2
|
||||
|
||||
/*
|
||||
/*
|
||||
*
|
||||
* HA
|
||||
* Variables and constants
|
||||
@ -102,46 +101,46 @@ ENTRY(_ZGVeN8v_exp2_skx)
|
||||
*/
|
||||
vmovups poly_coeff1+__svml_dexp2_data_internal_avx512(%rip), %zmm11
|
||||
|
||||
/* c6*r^2 + c5*r + c4 */
|
||||
/* c6*r^2 + c5*r + c4 */
|
||||
vfmadd213pd {rn-sae}, %zmm7, %zmm10, %zmm14
|
||||
|
||||
/*
|
||||
/*
|
||||
* Integer form of K+0.b1b2b3b4 in lower bits - call K_plus_f0
|
||||
* Mantisssa of normalized double precision FP: 1.b1b2...b52
|
||||
*/
|
||||
vaddpd {rd-sae}, %zmm3, %zmm0, %zmm4
|
||||
vandpd AbsMask+__svml_dexp2_data_internal_avx512(%rip), %zmm0, %zmm1
|
||||
|
||||
/* c6*r^3 + c5*r^2 + c4*r + c3 */
|
||||
/* c6*r^3 + c5*r^2 + c4*r + c3 */
|
||||
vfmadd213pd {rn-sae}, %zmm8, %zmm10, %zmm14
|
||||
vcmppd $29, {sae}, %zmm2, %zmm1, %k0
|
||||
|
||||
/* c6*r^4 + c5*r^3 + c4*r^2 + c3*r + c2 */
|
||||
/* c6*r^4 + c5*r^3 + c4*r^2 + c3*r + c2 */
|
||||
vfmadd213pd {rn-sae}, %zmm9, %zmm10, %zmm14
|
||||
kmovw %k0, %edx
|
||||
|
||||
/* c6*r^5 + c5*r^4 + c4*r^3 + c3*r^2 + c2*r + c1 */
|
||||
/* c6*r^5 + c5*r^4 + c4*r^3 + c3*r^2 + c2*r + c1 */
|
||||
vfmadd213pd {rn-sae}, %zmm11, %zmm10, %zmm14
|
||||
|
||||
/* Table value: 2^(0.b1b2b3b4) */
|
||||
/* Table value: 2^(0.b1b2b3b4) */
|
||||
vpandq _lIndexMask+__svml_dexp2_data_internal_avx512(%rip), %zmm4, %zmm5
|
||||
vpermt2pd Frac_PowerD0+64+__svml_dexp2_data_internal_avx512(%rip), %zmm5, %zmm13
|
||||
|
||||
/* T*r */
|
||||
/* T*r */
|
||||
vmulpd {rn-sae}, %zmm10, %zmm13, %zmm12
|
||||
|
||||
/* T + (T*r*(c6*r^5 + c5*r^4 + c4*r^3 + c3*r^2 + c2*r + c1)) */
|
||||
/* T + (T*r*(c6*r^5 + c5*r^4 + c4*r^3 + c3*r^2 + c2*r + c1)) */
|
||||
vfmadd213pd {rn-sae}, %zmm13, %zmm12, %zmm14
|
||||
|
||||
/* Scaling placed at the end to avoid accuracy loss when T*r*scale underflows */
|
||||
/* Scaling placed at the end to avoid accuracy loss when T*r*scale underflows */
|
||||
vscalefpd {rn-sae}, %zmm0, %zmm14, %zmm1
|
||||
testl %edx, %edx
|
||||
|
||||
/* Go to special inputs processing branch */
|
||||
/* Go to special inputs processing branch */
|
||||
jne L(SPECIAL_VALUES_BRANCH)
|
||||
# LOE rbx r12 r13 r14 r15 edx zmm0 zmm1
|
||||
|
||||
/* Restore registers
|
||||
/* Restore registers
|
||||
* and exit the function
|
||||
*/
|
||||
|
||||
@ -155,7 +154,7 @@ L(EXIT):
|
||||
cfi_def_cfa(6, 16)
|
||||
cfi_offset(6, -16)
|
||||
|
||||
/* Branch to process
|
||||
/* Branch to process
|
||||
* special inputs
|
||||
*/
|
||||
|
||||
@ -181,18 +180,18 @@ L(SPECIAL_VALUES_BRANCH):
|
||||
.cfi_escape 0x10, 0x0e, 0x0e, 0x38, 0x1c, 0x0d, 0xc0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0x40, 0xff, 0xff, 0xff, 0x22
|
||||
# LOE rbx 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 r15 r12d r13d
|
||||
|
||||
/* Special inputs
|
||||
/* Special inputs
|
||||
* processing loop
|
||||
*/
|
||||
|
||||
@ -200,7 +199,7 @@ L(SPECIAL_VALUES_LOOP):
|
||||
incl %r12d
|
||||
cmpl $8, %r12d
|
||||
|
||||
/* Check bits in range mask */
|
||||
/* Check bits in range mask */
|
||||
jl L(RANGEMASK_CHECK)
|
||||
# LOE rbx r15 r12d r13d
|
||||
|
||||
@ -212,7 +211,7 @@ L(SPECIAL_VALUES_LOOP):
|
||||
cfi_restore(14)
|
||||
vmovups 128(%rsp), %zmm1
|
||||
|
||||
/* Go to exit */
|
||||
/* Go to exit */
|
||||
jmp L(EXIT)
|
||||
/* DW_CFA_expression: r12 (r12) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -64; DW_OP_and; DW_OP_const4s: -176; DW_OP_plus) */
|
||||
.cfi_escape 0x10, 0x0c, 0x0e, 0x38, 0x1c, 0x0d, 0xc0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0x50, 0xff, 0xff, 0xff, 0x22
|
||||
@ -222,19 +221,19 @@ L(SPECIAL_VALUES_LOOP):
|
||||
.cfi_escape 0x10, 0x0e, 0x0e, 0x38, 0x1c, 0x0d, 0xc0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0x40, 0xff, 0xff, 0xff, 0x22
|
||||
# LOE rbx r12 r13 r14 r15 zmm1
|
||||
|
||||
/* Scalar math fucntion call
|
||||
/* Scalar math fucntion call
|
||||
* to process special input
|
||||
*/
|
||||
|
||||
L(SCALAR_MATH_CALL):
|
||||
movl %r12d, %r14d
|
||||
movsd 64(%rsp,%r14,8), %xmm0
|
||||
movsd 64(%rsp, %r14, 8), %xmm0
|
||||
call exp2@PLT
|
||||
# LOE rbx r14 r15 r12d r13d xmm0
|
||||
|
||||
movsd %xmm0, 128(%rsp,%r14,8)
|
||||
movsd %xmm0, 128(%rsp, %r14, 8)
|
||||
|
||||
/* Process special inputs in loop */
|
||||
/* Process special inputs in loop */
|
||||
jmp L(SPECIAL_VALUES_LOOP)
|
||||
# LOE rbx r15 r12d r13d
|
||||
END(_ZGVeN8v_exp2_skx)
|
||||
@ -259,7 +258,7 @@ typedef struct {
|
||||
} __svml_dexp2_data_internal_avx512;
|
||||
#endif
|
||||
__svml_dexp2_data_internal_avx512:
|
||||
/*== Frac_PowerD0 ==*/
|
||||
/* Frac_PowerD0 */
|
||||
.quad 0x3FF0000000000000
|
||||
.quad 0x3FF0B5586CF9890F
|
||||
.quad 0x3FF172B83C7D517B
|
||||
@ -277,17 +276,17 @@ __svml_dexp2_data_internal_avx512:
|
||||
.quad 0x3FFD5818DCFBA487
|
||||
.quad 0x3FFEA4AFA2A490DA
|
||||
.align 64
|
||||
.quad 0x3FE62E42FEFA398B, 0x3FE62E42FEFA398B, 0x3FE62E42FEFA398B, 0x3FE62E42FEFA398B, 0x3FE62E42FEFA398B, 0x3FE62E42FEFA398B, 0x3FE62E42FEFA398B, 0x3FE62E42FEFA398B /*== poly_coeff1 ==*/
|
||||
.quad 0x3FE62E42FEFA398B, 0x3FE62E42FEFA398B, 0x3FE62E42FEFA398B, 0x3FE62E42FEFA398B, 0x3FE62E42FEFA398B, 0x3FE62E42FEFA398B, 0x3FE62E42FEFA398B, 0x3FE62E42FEFA398B /* == poly_coeff1 == */
|
||||
.align 64
|
||||
.quad 0x3FCEBFBDFF84555A, 0x3FCEBFBDFF84555A, 0x3FCEBFBDFF84555A, 0x3FCEBFBDFF84555A, 0x3FCEBFBDFF84555A, 0x3FCEBFBDFF84555A, 0x3FCEBFBDFF84555A, 0x3FCEBFBDFF84555A /*== poly_coeff2 ==*/
|
||||
.quad 0x3FCEBFBDFF84555A, 0x3FCEBFBDFF84555A, 0x3FCEBFBDFF84555A, 0x3FCEBFBDFF84555A, 0x3FCEBFBDFF84555A, 0x3FCEBFBDFF84555A, 0x3FCEBFBDFF84555A, 0x3FCEBFBDFF84555A /* == poly_coeff2 == */
|
||||
.align 64
|
||||
.quad 0x3FAC6B08D4AD86B9, 0x3FAC6B08D4AD86B9, 0x3FAC6B08D4AD86B9, 0x3FAC6B08D4AD86B9, 0x3FAC6B08D4AD86B9, 0x3FAC6B08D4AD86B9, 0x3FAC6B08D4AD86B9, 0x3FAC6B08D4AD86B9 /*== poly_coeff3 ==*/
|
||||
.quad 0x3FAC6B08D4AD86B9, 0x3FAC6B08D4AD86B9, 0x3FAC6B08D4AD86B9, 0x3FAC6B08D4AD86B9, 0x3FAC6B08D4AD86B9, 0x3FAC6B08D4AD86B9, 0x3FAC6B08D4AD86B9, 0x3FAC6B08D4AD86B9 /* == poly_coeff3 == */
|
||||
.align 64
|
||||
.quad 0x3F83B2AD1B172252, 0x3F83B2AD1B172252, 0x3F83B2AD1B172252, 0x3F83B2AD1B172252, 0x3F83B2AD1B172252, 0x3F83B2AD1B172252, 0x3F83B2AD1B172252, 0x3F83B2AD1B172252 /*== poly_coeff4 ==*/
|
||||
.quad 0x3F83B2AD1B172252, 0x3F83B2AD1B172252, 0x3F83B2AD1B172252, 0x3F83B2AD1B172252, 0x3F83B2AD1B172252, 0x3F83B2AD1B172252, 0x3F83B2AD1B172252, 0x3F83B2AD1B172252 /* == poly_coeff4 == */
|
||||
.align 64
|
||||
.quad 0x3F55D7472713CD19, 0x3F55D7472713CD19, 0x3F55D7472713CD19, 0x3F55D7472713CD19, 0x3F55D7472713CD19, 0x3F55D7472713CD19, 0x3F55D7472713CD19, 0x3F55D7472713CD19 /*== poly_coeff5 ==*/
|
||||
.quad 0x3F55D7472713CD19, 0x3F55D7472713CD19, 0x3F55D7472713CD19, 0x3F55D7472713CD19, 0x3F55D7472713CD19, 0x3F55D7472713CD19, 0x3F55D7472713CD19, 0x3F55D7472713CD19 /* == poly_coeff5 == */
|
||||
.align 64
|
||||
.quad 0x3F24A1D7F526371B, 0x3F24A1D7F526371B, 0x3F24A1D7F526371B, 0x3F24A1D7F526371B, 0x3F24A1D7F526371B, 0x3F24A1D7F526371B, 0x3F24A1D7F526371B, 0x3F24A1D7F526371B /*== poly_coeff6 ==*/
|
||||
.quad 0x3F24A1D7F526371B, 0x3F24A1D7F526371B, 0x3F24A1D7F526371B, 0x3F24A1D7F526371B, 0x3F24A1D7F526371B, 0x3F24A1D7F526371B, 0x3F24A1D7F526371B, 0x3F24A1D7F526371B /* == poly_coeff6 == */
|
||||
.align 64
|
||||
.quad 0x42F8000000000000, 0x42F8000000000000, 0x42F8000000000000, 0x42F8000000000000, 0x42F8000000000000, 0x42F8000000000000, 0x42F8000000000000, 0x42F8000000000000 /* add_const */
|
||||
.align 64
|
||||
@ -297,5 +296,5 @@ __svml_dexp2_data_internal_avx512:
|
||||
.align 64
|
||||
.quad 0x000000000000000F, 0x000000000000000F, 0x000000000000000F, 0x000000000000000F, 0x000000000000000F, 0x000000000000000F, 0x000000000000000F, 0x000000000000000F /* _lIndexMask */
|
||||
.align 64
|
||||
.type __svml_dexp2_data_internal_avx512,@object
|
||||
.size __svml_dexp2_data_internal_avx512,.-__svml_dexp2_data_internal_avx512
|
||||
.type __svml_dexp2_data_internal_avx512, @object
|
||||
.size __svml_dexp2_data_internal_avx512, .-__svml_dexp2_data_internal_avx512
|
||||
|
||||
Loading…
Reference in New Issue
Block a user