x86_64: Fix svml_s_tanf4_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>

1 files changed, 2508 insertions, 2511 deletions

diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_tanf4_core_sse4.S b/sysdeps/x86_64/fpu/multiarch/svml_s_tanf4_core_sse4.S
index 9e3e672ecf..3dc82cae68 100644
--- a/sysdeps/x86_64/fpu/multiarch/svml_s_tanf4_core_sse4.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_s_tanf4_core_sse4.S
@@ -47,2554 +47,2551 @@
 
 /* Offsets for data table __svml_stan_data_internal
  */
-#define _sInvPI_uisa                  	0
-#define _sPI1_uisa                    	16
-#define _sPI2_uisa                    	32
-#define _sPI3_uisa                    	48
-#define _sPI2_ha_uisa                 	64
-#define _sPI3_ha_uisa                 	80
-#define Th_tbl_uisa                   	96
-#define Tl_tbl_uisa                   	224
-#define _sPC3_uisa                    	352
-#define _sPC5_uisa                    	368
-#define _sRangeReductionVal_uisa      	384
-#define _sInvPi                       	400
-#define _sSignMask                    	416
-#define _sAbsMask                     	432
-#define _sRangeVal                    	448
-#define _sRShifter                    	464
-#define _sOne                         	480
-#define _sRangeReductionVal           	496
-#define _sPI1                         	512
-#define _sPI2                         	528
-#define _sPI3                         	544
-#define _sPI4                         	560
-#define _sPI1_FMA                     	576
-#define _sPI2_FMA                     	592
-#define _sPI3_FMA                     	608
-#define _sP0                          	624
-#define _sP1                          	640
-#define _sQ0                          	656
-#define _sQ1                          	672
-#define _sQ2                          	688
-#define _sTwo                         	704
-#define _sCoeffs                      	720
+#define _sInvPI_uisa			0
+#define _sPI1_uisa			16
+#define _sPI2_uisa			32
+#define _sPI3_uisa			48
+#define _sPI2_ha_uisa			64
+#define _sPI3_ha_uisa			80
+#define Th_tbl_uisa			96
+#define Tl_tbl_uisa			224
+#define _sPC3_uisa			352
+#define _sPC5_uisa			368
+#define _sRangeReductionVal_uisa	384
+#define _sInvPi				400
+#define _sSignMask			416
+#define _sAbsMask			432
+#define _sRangeVal			448
+#define _sRShifter			464
+#define _sOne				480
+#define _sRangeReductionVal		496
+#define _sPI1				512
+#define _sPI2				528
+#define _sPI3				544
+#define _sPI4				560
+#define _sPI1_FMA			576
+#define _sPI2_FMA			592
+#define _sPI3_FMA			608
+#define _sP0				624
+#define _sP1				640
+#define _sQ0				656
+#define _sQ1				672
+#define _sQ2				688
+#define _sTwo				704
+#define _sCoeffs			720
 
 #include <sysdep.h>
 
-        .text
-	.section .text.sse4,"ax",@progbits
+	.section .text.sse4, "ax", @progbits
 ENTRY(_ZGVbN4v_tanf_sse4)
-        subq      $232, %rsp
-        cfi_def_cfa_offset(240)
-        movaps    %xmm0, %xmm13
-        movups    _sAbsMask+__svml_stan_data_internal(%rip), %xmm12
-
-/*
- * Legacy Code
- * Here HW FMA can be unavailable
- */
-        xorl      %eax, %eax
-        movaps    %xmm12, %xmm4
-        pxor      %xmm10, %xmm10
-        movups    _sInvPi+__svml_stan_data_internal(%rip), %xmm2
-        andps     %xmm13, %xmm4
-        mulps     %xmm4, %xmm2
-
-/* Range reduction */
-        movaps    %xmm4, %xmm1
-
-/*
- *
- * Main path (_LA_ and _EP_)
- *
- * Octant calculation
- */
-        movups    _sRShifter+__svml_stan_data_internal(%rip), %xmm3
-
-/* Large values check */
-        movaps    %xmm4, %xmm11
-        movups    _sPI1+__svml_stan_data_internal(%rip), %xmm5
-        andnps    %xmm13, %xmm12
-        movups    _sPI2+__svml_stan_data_internal(%rip), %xmm6
-        addps     %xmm3, %xmm2
-        cmpnleps  _sRangeReductionVal+__svml_stan_data_internal(%rip), %xmm11
-        movaps    %xmm2, %xmm8
-        movups    _sPI3+__svml_stan_data_internal(%rip), %xmm7
-        subps     %xmm3, %xmm8
-        movmskps  %xmm11, %edx
-        movups    _sPI4+__svml_stan_data_internal(%rip), %xmm9
-        mulps     %xmm8, %xmm5
-        mulps     %xmm8, %xmm6
-        mulps     %xmm8, %xmm7
-        subps     %xmm5, %xmm1
-        mulps     %xmm8, %xmm9
-        subps     %xmm6, %xmm1
-        movups    _sQ2+__svml_stan_data_internal(%rip), %xmm15
-
-/* Inversion mask and sign calculation */
-        movaps    %xmm2, %xmm5
-
-/* Rational approximation */
-        movups    _sP1+__svml_stan_data_internal(%rip), %xmm14
-        pslld     $30, %xmm2
-        cmpneqps  %xmm10, %xmm2
-        subps     %xmm7, %xmm1
-
-/* Exchanged numerator and denominator if necessary */
-        movaps    %xmm2, %xmm0
-        movaps    %xmm2, %xmm10
-        pslld     $31, %xmm5
-        subps     %xmm9, %xmm1
-        movaps    %xmm1, %xmm3
-        pxor      %xmm12, %xmm5
-        mulps     %xmm1, %xmm3
-        mulps     %xmm3, %xmm15
-        mulps     %xmm3, %xmm14
-        addps     _sQ1+__svml_stan_data_internal(%rip), %xmm15
-        addps     _sP0+__svml_stan_data_internal(%rip), %xmm14
-        mulps     %xmm15, %xmm3
-        mulps     %xmm14, %xmm1
-        addps     _sQ0+__svml_stan_data_internal(%rip), %xmm3
-        andnps    %xmm1, %xmm0
-        andps     %xmm3, %xmm10
-        andps     %xmm2, %xmm1
-        andnps    %xmm3, %xmm2
-        orps      %xmm10, %xmm0
-        orps      %xmm2, %xmm1
-
-/* Division */
-        divps     %xmm1, %xmm0
-
-/* Sign setting */
-        pxor      %xmm5, %xmm0
-
-/*
- *
- * End of main path (_LA_ and _EP_)
- */
-
-        testl     %edx, %edx
-
-/* Go to auxilary branch */
-        jne       L(AUX_BRANCH)
-                                # LOE rbx rbp r12 r13 r14 r15 eax xmm0 xmm4 xmm11 xmm12 xmm13
-
-/* Return from auxilary branch
- * for out of main path inputs
- */
+	subq	$232, %rsp
+	cfi_def_cfa_offset(240)
+	movaps	%xmm0, %xmm13
+	movups	_sAbsMask+__svml_stan_data_internal(%rip), %xmm12
+
+	/*
+	 * Legacy Code
+	 * Here HW FMA can be unavailable
+	 */
+	xorl	%eax, %eax
+	movaps	%xmm12, %xmm4
+	pxor	%xmm10, %xmm10
+	movups	_sInvPi+__svml_stan_data_internal(%rip), %xmm2
+	andps	%xmm13, %xmm4
+	mulps	%xmm4, %xmm2
+
+	/* Range reduction */
+	movaps	%xmm4, %xmm1
+
+	/*
+	 *
+	 * Main path (_LA_ and _EP_)
+	 *
+	 * Octant calculation
+	 */
+	movups	_sRShifter+__svml_stan_data_internal(%rip), %xmm3
+
+	/* Large values check */
+	movaps	%xmm4, %xmm11
+	movups	_sPI1+__svml_stan_data_internal(%rip), %xmm5
+	andnps	%xmm13, %xmm12
+	movups	_sPI2+__svml_stan_data_internal(%rip), %xmm6
+	addps	%xmm3, %xmm2
+	cmpnleps _sRangeReductionVal+__svml_stan_data_internal(%rip), %xmm11
+	movaps	%xmm2, %xmm8
+	movups	_sPI3+__svml_stan_data_internal(%rip), %xmm7
+	subps	%xmm3, %xmm8
+	movmskps %xmm11, %edx
+	movups	_sPI4+__svml_stan_data_internal(%rip), %xmm9
+	mulps	%xmm8, %xmm5
+	mulps	%xmm8, %xmm6
+	mulps	%xmm8, %xmm7
+	subps	%xmm5, %xmm1
+	mulps	%xmm8, %xmm9
+	subps	%xmm6, %xmm1
+	movups	_sQ2+__svml_stan_data_internal(%rip), %xmm15
+
+	/* Inversion mask and sign calculation */
+	movaps	%xmm2, %xmm5
+
+	/* Rational approximation */
+	movups	_sP1+__svml_stan_data_internal(%rip), %xmm14
+	pslld	$30, %xmm2
+	cmpneqps %xmm10, %xmm2
+	subps	%xmm7, %xmm1
+
+	/* Exchanged numerator and denominator if necessary */
+	movaps	%xmm2, %xmm0
+	movaps	%xmm2, %xmm10
+	pslld	$31, %xmm5
+	subps	%xmm9, %xmm1
+	movaps	%xmm1, %xmm3
+	pxor	%xmm12, %xmm5
+	mulps	%xmm1, %xmm3
+	mulps	%xmm3, %xmm15
+	mulps	%xmm3, %xmm14
+	addps	_sQ1+__svml_stan_data_internal(%rip), %xmm15
+	addps	_sP0+__svml_stan_data_internal(%rip), %xmm14
+	mulps	%xmm15, %xmm3
+	mulps	%xmm14, %xmm1
+	addps	_sQ0+__svml_stan_data_internal(%rip), %xmm3
+	andnps	%xmm1, %xmm0
+	andps	%xmm3, %xmm10
+	andps	%xmm2, %xmm1
+	andnps	%xmm3, %xmm2
+	orps	%xmm10, %xmm0
+	orps	%xmm2, %xmm1
+
+	/* Division */
+	divps	%xmm1, %xmm0
+
+	/* Sign setting */
+	pxor	%xmm5, %xmm0
+
+	/*
+	 *
+	 * End of main path (_LA_ and _EP_)
+	 */
+
+	testl	%edx, %edx
+
+	/* Go to auxilary branch */
+	jne	L(AUX_BRANCH)
+	# LOE rbx rbp r12 r13 r14 r15 eax xmm0 xmm4 xmm11 xmm12 xmm13
+
+	/* Return from auxilary branch
+	 * for out of main path inputs
+	 */
 
 L(AUX_BRANCH_RETURN):
-        testl     %eax, %eax
+	testl	%eax, %eax
 
-/* Go to special inputs processing branch */
-        jne       L(SPECIAL_VALUES_BRANCH)
-                                # LOE rbx rbp r12 r13 r14 r15 eax xmm0 xmm13
+	/* Go to special inputs processing branch */
+	jne	L(SPECIAL_VALUES_BRANCH)
+	# LOE rbx rbp r12 r13 r14 r15 eax xmm0 xmm13
 
-/* Restore registers
- * and exit the function
- */
+	/* Restore registers
+	 * and exit the function
+	 */
 
 L(EXIT):
-        addq      $232, %rsp
-        cfi_def_cfa_offset(8)
-        ret
-        cfi_def_cfa_offset(240)
+	addq	$232, %rsp
+	cfi_def_cfa_offset(8)
+	ret
+	cfi_def_cfa_offset(240)
 
-/* Branch to process
- * special inputs
- */
+	/* Branch to process
+	 * special inputs
+	 */
 
 L(SPECIAL_VALUES_BRANCH):
-        movups    %xmm13, 32(%rsp)
-        movups    %xmm0, 48(%rsp)
-                                # LOE rbx rbp r12 r13 r14 r15 eax xmm0
-
-        xorl      %edx, %edx
-        movq      %r12, 16(%rsp)
-        cfi_offset(12, -224)
-        movl      %edx, %r12d
-        movq      %r13, 8(%rsp)
-        cfi_offset(13, -232)
-        movl      %eax, %r13d
-        movq      %r14, (%rsp)
-        cfi_offset(14, -240)
-                                # LOE rbx rbp r15 r12d r13d
-
-/* Range mask
- * bits check
- */
+	movups	%xmm13, 32(%rsp)
+	movups	%xmm0, 48(%rsp)
+	# LOE rbx rbp r12 r13 r14 r15 eax xmm0
+
+	xorl	%edx, %edx
+	movq	%r12, 16(%rsp)
+	cfi_offset(12, -224)
+	movl	%edx, %r12d
+	movq	%r13, 8(%rsp)
+	cfi_offset(13, -232)
+	movl	%eax, %r13d
+	movq	%r14, (%rsp)
+	cfi_offset(14, -240)
+	# LOE rbx rbp 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 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
-
-/* 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    48(%rsp), %xmm0
-
-/* Go to exit */
-        jmp       L(EXIT)
-        cfi_offset(12, -224)
-        cfi_offset(13, -232)
-        cfi_offset(14, -240)
-                                # LOE rbx rbp r12 r13 r14 r15 xmm0
-
-/* Scalar math fucntion call
- * to process special input
- */
+	incl	%r12d
+	cmpl	$4, %r12d
+
+	/* 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	48(%rsp), %xmm0
+
+	/* Go to exit */
+	jmp	L(EXIT)
+	cfi_offset(12, -224)
+	cfi_offset(13, -232)
+	cfi_offset(14, -240)
+	# LOE rbx rbp r12 r13 r14 r15 xmm0
+
+	/* Scalar math fucntion call
+	 * to process special input
+	 */
 
 L(SCALAR_MATH_CALL):
-        movl      %r12d, %r14d
-        movss     32(%rsp,%r14,4), %xmm0
-        call      tanf@PLT
-                                # LOE rbx rbp r14 r15 r12d r13d xmm0
-
-        movss     %xmm0, 48(%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
-
-/* Auxilary branch
- * for out of main path inputs
- */
+	movl	%r12d, %r14d
+	movss	32(%rsp, %r14, 4), %xmm0
+	call	tanf@PLT
+	# LOE rbx rbp r14 r15 r12d r13d xmm0
 
-L(AUX_BRANCH):
-        movl      $2139095040, %eax
+	movss	%xmm0, 48(%rsp, %r14, 4)
 
-/*
- * Get the (2^a / 2pi) mod 1 values from the table.
- * Because doesn't have I-type gather, we need a trivial cast
- */
-        lea       __svml_stan_reduction_data_internal(%rip), %r8
-        movups    %xmm13, 64(%rsp)
+	/* 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
 
-/*
- * Also get the significand as an integer
- * NB: adding in the integer bit is wrong for denorms!
- * To make this work for denorms we should do something slightly different
- */
-        movl      $8388607, %r9d
-        movups    %xmm12, 80(%rsp)
-        movl      $8388608, %r10d
-        movups    %xmm11, 96(%rsp)
+	/* Auxilary branch
+	 * for out of main path inputs
+	 */
 
-/*
- * Break the P_xxx and m into 16-bit chunks ready for
- * the long multiplication via 16x16->32 multiplications
- */
-        movl      $65535, %r11d
-        movd      %eax, %xmm3
-        pshufd    $0, %xmm3, %xmm2
-        andps     %xmm2, %xmm13
-        cmpeqps   %xmm2, %xmm13
-        pand      %xmm4, %xmm2
-        psrld     $23, %xmm2
-        movdqa    %xmm2, %xmm12
-        pslld     $1, %xmm12
-        paddd     %xmm2, %xmm12
-        pslld     $2, %xmm12
-        pshufd    $1, %xmm12, %xmm10
-        pshufd    $2, %xmm12, %xmm11
-        pshufd    $3, %xmm12, %xmm14
-        movd      %xmm12, %edx
-        movd      %xmm10, %ecx
-        movd      %xmm11, %esi
-        movd      %r9d, %xmm11
-        movd      %xmm14, %edi
-        movd      4(%rdx,%r8), %xmm6
-        movd      4(%rcx,%r8), %xmm7
-        movd      4(%rsi,%r8), %xmm3
-        movl      $872415232, %r9d
-        movd      4(%rdi,%r8), %xmm5
-        punpckldq %xmm7, %xmm6
-        punpckldq %xmm5, %xmm3
-        movd      8(%rdi,%r8), %xmm10
-        movmskps  %xmm13, %eax
-        punpcklqdq %xmm3, %xmm6
-        movd      8(%rdx,%r8), %xmm3
-        movd      8(%rcx,%r8), %xmm2
-        movd      8(%rsi,%r8), %xmm13
-        punpckldq %xmm2, %xmm3
-        punpckldq %xmm10, %xmm13
-        punpcklqdq %xmm13, %xmm3
-        pshufd    $0, %xmm11, %xmm13
-        movdqa    %xmm3, %xmm2
-        movups    %xmm4, 48(%rsp)
-        pand      %xmm4, %xmm13
-        movd      %r10d, %xmm4
-        psrld     $16, %xmm2
-        movd      (%rdx,%r8), %xmm9
-
-/*
- * We want to incorporate the original sign now too.
- * Do it here for convenience in getting the right N value,
- * though we could wait right to the end if we were prepared
- * to modify the sign of N later too.
- * So get the appropriate sign mask now (or sooner).
- */
-        movl      $-2147483648, %edx
-        movd      (%rcx,%r8), %xmm8
-
-/*
- * Create floating-point high part, implicitly adding integer bit 1
- * Incorporate overall sign at this stage too.
- */
-        movl      $1065353216, %ecx
-        movd      (%rsi,%r8), %xmm15
-
-/*
- * Now round at the 2^-8 bit position for reduction mod pi/2^7
- * instead of the original 2pi (but still with the same 2pi scaling).
- * Use a shifter of 2^15 + 2^14.
- * The N we get is our final version; it has an offset of
- * 2^8 because of the implicit integer bit, and anyway for negative
- * starting value it's a 2s complement thing. But we need to mask
- * off the exponent part anyway so it's fine.
- */
-        movl      $1195376640, %esi
-        movd      (%rdi,%r8), %xmm1
-        movl      $511, %r10d
-        movups    %xmm0, 112(%rsp)
-        movd      %r11d, %xmm0
-        pshufd    $0, %xmm4, %xmm12
-        movdqa    %xmm2, %xmm4
-        punpckldq %xmm8, %xmm9
-        paddd     %xmm12, %xmm13
-        punpckldq %xmm1, %xmm15
-        movdqa    %xmm13, %xmm12
-        pshufd    $0, %xmm0, %xmm8
-        movdqa    %xmm6, %xmm0
-        punpcklqdq %xmm15, %xmm9
-        pand      %xmm8, %xmm13
-        movdqa    %xmm9, %xmm14
-        pand      %xmm8, %xmm9
-        movdqa    %xmm13, %xmm10
-        psrld     $16, %xmm14
-        movdqu    %xmm14, 128(%rsp)
-
-/* Now do the big multiplication and carry propagation */
-        movdqa    %xmm9, %xmm14
-        psrlq     $32, %xmm10
-        psrlq     $32, %xmm14
-        movdqa    %xmm13, %xmm15
-        movdqa    %xmm10, %xmm7
-        pmuludq   %xmm9, %xmm15
-        psrld     $16, %xmm0
-        pmuludq   %xmm14, %xmm7
-        movdqu    %xmm9, 144(%rsp)
-        psllq     $32, %xmm7
-        movdqu    .FLT_16(%rip), %xmm9
-        pand      %xmm8, %xmm6
-        pand      %xmm9, %xmm15
-        psrld     $16, %xmm12
-        movdqa    %xmm0, %xmm1
-        por       %xmm7, %xmm15
-        movdqa    %xmm13, %xmm7
-        pand      %xmm8, %xmm3
-        movdqu    %xmm0, 160(%rsp)
-        movdqa    %xmm12, %xmm11
-        movdqu    %xmm15, 208(%rsp)
-        psrlq     $32, %xmm1
-        pmuludq   %xmm0, %xmm7
-        movdqa    %xmm6, %xmm5
-        movdqa    %xmm10, %xmm15
-        movdqa    %xmm12, %xmm0
-        movdqu    %xmm14, 176(%rsp)
-        psrlq     $32, %xmm11
-        movdqu    %xmm1, 192(%rsp)
-        psrlq     $32, %xmm5
-        pmuludq   %xmm1, %xmm15
-        movdqa    %xmm13, %xmm1
-        pmuludq   %xmm3, %xmm0
-        pmuludq   %xmm6, %xmm1
-        pmuludq   %xmm12, %xmm6
-        movdqa    %xmm10, %xmm14
-        psrlq     $32, %xmm3
-        pmuludq   %xmm5, %xmm14
-        pand      %xmm9, %xmm1
-        pmuludq   %xmm11, %xmm3
-        pmuludq   %xmm11, %xmm5
-        psllq     $32, %xmm14
-        pand      %xmm9, %xmm0
-        psllq     $32, %xmm3
-        psrlq     $32, %xmm4
-        por       %xmm14, %xmm1
-        por       %xmm3, %xmm0
-        movdqa    %xmm12, %xmm14
-        movdqa    %xmm11, %xmm3
-        pmuludq   %xmm2, %xmm14
-        pand      %xmm9, %xmm7
-        pmuludq   %xmm4, %xmm3
-        pmuludq   %xmm13, %xmm2
-        pmuludq   %xmm10, %xmm4
-        pand      %xmm9, %xmm2
-        psllq     $32, %xmm4
-        psllq     $32, %xmm15
-        pand      %xmm9, %xmm14
-        psllq     $32, %xmm3
-        por       %xmm4, %xmm2
-        por       %xmm15, %xmm7
-        por       %xmm3, %xmm14
-        psrld     $16, %xmm2
-        pand      %xmm9, %xmm6
-        psllq     $32, %xmm5
-        movdqa    %xmm1, %xmm15
-        paddd     %xmm2, %xmm14
-        movdqa    %xmm7, %xmm2
-        por       %xmm5, %xmm6
-        psrld     $16, %xmm1
-        pand      %xmm8, %xmm2
-        paddd     %xmm1, %xmm6
-        movdqu    160(%rsp), %xmm1
-        paddd     %xmm6, %xmm2
-        movdqu    192(%rsp), %xmm6
-        psrld     $16, %xmm7
-        pmuludq   %xmm12, %xmm1
-        pand      %xmm8, %xmm15
-        pmuludq   %xmm11, %xmm6
-        pmuludq   144(%rsp), %xmm12
-        pmuludq   176(%rsp), %xmm11
-        pand      %xmm9, %xmm1
-        psllq     $32, %xmm6
-        por       %xmm6, %xmm1
-        psrld     $16, %xmm0
-        paddd     %xmm7, %xmm1
-        paddd     %xmm14, %xmm15
-        movdqu    128(%rsp), %xmm7
-        paddd     %xmm15, %xmm0
-        pmuludq   %xmm7, %xmm13
-        psrlq     $32, %xmm7
-        pmuludq   %xmm7, %xmm10
-        movdqa    %xmm0, %xmm14
-        pand      %xmm9, %xmm13
-        movdqu    208(%rsp), %xmm5
-        psrld     $16, %xmm14
-        paddd     %xmm2, %xmm14
-        movdqa    %xmm5, %xmm15
-        movdqa    %xmm14, %xmm3
-        pand      %xmm8, %xmm15
-        psrld     $16, %xmm3
-        paddd     %xmm1, %xmm15
-        psllq     $32, %xmm10
-        pand      %xmm9, %xmm12
-        psllq     $32, %xmm11
-        paddd     %xmm15, %xmm3
-        por       %xmm10, %xmm13
-        por       %xmm11, %xmm12
-        psrld     $16, %xmm5
-        movdqa    %xmm3, %xmm4
-        pand      %xmm8, %xmm13
-        paddd     %xmm5, %xmm12
-        psrld     $16, %xmm4
-        paddd     %xmm12, %xmm13
-        paddd     %xmm13, %xmm4
-        pand      %xmm8, %xmm3
-        pslld     $16, %xmm4
-        movd      %edx, %xmm9
-        movups    48(%rsp), %xmm15
-        paddd     %xmm3, %xmm4
-        pshufd    $0, %xmm9, %xmm7
-
-/* Assemble reduced argument from the pieces */
-        pand      %xmm8, %xmm0
-        movd      %ecx, %xmm8
-        pand      %xmm15, %xmm7
-        pshufd    $0, %xmm8, %xmm1
-        movdqa    %xmm4, %xmm5
-        psrld     $9, %xmm5
-        pxor      %xmm7, %xmm1
-        por       %xmm1, %xmm5
-        movd      %esi, %xmm6
-        pshufd    $0, %xmm6, %xmm3
-        movdqa    %xmm5, %xmm6
-        movl      $262143, %r8d
-
-/*
- * Create floating-point low and medium parts, respectively
- * lo_17, ... lo_0, 0, ..., 0
- * hi_8, ... hi_0, lo_31, ..., lo_18
- * then subtract off the implicitly added integer bits,
- * 2^-46 and 2^-23, respectively.
- * Put the original sign into all of them at this stage.
- */
-        movl      $679477248, %edi
-        movd      %r10d, %xmm13
-        pslld     $16, %xmm14
-        pshufd    $0, %xmm13, %xmm1
-        paddd     %xmm0, %xmm14
-        movd      %r9d, %xmm11
-        pand      %xmm4, %xmm1
-        movd      %r8d, %xmm9
-        movd      %edi, %xmm10
-        pshufd    $0, %xmm9, %xmm8
-        pslld     $14, %xmm1
-        pshufd    $0, %xmm10, %xmm0
-        pand      %xmm14, %xmm8
-        pshufd    $0, %xmm11, %xmm12
-        psrld     $18, %xmm14
-        pxor      %xmm7, %xmm0
-        pxor      %xmm12, %xmm7
-        por       %xmm14, %xmm1
-        pslld     $5, %xmm8
-        por       %xmm7, %xmm1
-
-/*
- * Now multiply those numbers all by 2 pi, reasonably accurately.
- * The top part uses 2pi = s2pi_lead + s2pi_trail, where
- * s2pi_lead has 12 significant bits.
- */
-        movl      $1086918619, %r11d
-
-/* Split RHi into 12-bit leading and trailing parts. */
-        movl      $-4096, %esi
-        por       %xmm0, %xmm8
-        movl      $1086918656, %edx
-        movl      $-1214941318, %ecx
-
-/*
- * If the magnitude of the input is <= 2^-20, then
- * just pass through the input, since no reduction will be needed and
- * the main path will only work accurately if the reduced argument is
- * about >= 2^-40 (which it is for all large pi multiples)
- */
-        movl      $2147483647, %edi
-        addps     %xmm3, %xmm6
-        subps     %xmm7, %xmm1
-        subps     %xm