crypto/bn256/cloudflare: fix asm for dynamic linking (#24476)

When using -buildmode=shared, R15 is clobbered by a global variable
access; use a different register instead.

Fixes: #24439

crypto/bn256/cloudflare/gfp_amd64.s

@@ -49,7 +49,7 @@ TEXT ·gfpNeg(SB),0,$0-16
 	SBBQ 24(DI), R11
 
 	MOVQ $0, AX
-	gfpCarry(R8,R9,R10,R11,AX, R12,R13,R14,R15,BX)
+	gfpCarry(R8,R9,R10,R11,AX, R12,R13,R14,CX,BX)
 
 	MOVQ c+0(FP), DI
 	storeBlock(R8,R9,R10,R11, 0(DI))
@@ -68,7 +68,7 @@ TEXT ·gfpAdd(SB),0,$0-24
 	ADCQ 24(SI), R11
 	ADCQ $0, R12
 
-	gfpCarry(R8,R9,R10,R11,R12, R13,R14,R15,AX,BX)
+	gfpCarry(R8,R9,R10,R11,R12, R13,R14,CX,AX,BX)
 
 	MOVQ c+0(FP), DI
 	storeBlock(R8,R9,R10,R11, 0(DI))
@@ -83,7 +83,7 @@ TEXT ·gfpSub(SB),0,$0-24
 	MOVQ ·p2+0(SB), R12
 	MOVQ ·p2+8(SB), R13
 	MOVQ ·p2+16(SB), R14
-	MOVQ ·p2+24(SB), R15
+	MOVQ ·p2+24(SB), CX
 	MOVQ $0, AX
 
 	SUBQ  0(SI), R8
@@ -94,12 +94,12 @@ TEXT ·gfpSub(SB),0,$0-24
 	CMOVQCC AX, R12
 	CMOVQCC AX, R13
 	CMOVQCC AX, R14
-	CMOVQCC AX, R15
+	CMOVQCC AX, CX
 
 	ADDQ R12, R8
 	ADCQ R13, R9
 	ADCQ R14, R10
-	ADCQ R15, R11
+	ADCQ CX, R11
 
 	MOVQ c+0(FP), DI
 	storeBlock(R8,R9,R10,R11, 0(DI))
@@ -115,7 +115,7 @@ TEXT ·gfpMul(SB),0,$160-24
 
 	mulBMI2(0(DI),8(DI),16(DI),24(DI), 0(SI))
 	storeBlock( R8, R9,R10,R11,  0(SP))
-	storeBlock(R12,R13,R14,R15, 32(SP))
+	storeBlock(R12,R13,R14,CX, 32(SP))
 	gfpReduceBMI2()
 	JMP end
 
@@ -125,5 +125,5 @@ nobmi2Mul:
 
 end:
 	MOVQ c+0(FP), DI
-	storeBlock(R12,R13,R14,R15, 0(DI))
+	storeBlock(R12,R13,R14,CX, 0(DI))
 	RET

crypto/bn256/cloudflare/mul_amd64.h

@@ -165,7 +165,7 @@
 	\
 	\ // Add the 512-bit intermediate to m*N
 	loadBlock(96+stack, R8,R9,R10,R11) \
-	loadBlock(128+stack, R12,R13,R14,R15) \
+	loadBlock(128+stack, R12,R13,R14,CX) \
 	\
 	MOVQ $0, AX \
 	ADDQ 0+stack, R8 \
@@ -175,7 +175,7 @@
 	ADCQ 32+stack, R12 \
 	ADCQ 40+stack, R13 \
 	ADCQ 48+stack, R14 \
-	ADCQ 56+stack, R15 \
+	ADCQ 56+stack, CX \
 	ADCQ $0, AX \
 	\
-	gfpCarry(R12,R13,R14,R15,AX, R8,R9,R10,R11,BX)
+	gfpCarry(R12,R13,R14,CX,AX, R8,R9,R10,R11,BX)

crypto/bn256/cloudflare/mul_bmi2_amd64.h

@@ -29,7 +29,7 @@
 	ADCQ $0, R14 \
 	\
 	MOVQ a2, DX \
-	MOVQ $0, R15 \
+	MOVQ $0, CX \
 	MULXQ 0+rb, AX, BX \
 	ADDQ AX, R10 \
 	ADCQ BX, R11 \
@@ -43,7 +43,7 @@
 	MULXQ 24+rb, AX, BX \
 	ADCQ AX, R13 \
 	ADCQ BX, R14 \
-	ADCQ $0, R15 \
+	ADCQ $0, CX \
 	\
 	MOVQ a3, DX \
 	MULXQ 0+rb, AX, BX \
@@ -52,13 +52,13 @@
 	MULXQ 16+rb, AX, BX \
 	ADCQ AX, R13 \
 	ADCQ BX, R14 \
-	ADCQ $0, R15 \
+	ADCQ $0, CX \
 	MULXQ 8+rb, AX, BX \
 	ADDQ AX, R12 \
 	ADCQ BX, R13 \
 	MULXQ 24+rb, AX, BX \
 	ADCQ AX, R14 \
-	ADCQ BX, R15
+	ADCQ BX, CX
 
 #define gfpReduceBMI2() \
 	\ // m = (T * N') mod R, store m in R8:R9:R10:R11
@@ -106,7 +106,7 @@
 	ADCQ 32(SP), R12 \
 	ADCQ 40(SP), R13 \
 	ADCQ 48(SP), R14 \
-	ADCQ 56(SP), R15 \
+	ADCQ 56(SP), CX \
 	ADCQ $0, AX \
 	\
-	gfpCarry(R12,R13,R14,R15,AX, R8,R9,R10,R11,BX)
+	gfpCarry(R12,R13,R14,CX,AX, R8,R9,R10,R11,BX)