MOVW	R16, (4*4)(R29)
	MOVW	R17, (4*5)(R29)
	MOVW	R18, (4*6)(R29)
	MOVW	R19, (4*7)(R29)
	MOVW	R20, (4*8)(R29)
	MOVW	R21, (4*9)(R29)
	MOVW	R22, (4*10)(R29)
	MOVW	R23, (4*11)(R29)
	MOVW	g, (4*12)(R29)
	MOVW	R31, (4*13)(R29)
#ifndef GOMIPS_softfloat
	MOVD	F20, (4*14)(R29)
	MOVD	F22, (4*14+8*1)(R29)
	MOVD	F24, (4*14+8*2)(R29)
	MOVD	F26, (4*14+8*3)(R29)
	MOVD	F28, (4*14+8*4)(R29)
	MOVD	F30, (4*14+8*5)(R29)
#endif

	MOVV	R10, 80(R29)
	MOVV	R11, 88(R29)
	MOVV	R12, 96(R29)
	MOVV	R13, 104(R29)
	MOVV	R14, 112(R29)
	MOVV	R15, 120(R29)
	MOVV	R16, 128(R29)
	MOVV	R17, 136(R29)
	MOVV	R18, 144(R29)
	MOVV	R19, 152(R29)
	// R20 already saved
	// R21 already saved.
	MOVV	R22, 160(R29)
	// R23 is tmp register.
	MOVV	R24, 168(R29)
	MOVV	R25, 176(R29)
	// R26 is reserved by kernel.
	// R27 is reserved by kernel.

	MOVV	runtime·vdsoClockgettimeSym(SB), R25
	BEQ	R25, fallback

	JAL	(R25)
	// see walltime for detail
	BEQ	R2, R0, finish
	MOVV	R0, runtime·vdsoClockgettimeSym(SB)
	MOVW	$1, R4 // CLOCK_MONOTONIC
	MOVV	$0(R29), R5
	JMP	fallback

finish:
	MOVV	0(R29), R3	// sec
	MOVV	8(R29), R5	// nsec

	MOVV	R16, R29	// restore SP
	// Restore vdsoPC, vdsoSP

	RET

	// In child, on new stack.
	// Check that SP is as we expect
	NOP	R29	// tell vet R29/SP changed - stop checking offsets
	MOVW	12(R29), R16
	MOVW	$1234, R1
	BEQ	R16, R1, 2(PC)
	MOVW	(R0), R0

	// Initialize m->procid to Linux tid
	MOVW	$SYS_gettid, R2
	SYSCALL

	MOVW	0(R29), R16	// m
	MOVW	4(R29), R17	// g
	MOVW	8(R29), R18	// fn

	BEQ	R16, nog
	BEQ	R17, nog

	MOVW	R2, m_procid(R16)

	JAL	runtime·cgocallback(SB)

	MOVV	(8*4)(R29), R16
	MOVV	(8*5)(R29), R17
	MOVV	(8*6)(R29), R18
	MOVV	(8*7)(R29), R19
	MOVV	(8*8)(R29), R20
	MOVV	(8*9)(R29), R21
	MOVV	(8*10)(R29), R22
	MOVV	(8*11)(R29), R23
	MOVV	(8*12)(R29), RSB
	MOVV	(8*13)(R29), g
	MOVV	(8*14)(R29), R31
#ifndef GOMIPS64_softfloat
	MOVD	(8*15)(R29), F24
	MOVD	(8*16)(R29), F25
	MOVD	(8*17)(R29), F26
	MOVD	(8*18)(R29), F27
	MOVD	(8*19)(R29), F28

	MOVW	R28, 96(R29)
	// R29 is SP.
	// R30 is g.
	// R31 is LR, which was saved by the prologue.

	CALL	runtime·wbBufFlush(SB)

	MOVW	4(R29), R20
	MOVW	8(R29), R21
	MOVW	12(R29), R3
	MOVW	16(R29), R4
	MOVW	20(R29), R5
	MOVW	24(R29), R6
	MOVW	28(R29), R7
	MOVW	32(R29), R8
	MOVW	36(R29), R9
	MOVW	40(R29), R10
	MOVW	44(R29), R11
	MOVW	48(R29), R12
	MOVW	52(R29), R13
	MOVW	56(R29), R14

	MOVW	a1+4(FP), R4
	MOVW	a2+8(FP), R5
	MOVW	a3+12(FP), R6
	MOVW	a4+16(FP), R7
	MOVW	a5+20(FP), R8
	MOVW	a6+24(FP), R9
	MOVW	a7+28(FP), R10
	MOVW	a8+32(FP), R11
	MOVW	R8, 16(R29)
	MOVW	R9, 20(R29)
	MOVW	R10, 24(R29)
	MOVW	R11, 28(R29)
	MOVW	trap+0(FP), R2	// syscall entry
	MOVW	R0, R3	// reset R3 to zero as 1-ret SYSCALL keeps it
	SYSCALL
	BEQ	R7, ok9
	MOVW	$-1, R1
	MOVW	R1, r1+40(FP)	// r1
	MOVW	R0, r2+44(FP)	// r2

	RET

TEXT runtime·usleep(SB),NOSPLIT,$24-4
	MOVWU	usec+0(FP), R3
	MOVV	R3, R5
	MOVW	$1000000, R4
	DIVVU	R4, R3
	MOVV	LO, R3
	MOVV	R3, 8(R29)		// tv_sec
	MOVW	$1000, R4
	MULVU	R3, R4
	MOVV	LO, R4
	SUBVU	R4, R5
	MOVV	R5, 16(R29)		// tv_nsec

	ADDV	$8, R29, R4		// arg 1 - rqtp
	MOVV	$0, R5			// arg 2 - rmtp
	MOVV	$91, R2			// sys_nanosleep
	SYSCALL
	RET

TEXT runtime·getthrid(SB),NOSPLIT,$0-4

package main

/*
#cgo CFLAGS: -Os

int foo_fpr() {
        asm volatile("":::"fr31","fr30","fr29","fr28");
}
int foo_gpr0() {
        asm volatile("":::"r30","r29","r28");
}
int foo_gpr1() {
        asm volatile("":::"fr31", "fr30","fr29","fr28","r30","r29","r28");
}
int foo_vr() {
        asm volatile("":::"v31","v30","v29","v28");
}
*/
import "C"

import "fmt"

func main() {
	C.foo_fpr()

	l.addSpecial(
		mov+" HI, R1\n"+mov+" R1, %d(R29)",
		mov+" %d(R29), R1\n"+mov+" R1, HI",
		regsize)
	l.addSpecial(
		mov+" LO, R1\n"+mov+" R1, %d(R29)",
		mov+" %d(R29), R1\n"+mov+" R1, LO",
		regsize)

	// Add floating point control/status register FCR31 (FCR0-FCR30 are irrelevant)
	var lfp = layout{sp: "R29", stack: l.stack}
	lfp.addSpecial(
		mov+" FCR31, R1\n"+mov+" R1, %d(R29)",
		mov+" %d(R29), R1\n"+mov+" R1, FCR31",