.global smp_trampoline_start
smp_trampoline_start:
        // The AP begins executing here with the following state:
        //      satp            = 0
        //      sstatus.SIE     = 0
        //      a0              = hartid
        //      a1              = struct trampoline_passed_info *
        //
        // All other registers are undefined.

        ld      a0, 16(a1)
        ld      t0, 8(a1)
        csrw    satp, t0

        // Tell the BSP we've started.
        li      t0, 1
        fence   rw, w
        sd      t0, (a1)

        // Zero all the things.
        // Preserve a0
        mv      a1, zero
        mv      a2, zero
        mv      a3, zero
        mv      a4, zero
        mv      a5, zero
        mv      a6, zero
        mv      a7, zero
        mv      s0, zero
        mv      s1, zero
        mv      s2, zero
        mv      s3, zero
        mv      s4, zero
        mv      s5, zero
        mv      s6, zero
        mv      s7, zero
        mv      s8, zero
        mv      s9, zero
        mv      s10, zero
        mv      s11, zero
        mv      t1, zero
        mv      t2, zero
        mv      t3, zero
        mv      t4, zero
        mv      t5, zero
        mv      t6, zero
        mv      tp, zero
        mv      ra, zero

        csrw    sie, zero
        csrw    stvec, zero

        // Wait for kernel to tell us where to go.
0:      .insn i 0x0F, 0, x0, x0, 0x010  // pause
        ld      t0, 24(a0)
        fence   r, rw
        beqz    t0, 0b

        // Load sp from reserved field of info struct
        ld      sp, 16(a0)

        jr      t0