I will soon get a Hinote Ultra loaner and might be in a better
position to advise you.  I use a NEC VERSA E.  I run a may 15 -current.

The suspend resume problem that I saw is a bug in hdcwait(). Below is
a copy of the code.  The problem is that on "errors" wdcwait returns a
1.  Some of the code that calls it; checks for < 0.  vs != 0.  So
there really is no timeout here.

I have an old ATA spec which is a little bit inconsistent.  If the
error bit set is really a MEDIA CHANGE, I guess the hardware is 
anticipating that you might have removed the hard drive around
the suspend/resume.  So this may be a feature.

NOTE: the inittodr(0) is a gross but really convenient hack.

#define new
int
wdcwait(wdc, mask)
	struct wdc_softc *wdc;
	int mask;
{
	int iobase = wdc->sc_iobase;
	int timeout = 0;
	u_char status;
	extern int cold;

#ifdef	new
	wdc->sc_error = 0;
#endif
	for (;;) {
		wdc->sc_status = status = inb(iobase+wd_status);
		if ((status & WDCS_BSY) == 0 && (status & mask) == mask)
			break;
		if (++timeout > WDCNDELAY)
			return -1;

		delay(WDCDELAY);
	}
	if (status & WDCS_ERR) {
		wdc->sc_error = inb(iobase+wd_error);
#ifdef	new
		if (wdc->sc_error & WDCE_MCR) {
			printf("wd: media change\n");
			inittodr(0);		/* hssss */
		}
		return -1;
#else
		return WDCS_ERR;
#endif
	}
#ifdef WDCNDELAY_DEBUG
	/* After autoconfig, there should be no long delays. */
	if (!cold && timeout > WDCNDELAY_DEBUG)
		printf("%s: warning: busy-wait took %dus\n",
		    wdc->sc_dev.dv_xname, WDCDELAY * timeout);
#endif
	return 0;
}
----------------------------------------------------------------------
Next:
I have also seen the seen the Fn+key presses do the wrong thing.  I am
not sure what is wrong here.

I can give you a dlink DE600 driver if that helps.

I've seen a 3c589 driver for linux.

Finally, I'd guess the battery drain is really a hardware problem.
In suspend/resume the machine should really be off.  I presume that
the battery normally lasts 2+ hours.  If it drains in 1 hour here
that would be really strange.