kernel: use gettime(9)/getuptime(9) in lieu of time_second(9)/time_uptime(9)

time_second(9) and time_uptime(9) are widely used in the kernel to
quickly get the system UTC or system uptime as a time_t.  However,
time_t is 64-bit everywhere, so it is not generally safe to use them
on 32-bit platforms: you have a split-read problem if your hardware
cannot perform atomic 64-bit reads.

This patch replaces time_second(9) with gettime(9), a safer successor
interface, throughout the kernel.  Similarly, time_uptime(9) is replaced
with getuptime(9).

There is a performance cost on 32-bit platforms in exchange for
eliminating the split-read problem: instead of two register reads you
now have a lockless read loop to pull the values from the timehands.
This is really not *too* bad in the grand scheme of things, but
compared to what we were doing before it is several times slower.

There is no performance cost on 64-bit (__LP64__) platforms.

With input from visa@, dlg@, and tedu@.

Several bugs squashed by visa@.

ok kettenis@

1 files changed, 2 insertions, 2 deletions

diff --git a/sys/dev/pci/mfii.c b/sys/dev/pci/mfii.c
index 6c399712763..3eadcdf81db 100644
--- a/sys/dev/pci/mfii.c
+++ b/sys/dev/pci/mfii.c
@@ -1,4 +1,4 @@
-/* $OpenBSD: mfii.c,v 1.70 2020/06/24 18:59:30 krw Exp $ */
+/* $OpenBSD: mfii.c,v 1.71 2020/06/24 22:03:41 cheloha Exp $ */
 
 /*
  * Copyright (c) 2012 David Gwynne <dlg@openbsd.org>
@@ -1966,7 +1966,7 @@ mfii_initialise_firmware(struct mfii_softc *sc)
 	htolem32(&iiq->system_request_frame_base_address_hi,
 	    MFII_DMA_DVA(sc->sc_requests) >> 32);
 
-	iiq->timestamp = htole64(time_uptime);
+	iiq->timestamp = htole64(getuptime());
 
 	ccb = scsi_io_get(&sc->sc_iopool, SCSI_NOSLEEP);
 	if (ccb == NULL) {