[IA64] ar.itc access must really be after xtime_lock.sequence has bee…

…n read

The ".acq" semantics of the load only apply w.r.t. other data access.
Reading the clock (ar.itc) isn't a data access so strange things can
happen here.  Specifically the read of ar.itc can be launched as soon
as the read of xtime_lock.sequence is ISSUED.  Since this may cache
miss, and that might cause a thread switch, and there may be cache
contention for the line containing xtime_lock, it may be a long time
before the actual value is returned, so the ar.itc value may be very
stale.

Move the consumption of r28 up before the read of ar.itc to make sure
that we really have got the current value of xtime_lock.sequence
before look at ar.itc.

Signed-off-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>

arch/ia64/kernel/fsys.S

@@ -247,6 +247,9 @@ ENTRY(fsys_gettimeofday)
 .time_redo:
 	.pred.rel.mutex p8,p9,p10
 	ld4.acq r28 = [r29]	// xtime_lock.sequence. Must come first for locking purposes
+	;;
+	and r28 = ~1,r28	// Make sequence even to force retry if odd
+	;;
 (p8)	mov r2 = ar.itc		// CPU_TIMER. 36 clocks latency!!!
 	add r22 = IA64_TIME_INTERPOLATOR_LAST_COUNTER_OFFSET,r20
 (p9)	ld8 r2 = [r30]		// readq(ti->address). Could also have latency issues..
@@ -284,7 +287,6 @@ EX(.fail_efault, probe.w.fault r31, 3)	// This takes 5 cycles and we have spare
 (p15)	ld8 r17 = [r19],-IA64_TIMESPEC_TV_NSEC_OFFSET
 (p7)	cmp.ne p7,p0 = r25,r3	// if cmpxchg not successful redo
 	// simulate tbit.nz.or p7,p0 = r28,0
-	and r28 = ~1,r28	// Make sequence even to force retry if odd
 	getf.sig r2 = f8
 	mf
 	add r8 = r8,r18		// Add time interpolator offset