Commit afdb4a5b authored by Helge Deller's avatar Helge Deller

parisc: Mark cr16 CPU clocksource unstable on all SMP machines

In commit c8c37359 ("parisc: Enhance detection of synchronous cr16
clocksources") I assumed that CPUs on the same physical core are syncronous.
While booting up the kernel on two different C8000 machines, one with a
dual-core PA8800 and one with a dual-core PA8900 CPU, this turned out to be
wrong. The symptom was that I saw a jump in the internal clocks printed to the
syslog and strange overall behaviour.  On machines which have 4 cores (2
dual-cores) the problem isn't visible, because the current logic already marked
the cr16 clocksource unstable in this case.

This patch now marks the cr16 interval timers unstable if we have more than one
CPU in the system, and it fixes this issue.

Fixes: c8c37359 ("parisc: Enhance detection of synchronous cr16 clocksources")
Signed-off-by: default avatarHelge Deller <deller@gmx.de>
Cc: <stable@vger.kernel.org> # v5.15+
parent 0f9fee4c
......@@ -249,30 +249,16 @@ void __init time_init(void)
static int __init init_cr16_clocksource(void)
{
/*
* The cr16 interval timers are not syncronized across CPUs on
* different sockets, so mark them unstable and lower rating on
* multi-socket SMP systems.
* The cr16 interval timers are not syncronized across CPUs, even if
* they share the same socket.
*/
if (num_online_cpus() > 1 && !running_on_qemu) {
int cpu;
unsigned long cpu0_loc;
cpu0_loc = per_cpu(cpu_data, 0).cpu_loc;
for_each_online_cpu(cpu) {
if (cpu == 0)
continue;
if ((cpu0_loc != 0) &&
(cpu0_loc == per_cpu(cpu_data, cpu).cpu_loc))
continue;
/* mark sched_clock unstable */
clear_sched_clock_stable();
clocksource_cr16.name = "cr16_unstable";
clocksource_cr16.flags = CLOCK_SOURCE_UNSTABLE;
clocksource_cr16.rating = 0;
break;
}
/* mark sched_clock unstable */
clear_sched_clock_stable();
clocksource_cr16.name = "cr16_unstable";
clocksource_cr16.flags = CLOCK_SOURCE_UNSTABLE;
clocksource_cr16.rating = 0;
}
/* register at clocksource framework */
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment