x86/tsc: Limit the adjust value further

Adjust value 0x80000000 and other values larger than that render the TSC
deadline timer disfunctional.

We have not yet any information about this from Intel, but experimentation
clearly proves that this is a 32/64 bit and sign extension issue.

If adjust values larger than that are actually required, which might be the
case for physical CPU hotplug, then we need to disable the deadline timer
on the affected package/CPUs and use the local APIC timer instead.

That requires some surgery in the APIC setup code, so we just limit the
ADJUST register value into the known to work range for now and revisit this
when Intel comes forth with proper information.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Roland Scheidegger <rscheidegger_lists@hispeed.ch>
Cc: Bruce Schlobohm <bruce.schlobohm@intel.com>
Cc: Kevin Stanton <kevin.b.stanton@intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Borislav Petkov <bp@alien8.de>

diff --git a/arch/x86/kernel/tsc_sync.c b/arch/x86/kernel/tsc_sync.c
index 1d8508fd15f7fbdb47640c9fe83b25810dd06af4..d0db011051a54212742680d3f25c2c52fb5c8ea6 100644 (file)
--- a/arch/x86/kernel/tsc_sync.c
+++ b/arch/x86/kernel/tsc_sync.c
@@ -73,9 +73,11 @@ static void tsc_sanitize_first_cpu(struct tsc_adjust *cur, s64 bootval,
         * the TSC is in sync with the already running cpus.
         *
         * But we always force positive ADJUST values. Otherwise the TSC
-        * deadline timer creates an interrupt storm. Sigh!
+        * deadline timer creates an interrupt storm. We also have to
+        * prevent values > 0x7FFFFFFF as those wreckage the timer as well.
         */
-       if ((bootcpu && bootval != 0) || (!bootcpu && bootval < 0)) {
+       if ((bootcpu && bootval != 0) || (!bootcpu && bootval < 0) ||
+           (bootval > 0x7FFFFFFF)) {
                pr_warn(FW_BUG "TSC ADJUST: CPU%u: %lld force to 0\n", cpu,
                        bootval);
                wrmsrl(MSR_IA32_TSC_ADJUST, 0);
@@ -448,13 +450,22 @@ retry:
         * that the warp is not longer detectable when the observed warp
         * value is used. In the worst case the adjustment needs to go
         * through a 3rd run for fine tuning.
-        *
-        * But we must make sure that the value doesn't become negative
-        * otherwise TSC deadline timer will create an interrupt storm.
         */
        cur->adjusted += cur_max_warp;
+
+       /*
+        * TSC deadline timer stops working or creates an interrupt storm
+        * with adjust values < 0 and > x07ffffff.
+        *
+        * To allow adjust values > 0x7FFFFFFF we need to disable the
+        * deadline timer and use the local APIC timer, but that requires
+        * more intrusive changes and we do not have any useful information
+        * from Intel about the underlying HW wreckage yet.
+        */
        if (cur->adjusted < 0)
                cur->adjusted = 0;
+       if (cur->adjusted > 0x7FFFFFFF)
+               cur->adjusted = 0x7FFFFFFF;
 
        pr_warn("TSC ADJUST compensate: CPU%u observed %lld warp. Adjust: %lld\n",
                cpu, cur_max_warp, cur->adjusted);