x86/panic: replace smp_send_stop() with kdump friendly version in pan…

…ic path

Daniel Walker reported problems which happens when
crash_kexec_post_notifiers kernel option is enabled
(https://lkml.org/lkml/2015/6/24/44).

In that case, smp_send_stop() is called before entering kdump routines
which assume other CPUs are still online.  As the result, for x86, kdump
routines fail to save other CPUs' registers and disable virtualization
extensions.

To fix this problem, call a new kdump friendly function,
crash_smp_send_stop(), instead of the smp_send_stop() when
crash_kexec_post_notifiers is enabled.  crash_smp_send_stop() is a weak
function, and it just call smp_send_stop().  Architecture codes should
override it so that kdump can work appropriately.  This patch only
provides x86-specific version.

For Xen's PV kernel, just keep the current behavior.

NOTES:

- Right solution would be to place crash_smp_send_stop() before
  __crash_kexec() invocation in all cases and remove smp_send_stop(), but
  we can't do that until all architectures implement own
  crash_smp_send_stop()

- crash_smp_send_stop()-like work is still needed by
  machine_crash_shutdown() because crash_kexec() can be called without
  entering panic()

Fixes: f06e515 (kernel/panic.c: add "crash_kexec_post_notifiers" option)
Link: http://lkml.kernel.org/r/20160810080948.11028.15344.stgit@sysi4-13.yrl.intra.hitachi.co.jp
Signed-off-by: Hidehiro Kawai <[email protected]>
Reported-by: Daniel Walker <[email protected]>
Cc: Dave Young <[email protected]>
Cc: Baoquan He <[email protected]>
Cc: Vivek Goyal <[email protected]>
Cc: Eric Biederman <[email protected]>
Cc: Masami Hiramatsu <[email protected]>
Cc: Daniel Walker <[email protected]>
Cc: Xunlei Pang <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: "H. Peter Anvin" <[email protected]>
Cc: Borislav Petkov <[email protected]>
Cc: David Vrabel <[email protected]>
Cc: Toshi Kani <[email protected]>
Cc: Ralf Baechle <[email protected]>
Cc: David Daney <[email protected]>
Cc: Aaro Koskinen <[email protected]>
Cc: "Steven J. Hill" <[email protected]>
Cc: Corey Minyard <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>

arch/x86/include/asm/kexec.h

@@ -210,6 +210,7 @@ struct kexec_entry64_regs {
 
 typedef void crash_vmclear_fn(void);
 extern crash_vmclear_fn __rcu *crash_vmclear_loaded_vmcss;
+extern void kdump_nmi_shootdown_cpus(void);
 
 #endif /* __ASSEMBLY__ */
 

arch/x86/include/asm/smp.h

@@ -47,6 +47,7 @@ struct smp_ops {
 	void (*smp_cpus_done)(unsigned max_cpus);
 
 	void (*stop_other_cpus)(int wait);
+	void (*crash_stop_other_cpus)(void);
 	void (*smp_send_reschedule)(int cpu);
 
 	int (*cpu_up)(unsigned cpu, struct task_struct *tidle);

arch/x86/kernel/crash.c

@@ -133,15 +133,31 @@ static void kdump_nmi_callback(int cpu, struct pt_regs *regs)
 	disable_local_APIC();
 }
 
-static void kdump_nmi_shootdown_cpus(void)
+void kdump_nmi_shootdown_cpus(void)
 {
 	nmi_shootdown_cpus(kdump_nmi_callback);
 
 	disable_local_APIC();
 }
 
+/* Override the weak function in kernel/panic.c */
+void crash_smp_send_stop(void)
+{
+	static int cpus_stopped;
+
+	if (cpus_stopped)
+		return;
+
+	if (smp_ops.crash_stop_other_cpus)
+		smp_ops.crash_stop_other_cpus();
+	else
+		smp_send_stop();
+
+	cpus_stopped = 1;
+}
+
 #else
-static void kdump_nmi_shootdown_cpus(void)
+void crash_smp_send_stop(void)
 {
 	/* There are no cpus to shootdown */
 }
@@ -160,7 +176,7 @@ void native_machine_crash_shutdown(struct pt_regs *regs)
 	/* The kernel is broken so disable interrupts */
 	local_irq_disable();
 
-	kdump_nmi_shootdown_cpus();
+	crash_smp_send_stop();
 
 	/*
 	 * VMCLEAR VMCSs loaded on this cpu if needed.

arch/x86/kernel/smp.c

@@ -32,6 +32,8 @@
 #include <asm/nmi.h>
 #include <asm/mce.h>
 #include <asm/trace/irq_vectors.h>
+#include <asm/kexec.h>
+
 /*
  *	Some notes on x86 processor bugs affecting SMP operation:
  *
@@ -342,6 +344,9 @@ struct smp_ops smp_ops = {
 	.smp_cpus_done		= native_smp_cpus_done,
 
 	.stop_other_cpus	= native_stop_other_cpus,
+#if defined(CONFIG_KEXEC_CORE)
+	.crash_stop_other_cpus	= kdump_nmi_shootdown_cpus,
+#endif
 	.smp_send_reschedule	= native_smp_send_reschedule,
 
 	.cpu_up			= native_cpu_up,

kernel/panic.c

@@ -71,6 +71,32 @@ void __weak nmi_panic_self_stop(struct pt_regs *regs)
 	panic_smp_self_stop();
 }
 
+/*
+ * Stop other CPUs in panic.  Architecture dependent code may override this
+ * with more suitable version.  For example, if the architecture supports
+ * crash dump, it should save registers of each stopped CPU and disable
+ * per-CPU features such as virtualization extensions.
+ */
+void __weak crash_smp_send_stop(void)
+{
+	static int cpus_stopped;
+
+	/*
+	 * This function can be called twice in panic path, but obviously
+	 * we execute this only once.
+	 */
+	if (cpus_stopped)
+		return;
+
+	/*
+	 * Note smp_send_stop is the usual smp shutdown function, which
+	 * unfortunately means it may not be hardened to work in a panic
+	 * situation.
+	 */
+	smp_send_stop();
+	cpus_stopped = 1;
+}
+
 atomic_t panic_cpu = ATOMIC_INIT(PANIC_CPU_INVALID);
 
 /*
@@ -164,14 +190,21 @@ void panic(const char *fmt, ...)
 	if (!_crash_kexec_post_notifiers) {
 		printk_nmi_flush_on_panic();
 		__crash_kexec(NULL);
-	}
 
-	/*
-	 * Note smp_send_stop is the usual smp shutdown function, which
-	 * unfortunately means it may not be hardened to work in a panic
-	 * situation.
-	 */
-	smp_send_stop();
+		/*
+		 * Note smp_send_stop is the usual smp shutdown function, which
+		 * unfortunately means it may not be hardened to work in a
+		 * panic situation.
+		 */
+		smp_send_stop();
+	} else {
+		/*
+		 * If we want to do crash dump after notifier calls and
+		 * kmsg_dump, we will need architecture dependent extra
+		 * works in addition to stopping other CPUs.
+		 */
+		crash_smp_send_stop();
+	}
 
 	/*
 	 * Run any panic handlers, including those that might need to