powerpc/mce: Avoid using irq_work_queue() in realmode

In realmode mce handler we use irq_work_queue() to defer
the processing of mce events, irq_work_queue() can only
be called when translation is enabled because it touches
memory outside RMA, hence we enable translation before
calling irq_work_queue and disable on return, though it
is not safe to do in realmode.

To avoid this, program the decrementer and call the event
processing functions from timer handler.

Signed-off-by: Ganesh Goudar <[e-mail]>
Signed-off-by: Michael Ellerman <[e-mail]>
Link: https://lore.kernel.org/r/[e-mail]

arch/powerpc/include/asm/machdep.h

···
 	/* Called during machine check exception to retrive fixup address. */
 	bool		(*mce_check_early_recovery)(struct pt_regs *regs);
 
+	void            (*machine_check_log_err)(void);
+
 	/* Motherboard/chipset features. This is a kind of general purpose
 	 * hook used to control some machine specific features (like reset
 	 * lines, chip power control, etc...).

arch/powerpc/include/asm/mce.h

···
 unsigned long addr_to_pfn(struct pt_regs *regs, unsigned long addr);
 extern void mce_common_process_ue(struct pt_regs *regs,
 				  struct mce_error_info *mce_err);
+void mce_irq_work_queue(void);
 int mce_register_notifier(struct notifier_block *nb);
 int mce_unregister_notifier(struct notifier_block *nb);
+
+#ifdef CONFIG_PPC_BOOK3S_64
+void mce_run_irq_context_handlers(void);
+#else
+static inline void mce_run_irq_context_handlers(void) { };
+#endif /* CONFIG_PPC_BOOK3S_64 */
+
+#ifdef CONFIG_PPC_BOOK3S_64
+void set_mce_pending_irq_work(void);
+void clear_mce_pending_irq_work(void);
+#endif /* CONFIG_PPC_BOOK3S_64 */
+
 #ifdef CONFIG_PPC_BOOK3S_64
 void flush_and_reload_slb(void);
 void flush_erat(void);

arch/powerpc/include/asm/paca.h

···
 #endif
 #ifdef CONFIG_PPC_BOOK3S_64
 	struct mce_info *mce_info;
+	u8 mce_pending_irq_work;
 #endif /* CONFIG_PPC_BOOK3S_64 */
 } ____cacheline_aligned;
 

arch/powerpc/kernel/mce.c

···
 
 #include "setup.h"
 
-static void machine_check_process_queued_event(struct irq_work *work);
-static void machine_check_ue_irq_work(struct irq_work *work);
 static void machine_check_ue_event(struct machine_check_event *evt);
 static void machine_process_ue_event(struct work_struct *work);
-
-static struct irq_work mce_event_process_work = {
-        .func = machine_check_process_queued_event,
-};
-
-static struct irq_work mce_ue_event_irq_work = {
-	.func = machine_check_ue_irq_work,
-};
 
 static DECLARE_WORK(mce_ue_event_work, machine_process_ue_event);
 
···
 	default:
 		break;
 	}
+}
+
+void mce_irq_work_queue(void)
+{
+	/* Raise decrementer interrupt */
+	arch_irq_work_raise();
+	set_mce_pending_irq_work();
 }
 
 /*
···
 	get_mce_event(NULL, true);
 }
 
-static void machine_check_ue_irq_work(struct irq_work *work)
+static void machine_check_ue_work(void)
 {
 	schedule_work(&mce_ue_event_work);
 }
···
 	       evt, sizeof(*evt));
 
 	/* Queue work to process this event later. */
-	irq_work_queue(&mce_ue_event_irq_work);
+	mce_irq_work_queue();
 }
 
 /*
···
 {
 	int index;
 	struct machine_check_event evt;
-	unsigned long msr;
 
 	if (!get_mce_event(&evt, MCE_EVENT_RELEASE))
 		return;
···
 	memcpy(&local_paca->mce_info->mce_event_queue[index],
 	       &evt, sizeof(evt));
 
-	/*
-	 * Queue irq work to process this event later. Before
-	 * queuing the work enable translation for non radix LPAR,
-	 * as irq_work_queue may try to access memory outside RMO
-	 * region.
-	 */
-	if (!radix_enabled() && firmware_has_feature(FW_FEATURE_LPAR)) {
-		msr = mfmsr();
-		mtmsr(msr | MSR_IR | MSR_DR);
-		irq_work_queue(&mce_event_process_work);
-		mtmsr(msr);
-	} else {
-		irq_work_queue(&mce_event_process_work);
-	}
+	mce_irq_work_queue();
 }
 
 void mce_common_process_ue(struct pt_regs *regs,
···
  * process pending MCE event from the mce event queue. This function will be
  * called during syscall exit.
  */
-static void machine_check_process_queued_event(struct irq_work *work)
+static void machine_check_process_queued_event(void)
 {
 	int index;
 	struct machine_check_event *evt;
···
 		}
 		machine_check_print_event_info(evt, false, false);
 		local_paca->mce_info->mce_queue_count--;
+	}
+}
+
+void set_mce_pending_irq_work(void)
+{
+	local_paca->mce_pending_irq_work = 1;
+}
+
+void clear_mce_pending_irq_work(void)
+{
+	local_paca->mce_pending_irq_work = 0;
+}
+
+void mce_run_irq_context_handlers(void)
+{
+	if (unlikely(local_paca->mce_pending_irq_work)) {
+		if (ppc_md.machine_check_log_err)
+			ppc_md.machine_check_log_err();
+		machine_check_process_queued_event();
+		machine_check_ue_work();
+		clear_mce_pending_irq_work();
 	}
 }
 

arch/powerpc/kernel/time.c

···
 #include <asm/vdso_datapage.h>
 #include <asm/firmware.h>
 #include <asm/asm-prototypes.h>
+#include <asm/mce.h>
 
 /* powerpc clocksource/clockevent code */
 
···
 
 	if (test_irq_work_pending()) {
 		clear_irq_work_pending();
+		mce_run_irq_context_handlers();
 		irq_work_run();
 	}
 

arch/powerpc/platforms/pseries/pseries.h

···
 extern int pSeries_system_reset_exception(struct pt_regs *regs);
 extern int pSeries_machine_check_exception(struct pt_regs *regs);
 extern long pseries_machine_check_realmode(struct pt_regs *regs);
+void pSeries_machine_check_log_err(void);
 
 #ifdef CONFIG_SMP
 extern void smp_init_pseries(void);

arch/powerpc/platforms/pseries/ras.c

···
 
 static int ras_check_exception_token;
 
-static void mce_process_errlog_event(struct irq_work *work);
-static struct irq_work mce_errlog_process_work = {
-	.func = mce_process_errlog_event,
-};
-
 #define EPOW_SENSOR_TOKEN	9
 #define EPOW_SENSOR_INDEX	0
 
···
 	struct pseries_errorlog *pseries_log;
 	struct pseries_mc_errorlog *mce_log = NULL;
 	int disposition = rtas_error_disposition(errp);
-	unsigned long msr;
 	u8 error_type;
 
 	if (!rtas_error_extended(errp))
···
 	error_type = mce_log->error_type;
 
 	disposition = mce_handle_err_realmode(disposition, error_type);
-
-	/*
-	 * Enable translation as we will be accessing per-cpu variables
-	 * in save_mce_event() which may fall outside RMO region, also
-	 * leave it enabled because subsequently we will be queuing work
-	 * to workqueues where again per-cpu variables accessed, besides
-	 * fwnmi_release_errinfo() crashes when called in realmode on
-	 * pseries.
-	 * Note: All the realmode handling like flushing SLB entries for
-	 *       SLB multihit is done by now.
-	 */
 out:
-	msr = mfmsr();
-	mtmsr(msr | MSR_IR | MSR_DR);
-
 	disposition = mce_handle_err_virtmode(regs, errp, mce_log,
 					      disposition);
-
-	/*
-	 * Queue irq work to log this rtas event later.
-	 * irq_work_queue uses per-cpu variables, so do this in virt
-	 * mode as well.
-	 */
-	irq_work_queue(&mce_errlog_process_work);
-
-	mtmsr(msr);
-
 	return disposition;
 }
 
 /*
  * Process MCE rtas errlog event.
  */
-static void mce_process_errlog_event(struct irq_work *work)
+void pSeries_machine_check_log_err(void)
 {
 	struct rtas_error_log *err;
 

arch/powerpc/platforms/pseries/setup.c

···
 	.system_reset_exception = pSeries_system_reset_exception,
 	.machine_check_early	= pseries_machine_check_realmode,
 	.machine_check_exception = pSeries_machine_check_exception,
+	.machine_check_log_err	= pSeries_machine_check_log_err,
 #ifdef CONFIG_KEXEC_CORE
 	.machine_kexec          = pSeries_machine_kexec,
 	.kexec_cpu_down         = pseries_kexec_cpu_down,