Detect and mitigate uncontrolled ACPI GPE interrupt storms on OpenBSD (7.3 through 7.8-stable)

combined78.diff

Index: src/sys/dev/acpi/acpi.c
===================================================================
RCS file: /cvs/src/sys/dev/acpi/acpi.c,v
diff -u -p -r1.454 acpi.c
--- src/sys/dev/acpi/acpi.c	20 Sep 2025 17:43:28 -0000	1.454
+++ src/sys/dev/acpi/acpi.c	24 Oct 2025 20:02:22 -0000
@@ -45,6 +45,9 @@
 
 #include <machine/apmvar.h>
 
+#define GPE_RATE_MIN_CYCLE	5		/* seconds    */
+#define GPE_RATE_MAX		1000	/* per second */
+
 #include "wd.h"
 
 extern int cpu_suspended;
@@ -95,6 +98,8 @@ void	acpi_disable_allgpes(struct acpi_so
 struct gpe_block *acpi_find_gpe(struct acpi_softc *, int);
 void	acpi_enable_onegpe(struct acpi_softc *, int);
 int	acpi_gpe(struct acpi_softc *, int, void *);
+void	acpi_init_gpe_rate(struct acpi_softc *, int);
+int	acpi_gpe_rate(struct acpi_softc *, int);
 
 void	acpi_enable_rungpes(struct acpi_softc *);
 
@@ -2280,6 +2285,7 @@ acpi_enable_onegpe(struct acpi_softc *sc
 	dnprintf(50, "enabling GPE %.2x (current: %sabled) %.2x\n",
 	    gpe, (en & mask) ? "en" : "dis", en);
 	acpi_write_pmreg(sc, ACPIREG_GPE_EN, gpe>>3, en | mask);
+	acpi_init_gpe_rate(sc, gpe);
 }
 
 /* Clear all GPEs */
@@ -2358,7 +2364,40 @@ acpi_gpe(struct acpi_softc *sc, int gpe,
 	if (sc->gpe_table[gpe].flags & GPE_LEVEL)
 		acpi_write_pmreg(sc, ACPIREG_GPE_STS, gpe>>3, mask);
 	en = acpi_read_pmreg(sc, ACPIREG_GPE_EN,  gpe>>3);
-	acpi_write_pmreg(sc, ACPIREG_GPE_EN,  gpe>>3, en | mask);
+	/* Re-enable if GPE rate passes, otherwise leave disabled */
+	if (!acpi_gpe_rate(sc, gpe))
+		acpi_write_pmreg(sc, ACPIREG_GPE_EN,  gpe>>3, en | mask);
+	return (0);
+}
+
+void
+acpi_init_gpe_rate(struct acpi_softc *sc, int gpe)
+{
+	sc->gpe_table[gpe].rate_start = getuptime();
+	sc->gpe_table[gpe].rate_count = 0;
+}
+
+int
+acpi_gpe_rate(struct acpi_softc *sc, int gpe)
+{
+	struct gpe_block *pgpe = &sc->gpe_table[gpe];
+	time_t cycle;
+
+	pgpe->rate_count++;
+	dnprintf(10, "rate GPE %.2x start %lld elapsed %lld count %zu\n", gpe,
+	    pgpe->rate_start, getuptime() - pgpe->rate_start, pgpe->rate_count);
+
+	cycle = getuptime() - pgpe->rate_start;
+	if (cycle >= GPE_RATE_MIN_CYCLE) {
+		if (pgpe->rate_count > (GPE_RATE_MAX * cycle)) {
+			printf("uncontrolled GPE storm %lld/s, disabling GPE %.2x\n",
+			    pgpe->rate_count / cycle, gpe);
+			return (1);
+		}
+
+		/* Reset and start a new cycle */
+		acpi_init_gpe_rate(sc, gpe);
+	}
 	return (0);
 }
 
Index: src/sys/dev/acpi/acpivar.h
===================================================================
RCS file: /cvs/src/sys/dev/acpi/acpivar.h,v
diff -u -p -r1.138 acpivar.h
--- src/sys/dev/acpi/acpivar.h	20 Sep 2025 17:43:28 -0000	1.138
+++ src/sys/dev/acpi/acpivar.h	24 Oct 2025 20:02:22 -0000
@@ -181,6 +181,9 @@ struct gpe_block {
 	void *arg;
 	int   active;
 	int   flags;
+
+	time_t rate_start;
+	size_t rate_count;
 };
 
 struct acpi_devlist {
Index: src/sys/kern/sched_bsd.c
===================================================================
RCS file: /cvs/src/sys/kern/sched_bsd.c,v
diff -u -p -r1.105 sched_bsd.c
--- src/sys/kern/sched_bsd.c	25 Sep 2025 08:46:50 -0000	1.105
+++ src/sys/kern/sched_bsd.c	24 Oct 2025 20:02:23 -0000
@@ -594,11 +594,6 @@ setperf_auto(void *v)
 	if (cpu_setperf == NULL)
 		return;
 
-	if (current_perfpolicy() == PERFPOL_HIGH) {
-		speedup = 1;
-		goto faster;
-	}
-
 	if (!idleticks)
 		if (!(idleticks = mallocarray(ncpusfound, sizeof(*idleticks),
 		    M_DEVBUF, M_NOWAIT | M_ZERO)))
@@ -642,7 +637,6 @@ setperf_auto(void *v)
 		downbeats++;
 
 	if (speedup && perflevel != 100) {
-faster:
 		perflevel = 100;
 		cpu_setperf(perflevel);
 	} else if (!speedup && perflevel != 0 && --downbeats <= 0) {

Also re-enables dormant frequency scaling code in sched_bsd.c

Update source (from https://www.openbsd.org/anoncvs.html#updating)

$ cd /usr/src
$ cvs -d [email protected]:/cvs -q update -Pd -rOPENBSD_7_8

Apply this patch (download gist to working directory first)

$ patch -d /usr --forward < combined78.diff

Build modded objects

# cd /sys/arch/amd64/compile/GENERIC.MP
# make clean
# make obj
# make config
# make acpi.o sched_bsd.o

Replace stock relink objects and link a new kernel

# cd /usr/obj/sys/arch/amd64/compile/GENERIC.MP
# cp -iv acpi.o sched_bsd.o /usr/share/relink/kernel/GENERIC.MP/
# /usr/libexec/reorder_kernel

Reboot to use
Check dmesg for line similar to uncontrolled GPE storm 5634/s, disabling GPE 6f and much-reduced CPU usage in system process acpi0
Check sysctl hw.sensors for frequency scaling