From 8272ea32c72a9b6974df8a91bd1bc3c0e2aa883f Mon Sep 17 00:00:00 2001
From: cheloha <cheloha@openbsd.org>
Date: Tue, 9 Aug 2022 04:49:08 +0000
Subject: [PATCH] riscv64: trigger deferred timer interrupts from splx(9)

In order to move riscv64 to a machine-independent clock interrupt
subsystem, the riscv64 clock interrupt code needs to function without
any specific knowledge of the clock interrupt schedule.

The easiest way to achieve this (as we just did with powerpc and
powerpc64) is, if the timer interrupt fires while the CPU is at or
above IPL_CLOCK, defer clock interrupt work until the the timer
interrupt is logically unmasked in splx(9).

In particular, trigger the timer interrupt from plic_setipl() so the
interrupt, if any, is pending before we handle soft interrupts.

Because we're no longer deferring work until the next tick, we don't
need to count pending statclock ticks in struct cpu_info.

kettenis@ notes that the timer triggering code should be refactored
into more generic code when we add support for a non-plic(4) riscv64
machine.

Graciously fixed, compiled, and tested by jca@.

Link: https://marc.info/?l=openbsd-tech&m=165931635410276&w=2

ok kettenis@
---
 sys/arch/riscv64/dev/plic.c      |  7 ++++-
 sys/arch/riscv64/include/cpu.h   |  4 +--
 sys/arch/riscv64/riscv64/clock.c | 53 +++++++++++++++++---------------
 3 files changed, 37 insertions(+), 27 deletions(-)

diff --git a/sys/arch/riscv64/dev/plic.c b/sys/arch/riscv64/dev/plic.c
index 6a863fedba3..755fd9262e7 100644
--- a/sys/arch/riscv64/dev/plic.c
+++ b/sys/arch/riscv64/dev/plic.c
@@ -1,4 +1,4 @@
-/*	$OpenBSD: plic.c,v 1.10 2022/04/06 18:59:27 naddy Exp $	*/
+/*	$OpenBSD: plic.c,v 1.11 2022/08/09 04:49:08 cheloha Exp $	*/
 
 /*
  * Copyright (c) 2020, Mars Li <mengshi.li.mars@gmail.com>
@@ -27,6 +27,7 @@
 #include <machine/bus.h>
 #include <machine/fdt.h>
 #include <machine/cpu.h>
+#include <machine/sbi.h>
 #include "riscv64/dev/riscv_cpu_intc.h"
 
 #include <dev/ofw/openfirm.h>
@@ -557,6 +558,10 @@ plic_setipl(int new)
 	/* higher values are higher priority */
 	plic_set_threshold(ci->ci_cpuid, new);
 
+	/* trigger deferred timer interrupt if cpl is now low enough */
+	if (ci->ci_timer_deferred && new < IPL_CLOCK)
+		sbi_set_timer(0);
+
 	intr_restore(sie);
 }
 
diff --git a/sys/arch/riscv64/include/cpu.h b/sys/arch/riscv64/include/cpu.h
index 4c32f35fdff..a455820bb7c 100644
--- a/sys/arch/riscv64/include/cpu.h
+++ b/sys/arch/riscv64/include/cpu.h
@@ -1,4 +1,4 @@
-/*	$OpenBSD: cpu.h,v 1.12 2022/06/10 21:34:15 jca Exp $	*/
+/*	$OpenBSD: cpu.h,v 1.13 2022/08/09 04:49:08 cheloha Exp $	*/
 
 /*
  * Copyright (c) 2019 Mike Larkin <mlarkin@openbsd.org>
@@ -92,7 +92,7 @@ struct cpu_info {
 	uint64_t		ci_lasttb;
 	uint64_t		ci_nexttimerevent;
 	uint64_t		ci_nextstatevent;
-	int			ci_statspending;
+	volatile int		ci_timer_deferred;
 
 	uint32_t		ci_cpl;
 	uint32_t		ci_ipending;
diff --git a/sys/arch/riscv64/riscv64/clock.c b/sys/arch/riscv64/riscv64/clock.c
index 15b2eaf8c5d..75643f998f5 100644
--- a/sys/arch/riscv64/riscv64/clock.c
+++ b/sys/arch/riscv64/riscv64/clock.c
@@ -1,4 +1,4 @@
-/*	$OpenBSD: clock.c,v 1.3 2021/07/24 22:41:09 jca Exp $	*/
+/*	$OpenBSD: clock.c,v 1.4 2022/08/09 04:49:08 cheloha Exp $	*/
 
 /*
  * Copyright (c) 2020 Mark Kettenis <kettenis@openbsd.org>
@@ -21,6 +21,7 @@
 #include <sys/kernel.h>
 #include <sys/systm.h>
 #include <sys/evcount.h>
+#include <sys/stdint.h>
 #include <sys/timetc.h>
 
 #include <machine/cpufunc.h>
@@ -105,6 +106,17 @@ clock_intr(void *frame)
 	int nstats;
 	int s;
 
+	/*
+	 * If the clock interrupt is masked, defer all clock interrupt
+	 * work until the clock interrupt is unmasked from splx(9).
+	 */
+	if (ci->ci_cpl >= IPL_CLOCK) {
+		ci->ci_timer_deferred = 1;
+		sbi_set_timer(UINT64_MAX);
+		return 0;
+	}
+	ci->ci_timer_deferred = 0;
+
 	/*
 	 * Based on the actual time delay since the last clock interrupt,
 	 * we arrange for earlier interrupt next time.
@@ -132,31 +144,24 @@ clock_intr(void *frame)
 
 	sbi_set_timer(nextevent);
 
-	if (ci->ci_cpl >= IPL_CLOCK) {
-		ci->ci_statspending += nstats;
-	} else {
-		nstats += ci->ci_statspending;
-		ci->ci_statspending = 0;
-
-		s = splclock();
-		intr_enable();
-
-		/*
-		 * Do standard timer interrupt stuff.
-		 */
-		while (ci->ci_lasttb < prevtb) {
-			ci->ci_lasttb += tick_increment;
-			clock_count.ec_count++;
-			hardclock((struct clockframe *)frame);
-		}
-
-		while (nstats-- > 0)
-			statclock((struct clockframe *)frame);
-
-		intr_disable();
-		splx(s);
+	s = splclock();
+	intr_enable();
+
+	/*
+	 * Do standard timer interrupt stuff.
+	 */
+	while (ci->ci_lasttb < prevtb) {
+		ci->ci_lasttb += tick_increment;
+		clock_count.ec_count++;
+		hardclock((struct clockframe *)frame);
 	}
 
+	while (nstats-- > 0)
+		statclock((struct clockframe *)frame);
+
+	intr_disable();
+	splx(s);
+
 	return 0;
 }
 
-- 
2.20.1