From: Paolo Bonzini <pbonzini@redhat.com>
To: qemu-devel@nongnu.org
Subject: [PATCH 7/7] hpet: avoid timer storms on periodic timers
Date: Mon, 22 Jul 2024 14:05:41 +0200 [thread overview]
Message-ID: <20240722120541.70790-8-pbonzini@redhat.com> (raw)
In-Reply-To: <20240722120541.70790-1-pbonzini@redhat.com>
If the period is set to a value that is too low, there could be no
time left to run the rest of QEMU. Do not trigger interrupts faster
than 1 MHz.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
hw/timer/hpet.c | 13 +++++++++++--
1 file changed, 11 insertions(+), 2 deletions(-)
diff --git a/hw/timer/hpet.c b/hw/timer/hpet.c
index 1654b7cb8b8..471950adef1 100644
--- a/hw/timer/hpet.c
+++ b/hw/timer/hpet.c
@@ -59,6 +59,7 @@ typedef struct HPETTimer { /* timers */
uint8_t wrap_flag; /* timer pop will indicate wrap for one-shot 32-bit
* mode. Next pop will be actual timer expiration.
*/
+ uint64_t last; /* last value armed, to avoid timer storms */
} HPETTimer;
struct HPETState {
@@ -266,6 +267,7 @@ static int hpet_post_load(void *opaque, int version_id)
for (i = 0; i < s->num_timers; i++) {
HPETTimer *t = &s->timer[i];
t->cmp64 = hpet_calculate_cmp64(t, s->hpet_counter, t->cmp);
+ t->last = qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL) - NANOSECONDS_PER_SECOND;
}
/* Recalculate the offset between the main counter and guest time */
if (!s->hpet_offset_saved) {
@@ -364,8 +366,15 @@ static const VMStateDescription vmstate_hpet = {
static void hpet_arm(HPETTimer *t, uint64_t tick)
{
- /* FIXME: Clamp period to reasonable min value? */
- timer_mod(t->qemu_timer, hpet_get_ns(t->state, tick));
+ uint64_t ns = hpet_get_ns(t->state, tick);
+
+ /* Clamp period to reasonable min value (1 us) */
+ if (timer_is_periodic(t) && ns - t->last < 1000) {
+ ns = t->last + 1000;
+ }
+
+ t->last = ns;
+ timer_mod(t->qemu_timer, ns);
}
/*
--
2.45.2
prev parent reply other threads:[~2024-07-22 12:07 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-07-22 12:05 [PATCH 0/7] hpet: fixes for 64-bit mode and interrupt status registers Paolo Bonzini
2024-07-22 12:05 ` [PATCH 1/7] hpet: fix and cleanup persistence of interrupt status Paolo Bonzini
2024-07-22 12:05 ` [PATCH 2/7] hpet: ignore high bits of comparator in 32-bit mode Paolo Bonzini
2024-07-22 12:05 ` [PATCH 3/7] hpet: remove unnecessary variable "index" Paolo Bonzini
2024-07-22 12:05 ` [PATCH 4/7] hpet: place read-only bits directly in "new_val" Paolo Bonzini
2024-07-22 12:05 ` [PATCH 5/7] hpet: accept 64-bit reads and writes Paolo Bonzini
2024-07-22 12:05 ` [PATCH 6/7] hpet: store full 64-bit target value of the counter Paolo Bonzini
2024-07-22 12:05 ` Paolo Bonzini [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20240722120541.70790-8-pbonzini@redhat.com \
--to=pbonzini@redhat.com \
--cc=qemu-devel@nongnu.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).