From: Kuan-Wei Chiu <visitorckw@gmail.com>
To: daniel.lezcano@kernel.org, tglx@kernel.org
Cc: pjw@kernel.org, palmer@dabbelt.com, aou@eecs.berkeley.edu,
alex@ghiti.fr, geert@linux-m68k.org, laurent@vivier.eu,
marscheng@google.com, eleanor15x@gmail.com,
jserv@ccns.ncku.edu.tw, linux-kernel@vger.kernel.org,
linux-riscv@lists.infradead.org, linux-m68k@lists.linux-m68k.org,
Kuan-Wei Chiu <visitorckw@gmail.com>
Subject: [PATCH 1/8] clocksource/drivers/timer-goldfish: Fix latch corruption with spinlock
Date: Thu, 5 Nov 2026 16:16:55 +0000 [thread overview]
Message-ID: <20261105161702.3515900-2-visitorckw@gmail.com> (raw)
In-Reply-To: <20261105161702.3515900-1-visitorckw@gmail.com>
Reading TIMER_TIME_LOW latches the upper 32 bits of the current time
into TIMER_TIME_HIGH. Because goldfish_timer_read() reads these two
registers without synchronization, a concurrent reader from an interrupt
or another CPU can read TIMER_TIME_LOW in between and overwrite the
latched high bits.
Protect the register read sequence with a raw spinlock.
Fixes: c92e7ef16400 ("clocksource/drivers: Add a goldfish-timer clocksource")
Signed-off-by: Kuan-Wei Chiu <visitorckw@gmail.com>
---
drivers/clocksource/timer-goldfish.c | 6 ++++++
1 file changed, 6 insertions(+)
diff --git a/drivers/clocksource/timer-goldfish.c b/drivers/clocksource/timer-goldfish.c
index 3fe12c7ee7b7..cc02b012ff76 100644
--- a/drivers/clocksource/timer-goldfish.c
+++ b/drivers/clocksource/timer-goldfish.c
@@ -6,6 +6,7 @@
#include <linux/clockchips.h>
#include <linux/module.h>
#include <linux/slab.h>
+#include <linux/spinlock.h>
#include <linux/goldfish.h>
#include <clocksource/timer-goldfish.h>
@@ -14,6 +15,7 @@ struct goldfish_timer {
struct clock_event_device ced;
struct resource res;
void __iomem *base;
+ raw_spinlock_t lock;
};
static struct goldfish_timer *ced_to_gf(struct clock_event_device *ced)
@@ -30,6 +32,7 @@ static u64 goldfish_timer_read(struct clocksource *cs)
{
struct goldfish_timer *timerdrv = cs_to_gf(cs);
void __iomem *base = timerdrv->base;
+ unsigned long flags;
u32 time_low, time_high;
u64 ticks;
@@ -37,8 +40,10 @@ static u64 goldfish_timer_read(struct clocksource *cs)
* time_low: get low bits of current time and update time_high
* time_high: get high bits of time at last time_low read
*/
+ raw_spin_lock_irqsave(&timerdrv->lock, flags);
time_low = gf_ioread32(base + TIMER_TIME_LOW);
time_high = gf_ioread32(base + TIMER_TIME_HIGH);
+ raw_spin_unlock_irqrestore(&timerdrv->lock, flags);
ticks = ((u64)time_high << 32) | time_low;
@@ -107,6 +112,7 @@ int __init goldfish_timer_init(int irq, void __iomem *base)
return -ENOMEM;
timerdrv->base = base;
+ raw_spin_lock_init(&timerdrv->lock);
timerdrv->ced = (struct clock_event_device){
.name = "goldfish_timer",
--
2.56.0.398.g882cbad275-goog
next prev parent reply other threads:[~2026-11-05 16:17 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-11-05 16:16 [PATCH 0/8] clocksource/drivers/timer-goldfish: Fix bugs and add DT support Kuan-Wei Chiu
2026-11-05 16:16 ` Kuan-Wei Chiu [this message]
2026-11-05 16:16 ` [PATCH 2/8] clocksource/drivers/timer-goldfish: Fix oneshot mode spurious interrupt Kuan-Wei Chiu
2026-11-05 16:16 ` [PATCH 3/8] clocksource/drivers/timer-goldfish: Fix resource leaks in init Kuan-Wei Chiu
2026-11-05 16:16 ` [PATCH 4/8] clocksource/drivers/timer-goldfish: Set CLOCK_SOURCE_IS_CONTINUOUS flag Kuan-Wei Chiu
2026-11-05 16:16 ` [PATCH 5/8] clocksource/drivers/timer-goldfish: Drop redundant max_idle_ns setting Kuan-Wei Chiu
2026-11-05 16:17 ` [PATCH 6/8] clocksource/drivers/timer-goldfish: Set clock_event_device cpumask Kuan-Wei Chiu
2026-11-05 16:17 ` [PATCH 7/8] clocksource/drivers/timer-goldfish: Add device tree probing support Kuan-Wei Chiu
2026-11-05 16:17 ` [PATCH 8/8] MAINTAINERS: Add entry for goldfish timer driver Kuan-Wei Chiu
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=20261105161702.3515900-2-visitorckw@gmail.com \
--to=visitorckw@gmail.com \
--cc=alex@ghiti.fr \
--cc=aou@eecs.berkeley.edu \
--cc=daniel.lezcano@kernel.org \
--cc=eleanor15x@gmail.com \
--cc=geert@linux-m68k.org \
--cc=jserv@ccns.ncku.edu.tw \
--cc=laurent@vivier.eu \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-m68k@lists.linux-m68k.org \
--cc=linux-riscv@lists.infradead.org \
--cc=marscheng@google.com \
--cc=palmer@dabbelt.com \
--cc=pjw@kernel.org \
--cc=tglx@kernel.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
all inboxes | Powered by JetHome®