From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: stable@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
patches@lists.linux.dev, Liuye <liu.yeC@h3c.com>,
Douglas Anderson <dianders@chromium.org>,
Daniel Thompson <daniel.thompson@linaro.org>
Subject: [PATCH 4.19 17/18] serial: kgdboc: Fix NMI-safety problems from keyboard reset code
Date: Thu, 23 May 2024 15:12:40 +0200 [thread overview]
Message-ID: <20240523130326.392944420@linuxfoundation.org> (raw)
In-Reply-To: <20240523130325.727602650@linuxfoundation.org>
4.19-stable review patch. If anyone has any objections, please let me know.
------------------
From: Daniel Thompson <daniel.thompson@linaro.org>
commit b2aba15ad6f908d1a620fd97f6af5620c3639742 upstream.
Currently, when kdb is compiled with keyboard support, then we will use
schedule_work() to provoke reset of the keyboard status. Unfortunately
schedule_work() gets called from the kgdboc post-debug-exception
handler. That risks deadlock since schedule_work() is not NMI-safe and,
even on platforms where the NMI is not directly used for debugging, the
debug trap can have NMI-like behaviour depending on where breakpoints
are placed.
Fix this by using the irq work system, which is NMI-safe, to defer the
call to schedule_work() to a point when it is safe to call.
Reported-by: Liuye <liu.yeC@h3c.com>
Closes: https://lore.kernel.org/all/20240228025602.3087748-1-liu.yeC@h3c.com/
Cc: stable@vger.kernel.org
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Link: https://lore.kernel.org/r/20240424-kgdboc_fix_schedule_work-v2-1-50f5a490aec5@linaro.org
Signed-off-by: Daniel Thompson <daniel.thompson@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
drivers/tty/serial/kgdboc.c | 30 +++++++++++++++++++++++++++++-
1 file changed, 29 insertions(+), 1 deletion(-)
--- a/drivers/tty/serial/kgdboc.c
+++ b/drivers/tty/serial/kgdboc.c
@@ -16,6 +16,7 @@
#include <linux/console.h>
#include <linux/vt_kern.h>
#include <linux/input.h>
+#include <linux/irq_work.h>
#include <linux/module.h>
#define MAX_CONFIG_LEN 40
@@ -35,6 +36,25 @@ static int kgdboc_use_kms; /* 1 if we u
static struct tty_driver *kgdb_tty_driver;
static int kgdb_tty_line;
+/*
+ * When we leave the debug trap handler we need to reset the keyboard status
+ * (since the original keyboard state gets partially clobbered by kdb use of
+ * the keyboard).
+ *
+ * The path to deliver the reset is somewhat circuitous.
+ *
+ * To deliver the reset we register an input handler, reset the keyboard and
+ * then deregister the input handler. However, to get this done right, we do
+ * have to carefully manage the calling context because we can only register
+ * input handlers from task context.
+ *
+ * In particular we need to trigger the action from the debug trap handler with
+ * all its NMI and/or NMI-like oddities. To solve this the kgdboc trap exit code
+ * (the "post_exception" callback) uses irq_work_queue(), which is NMI-safe, to
+ * schedule a callback from a hardirq context. From there we have to defer the
+ * work again, this time using schedule_work(), to get a callback using the
+ * system workqueue, which runs in task context.
+ */
#ifdef CONFIG_KDB_KEYBOARD
static int kgdboc_reset_connect(struct input_handler *handler,
struct input_dev *dev,
@@ -86,10 +106,17 @@ static void kgdboc_restore_input_helper(
static DECLARE_WORK(kgdboc_restore_input_work, kgdboc_restore_input_helper);
+static void kgdboc_queue_restore_input_helper(struct irq_work *unused)
+{
+ schedule_work(&kgdboc_restore_input_work);
+}
+
+static DEFINE_IRQ_WORK(kgdboc_restore_input_irq_work, kgdboc_queue_restore_input_helper);
+
static void kgdboc_restore_input(void)
{
if (likely(system_state == SYSTEM_RUNNING))
- schedule_work(&kgdboc_restore_input_work);
+ irq_work_queue(&kgdboc_restore_input_irq_work);
}
static int kgdboc_register_kbd(char **cptr)
@@ -120,6 +147,7 @@ static void kgdboc_unregister_kbd(void)
i--;
}
}
+ irq_work_sync(&kgdboc_restore_input_irq_work);
flush_work(&kgdboc_restore_input_work);
}
#else /* ! CONFIG_KDB_KEYBOARD */
next prev parent reply other threads:[~2024-05-23 13:15 UTC|newest]
Thread overview: 24+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-05-23 13:12 [PATCH 4.19 00/18] 4.19.315-rc1 review Greg Kroah-Hartman
2024-05-23 13:12 ` [PATCH 4.19 01/18] Revert "selftests: mm: fix map_hugetlb failure on 64K page size systems" Greg Kroah-Hartman
2024-05-23 13:12 ` [PATCH 4.19 02/18] dm: limit the number of targets and parameter size area Greg Kroah-Hartman
2024-05-23 13:12 ` [PATCH 4.19 03/18] btrfs: add missing mutex_unlock in btrfs_relocate_sys_chunks() Greg Kroah-Hartman
2024-05-23 13:12 ` [PATCH 4.19 04/18] tracing: Simplify creation and deletion of synthetic events Greg Kroah-Hartman
2024-05-23 13:12 ` [PATCH 4.19 05/18] tracing: Add unified dynamic event framework Greg Kroah-Hartman
2024-05-23 13:12 ` [PATCH 4.19 06/18] tracing: Use dyn_event framework for synthetic events Greg Kroah-Hartman
2024-05-23 13:12 ` [PATCH 4.19 07/18] tracing: Remove unneeded synth_event_mutex Greg Kroah-Hartman
2024-05-23 13:12 ` [PATCH 4.19 08/18] tracing: Consolidate trace_add/remove_event_call back to the nolock functions Greg Kroah-Hartman
2024-05-23 13:12 ` [PATCH 4.19 09/18] string.h: Add str_has_prefix() helper function Greg Kroah-Hartman
2024-05-23 13:12 ` [PATCH 4.19 10/18] tracing: Use str_has_prefix() helper for histogram code Greg Kroah-Hartman
2024-05-23 13:12 ` [PATCH 4.19 11/18] tracing: Use str_has_prefix() instead of using fixed sizes Greg Kroah-Hartman
2024-05-23 13:12 ` [PATCH 4.19 12/18] tracing: Have the historgram use the result of str_has_prefix() for len of prefix Greg Kroah-Hartman
2024-05-23 13:12 ` [PATCH 4.19 13/18] tracing: Refactor hist trigger action code Greg Kroah-Hartman
2024-05-23 13:12 ` [PATCH 4.19 14/18] tracing: Split up onmatch action data Greg Kroah-Hartman
2024-05-23 13:12 ` [PATCH 4.19 15/18] tracing: Generalize hist trigger onmax and save action Greg Kroah-Hartman
2024-05-23 13:12 ` [PATCH 4.19 16/18] tracing: Remove unnecessary var_ref destroy in track_data_destroy() Greg Kroah-Hartman
2024-05-23 13:12 ` Greg Kroah-Hartman [this message]
2024-05-23 13:12 ` [PATCH 4.19 18/18] docs: kernel_include.py: Cope with docutils 0.21 Greg Kroah-Hartman
2024-05-24 6:55 ` [PATCH 4.19 00/18] 4.19.315-rc1 review Harshit Mogalapalli
2024-05-24 11:23 ` Pavel Machek
2024-05-24 13:46 ` Anders Roxell
2024-05-24 15:19 ` Jon Hunter
2024-05-24 16:26 ` Shuah Khan
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=20240523130326.392944420@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=daniel.thompson@linaro.org \
--cc=dianders@chromium.org \
--cc=liu.yeC@h3c.com \
--cc=patches@lists.linux.dev \
--cc=stable@vger.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