pps: Fix a use-after-free

commit c79a39d upstream.

On a board running ntpd and gpsd, I'm seeing a consistent use-after-free
in sys_exit() from gpsd when rebooting:

    pps pps1: removed
    ------------[ cut here ]------------
    kobject: '(null)' (00000000db4bec24): is not initialized, yet kobject_put() is being called.
    WARNING: CPU: 2 PID: 440 at lib/kobject.c:734 kobject_put+0x120/0x150
    CPU: 2 UID: 299 PID: 440 Comm: gpsd Not tainted 6.11.0-rc6-00308-gb31c44928842 #1
    Hardware name: Raspberry Pi 4 Model B Rev 1.1 (DT)
    pstate: 60000005 (nZCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
    pc : kobject_put+0x120/0x150
    lr : kobject_put+0x120/0x150
    sp : ffffffc0803d3ae0
    x29: ffffffc0803d3ae0 x28: ffffff8042dc9738 x27: 0000000000000001
    x26: 0000000000000000 x25: ffffff8042dc9040 x24: ffffff8042dc9440
    x23: ffffff80402a4620 x22: ffffff8042ef4bd0 x21: ffffff80405cb600
    x20: 000000000008001b x19: ffffff8040b3b6e0 x18: 0000000000000000
    x17: 0000000000000000 x16: 0000000000000000 x15: 696e6920746f6e20
    x14: 7369203a29343263 x13: 205d303434542020 x12: 0000000000000000
    x11: 0000000000000000 x10: 0000000000000000 x9 : 0000000000000000
    x8 : 0000000000000000 x7 : 0000000000000000 x6 : 0000000000000000
    x5 : 0000000000000000 x4 : 0000000000000000 x3 : 0000000000000000
    x2 : 0000000000000000 x1 : 0000000000000000 x0 : 0000000000000000
    Call trace:
     kobject_put+0x120/0x150
     cdev_put+0x20/0x3c
     __fput+0x2c4/0x2d8
     ____fput+0x1c/0x38
     task_work_run+0x70/0xfc
     do_exit+0x2a0/0x924
     do_group_exit+0x34/0x90
     get_signal+0x7fc/0x8c0
     do_signal+0x128/0x13b4
     do_notify_resume+0xdc/0x160
     el0_svc+0xd4/0xf8
     el0t_64_sync_handler+0x140/0x14c
     el0t_64_sync+0x190/0x194
    ---[ end trace 0000000000000000 ]---

...followed by more symptoms of corruption, with similar stacks:

    refcount_t: underflow; use-after-free.
    kernel BUG at lib/list_debug.c:62!
    Kernel panic - not syncing: Oops - BUG: Fatal exception

This happens because pps_device_destruct() frees the pps_device with the
embedded cdev immediately after calling cdev_del(), but, as the comment
above cdev_del() notes, fops for previously opened cdevs are still
callable even after cdev_del() returns. I think this bug has always
been there: I can't explain why it suddenly started happening every time
I reboot this particular board.

In commit d953e0e ("pps: Fix a use-after free bug when
unregistering a source."), George Spelvin suggested removing the
embedded cdev. That seems like the simplest way to fix this, so I've
implemented his suggestion, using __register_chrdev() with pps_idr
becoming the source of truth for which minor corresponds to which
device.

But now that pps_idr defines userspace visibility instead of cdev_add(),
we need to be sure the pps->dev refcount can't reach zero while
userspace can still find it again. So, the idr_remove() call moves to
pps_unregister_cdev(), and pps_idr now holds a reference to pps->dev.

    pps_core: source serial1 got cdev (251:1)
    <...>
    pps pps1: removed
    pps_core: unregistering pps1
    pps_core: deallocating pps1

Fixes: d953e0e ("pps: Fix a use-after free bug when unregistering a source.")
Cc: stable@vger.kernel.org
Signed-off-by: Calvin Owens <calvin@wbinvd.org>
Reviewed-by: Michal Schmidt <mschmidt@redhat.com>
Link: https://lore.kernel.org/r/a17975fd5ae99385791929e563f72564edbcf28f.1731383727.git.calvin@wbinvd.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

[ Resolving merge conflict by keeping the original code only changing
the reference in dbg_*() calls. ]

Signed-off-by: Stanislav Uschakow <suschako@amazon.de>

Author: jcalvinowens

drivers/pps/clients/pps-gpio.c

@@ -174,7 +174,7 @@ static int pps_gpio_probe(struct platform_device *pdev)
 	}
 
 	platform_set_drvdata(pdev, data);
-	dev_info(data->pps->dev, "Registered IRQ %d as PPS source\n",
+	dev_info(&data->pps->dev, "Registered IRQ %d as PPS source\n",
 		 data->irq);
 
 	return 0;

drivers/pps/clients/pps-ktimer.c

@@ -70,7 +70,7 @@ static struct pps_source_info pps_ktimer_info = {
 
 static void __exit pps_ktimer_exit(void)
 {
-	dev_info(pps->dev, "ktimer PPS source unregistered\n");
+	dev_dbg(&pps->dev, "ktimer PPS source unregistered\n");
 
 	del_timer_sync(&ktimer);
 	pps_unregister_source(pps);
@@ -88,7 +88,7 @@ static int __init pps_ktimer_init(void)
 	setup_timer(&ktimer, pps_ktimer_event, 0);
 	mod_timer(&ktimer, jiffies + HZ);
 
-	dev_info(pps->dev, "ktimer PPS source registered\n");
+	dev_dbg(&pps->dev, "ktimer PPS source registered\n");
 
 	return 0;
 }

drivers/pps/clients/pps-ldisc.c

@@ -48,7 +48,7 @@ static void pps_tty_dcd_change(struct tty_struct *tty, unsigned int status)
 	pps_event(pps, &ts, status ? PPS_CAPTUREASSERT :
 			PPS_CAPTURECLEAR, NULL);
 
-	dev_dbg(pps->dev, "PPS %s at %lu\n",
+	dev_dbg(&pps->dev, "PPS %s at %lu\n",
 			status ? "assert" : "clear", jiffies);
 }
 
@@ -85,7 +85,7 @@ static int pps_tty_open(struct tty_struct *tty)
 		goto err_unregister;
 	}
 
-	dev_info(pps->dev, "source \"%s\" added\n", info.path);
+	dev_dbg(&pps->dev, "source \"%s\" added\n", info.path);
 
 	return 0;
 
@@ -105,7 +105,7 @@ static void pps_tty_close(struct tty_struct *tty)
 	if (WARN_ON(!pps))
 		return;
 
-	dev_info(pps->dev, "removed\n");
+	dev_info(&pps->dev, "removed\n");
 	pps_unregister_source(pps);
 }
 

drivers/pps/clients/pps_parport.c

@@ -95,7 +95,7 @@ static void parport_irq(void *handle)
 	/* check the signal (no signal means the pulse is lost this time) */
 	if (!signal_is_set(port)) {
 		local_irq_restore(flags);
-		dev_err(dev->pps->dev, "lost the signal\n");
+		dev_err(&dev->pps->dev, "lost the signal\n");
 		goto out_assert;
 	}
 
@@ -112,7 +112,7 @@ static void parport_irq(void *handle)
 	/* timeout */
 	dev->cw_err++;
 	if (dev->cw_err >= CLEAR_WAIT_MAX_ERRORS) {
-		dev_err(dev->pps->dev, "disabled clear edge capture after %d"
+		dev_err(&dev->pps->dev, "disabled clear edge capture after %d"
 				" timeouts\n", dev->cw_err);
 		dev->cw = 0;
 		dev->cw_err = 0;

drivers/pps/kapi.c

@@ -55,7 +55,7 @@ static void pps_add_offset(struct pps_ktime *ts, struct pps_ktime *offset)
 static void pps_echo_client_default(struct pps_device *pps, int event,
 		void *data)
 {
-	dev_info(pps->dev, "echo %s %s\n",
+	dev_info(&pps->dev, "echo %s %s\n",
 		event & PPS_CAPTUREASSERT ? "assert" : "",
 		event & PPS_CAPTURECLEAR ? "clear" : "");
 }
@@ -125,7 +125,7 @@ struct pps_device *pps_register_source(struct pps_source_info *info,
 		goto kfree_pps;
 	}
 
-	dev_info(pps->dev, "new PPS source %s\n", info->name);
+	dev_dbg(&pps->dev, "new PPS source %s\n", info->name);
 
 	return pps;
 
@@ -179,7 +179,7 @@ void pps_event(struct pps_device *pps, struct pps_event_time *ts, int event,
 	/* check event type */
 	BUG_ON((event & (PPS_CAPTUREASSERT | PPS_CAPTURECLEAR)) == 0);
 
-	dev_dbg(pps->dev, "PPS event at %lld.%09ld\n",
+	dev_dbg(&pps->dev, "PPS event at %lld.%09ld\n",
 			(s64)ts->ts_real.tv_sec, ts->ts_real.tv_nsec);
 
 	timespec_to_pps_ktime(&ts_real, ts->ts_real);
@@ -201,7 +201,7 @@ void pps_event(struct pps_device *pps, struct pps_event_time *ts, int event,
 		/* Save the time stamp */
 		pps->assert_tu = ts_real;
 		pps->assert_sequence++;
-		dev_dbg(pps->dev, "capture assert seq #%u\n",
+		dev_dbg(&pps->dev, "capture assert seq #%u\n",
 			pps->assert_sequence);
 
 		captured = ~0;
@@ -215,7 +215,7 @@ void pps_event(struct pps_device *pps, struct pps_event_time *ts, int event,
 		/* Save the time stamp */
 		pps->clear_tu = ts_real;
 		pps->clear_sequence++;
-		dev_dbg(pps->dev, "capture clear seq #%u\n",
+		dev_dbg(&pps->dev, "capture clear seq #%u\n",
 			pps->clear_sequence);
 
 		captured = ~0;

drivers/pps/kc.c

@@ -56,11 +56,11 @@ int pps_kc_bind(struct pps_device *pps, struct pps_bind_args *bind_args)
 			pps_kc_hardpps_mode = 0;
 			pps_kc_hardpps_dev = NULL;
 			spin_unlock_irq(&pps_kc_hardpps_lock);
-			dev_info(pps->dev, "unbound kernel"
+			dev_info(&pps->dev, "unbound kernel"
 					" consumer\n");
 		} else {
 			spin_unlock_irq(&pps_kc_hardpps_lock);
-			dev_err(pps->dev, "selected kernel consumer"
+			dev_err(&pps->dev, "selected kernel consumer"
 					" is not bound\n");
 			return -EINVAL;
 		}
@@ -70,11 +70,11 @@ int pps_kc_bind(struct pps_device *pps, struct pps_bind_args *bind_args)
 			pps_kc_hardpps_mode = bind_args->edge;
 			pps_kc_hardpps_dev = pps;
 			spin_unlock_irq(&pps_kc_hardpps_lock);
-			dev_info(pps->dev, "bound kernel consumer: "
+			dev_info(&pps->dev, "bound kernel consumer: "
 				"edge=0x%x\n", bind_args->edge);
 		} else {
 			spin_unlock_irq(&pps_kc_hardpps_lock);
-			dev_err(pps->dev, "another kernel consumer"
+			dev_err(&pps->dev, "another kernel consumer"
 					" is already bound\n");
 			return -EINVAL;
 		}
@@ -96,7 +96,7 @@ void pps_kc_remove(struct pps_device *pps)
 		pps_kc_hardpps_mode = 0;
 		pps_kc_hardpps_dev = NULL;
 		spin_unlock_irq(&pps_kc_hardpps_lock);
-		dev_info(pps->dev, "unbound kernel consumer"
+		dev_info(&pps->dev, "unbound kernel consumer"
 				" on device removal\n");
 	} else
 		spin_unlock_irq(&pps_kc_hardpps_lock);