public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: syzbot <syzbot+f342ea16c9d06d80b585@syzkaller.appspotmail.com>
To: linux-kernel@vger.kernel.org, syzkaller-bugs@googlegroups.com
Subject: Re: [syzbot] Re: [syzbot] [usb?] INFO: task hung in usb_port_suspend
Date: Sun, 13 Oct 2024 18:24:32 -0700	[thread overview]
Message-ID: <670c72d0.050a0220.4cbc0.003d.GAE@google.com> (raw)
In-Reply-To: <6709234e.050a0220.3e960.0011.GAE@google.com>

For archival purposes, forwarding an incoming command email to
linux-kernel@vger.kernel.org, syzkaller-bugs@googlegroups.com.

***

Subject: Re: [syzbot] [usb?] INFO: task hung in usb_port_suspend
Author: stern@rowland.harvard.edu

On Sun, Oct 13, 2024 at 01:34:05PM -0700, syzbot wrote:
> Hello,
> 
> syzbot has tested the proposed patch but the reproducer is still triggering an issue:

Okay, that's more like it.  This exercise has focused my mind on one 
particular spot in the code, and I believe I see the problem.  The 
driver needs to do a more careful job keeping track of whether the 
hrtimer callback is pending; neither hrtimer_active() nor 
dum_hcd->rh_state is quite the right thing to test.  In particular, the 
root hub can be in the DUMMY_RH_RUNNING state without the timer being 
active.

This patch adds a flag for a pending timer callback, on top of all the 
other debugging material.  Let's see if it fixes the problem.

Alan Stern

#syz test: https://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb.git
usb-testing

Index: usb-devel/drivers/usb/gadget/udc/dummy_hcd.c
===================================================================
--- usb-devel.orig/drivers/usb/gadget/udc/dummy_hcd.c
+++ usb-devel/drivers/usb/gadget/udc/dummy_hcd.c
@@ -50,7 +50,7 @@
 #define POWER_BUDGET	500	/* in mA; use 8 for low-power port testing */
 #define POWER_BUDGET_3	900	/* in mA */
 
-#define DUMMY_TIMER_INT_NSECS	125000 /* 1 microframe */
+#define DUMMY_INT_KTIME	ns_to_ktime(125000)	 /* 1 microframe */
 
 static const char	driver_name[] = "dummy_hcd";
 static const char	driver_desc[] = "USB Host+Gadget Emulator";
@@ -254,9 +254,12 @@ struct dummy_hcd {
 	u32				stream_en_ep;
 	u8				num_stream[30 / 2];
 
+	unsigned			timer_pending:1;
 	unsigned			active:1;
 	unsigned			old_active:1;
 	unsigned			resuming:1;
+
+	bool				alanflag;
 };
 
 struct dummy {
@@ -1303,9 +1306,11 @@ static int dummy_urb_enqueue(
 		urb->error_count = 1;		/* mark as a new urb */
 
 	/* kick the scheduler, it'll do the rest */
-	if (!hrtimer_active(&dum_hcd->timer))
-		hrtimer_start(&dum_hcd->timer, ns_to_ktime(DUMMY_TIMER_INT_NSECS),
+	if (!dum_hcd->timer_pending) {
+		hrtimer_start(&dum_hcd->timer, DUMMY_INT_KTIME,
 				HRTIMER_MODE_REL_SOFT);
+		dum_hcd->timer_pending = 1;
+	}
 
  done:
 	spin_unlock_irqrestore(&dum_hcd->dum->lock, flags);
@@ -1324,10 +1329,17 @@ static int dummy_urb_dequeue(struct usb_
 	spin_lock_irqsave(&dum_hcd->dum->lock, flags);
 
 	rc = usb_hcd_check_unlink_urb(hcd, urb, status);
-	if (!rc && dum_hcd->rh_state != DUMMY_RH_RUNNING &&
-			!list_empty(&dum_hcd->urbp_list))
-		hrtimer_start(&dum_hcd->timer, ns_to_ktime(0), HRTIMER_MODE_REL_SOFT);
-
+	if (!rc && !dum_hcd->timer_pending &&
+			!list_empty(&dum_hcd->urbp_list)) {
+		dev_info(dummy_dev(dum_hcd), "Dequeue restart %p\n", urb);
+		hrtimer_start(&dum_hcd->timer, DUMMY_INT_KTIME,
+				HRTIMER_MODE_REL_SOFT);
+		dum_hcd->timer_pending = 1;
+	} else {
+		dev_info(dummy_dev(dum_hcd), "Dequeue norestart: %d %p\n",
+				rc, urb);
+	}
+	dum_hcd->alanflag = true;
 	spin_unlock_irqrestore(&dum_hcd->dum->lock, flags);
 	return rc;
 }
@@ -1813,6 +1825,9 @@ static enum hrtimer_restart dummy_timer(
 
 	/* look at each urb queued by the host side driver */
 	spin_lock_irqsave(&dum->lock, flags);
+	dum_hcd->timer_pending = 0;
+	if (dum_hcd->alanflag)
+		dev_info(dummy_dev(dum_hcd), "Timer handler\n");
 
 	if (!dum_hcd->udev) {
 		dev_err(dummy_dev(dum_hcd),
@@ -1984,6 +1999,8 @@ return_urb:
 			ep->already_seen = ep->setup_stage = 0;
 
 		usb_hcd_unlink_urb_from_ep(dummy_hcd_to_hcd(dum_hcd), urb);
+		if (dum_hcd->alanflag)
+			dev_info(dummy_dev(dum_hcd), "Giveback %p\n", urb);
 		spin_unlock(&dum->lock);
 		usb_hcd_giveback_urb(dummy_hcd_to_hcd(dum_hcd), urb, status);
 		spin_lock(&dum->lock);
@@ -1995,11 +2012,12 @@ return_urb:
 		usb_put_dev(dum_hcd->udev);
 		dum_hcd->udev = NULL;
 	} else if (dum_hcd->rh_state == DUMMY_RH_RUNNING) {
-		/* want a 1 msec delay here */
-		hrtimer_start(&dum_hcd->timer, ns_to_ktime(DUMMY_TIMER_INT_NSECS),
+		hrtimer_start(&dum_hcd->timer, DUMMY_INT_KTIME,
 				HRTIMER_MODE_REL_SOFT);
+		dum_hcd->timer_pending = 1;
 	}
 
+	dum_hcd->alanflag = false;
 	spin_unlock_irqrestore(&dum->lock, flags);
 
 	return HRTIMER_NORESTART;
@@ -2390,8 +2408,11 @@ static int dummy_bus_resume(struct usb_h
 	} else {
 		dum_hcd->rh_state = DUMMY_RH_RUNNING;
 		set_link_state(dum_hcd);
-		if (!list_empty(&dum_hcd->urbp_list))
-			hrtimer_start(&dum_hcd->timer, ns_to_ktime(0), HRTIMER_MODE_REL_SOFT);
+		if (!list_empty(&dum_hcd->urbp_list)) {
+			hrtimer_start(&dum_hcd->timer, DUMMY_INT_KTIME,
+					HRTIMER_MODE_REL_SOFT);
+			dum_hcd->timer_pending = 1;
+		}
 		hcd->state = HC_STATE_RUNNING;
 	}
 	spin_unlock_irq(&dum_hcd->dum->lock);
@@ -2522,6 +2543,7 @@ static void dummy_stop(struct usb_hcd *h
 	struct dummy_hcd	*dum_hcd = hcd_to_dummy_hcd(hcd);
 
 	hrtimer_cancel(&dum_hcd->timer);
+	dum_hcd->timer_pending = 0;
 	device_remove_file(dummy_dev(dum_hcd), &dev_attr_urbs);
 	dev_info(dummy_dev(dum_hcd), "stopped\n");
 }

  parent reply	other threads:[~2024-10-14  1:24 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-10-11 13:08 [syzbot] [usb?] INFO: task hung in usb_port_suspend syzbot
2024-10-11 14:08 ` Alan Stern
2024-10-11 14:35   ` syzbot
2024-10-11 14:55     ` Alan Stern
2024-10-11 15:00       ` syzbot
2024-10-11 15:17         ` Alan Stern
2024-10-11 15:45           ` syzbot
2024-10-12  0:48             ` Alan Stern
2024-10-12  1:14               ` syzbot
2024-10-13  1:09                 ` Alan Stern
2024-10-13  2:10                   ` syzbot
2024-10-13  2:43                     ` Alan Stern
2024-10-13  3:05                       ` syzbot
2024-10-13 14:30                         ` Alan Stern
2024-10-13 15:02                           ` syzbot
2024-10-13 15:45                             ` Alan Stern
2024-10-13 16:14                               ` syzbot
2024-10-13 18:02                                 ` Alan Stern
2024-10-13 18:38                                   ` syzbot
2024-10-13 19:24                                     ` Alan Stern
2024-10-13 20:34                                       ` syzbot
2024-10-14  1:24 ` syzbot [this message]
2024-10-20 16:38 ` syzbot
2024-10-21  8:04   ` Oliver Neukum
2024-10-21 13:37     ` Alan Stern
2024-10-22 11:42     ` Hillf Danton
2024-10-22 10:46 ` Hillf Danton
2024-10-22 11:29   ` syzbot

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=670c72d0.050a0220.4cbc0.003d.GAE@google.com \
    --to=syzbot+f342ea16c9d06d80b585@syzkaller.appspotmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=syzkaller-bugs@googlegroups.com \
    /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