linux-kernel - Re: uvcvideo causes ehci

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Thu, 22 Oct 2009 10:05:23 -0400 (EDT)
From:	Alan Stern <stern@...land.harvard.edu>
To:	Ozan Çağlayan <ozan@...dus.org.tr>
cc:	Laurent Pinchart <laurent.pinchart@...asonboard.com>,
	<linux-media@...r.kernel.org>,
	linux-kernel <linux-kernel@...r.kernel.org>,
	USB list <linux-usb@...r.kernel.org>
Subject: Re: uvcvideo causes ehci_hcd to halt

On Thu, 22 Oct 2009, [UTF-8] Ozan ÃaÄlayan wrote:

> Here's the outputs from /sys/kernel/debug/usb/ehci:
> 
> periodic:
> ----------------
> size = 1024
>    1:  qh1024-0001/f6ffe280 (h2 ep2 [1/0] q0 p8)

There's something odd about this.  I'd like to see this file again, 
after the patch below has been applied.

> registers:
> ----------------
> bus pci, device 0000:00:1d.7
> EHCI Host Controller
> EHCI 1.00, hcd state 0
> ownership 00000001
> SMI sts/enable 0x80080000
> structural params 0x00104208
> capability params 0x00016871
> status 6008 Periodic Recl FLR
> command 010000 (park)=0 ithresh=1 period=1024 HALT
> intrenable 00
> uframe 2fa0
> port 1 status 001000 POWER sig=se0
> port 2 status 001000 POWER sig=se0
> port 3 status 001000 POWER sig=se0
> port 4 status 001000 POWER sig=se0
> port 5 status 001005 POWER sig=se0 PE CONNECT
> port 6 status 001005 POWER sig=se0 PE CONNECT
> port 7 status 001000 POWER sig=se0
> port 8 status 001000 POWER sig=se0
> irq normal 60 err 0 reclaim 16 (lost 0)
> complete 60 unlink 1

This confirms that the periodic schedule was never disabled.

So first get another copy of the "periodic" file with the patch below.  
Then try changing the enable_periodic() routine in ehci-sched.c: Add

	udelay(2000);

just before the final "return 0;" line.  Let's see if that prevents the 
problem from occurring.

Alan Stern



--- usb-2.6.orig/drivers/usb/host/ehci-dbg.c
+++ usb-2.6/drivers/usb/host/ehci-dbg.c
@@ -591,18 +591,21 @@ static ssize_t fill_periodic_buffer(stru
 							qtd->hw_token) >> 8)) {
 						case 0: type = "out"; continue;
 						case 1: type = "in"; continue;
+						case 2: type = "?2"; continue;
+						case 3: type = "?3"; continue;
 						}
 					}
 
 					temp = scnprintf (next, size,
 						" (%c%d ep%d%s "
-						"[%d/%d] q%d p%d)",
+						"[%d/%d] q%d p%d) %08x",
 						speed_char (scratch),
 						scratch & 0x007f,
 						(scratch >> 8) & 0x000f, type,
 						p.qh->usecs, p.qh->c_usecs,
 						temp,
-						0x7ff & (scratch >> 16));
+						0x7ff & (scratch >> 16),
+						hc32_to_cpu(ehci, qtd->hw_token));
 
 					if (seen_count < DBG_SCHED_LIMIT)
 						seen [seen_count++].qh = p.qh;

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/