Test patch for ub and double registration

Test patch for ub and double registration

[Pete Zaitcev]

Hello,

here's a patch to fix the double kobject registration problem with the ub.
One little problem here is that I do not have a device which fails this
way, so I would like owners of such devices to give it a try.

The latest victim of this is Fabio Coatti. It should be noted that this
fix only (supposed to) prevents oopses on deregistration. If the device
doesn't work generally (for example, requires START STOP UNIT), it won't
help that.

Clemens' bug report wasn't clear enough. It may be this, may be something
else.

The flash key given on KS does the same, but we worked around that
particular units with Ben Herrenschmidt, so they are not good test cases
anymore.

-- Pete

diff -urp -X dontdiff linux-2.6.10-rc1/drivers/block/ub.c linux-2.6.10-rc1-ub/drivers/block/ub.c
--- linux-2.6.10-rc1/drivers/block/ub.c	2004-10-28 09:46:38.000000000 -0700
+++ linux-2.6.10-rc1-ub/drivers/block/ub.c	2004-11-01 12:35:10.000000000 -0800
@@ -188,7 +188,7 @@ struct ub_capacity {
  */
 
 #define SCMD_ST_HIST_SZ   8
-#define SCMD_TRACE_SZ    15	/* No more than 256 (trace_index) */
+#define SCMD_TRACE_SZ    63		/* Less than 4KB of 61-byte lines */
 
 struct ub_scsi_cmd_trace {
 	int hcur;
@@ -267,6 +267,7 @@ struct ub_dev {
 	int changed;			/* Media was changed */
 	int removable;
 	int readonly;
+	int first_open;			/* Kludge. See ub_bd_open. */
 	char name[8];
 	struct usb_device *dev;
 	struct usb_interface *intf;
@@ -959,9 +957,12 @@ static void ub_scsi_urb_compl(struct ub_
 			ub_cmdtr_state(sc, cmd);
 			return;
 		}
-		if (urb->status != 0)
+		if (urb->status != 0) {
+			printk("ub: cmd #%d cmd status (%d)\n", cmd->tag, urb->status); /* P3 */
 			goto Bad_End;
+		}
 		if (urb->actual_length != US_BULK_CB_WRAP_LEN) {
+			printk("ub: cmd #%d xferred %d\n", cmd->tag, urb->actual_length); /* P3 */
 			/* XXX Must do reset here to unconfuse the device */
 			goto Bad_End;
 		}
@@ -1428,6 +1424,26 @@ static int ub_bd_open(struct inode *inod
 	sc->openc++;
 	spin_unlock_irqrestore(&ub_lock, flags);
 
+	/*
+	 * This is a workaround for a specific problem in our block layer.
+	 * In 2.6.9, register_disk duplicates the code from rescan_partitions.
+	 * However, if we do add_disk with a device which persistently reports
+	 * a changed media, add_disk calls register_disk, which does do_open,
+	 * which will call rescan_paritions for changed media. After that,
+	 * register_disk attempts to do it all again and causes double kobject
+	 * registration and a eventually an oops on module removal.
+	 *
+	 * The bottom line is, Al Viro says that we should not allow
+	 * bdev->bd_invalidated to be set when doing add_disk no matter what.
+	 */
+	if (sc->first_open) {
+		if (sc->changed) {
+			sc->first_open = 0;
+			rc = -ENOMEDIUM;
+			goto err_open;
+		}
+	}
+
 	if (sc->removable || sc->readonly)
 		check_disk_change(inode->i_bdev);
 
@@ -1467,6 +1483,8 @@ static int ub_bd_release(struct inode *i
 
 	spin_lock_irqsave(&ub_lock, flags);
 	--sc->openc;
+	if (sc->openc == 0)
+		sc->first_open = 0;
 	if (sc->openc == 0 && atomic_read(&sc->poison))
 		ub_cleanup(sc);
 	spin_unlock_irqrestore(&ub_lock, flags);

@@ -1919,6 +1933,8 @@ static int ub_probe(struct usb_interface
 	}
 
 	sc->removable = 1;		/* XXX Query this from the device */
+	sc->changed = 1;		/* ub_revalidate clears only */
+	sc->first_open = 1;
 
 	ub_revalidate(sc);
 	/* This is pretty much a long term P3 */

Re: Test patch for ub and double registration

[Fabio Coatti]

Alle 01:44, martedì 2 novembre 2004, Pete Zaitcev ha scritto:
> Hello,
>
> here's a patch to fix the double kobject registration problem with the ub.
> One little problem here is that I do not have a device which fails this
> way, so I would like owners of such devices to give it a try.
>
> The latest victim of this is Fabio Coatti. It should be noted that this
> fix only (supposed to) prevents oopses on deregistration. If the device
> doesn't work generally (for example, requires START STOP UNIT), it won't
> help that.

Ill'try this patch in a few hours, report will follow ASAP; 

thanks.

-- 
Fabio "Cova" Coatti    http://members.ferrara.linux.it/cova     
Ferrara Linux Users Group           http://ferrara.linux.it
GnuPG fp:9765 A5B6 6843 17BC A646  BE8C FA56 373A 5374 C703
Old SysOps never die... they simply forget their password.

Re: Test patch for ub and double registration

[Fabio Coatti]

Alle 01:44, martedì 02 novembre 2004, Pete Zaitcev ha scritto:
> Hello,
>
> here's a patch to fix the double kobject registration problem with the ub.
> One little problem here is that I do not have a device which fails this
> way, so I would like owners of such devices to give it a try.
>
> The latest victim of this is Fabio Coatti. It should be noted that this
> fix only (supposed to) prevents oopses on deregistration. If the device
> doesn't work generally (for example, requires START STOP UNIT), it won't
> help that.

I've tried (under 2.6.10-rc1-mm1, that was the kernel that showed me problems) 
and in fact I can plug/unplug without problems usb flash drive, but there is 
a difference with previous behaviour (apart from crash, of course :) )
here the log from patched version of ub.c:

=================================================
Nov  2 22:44:20 kefk kernel: ehci_hcd 0000:00:1d.7: GetStatus port 3 status 
001803 POWER sig=j  CSC CONNECT
Nov  2 22:44:20 kefk kernel: hub 5-0:1.0: port 3, status 0501, change 0001, 
480 Mb/s
Nov  2 22:44:21 kefk kernel: hub 5-0:1.0: debounce: port 3: total 100ms stable 
100ms status 0x501
Nov  2 22:44:21 kefk kernel: ehci_hcd 0000:00:1d.7: port 3 high speed
Nov  2 22:44:21 kefk kernel: ehci_hcd 0000:00:1d.7: GetStatus port 3 status 
001005 POWER sig=se0  PE CONNECT
Nov  2 22:44:21 kefk kernel: usb 5-3: new high speed USB device using ehci_hcd 
and address 6
Nov  2 22:44:21 kefk kernel: ehci_hcd 0000:00:1d.7: port 3 high speed
Nov  2 22:44:21 kefk kernel: ehci_hcd 0000:00:1d.7: GetStatus port 3 status 
001005 POWER sig=se0  PE CONNECT
Nov  2 22:44:21 kefk kernel: usb 5-3: ep0 maxpacket = 64
Nov  2 22:44:21 kefk kernel: usb 5-3: new device strings: Mfr=1, Product=2, 
SerialNumber=3
Nov  2 22:44:21 kefk kernel: usb 5-3: default language 0x0409
Nov  2 22:44:21 kefk kernel: usb 5-3: Product: Mass storage
Nov  2 22:44:21 kefk kernel: usb 5-3: Manufacturer: USB
Nov  2 22:44:21 kefk kernel: usb 5-3: SerialNumber: 142E19413C2FCA34
Nov  2 22:44:21 kefk kernel: usb 5-3: hotplug
Nov  2 22:44:21 kefk kernel: usb 5-3: adding 5-3:1.0 (config #1, interface 0)
Nov  2 22:44:21 kefk kernel: usb 5-3:1.0: hotplug
Nov  2 22:44:21 kefk kernel: ub 5-3:1.0: usb_probe_interface
Nov  2 22:44:21 kefk kernel: ub 5-3:1.0: usb_probe_interface - got id
Nov  2 22:44:21 kefk kernel: uba: device 6 capacity nsec 50 bsize 512
(nothing more)
=================================================

but at this point no device is created in /dev, while previous behaviour was 
different:
(snip)
Oct 28 00:32:22 kefk kernel: ub 5-3:1.0: usb_probe_interface
Oct 28 00:32:22 kefk kernel: ub 5-3:1.0: usb_probe_interface - got id
Oct 28 00:32:22 kefk kernel: uba: device 4 capacity nsec 50 bsize 512
Oct 28 00:32:22 kefk kernel: uba: made changed
Oct 28 00:32:22 kefk kernel: uba: device 4 capacity nsec 1024000 bsize 512
Oct 28 00:32:22 kefk kernel: uba: device 4 capacity nsec 1024000 bsize 512
Oct 28 00:32:22 kefk kernel:  uba: uba1
Oct 28 00:32:22 kefk kernel:  uba: uba1
Oct 28 00:32:22 kefk kernel: kobject_register failed for uba1 (-17)
Oct 28 00:32:22 kefk kernel:  [<c01f1fd7>] kobject_register+0x51/0x5f
Oct 28 00:32:22 kefk kernel:  [<c0184720>] add_partition+0xbb/0xf0
(snip)

i.e. no lines like

uba: device 4 capacity nsec 1024000 bsize 512
uba: uba1


Apart that, removal of the usb device causes no harm to the system, log here:
Nov  2 22:55:47 kefk kernel: ehci_hcd 0000:00:1d.7: GetStatus port 3 status 
001002 POWER sig=se0  CSC
Nov  2 22:55:47 kefk kernel: hub 5-0:1.0: port 3, status 0100, change 0001, 12 
Mb/s
Nov  2 22:55:47 kefk kernel: usb 5-3: USB disconnect, address 6
Nov  2 22:55:47 kefk kernel: usb 5-3: usb_disable_device nuking all URBs
Nov  2 22:55:47 kefk kernel: usb 5-3: unregistering interface 5-3:1.0
Nov  2 22:55:47 kefk kernel: usb 5-3:1.0: hotplug
Nov  2 22:55:47 kefk kernel: usb 5-3: unregistering device
Nov  2 22:55:47 kefk kernel: usb 5-3: hotplug
Nov  2 22:55:47 kefk kernel: hub 5-0:1.0: debounce: port 3: total 100ms stable 
100ms status 0x100


-- 
Fabio Coatti       http://members.ferrara.linux.it/cova     
Ferrara Linux Users Group           http://ferrara.linux.it
GnuPG fp:9765 A5B6 6843 17BC A646  BE8C FA56 373A 5374 C703
Old SysOps never die... they simply forget their password.

Re: Test patch for ub and double registration

[Pete Zaitcev]

On Tue, 2 Nov 2004 22:57:24 +0100, Fabio Coatti <cova@ferrara.linux.it> wrote:

Dear Fabio, thank you for the test. Indeed, this is wrong:

> i.e. no lines like
> 
> uba: device 4 capacity nsec 1024000 bsize 512
> uba: uba1

Previously, it would retry capacity reading, quite by accident: the block
level did revalidation several times when faced with unexpected responses
from ub. The last one succeeded.

I'd like to secure one last favour from you. Please do this for me:
1. Connect the thing
2. Run
find /sys -name diag | xargs cat | mail -s "Flavio's ub diag" zaitcev@redhat.com

Thanks a lot,
-- Pete

Re: Test patch for ub and double registration

[Fabio Coatti]

Alle 00:10, mercoledì 03 novembre 2004, Pete Zaitcev ha scritto:

> I'd like to secure one last favour from you. Please do this for me:
> 1. Connect the thing
> 2. Run
> find /sys -name diag | xargs cat | mail -s "Flavio's ub diag"
> zaitcev@redhat.com

uhm.. maybe the mail was not so interesting.. :)

[root@kefk root]# find /sys -name diag | xargs cat | mail -s "Flavio's ub 
diag" zaitcev@redhat.com
find: /sys/devices/system/timer: No such file or directory
Null message body; hope that's ok

After inserting the device, I've got this:
============
Nov  3 21:36:25 kefk kernel: ehci_hcd 0000:00:1d.7: GetStatus port 3 status 
001803 POWER sig=j  CSC CONNECT
Nov  3 21:36:25 kefk kernel: hub 5-0:1.0: port 3, status 0501, change 0001, 
480 Mb/s
Nov  3 21:36:25 kefk kernel: hub 5-0:1.0: debounce: port 3: total 100ms stable 
100ms status 0x501
Nov  3 21:36:25 kefk kernel: ehci_hcd 0000:00:1d.7: port 3 high speed
Nov  3 21:36:25 kefk kernel: ehci_hcd 0000:00:1d.7: GetStatus port 3 status 
001005 POWER sig=se0  PE CONNECT
Nov  3 21:36:25 kefk kernel: usb 5-3: new high speed USB device using ehci_hcd 
and address 8
Nov  3 21:36:25 kefk kernel: ehci_hcd 0000:00:1d.7: devpath 3 ep0in 3strikes
Nov  3 21:36:26 kefk kernel: ehci_hcd 0000:00:1d.7: port 3 full speed --> 
companion
Nov  3 21:36:26 kefk kernel: ehci_hcd 0000:00:1d.7: GetStatus port 3 status 
003801 POWER OWNER sig=j  CONNECT
Nov  3 21:36:26 kefk kernel: uhci_hcd 0000:00:1d.1: wakeup_hc
Nov  3 21:36:26 kefk kernel: uhci_hcd 0000:00:1d.1: port 1 portsc 0083,00
Nov  3 21:36:26 kefk kernel: hub 2-0:1.0: port 1, status 0101, change 0001, 12 
Mb/s
Nov  3 21:36:26 kefk kernel: hub 2-0:1.0: debounce: port 1: total 100ms stable 
100ms status 0x101
Nov  3 21:36:26 kefk kernel: usb 2-1: new full speed USB device using uhci_hcd 
and address 4
Nov  3 21:36:26 kefk kernel: uhci_hcd 0000:00:1d.1: uhci_result_control: 
failed with status 440000
Nov  3 21:36:26 kefk kernel: [f7a6c240] link (37a6c1b2) element (37a6b040)
Nov  3 21:36:26 kefk kernel:   0: [f7a6b040] link (37a6b080) e0 Stalled 
CRC/Timeo Length=7 MaxLen=7 DT0 EndPt=0 Dev=0, PID=2d(SETUP) (buf=37c882a0)
Nov  3 21:36:26 kefk kernel:   1: [f7a6b080] link (37a6b0c0) e3 SPD Active 
Length=0 MaxLen=3f DT1 EndPt=0 Dev=0, PID=69(IN) (buf=1cd988c0)
Nov  3 21:36:26 kefk kernel:   2: [f7a6b0c0] link (00000001) e3 IOC Active 
Length=0 MaxLen=7ff DT1 EndPt=0 Dev=0, PID=e1(OUT) (buf=00000000)
Nov  3 21:36:26 kefk kernel:
Nov  3 21:36:26 kefk kernel: usb 2-1: device descriptor read/64, error -71
Nov  3 21:36:26 kefk kernel: uhci_hcd 0000:00:1d.1: uhci_result_control: 
failed with status 440000
Nov  3 21:36:26 kefk kernel: [f7a6c240] link (37a6c1b2) element (37a6b040)
Nov  3 21:36:26 kefk kernel:   0: [f7a6b040] link (37a6b080) e0 Stalled 
CRC/Timeo Length=7 MaxLen=7 DT0 EndPt=0 Dev=0, PID=2d(SETUP) (buf=377657c0)
Nov  3 21:36:26 kefk kernel:   1: [f7a6b080] link (37a6b0c0) e3 SPD Active 
Length=0 MaxLen=3f DT1 EndPt=0 Dev=0, PID=69(IN) (buf=1cd988c0)
Nov  3 21:36:26 kefk kernel:   2: [f7a6b0c0] link (00000001) e3 IOC Active 
Length=0 MaxLen=7ff DT1 EndPt=0 Dev=0, PID=e1(OUT) (buf=00000000)
Nov  3 21:36:26 kefk kernel:
Nov  3 21:36:26 kefk kernel: usb 2-1: device descriptor read/64, error -71
Nov  3 21:36:27 kefk kernel: usb 2-1: new full speed USB device using uhci_hcd 
and address 5
Nov  3 21:36:27 kefk kernel: uhci_hcd 0000:00:1d.1: uhci_result_control: 
failed with status 440000
Nov  3 21:36:27 kefk kernel: [f7a6c240] link (37a6c1b2) element (37a6b040)
Nov  3 21:36:27 kefk kernel:   0: [f7a6b040] link (37a6b080) e0 Stalled 
CRC/Timeo Length=7 MaxLen=7 DT0 EndPt=0 Dev=0, PID=2d(SETUP) (buf=377657c0)
Nov  3 21:36:27 kefk kernel:   1: [f7a6b080] link (37a6b0c0) e3 SPD Active 
Length=0 MaxLen=3f DT1 EndPt=0 Dev=0, PID=69(IN) (buf=1cd988c0)
Nov  3 21:36:27 kefk kernel:   2: [f7a6b0c0] link (00000001) e3 IOC Active 
Length=0 MaxLen=7ff DT1 EndPt=0 Dev=0, PID=e1(OUT) (buf=00000000)
Nov  3 21:36:27 kefk kernel:
Nov  3 21:36:27 kefk kernel: usb 2-1: device descriptor read/64, error -71
Nov  3 21:36:27 kefk kernel: uhci_hcd 0000:00:1d.1: uhci_result_control: 
failed with status 440000
Nov  3 21:36:27 kefk kernel: [f7a6c240] link (37a6c1b2) element (37a6b040)
Nov  3 21:36:27 kefk kernel:   0: [f7a6b040] link (37a6b080) e0 Stalled 
CRC/Timeo Length=7 MaxLen=7 DT0 EndPt=0 Dev=0, PID=2d(SETUP) (buf=377657c0)
Nov  3 21:36:27 kefk kernel:   1: [f7a6b080] link (37a6b0c0) e3 SPD Active 
Length=0 MaxLen=3f DT1 EndPt=0 Dev=0, PID=69(IN) (buf=1cd988c0)
Nov  3 21:36:27 kefk kernel:   2: [f7a6b0c0] link (00000001) e3 IOC Active 
Length=0 MaxLen=7ff DT1 EndPt=0 Dev=0, PID=e1(OUT) (buf=00000000)
Nov  3 21:36:27 kefk kernel:
Nov  3 21:36:27 kefk kernel: usb 2-1: device descriptor read/64, error -71
=================

But I don't know why now usb 2-1 device is called (the machine is in the same 
state of last try, no reboots).




-- 
Fabio Coatti       http://members.ferrara.linux.it/cova     
Ferrara Linux Users Group           http://ferrara.linux.it
GnuPG fp:9765 A5B6 6843 17BC A646  BE8C FA56 373A 5374 C703
Old SysOps never die... they simply forget their password.

Re: Test patch for ub and double registration

[Pete Zaitcev]

On Wed, 3 Nov 2004 21:43:01 +0100, Fabio Coatti <cova@ferrara.linux.it> wrote:

> Nov  3 21:36:26 kefk kernel: usb 2-1: new full speed USB device using uhci_hcd 
> and address 4
> Nov  3 21:36:26 kefk kernel: uhci_hcd 0000:00:1d.1: uhci_result_control: 
> failed with status 440000
> Nov  3 21:36:26 kefk kernel: [f7a6c240] link (37a6c1b2) element (37a6b040)
> Nov  3 21:36:26 kefk kernel:   0: [f7a6b040] link (37a6b080) e0 Stalled 
> CRC/Timeo Length=7 MaxLen=7 DT0 EndPt=0 Dev=0, PID=2d(SETUP) (buf=37c882a0)
> Nov  3 21:36:26 kefk kernel:   1: [f7a6b080] link (37a6b0c0) e3 SPD Active 
> Length=0 MaxLen=3f DT1 EndPt=0 Dev=0, PID=69(IN) (buf=1cd988c0)
> Nov  3 21:36:26 kefk kernel:   2: [f7a6b0c0] link (00000001) e3 IOC Active 
> Length=0 MaxLen=7ff DT1 EndPt=0 Dev=0, PID=e1(OUT) (buf=00000000)
> Nov  3 21:36:26 kefk kernel:
> Nov  3 21:36:26 kefk kernel: usb 2-1: device descriptor read/64, error -71

So, this fails even before getting to ub. I don't know how to help this,
but perhaps Greg has any ideas. I'm afraid I would just reboot if weird
things like this start happening. I hope your device hasn't just died
suddenly...

-- Pete

Re: Test patch for ub and double registration

[Fabio Coatti]

Alle 21:56, mercoledì 03 novembre 2004, hai scritto:

> > Nov  3 21:36:26 kefk kernel:   2: [f7a6b0c0] link (00000001) e3 IOC
> > Active Length=0 MaxLen=7ff DT1 EndPt=0 Dev=0, PID=e1(OUT) (buf=00000000)
> > Nov  3 21:36:26 kefk kernel:
> > Nov  3 21:36:26 kefk kernel: usb 2-1: device descriptor read/64, error
> > -71
>
> So, this fails even before getting to ub. I don't know how to help this,
> but perhaps Greg has any ideas. I'm afraid I would just reboot if weird
> things like this start happening. I hope your device hasn't just died
> suddenly...

well, the flash key is working just fine (tried with mdk 10.1 cooker, kernel 
2.6.8.1-12mdk); i've rebooted my box and the behaviour was exactly the same, 
but after some minutes, it worked (more or less).
Details:
First insertion after reboot (tried also by changing usb port, no differences)
=============
Nov  4 00:28:16 kefk kernel: ehci_hcd 0000:00:1d.7: GetStatus port 3 status 
001803 POWER sig=j  CSC CONNECT
Nov  4 00:28:16 kefk kernel: hub 5-0:1.0: port 3, status 0501, change 0001, 
480 Mb/s
Nov  4 00:28:16 kefk kernel: hub 5-0:1.0: debounce: port 3: total 100ms stable 
100ms status 0x501
Nov  4 00:28:16 kefk kernel: ehci_hcd 0000:00:1d.7: port 3 high speed
Nov  4 00:28:16 kefk kernel: ehci_hcd 0000:00:1d.7: GetStatus port 3 status 
001005 POWER sig=se0  PE CONNECT
Nov  4 00:28:16 kefk kernel: usb 5-3: new high speed USB device using ehci_hcd 
and address 4
Nov  4 00:28:16 kefk kernel: ehci_hcd 0000:00:1d.7: devpath 3 ep0in 3strikes
Nov  4 00:28:16 kefk kernel: ehci_hcd 0000:00:1d.7: port 3 full speed --> 
companion
Nov  4 00:28:16 kefk kernel: ehci_hcd 0000:00:1d.7: GetStatus port 3 status 
003801 POWER OWNER sig=j  CONNECT
Nov  4 00:28:16 kefk kernel: uhci_hcd 0000:00:1d.1: wakeup_hc
Nov  4 00:28:16 kefk kernel: uhci_hcd 0000:00:1d.1: port 1 portsc 0093,00
Nov  4 00:28:16 kefk kernel: hub 2-0:1.0: port 1, status 0101, change 0001, 12 
Mb/s
Nov  4 00:28:17 kefk kernel: hub 2-0:1.0: debounce: port 1: total 100ms stable 
100ms status 0x101
Nov  4 00:28:17 kefk kernel: usb 2-1: new full speed USB device using uhci_hcd 
and address 2
Nov  4 00:28:17 kefk kernel: uhci_hcd 0000:00:1d.1: uhci_result_control: 
failed with status 440000
Nov  4 00:28:17 kefk kernel: [f79ab240] link (379ab1b2) element (3791e040)
Nov  4 00:28:17 kefk kernel:   0: [f791e040] link (3791e080) e0 Stalled 
CRC/Timeo Length=7 MaxLen=7 DT0 EndPt=0 Dev=0, PID=2d(SETUP) (buf=37cf9c00)
Nov  4 00:28:17 kefk kernel:   1: [f791e080] link (3791e0c0) e3 SPD Active 
Length=0 MaxLen=3f DT1 EndPt=0 Dev=0, PID=69(IN) (buf=347e2300)
Nov  4 00:28:17 kefk kernel:   2: [f791e0c0] link (00000001) e3 IOC Active 
Length=0 MaxLen=7ff DT1 EndPt=0 Dev=0, PID=e1(OUT) (buf=00000000)
==================

Here i don't understand why i get this:
usb 5-3: new high speed USB device using ehci_hcd and address 4
and few lines below this:
usb 2-1: new full speed USB device using uhci_hcd and address 2
when I've inserted only one device :)

After this, I've removed an plugged again the key, with a different result:

==============
ov  4 00:37:16 kefk kernel: ehci_hcd 0000:00:1d.7: GetStatus port 3 status 
001803 POWER sig=j  CSC CONNECT
Nov  4 00:37:16 kefk kernel: hub 5-0:1.0: port 3, status 0501, change 0001, 
480 Mb/s
Nov  4 00:37:16 kefk kernel: hub 5-0:1.0: debounce: port 3: total 100ms stable 
100ms status 0x501
Nov  4 00:37:16 kefk kernel: ehci_hcd 0000:00:1d.7: port 3 high speed
Nov  4 00:37:16 kefk kernel: ehci_hcd 0000:00:1d.7: GetStatus port 3 status 
001005 POWER sig=se0  PE CONNECT
Nov  4 00:37:16 kefk kernel: usb 5-3: new high speed USB device using ehci_hcd 
and address 7
Nov  4 00:37:16 kefk kernel: ehci_hcd 0000:00:1d.7: port 3 high speed
Nov  4 00:37:16 kefk kernel: ehci_hcd 0000:00:1d.7: GetStatus port 3 status 
001005 POWER sig=se0  PE CONNECT
Nov  4 00:37:16 kefk kernel: usb 5-3: ep0 maxpacket = 64
Nov  4 00:37:16 kefk kernel: usb 5-3: new device strings: Mfr=1, Product=2, 
SerialNumber=3
Nov  4 00:37:16 kefk kernel: usb 5-3: default language 0x0409
Nov  4 00:37:16 kefk kernel: usb 5-3: Product: Mass storage
Nov  4 00:37:16 kefk kernel: usb 5-3: Manufacturer: USB
Nov  4 00:37:16 kefk kernel: usb 5-3: SerialNumber: 142E19413C2FCA34
Nov  4 00:37:16 kefk kernel: usb 5-3: hotplug
Nov  4 00:37:16 kefk kernel: usb 5-3: adding 5-3:1.0 (config #1, interface 0)
Nov  4 00:37:16 kefk kernel: usb 5-3:1.0: hotplug
Nov  4 00:37:16 kefk kernel: ub: sizeof ub_scsi_cmd 64 ub_dev 2504
Nov  4 00:37:16 kefk kernel: ub 5-3:1.0: usb_probe_interface
Nov  4 00:37:16 kefk kernel: ub 5-3:1.0: usb_probe_interface - got id
Nov  4 00:37:16 kefk kernel: uba: device 7 capacity nsec 50 bsize 512
Nov  4 00:37:16 kefk kernel: usbcore: registered new driver ub
===============

Note, no usb 2-1 related activity, and  I've been able to see this:

[root@kefk root]# find /sys -name diag
/sys/devices/pci0000:00/0000:00:1d.7/usb5/5-3/5-3:1.0/diag
find: /sys/devices/system/timer: No such file or directory

(/sys/devices/system/timer is a completely diffent story, it seems that timer 
is prepended with non-printable character)

I've tried several times to plug the flash key on a different system, no 
problems at all, and all usb ports of my box are working fine, tested with 
printer, scanner, bluetooth dongles, and so on..




-- 
Fabio Coatti       http://members.ferrara.linux.it/cova     
Ferrara Linux Users Group           http://ferrara.linux.it
GnuPG fp:9765 A5B6 6843 17BC A646  BE8C FA56 373A 5374 C703
Old SysOps never die... they simply forget their password.