public inbox for linux-usb@vger.kernel.org
 help / color / mirror / Atom feed
* [BUG] UAS protocol hangs on Realtek RTL9210B (0bda:9210, bcdDevice f0.01)
@ 2026-03-08 14:10 Nils Reuter
  2026-03-08 16:16 ` Greg KH
  0 siblings, 1 reply; 5+ messages in thread
From: Nils Reuter @ 2026-03-08 14:10 UTC (permalink / raw)
  To: linux-usb

Hello,

I would like to report a persistent UAS protocol issue with a Realtek 
RTL9210B based M.2 enclosure.

*Hardware Environment:*

  *

    Enclosure: Ugreen SKU 15813 (Model CM767)

  *

    Bridge Chip: Realtek RTL9210B (ID 0bda:9210, bcdDevice f0.01)

  *

    Drive: Transcend TS512GMTS430S (SATA M.2)

  *

    System: Debian / OpenMediaVault (Linux omv 6.18.12+deb13-amd64)

  *

    Kernel: 6.18.12-amd64

*Problem Description:* Under heavy I/O load (e.g., backup operations), 
the device experience UAS protocol hangs. The kernel log shows multiple 
|uas_eh_abort_handler| events, followed by SCSI read/write errors and a 
device reset (error -71). Eventually, the device is offlined or 
remounted as read-only.

SMART logs show incrementing UDMA CRC errors (Attribute 199) during 
these events, even though the hardware (cable/SSD) tests fine on other 
platforms.

*Logs:*

|[timestamp] sd 1:0:0:0: [sdb] tag#18 uas_eh_abort_handler 0 uas-tag 15 
inflight: CMD IN [timestamp] usb 2-1-port3: cannot reset (err = -71) 
[timestamp] sd 1:0:0:0: Device offlined - not ready after error recovery |

*Workaround:* The device works perfectly stable when UAS is disabled via 
usb-storage quirks: |options usb-storage quirks=0bda:9210:u|

*Request:* Please consider adding this device (0bda:9210 with bcdDevice 
f0.01) to the kernel's UAS quirks list to ensure out-of-the-box 
stability for other users.

Best regards, Nils Reuter


-- 
Nils Reuter
Houbirggasse 7
90482 Nürnberg

nils.reuter@gmx.net


^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [BUG] UAS protocol hangs on Realtek RTL9210B (0bda:9210, bcdDevice f0.01)
  2026-03-08 14:10 [BUG] UAS protocol hangs on Realtek RTL9210B (0bda:9210, bcdDevice f0.01) Nils Reuter
@ 2026-03-08 16:16 ` Greg KH
  2026-03-08 19:37   ` [PATCH] USB: storage: UAS: ignore UAS on Realtek RTL9210B with firmware f0.01 Nils Reuter
  0 siblings, 1 reply; 5+ messages in thread
From: Greg KH @ 2026-03-08 16:16 UTC (permalink / raw)
  To: Nils Reuter; +Cc: linux-usb

On Sun, Mar 08, 2026 at 03:10:54PM +0100, Nils Reuter wrote:
> Hello,
> 
> I would like to report a persistent UAS protocol issue with a Realtek
> RTL9210B based M.2 enclosure.
> 
> *Hardware Environment:*
> 
>  *
> 
>    Enclosure: Ugreen SKU 15813 (Model CM767)
> 
>  *
> 
>    Bridge Chip: Realtek RTL9210B (ID 0bda:9210, bcdDevice f0.01)
> 
>  *
> 
>    Drive: Transcend TS512GMTS430S (SATA M.2)
> 
>  *
> 
>    System: Debian / OpenMediaVault (Linux omv 6.18.12+deb13-amd64)
> 
>  *
> 
>    Kernel: 6.18.12-amd64
> 
> *Problem Description:* Under heavy I/O load (e.g., backup operations), the
> device experience UAS protocol hangs. The kernel log shows multiple
> |uas_eh_abort_handler| events, followed by SCSI read/write errors and a
> device reset (error -71). Eventually, the device is offlined or remounted as
> read-only.
> 
> SMART logs show incrementing UDMA CRC errors (Attribute 199) during these
> events, even though the hardware (cable/SSD) tests fine on other platforms.
> 
> *Logs:*
> 
> |[timestamp] sd 1:0:0:0: [sdb] tag#18 uas_eh_abort_handler 0 uas-tag 15
> inflight: CMD IN [timestamp] usb 2-1-port3: cannot reset (err = -71)
> [timestamp] sd 1:0:0:0: Device offlined - not ready after error recovery |
> 
> *Workaround:* The device works perfectly stable when UAS is disabled via
> usb-storage quirks: |options usb-storage quirks=0bda:9210:u|
> 
> *Request:* Please consider adding this device (0bda:9210 with bcdDevice
> f0.01) to the kernel's UAS quirks list to ensure out-of-the-box stability
> for other users.

Can you submit a patch that adds this quirk to the kernel so you get
credit for adding it?

thanks,

greg k-h

^ permalink raw reply	[flat|nested] 5+ messages in thread

* [PATCH] USB: storage: UAS: ignore UAS on Realtek RTL9210B with firmware f0.01
  2026-03-08 16:16 ` Greg KH
@ 2026-03-08 19:37   ` Nils Reuter
  2026-03-11 12:35     ` Greg KH
  0 siblings, 1 reply; 5+ messages in thread
From: Nils Reuter @ 2026-03-08 19:37 UTC (permalink / raw)
  To: Greg KH; +Cc: linux-usb

Hi Greg,

Thanks for the feedback. As requested, here is the patch to add the UAS
quirk for the Realtek RTL9210B (bcdDevice f0.01).

Regards,
Nils

---

From cb1f8118e802be51a06f4564bb01f91110a51b53 Mon Sep 17 00:00:00 2001
From: Nils Reuter <nils.reuter@gmx.net>
Date: Sun, 8 Mar 2026 19:39:21 +0100
Subject: [PATCH] USB: storage: UAS: ignore UAS on Realtek RTL9210B with
 firmware f0.01

The Realtek RTL9210B (0bda:9210) with firmware version f0.01 suffers
from UAS protocol hangs under heavy I/O load, leading to device
resets and read-only remounts. Disabling UAS via the IGNORE_UAS
quirk stabilizes the connection.

Signed-off-by: Nils Reuter <nils.reuter@gmx.net>
---
 drivers/usb/storage/unusual_uas.h | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/usb/storage/unusual_uas.h b/drivers/usb/storage/unusual_uas.h
index 939a98c2d..f17469f8c 100644
--- a/drivers/usb/storage/unusual_uas.h
+++ b/drivers/usb/storage/unusual_uas.h
@@ -90,6 +90,13 @@ UNUSUAL_DEV(0x0bc2, 0x331a, 0x0000, 0x9999,
 		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_NO_REPORT_LUNS),
 
+/* Reported-by: Nils Reuter <nils.reuter@gmx.net> */
+UNUSUAL_DEV(0x0bda, 0x9210, 0xf001, 0xf001,
+		"Realtek",
+		"RTL9210B",
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
+		US_FL_IGNORE_UAS),
+
 /* Reported-by: Oliver Neukum <oneukum@suse.com> */
 UNUSUAL_DEV(0x125f, 0xa94a, 0x0160, 0x0160,
 		"ADATA",
-- 
2.53.0

---

^ permalink raw reply related	[flat|nested] 5+ messages in thread

* Re: [PATCH] USB: storage: UAS: ignore UAS on Realtek RTL9210B with firmware f0.01
  2026-03-08 19:37   ` [PATCH] USB: storage: UAS: ignore UAS on Realtek RTL9210B with firmware f0.01 Nils Reuter
@ 2026-03-11 12:35     ` Greg KH
  2026-03-11 21:59       ` Nils Reuter
  0 siblings, 1 reply; 5+ messages in thread
From: Greg KH @ 2026-03-11 12:35 UTC (permalink / raw)
  To: Nils Reuter; +Cc: linux-usb

On Sun, Mar 08, 2026 at 08:37:28PM +0100, Nils Reuter wrote:
> Hi Greg,
> 
> Thanks for the feedback. As requested, here is the patch to add the UAS
> quirk for the Realtek RTL9210B (bcdDevice f0.01).
> 
> Regards,
> Nils
> 
> ---
> 
> From cb1f8118e802be51a06f4564bb01f91110a51b53 Mon Sep 17 00:00:00 2001
> From: Nils Reuter <nils.reuter@gmx.net>
> Date: Sun, 8 Mar 2026 19:39:21 +0100
> Subject: [PATCH] USB: storage: UAS: ignore UAS on Realtek RTL9210B with
>  firmware f0.01
> 
> The Realtek RTL9210B (0bda:9210) with firmware version f0.01 suffers
> from UAS protocol hangs under heavy I/O load, leading to device
> resets and read-only remounts. Disabling UAS via the IGNORE_UAS
> quirk stabilizes the connection.
> 
> Signed-off-by: Nils Reuter <nils.reuter@gmx.net>
> ---
>  drivers/usb/storage/unusual_uas.h | 7 +++++++
>  1 file changed, 7 insertions(+)
> 
> diff --git a/drivers/usb/storage/unusual_uas.h b/drivers/usb/storage/unusual_uas.h
> index 939a98c2d..f17469f8c 100644
> --- a/drivers/usb/storage/unusual_uas.h
> +++ b/drivers/usb/storage/unusual_uas.h
> @@ -90,6 +90,13 @@ UNUSUAL_DEV(0x0bc2, 0x331a, 0x0000, 0x9999,
>  		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
>  		US_FL_NO_REPORT_LUNS),
>  
> +/* Reported-by: Nils Reuter <nils.reuter@gmx.net> */
> +UNUSUAL_DEV(0x0bda, 0x9210, 0xf001, 0xf001,
> +		"Realtek",
> +		"RTL9210B",
> +		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
> +		US_FL_IGNORE_UAS),

This is a VERY heavy hammer for this type of device.  You say that the
device just starts to time out, which implies it overheats or something
else?  Are you sure that slowing it down like this is the best overall
solution?  What does other operating systems do for this device?

thanks,

greg k-h

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH] USB: storage: UAS: ignore UAS on Realtek RTL9210B with firmware f0.01
  2026-03-11 12:35     ` Greg KH
@ 2026-03-11 21:59       ` Nils Reuter
  0 siblings, 0 replies; 5+ messages in thread
From: Nils Reuter @ 2026-03-11 21:59 UTC (permalink / raw)
  To: Greg KH; +Cc: linux-usb


On 11/03/2026 13:35, Greg KH wrote:
> This is a VERY heavy hammer for this type of device.  You say that the
> device just starts to time out, which implies it overheats or something
> else?  Are you sure that slowing it down like this is the best overall
> solution?  What does other operating systems do for this device?
>
> thanks,
>
> greg k-h

Hi Greg,

Please withdraw this patch.

Extensive stress testing with fio confirmed that the device is stable at high temperatures under UAS after I improved the hardware cooling. The initial hangs were likely thermal-related; I cannot confirm with certainty that the firmware version (f0.01) was the root cause.

Apologies for the noise.

Best regards,
Nils



^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2026-03-11 21:59 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-03-08 14:10 [BUG] UAS protocol hangs on Realtek RTL9210B (0bda:9210, bcdDevice f0.01) Nils Reuter
2026-03-08 16:16 ` Greg KH
2026-03-08 19:37   ` [PATCH] USB: storage: UAS: ignore UAS on Realtek RTL9210B with firmware f0.01 Nils Reuter
2026-03-11 12:35     ` Greg KH
2026-03-11 21:59       ` Nils Reuter

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox