New: SuperSpeed+ speed reporting wrong from Asmedia ASM3242

https://bugzilla.kernel.org/show_bug.cgi?id=217434

            Bug ID: 217434
           Summary: SuperSpeed+ speed reporting wrong from Asmedia ASM3242
           Product: Drivers
           Version: 2.5
          Hardware: All
                OS: Linux
            Status: NEW
          Severity: normal
          Priority: P3
         Component: USB
          Assignee: [email protected]
          Reporter: [email protected]
        Regression: No

I have a PCIe plugin card with an Asmedia ASM3242 controller, PCI ID 1b21:3242, that supports a 20Gbps transfer rate. SuperSpeed+ devices do connect at 10 and 20Gbps where supported, but lsusb -t always reports that they are connected at 5Gbps.

The devices I've tested this with are:

- 10Gbps USB hub
- RTL9210B based NVME enclosure, 10Gbps
- ASM236X based NVME enclosure, 20Gbps

I believe the NVME enclosures are connecting at the stated speeds from observing their measured transfer rates (which would be impossible otherwise).

This issue has been discussed already on linux-usb:

[1] https://marc.info/?l=linux-usb&m=161527212127481&w=2

and later when I encountered it:

[2] https://marc.info/?l=linux-usb&m=168046080403282&w=2

I applied Thinh's patch from the discussion under [1] ('capture the device speed
from sublink speed notification') to Linux 6.2.9 and it corrected the behaviour, but as noted in [2] this was a test-of-concept patch and a proper xHCI quirk is needed. Unfortunately this is somewhat beyond this reporter's ability -- hence I've made this KBZ entry, mainly so the issue doesn't get lost.

You may reply to this email to add a comment.

You are receiving this mail because:
You are watching the assignee of the bug.

https://bugzilla.kernel.org/show_bug.cgi?id=217434

--- Comment #1 from James Ettle ([email protected]) ---
The same seems to apply to the ASM2142/3142 controller (1b21:2142). This is a 10Gbps chip, and devices do seem to connect at 10Gbps where supported, but this is reported as 5Gbps.

You may reply to this email to add a comment.

You are receiving this mail because:
You are watching the assignee of the bug.