Skip to contentSkip to conversation

New: USB errors: uhci_hcd 0000:01:00.4: swiotlb buffer is full (sz: 8 bytes) ; iommu=off

Bug 207147 [email protected] 1 participant 4 messages Apr 7, 2020 USB

lore
Share & tools
lore

[email protected]

author
#0 lore raw

https://bugzilla.kernel.org/show_bug.cgi?id=207147

            Bug ID: 207147
           Summary: USB errors: uhci_hcd 0000:01:00.4: swiotlb buffer is
                    full (sz: 8 bytes) ; iommu=off
           Product: Drivers
           Version: 2.5
    Kernel Version: 5.6.0.rc6
          Hardware: All
                OS: Linux
              Tree: Mainline
            Status: NEW
          Severity: normal
          Priority: P1
         Component: USB
          Assignee: [email protected]
          Reporter: [email protected]
        Regression: No

Platform:

HP Gen9 and Gen10 servers.

This USB device Bus 03 ( usb3 ) is associated with the HP iLO subsystem.

When iommu=on the device and virtual keyboard and mouse works :

lsusb -t

/:  Bus 05.Port 1: Dev 1, Class=root_hub, Driver=xhci_hcd/6p, 5000M
/:  Bus 04.Port 1: Dev 1, Class=root_hub, Driver=xhci_hcd/15p, 480M
    |__ Port 3: Dev 2, If 0, Class=Hub, Driver=hub/2p, 480M
/:  Bus 03.Port 1: Dev 1, Class=root_hub, Driver=uhci_hcd/2p, 12M
    |__ Port 1: Dev 2, If 0, Class=Human Interface Device, Driver=usbhid, 12M
    |__ Port 1: Dev 2, If 1, Class=Human Interface Device, Driver=usbhid, 12M
/:  Bus 02.Port 1: Dev 1, Class=root_hub, Driver=ehci-pci/2p, 480M
    |__ Port 1: Dev 2, If 0, Class=Hub, Driver=hub/8p, 480M
/:  Bus 01.Port 1: Dev 1, Class=root_hub, Driver=ehci-pci/2p, 480M
    |__ Port 1: Dev 2, If 0, Class=Hub, Driver=hub/6p, 480M

When iommu=off the keyboard and mouse are not enumerated:

lsusb -t

/:  Bus 05.Port 1: Dev 1, Class=root_hub, Driver=xhci_hcd/6p, 5000M
/:  Bus 04.Port 1: Dev 1, Class=root_hub, Driver=xhci_hcd/15p, 480M
    |__ Port 3: Dev 2, If 0, Class=Hub, Driver=hub/2p, 480M
/:  Bus 03.Port 1: Dev 1, Class=root_hub, Driver=uhci_hcd/2p, 12M
/:  Bus 02.Port 1: Dev 1, Class=root_hub, Driver=ehci-pci/2p, 480M
/:  Bus 01.Port 1: Dev 1, Class=root_hub, Driver=ehci-pci/2p, 480M

Device info :

[    4.233918] pci 0000:01:00.4: [103c:3300] type 00 class 0x0c0300
[    4.235902] pci 0000:01:00.4: reg 0x20: [io  0x1300-0x131f]
[    8.172229] uhci_hcd 0000:01:00.4: UHCI Host Controller
[    8.185583] clocksource: tsc: mask: 0xffffffffffffffff max_cycles:
0x1e3c373503e, max_idle_ns: 440795250101 ns
[    8.198841] uhci_hcd 0000:01:00.4: new USB bus registered, assigned bus
number 3
[    8.228336] clocksource: Switched to clocksource tsc
[    8.228355] uhci_hcd 0000:01:00.4: detected 8 ports
[    8.255642] uhci_hcd 0000:01:00.4: port count misdetected? forcing to 2
ports
[    8.269807] uhci_hcd 0000:01:00.4: supports USB remote wakeup
[    8.269833] uhci_hcd 0000:01:00.4: irq 19, io base 0x00001300
[    8.284303] usb usb3: default language 0x0409
[    8.284319] usb usb3: udev 1, busnum 3, minor = 256
[    8.284322] usb usb3: New USB device found, idVendor=1d6b, idProduct=0001,
bcdDevice= 5.06
[    8.299961] usb usb3: New USB device strings: Mfr=3, Product=2,
SerialNumber=1
[    8.314767] usb usb3: Product: UHCI Host Controller
[    8.328498] usb usb3: Manufacturer: Linux 5.6.0-rc6+ uhci_hcd
[    8.342875] usb usb3: SerialNumber: 0000:01:00.4
[    8.853062] usb usb4-port3: status 0101, change 0000, 12 Mb/s
[    8.853847] usb 3-1: new full-speed USB device number 2 using uhci_hcd
[    8.869995] uhci_hcd 0000:01:00.4: swiotlb buffer is full (sz: 8 bytes),
total 0 (slots), used 0 (slots)
[    8.886593] uhci_hcd 0000:01:00.4: swiotlb buffer is full (sz: 8 bytes),
total 0 (slots), used 0 (slots)
[    8.903838] uhci_hcd 0000:01:00.4: swiotlb buffer is full (sz: 8 bytes),
total 0 (slots), used 0 (slots)
[    9.074128] ehci-pci 0000:00:1d.0: swiotlb buffer is full (sz: 8 bytes),
total 0 (slots), used 0 (slots)
[    9.090870] usb 1-1: device not accepting address 2, error -11
[    9.105875] hub 5-0:1.0: state 7 ports 6 chg 0000 evt 0000
[    9.105948] hub 5-0:1.0: hub_suspend
[    9.105957] usb usb5: bus auto-suspend, wakeup 1
[    9.142840] usb 3-1: device descriptor read/64, error -11

When iommu=on , the devices are enervated :

[    9.149664] uhci_hcd 0000:01:00.4: new USB bus registered, assigned bus
number 3
[    9.163434] uhci_hcd 0000:01:00.4: detected 8 ports
[    9.176561] uhci_hcd 0000:01:00.4: port count misdetected? forcing to 2
ports
[    9.190639] uhci_hcd 0000:01:00.4: supports USB remote wakeup
[    9.190655] uhci_hcd 0000:01:00.4: irq 19, io base 0x00001300
[    9.204506] uhci_hcd 0000:01:00.4: DMAR: Setting identity map [0x791dc000 -
0x791ddfff]
[    9.220241] uhci_hcd 0000:01:00.4: DMAR: Setting identity map [0x791cb000 -
0x791dbfff]
[    9.233903] uhci_hcd 0000:01:00.4: DMAR: Setting identity map [0x791de000 -
0x791f3fff]
[    9.247398] uhci_hcd 0000:01:00.4: DMAR: Setting identity map [0x791f4000 -
0x791f7fff]
[    9.260767] uhci_hcd 0000:01:00.4: DMAR: 32bit DMA uses non-identity mapping
[    9.274196] usb usb3: default language 0x0409
[    9.274211] usb usb3: udev 1, busnum 3, minor = 256
[    9.274214] usb usb3: New USB device found, idVendor=1d6b, idProduct=0001,
bcdDevice= 5.06
[    9.288716] usb usb3: New USB device strings: Mfr=3, Product=2,
SerialNumber=1
[    9.302288] usb usb3: Product: UHCI Host Controller
[    9.315276] usb usb3: Manufacturer: Linux 5.6.0-rc6+ uhci_hcd
[    9.329689] usb usb3: SerialNumber: 0000:01:00.4
[    9.342071] usb 1-1: new high-speed USB device number 2 using ehci-pci
[    9.356176] usb usb2-port1: status 0501 change 0001
[    9.356199] usb usb3: usb_probe_device
[    9.356201] usb usb3: configuration #1 chosen from 1 choice
[    9.356208] usb usb3: adding 3-0:1.0 (config #1, interface 0)
[   10.512162] usbhid 3-1:1.0: usb_probe_interface
[   10.512164] usbhid 3-1:1.0: usb_probe_interface - got id
[   10.513820] input: BMC Virtual Keyboard  as
/devices/pci0000:00/0000:00:1c.2/0000:01:00.4/usb3/3-1/3-1:1.0/0003:03F0:7029.0001/input/input1
[   10.595493] hid-generic 0003:03F0:7029.0001: input,hidraw0: USB HID v1.01
Keyboard [BMC Virtual Keyboard ] on usb-0000:01:00.4-1/input0
[   10.627314] usbhid 3-1:1.1: usb_probe_interface
[   10.627315] usbhid 3-1:1.1: usb_probe_interface - got id
[   10.629630] input: BMC Virtual Keyboard  as
/devices/pci0000:00/0000:00:1c.2/0000:01:00.4/usb3/3-1/3-1:1.1/0003:03F0:7029.0002/input/input2
[   10.662296] hid-generic 0003:03F0:7029.0002: input,hidraw1: USB HID v1.01
Mouse [BMC Virtual Keyboard ] on usb-0000:01:00.4-1/input1
[   10.696057] usbcore: registered new interface driver usbhid

( Complete dmesg will be attached )

Signature · 2 lines
You are receiving this mail because:
You are watching the assignee of the bug.
author
#2 lore raw

[Bug 207147] USB errors: uhci_hcd 0000:01:00.4: swiotlb buffer is full (sz: 8 bytes) ; iommu=off

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=207147

--- Comment #3 from [email protected] --- Hi Alan,

wrt to comment 2.

If that is the case that iommu=on is needed on that is fine. I just wanted to report my observation.

FWIIW - 4.14 Oracle UEK5 kernels and 4.18 Redhat kernels do not require iommu=on to have the HP iLo devices operate. This is an operational difference between the 4.x and 5.x kernels.

Our own Oracle servers do not exhibit the same failures as HP Gen9 and Gen10 do. They operation fine regardless of the iommu setting(s).

Signature · 2 lines
You are receiving this mail because:
You are watching the assignee of the bug.
author
#3 lore raw

[Bug 207147] USB errors: uhci_hcd 0000:01:00.4: swiotlb buffer is full (sz: 8 bytes) ; iommu=off

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=207147

Alan Stern ([email protected]) changed:

What |Removed |Added
----------------------------------------------------------------------------
CC| |[email protected]

--- Comment #4 from Alan Stern ([email protected]) ---
I would appreciate if you would carry out a git bisection to find the commit that changed the behavior between 4.18 and 5.6-rc6. I suspect that commit will turn out to affect a subsystem completely different from USB (like your previous bug report, where the important commit affected the iommu subsystem).

Signature · 2 lines
You are receiving this mail because:
You are watching the assignee of the bug.