Re: NEC uPD720200 xHCI Controller dies when Runtime PM enabled
Hello, On Mon, Aug 1, 2016 at 1:44 PM, Durval Menezes wrote: > Hi Mike, > > On Mon, Aug 1, 2016 at 12:05 PM, Mike Murdoch > wrote: > > On 2016-08-01 13:57, Durval Menezes wrote: > > > Hi Mathias, > > > > > > On Mon, Aug 1, 2016 at 8:20 AM, Mathias Nyman > > > wrote: > > >>> On 01.08.2016 13:15, Durval Menezes wrote: > > >>> Hello Mike, Mathias, list, > > >>> > > >>> On 06.02.2016 19:08, Mike Murdoch wrote: > > >>> Bug ID: 111251 > > >>> > > >>> I have a NEC uPD720200 USB3.0 controller in a Thinkpad W520 laptop on > > >>> kernel 4.4.1-gentoo. > > >>> > > >>> 0e:00.0 USB controller: NEC Corporation uPD720200 USB 3.0 Host > > >>> Controller (rev 04) (prog-if 30 [XHCI]) > > >>> Subsystem: Lenovo uPD720200 USB 3.0 Host Controller > > >>> Flags: bus master, fast devsel, latency 0 > > >>> Memory at f380 (64-bit, non-prefetchable) [size=8K] > > >>> Capabilities: [50] Power Management version 3 > > >>> Capabilities: [70] MSI: Enable- Count=1/8 Maskable- 64bit+ > > >>> Capabilities: [90] MSI-X: Enable+ Count=8 Masked- > > >>> Capabilities: [a0] Express Endpoint, MSI 00 > > >>> Capabilities: [100] Advanced Error Reporting > > >>> Capabilities: [140] Device Serial Number ff-ff-ff-ff-ff-ff-ff-ff > > >>> Capabilities: [150] Latency Tolerance Reporting > > >>> Kernel driver in use: xhci_hcd > > >>> Kernel modules: xhci_pci > > >>> > > >>> When runtime power control for this controller is disabled > > >>> (/sys/bus/pci/devices/:0e:00.0/power/control = on), the controller > > >>> works fine and reaches over 120MB/s transfer rates. > > >>> > > >>> When runtime power control for this controller is enabled > > >>> (/sys/bus/pci/devices/:0e:00.0/power/control = auto), two effects > > >>> can be observed: > > >>> > > >>> - Transfer rates are much lower at around 30MB/s > > >>> - During transfers, the controller dies after a couple of seconds: > > >>> > > >>> I found this message in the list archives, and I have the exact same > > >>> issues on exactly the same hardware (Thinkpad W520 laptop with the same > > >>> USB3 controller showing on lspci -v); otherwise, I'm running distro > > >>> kernel > > >>> 2.6.32-573.7.1.el6.x86_64 on a Springdale Linux 6.7 (RHEL6) install. > > >>> > > >>> I just verified that my controller's PM was set by default to "auto": > > >>> cat /sys/bus/pci/devices/\:0e\:00.0/power/control > > >>> auto > > >>> I have now set it to "on" and will test whether this will work around > > >>> the issue (I'm waiting for my USB3.0 "heavy duty" disk docks to be > > >>> released from another system that is using them right now). The docks (actually a 4-disk Mediasonic Probox enclosure, and a single-disk USpeed SATA-to-USB adapter) have returned; I've rebooted my machine (to make sure I was starting from as clean a slate as possible), then (before plugging anything) set the controller's PM to "on" (ie, no power management) with the "echo" command above, and confirmed it with the "cat" command above. Then I tried to plug first the adapter to each of the two ports controlled by the uPD720200; the result for each attempt (as recorded on syslog with level debug) was just: Aug 2 11:45:16 localhost kernel: hub 3-0:1.0: unable to enumerate USB device on port 1 For the lower port, and Aug 2 11:54:19 localhost kernel: hub 3-0:1.0: unable to enumerate USB device on port 2 For the upper port. To confirm that the adapter is working, I connected it to the "combo" USB/eSATA adapter (which on the W520 is right besides the two uPD720200 ports) and it worked great (albeit limited to USB2.1, as this is this port's type): Aug 2 11:56:46 localhost kernel: usb 2-1.2: new high speed USB device number 3 using ehci_hcd Aug 2 11:56:47 localhost kernel: usb 2-1.2: New USB device found, idVendor=174c, idProduct=5106 Aug 2 11:56:47 localhost kernel: usb 2-1.2: New USB device strings: Mfr=2, Product=3, SerialNumber=1 Aug 2 11:56:47 localhost kernel: usb 2-1.2: Product: AS2105 Aug 2 11:56:47 localhost kernel: usb 2-1.2: M
Re: NEC uPD720200 xHCI Controller dies when Runtime PM enabled
Hi Mike, On Mon, Aug 1, 2016 at 12:05 PM, Mike Murdoch wrote: > On 2016-08-01 13:57, Durval Menezes wrote: > > Hi Mathias, > > > > On Mon, Aug 1, 2016 at 8:20 AM, Mathias Nyman > > wrote: > >>> On 01.08.2016 13:15, Durval Menezes wrote: > >>> Hello Mike, Mathias, list, > >>> > >>> On 06.02.2016 19:08, Mike Murdoch wrote: > >>> Bug ID: 111251 > >>> > >>> I have a NEC uPD720200 USB3.0 controller in a Thinkpad W520 laptop on > >>> kernel 4.4.1-gentoo. > >>> > >>> 0e:00.0 USB controller: NEC Corporation uPD720200 USB 3.0 Host > >>> Controller (rev 04) (prog-if 30 [XHCI]) > >>> Subsystem: Lenovo uPD720200 USB 3.0 Host Controller > >>> Flags: bus master, fast devsel, latency 0 > >>> Memory at f380 (64-bit, non-prefetchable) [size=8K] > >>> Capabilities: [50] Power Management version 3 > >>> Capabilities: [70] MSI: Enable- Count=1/8 Maskable- 64bit+ > >>> Capabilities: [90] MSI-X: Enable+ Count=8 Masked- > >>> Capabilities: [a0] Express Endpoint, MSI 00 > >>> Capabilities: [100] Advanced Error Reporting > >>> Capabilities: [140] Device Serial Number ff-ff-ff-ff-ff-ff-ff-ff > >>> Capabilities: [150] Latency Tolerance Reporting > >>> Kernel driver in use: xhci_hcd > >>> Kernel modules: xhci_pci > >>> > >>> When runtime power control for this controller is disabled > >>> (/sys/bus/pci/devices/:0e:00.0/power/control = on), the controller > >>> works fine and reaches over 120MB/s transfer rates. > >>> > >>> When runtime power control for this controller is enabled > >>> (/sys/bus/pci/devices/:0e:00.0/power/control = auto), two effects > >>> can be observed: > >>> > >>> - Transfer rates are much lower at around 30MB/s > >>> - During transfers, the controller dies after a couple of seconds: > >>> > >>> I found this message in the list archives, and I have the exact same > >>> issues on exactly the same hardware (Thinkpad W520 laptop with the same > >>> USB3 controller showing on lspci -v); otherwise, I'm running distro kernel > >>> 2.6.32-573.7.1.el6.x86_64 on a Springdale Linux 6.7 (RHEL6) install. > >>> > >>> I just verified that my controller's PM was set by default to "auto": > >>> cat /sys/bus/pci/devices/\:0e\:00.0/power/control > >>> auto > >>> I have now set it to "on" and will test whether this will work around > >>> the issue (I'm waiting for my USB3.0 "heavy duty" disk docks to be > >>> released from another system that is using them right now). > >>> > >>> I have one question for Mike: have you upgraded your uPD720200 controller > >>> firmware (as per [1], [2]) or are you still running stock? > >>> > >>> Also, one question for Mathias: do you know whether your patches at [3] > >>> can be applied to kernel 2.6.32? > >> The last patch in [3] is faulty. So don't use the patches from the mail. > >> > >> I just force updated that branch, so if you like you can try to backport > >> patches from: > >> > >> git://git.kernel.org/pub/scm/linux/kernel/git/mnyman/xhci.git > >> bug_usb3_enum_rtresume > >> > >> only 2 patches are relevant: > >> > >> 8caabe9 xhci: Don't suspend the xhci bus it there is a pending event. > >> 4427456 xhci: resume USB 3 roothub first > > Thanks Mathias. Now I only need Mike's response concerning the firmware > > in order to proceed. > > > No, I haven't tried updating the firmware. Feel free to give it a go, > I'm curious if it'll make a difference. As long as the patch (or the workaround) allow me to avoid the issue, I'd rather let sleeping dragons lie (windows-only update procedure, etc) :-) > As for the patches. All three of them did fix this bug, but introduced > other problems (I don't remember details, sorry). As Mathias said, the > last one is faulty. However, using only the first two patches is *not* > enough to completely fix this bug (I verified it just now). No prob, Mathias sent me a reference to the relevant (and presumably working) patches, I will first try the "disable PM mode" workaround and later (possibly *much* later) to backport the patches to my kernel. > Unfortunately I don't have the time to do much testing. The Thinkpad is > used by someone else and I only have access to it on the weekends. A > workaround is to just disable runtime power management. Thanks for the feedback. > Let me know how things work for you! Will do: will post a progress report (on the workaround) later today to the list and, as you asked, directly to your email too. Thanks again, -- Durval Menezes (durval AT tmp DOT com DOT br, http://www.tmp.com.br/) -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: NEC uPD720200 xHCI Controller dies when Runtime PM enabled
Hi Mathias, On Mon, Aug 1, 2016 at 8:20 AM, Mathias Nyman wrote: > > On 01.08.2016 13:15, Durval Menezes wrote: > > Hello Mike, Mathias, list, > > > > On 06.02.2016 19:08, Mike Murdoch wrote: > > Bug ID: 111251 > > > > I have a NEC uPD720200 USB3.0 controller in a Thinkpad W520 laptop on > > kernel 4.4.1-gentoo. > > > > 0e:00.0 USB controller: NEC Corporation uPD720200 USB 3.0 Host > > Controller (rev 04) (prog-if 30 [XHCI]) > > Subsystem: Lenovo uPD720200 USB 3.0 Host Controller > > Flags: bus master, fast devsel, latency 0 > > Memory at f380 (64-bit, non-prefetchable) [size=8K] > > Capabilities: [50] Power Management version 3 > > Capabilities: [70] MSI: Enable- Count=1/8 Maskable- 64bit+ > > Capabilities: [90] MSI-X: Enable+ Count=8 Masked- > > Capabilities: [a0] Express Endpoint, MSI 00 > > Capabilities: [100] Advanced Error Reporting > > Capabilities: [140] Device Serial Number ff-ff-ff-ff-ff-ff-ff-ff > > Capabilities: [150] Latency Tolerance Reporting > > Kernel driver in use: xhci_hcd > > Kernel modules: xhci_pci > > > > When runtime power control for this controller is disabled > > (/sys/bus/pci/devices/:0e:00.0/power/control = on), the controller > > works fine and reaches over 120MB/s transfer rates. > > > > When runtime power control for this controller is enabled > > (/sys/bus/pci/devices/:0e:00.0/power/control = auto), two effects > > can be observed: > > > > - Transfer rates are much lower at around 30MB/s > > - During transfers, the controller dies after a couple of seconds: > > > > I found this message in the list archives, and I have the exact same > > issues on exactly the same hardware (Thinkpad W520 laptop with the same > > USB3 controller showing on lspci -v); otherwise, I'm running distro kernel > > 2.6.32-573.7.1.el6.x86_64 on a Springdale Linux 6.7 (RHEL6) install. > > > > I just verified that my controller's PM was set by default to "auto": > > cat /sys/bus/pci/devices/\:0e\:00.0/power/control > > auto > > I have now set it to "on" and will test whether this will work around > > the issue (I'm waiting for my USB3.0 "heavy duty" disk docks to be > > released from another system that is using them right now). > > > > I have one question for Mike: have you upgraded your uPD720200 controller > > firmware (as per [1], [2]) or are you still running stock? > > > > Also, one question for Mathias: do you know whether your patches at [3] > > can be applied to kernel 2.6.32? > > The last patch in [3] is faulty. So don't use the patches from the mail. > > I just force updated that branch, so if you like you can try to backport > patches from: > > git://git.kernel.org/pub/scm/linux/kernel/git/mnyman/xhci.git > bug_usb3_enum_rtresume > > only 2 patches are relevant: > > 8caabe9 xhci: Don't suspend the xhci bus it there is a pending event. > 4427456 xhci: resume USB 3 roothub first Thanks Mathias. Now I only need Mike's response concerning the firmware in order to proceed. Cheers, -- Durval Menezes (durval AT tmp DOT com DOT br, http://www.tmp.com.br/) -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: NEC uPD720200 xHCI Controller dies when Runtime PM enabled
Hello Mike, Mathias, list, On 06.02.2016 19:08, Mike Murdoch wrote: > Bug ID: 111251 > > I have a NEC uPD720200 USB3.0 controller in a Thinkpad W520 laptop on > kernel 4.4.1-gentoo. > > 0e:00.0 USB controller: NEC Corporation uPD720200 USB 3.0 Host > Controller (rev 04) (prog-if 30 [XHCI]) > Subsystem: Lenovo uPD720200 USB 3.0 Host Controller > Flags: bus master, fast devsel, latency 0 > Memory at f380 (64-bit, non-prefetchable) [size=8K] > Capabilities: [50] Power Management version 3 > Capabilities: [70] MSI: Enable- Count=1/8 Maskable- 64bit+ > Capabilities: [90] MSI-X: Enable+ Count=8 Masked- > Capabilities: [a0] Express Endpoint, MSI 00 > Capabilities: [100] Advanced Error Reporting > Capabilities: [140] Device Serial Number ff-ff-ff-ff-ff-ff-ff-ff > Capabilities: [150] Latency Tolerance Reporting > Kernel driver in use: xhci_hcd > Kernel modules: xhci_pci > > When runtime power control for this controller is disabled > (/sys/bus/pci/devices/:0e:00.0/power/control = on), the controller > works fine and reaches over 120MB/s transfer rates. > > When runtime power control for this controller is enabled > (/sys/bus/pci/devices/:0e:00.0/power/control = auto), two effects > can be observed: > > - Transfer rates are much lower at around 30MB/s > - During transfers, the controller dies after a couple of seconds: I found this message in the list archives, and I have the exact same issues on exactly the same hardware (Thinkpad W520 laptop with the same USB3 controller showing on lspci -v); otherwise, I'm running distro kernel 2.6.32-573.7.1.el6.x86_64 on a Springdale Linux 6.7 (RHEL6) install. I just verified that my controller's PM was set by default to "auto": cat /sys/bus/pci/devices/\:0e\:00.0/power/control auto I have now set it to "on" and will test whether this will work around the issue (I'm waiting for my USB3.0 "heavy duty" disk docks to be released from another system that is using them right now). I have one question for Mike: have you upgraded your uPD720200 controller firmware (as per [1], [2]) or are you still running stock? Also, one question for Mathias: do you know whether your patches at [3] can be applied to kernel 2.6.32? References [1] https://forums.lenovo.com/t5/ThinkPad-P-and-W-Series-Mobile/Anyone-updated-their-W520-USB-3-0-firmware/td-p/1164719 [2] http://pete.akeo.ie/2011/10/flashing-necrenesas-usb-30.html [3] http://marc.info/?l=linux-usb&m=145684596900873&w=2 Cheers, -- Durval Menezes (durval AT tmp DOT com DOT br, http://www.tmp.com.br/) -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
