| Message ID | 20200224062145.25785-1-jckuo@nvidia.com (mailing list archive) |
|---|---|
| State | Mainlined |
| Commit | cbb23d5572f0186f193b1af65a7994f2d40b1d43 |
| Headers | show |
| Series | usb: host: xhci-tegra: Tegra186/Tegra194 LPM | expand |
On 24.2.2020 8.21, JC Kuo wrote: > Tegra186 and Tegra194 xHC supports USB 3.0 LPM. This commit enables > XHCI_LPM_SUPPORT quirk for Tegra186 and Tegra194. > > Signed-off-by: JC Kuo <jckuo@nvidia.com> Thanks, added to queue -Mathias
On Mon, Feb 24, 2020 at 02:21:45PM +0800, JC Kuo wrote: > Tegra186 and Tegra194 xHC supports USB 3.0 LPM. This commit enables > XHCI_LPM_SUPPORT quirk for Tegra186 and Tegra194. > > Signed-off-by: JC Kuo <jckuo@nvidia.com> > --- > drivers/usb/host/xhci-tegra.c | 7 +++++++ > 1 file changed, 7 insertions(+) I see that Mathias has already queued this up, but for the record: Acked-by: Thierry Reding <treding@nvidia.com> JC, is there some way that we can test this? I see that there are some sysfs files that can control LPM enablement on a per-device basis, but is there some way to check that this works as expected? Or do we just assume everything is fine as long as all the devices continue to operate properly? Perhaps there are some state transition counters or something that would indicate that devices are properly transitioning? Thierry
Hi Thierry,
Yes, it can be verified with a LPM capable device. For example, a VIA USB 3.0
hub is connected to Jetson-Xavier. "lsusb -v" output [1] shows the device
supports LPM and the host has enabled U1/U2 states for the device. If host LPM
is disabled, there will be no "U1 Enabled" and "U2 Enabled" strings in "Device
Status" section.
To check LPM operation, disconnect all USB 3.0 devices from the hub and disable
runtime PM for the super-speed portion of the hub, so that it won't be
auto-suspended.
root@tegra-ubuntu:~# echo on > /sys/bus/usb/devices/2-4/power/control
Since there is no data transaction for the hub, link will enter U2 soon. This
can be checked by reading XHCI.PORTSC register. In below, PLS (Port Link State)
field is U2.
root@tegra-ubuntu:~# devmem 0x3610450
0x00001243
Thanks,
JC
[1] lsusb -v -d 2109:0820
Bus 002 Device 002: ID 2109:0820 VIA Labs, Inc.
Device Descriptor:
bLength 18
bDescriptorType 1
bcdUSB 3.10
bDeviceClass 9 Hub
bDeviceSubClass 0 Unused
bDeviceProtocol 3
bMaxPacketSize0 9
idVendor 0x2109 VIA Labs, Inc.
idProduct 0x0820
bcdDevice 4.e3
iManufacturer 1 VIA Labs, Inc.
iProduct 2 USB3.0 Hub
iSerial 0
bNumConfigurations 1
...
Binary Object Store Descriptor:
bLength 5
bDescriptorType 15
wTotalLength 42
bNumDeviceCaps 3
USB 2.0 Extension Device Capability:
bLength 7
bDescriptorType 16
bDevCapabilityType 2
bmAttributes 0x00000002
Link Power Management (LPM) Supported
SuperSpeed USB Device Capability:
bLength 10
bDescriptorType 16
bDevCapabilityType 3
bmAttributes 0x00
wSpeedsSupported 0x000e
Device can operate at Full Speed (12Mbps)
Device can operate at High Speed (480Mbps)
Device can operate at SuperSpeed (5Gbps)
bFunctionalitySupport 1
Lowest fully-functional device speed is Full Speed (12Mbps)
bU1DevExitLat 4 micro seconds
bU2DevExitLat 231 micro seconds
Container ID Device Capability:
bLength 20
bDescriptorType 16
bDevCapabilityType 4
bReserved 0
ContainerID {5cf3ee30-d507-4925-b001-802d79434c30}
Device Status: 0x000d
Self Powered
U1 Enabled
U2 Enabled
On 2/24/20 8:51 PM, Thierry Reding wrote:
> On Mon, Feb 24, 2020 at 02:21:45PM +0800, JC Kuo wrote:
>> Tegra186 and Tegra194 xHC supports USB 3.0 LPM. This commit enables
>> XHCI_LPM_SUPPORT quirk for Tegra186 and Tegra194.
>>
>> Signed-off-by: JC Kuo <jckuo@nvidia.com>
>> ---
>> drivers/usb/host/xhci-tegra.c | 7 +++++++
>> 1 file changed, 7 insertions(+)
>
> I see that Mathias has already queued this up, but for the record:
>
> Acked-by: Thierry Reding <treding@nvidia.com>
>
> JC, is there some way that we can test this? I see that there are some
> sysfs files that can control LPM enablement on a per-device basis, but
> is there some way to check that this works as expected? Or do we just
> assume everything is fine as long as all the devices continue to operate
> properly? Perhaps there are some state transition counters or something
> that would indicate that devices are properly transitioning?
>
> Thierry
>
On 26.2.2020 10.12, JC Kuo wrote: > Hi Thierry, > Yes, it can be verified with a LPM capable device. For example, a VIA USB 3.0 > hub is connected to Jetson-Xavier. "lsusb -v" output [1] shows the device > supports LPM and the host has enabled U1/U2 states for the device. If host LPM > is disabled, there will be no "U1 Enabled" and "U2 Enabled" strings in "Device > Status" section. > > To check LPM operation, disconnect all USB 3.0 devices from the hub and disable > runtime PM for the super-speed portion of the hub, so that it won't be > auto-suspended. > root@tegra-ubuntu:~# echo on > /sys/bus/usb/devices/2-4/power/control > > Since there is no data transaction for the hub, link will enter U2 soon. This > can be checked by reading XHCI.PORTSC register. In below, PLS (Port Link State) > field is U2. > root@tegra-ubuntu:~# devmem 0x3610450 > 0x00001243 A more human friendly way to read portsc registers: # cat /sys/kernel/debug/usb/xhci/0000\:00\:14.0/ports/port02/portsc Powered Connected Enabled Link:U0 PortSpeed:3 Change: Wake: -Mathias
Hi Mathias, Thanks for the tip. It's really helpful. root@tegra-ubuntu:/d/usb/xhci/3610000.usb/ports/port04# cat portsc Powered Connected Enabled Link:U2 PortSpeed:4 Change: Wake: -JC On 2/26/20 5:32 PM, Mathias Nyman wrote: > On 26.2.2020 10.12, JC Kuo wrote: >> Hi Thierry, >> Yes, it can be verified with a LPM capable device. For example, a VIA USB 3.0 >> hub is connected to Jetson-Xavier. "lsusb -v" output [1] shows the device >> supports LPM and the host has enabled U1/U2 states for the device. If host LPM >> is disabled, there will be no "U1 Enabled" and "U2 Enabled" strings in "Device >> Status" section. >> >> To check LPM operation, disconnect all USB 3.0 devices from the hub and disable >> runtime PM for the super-speed portion of the hub, so that it won't be >> auto-suspended. >> root@tegra-ubuntu:~# echo on > /sys/bus/usb/devices/2-4/power/control >> >> Since there is no data transaction for the hub, link will enter U2 soon. This >> can be checked by reading XHCI.PORTSC register. In below, PLS (Port Link State) >> field is U2. >> root@tegra-ubuntu:~# devmem 0x3610450 >> 0x00001243 > > A more human friendly way to read portsc registers: > > # cat /sys/kernel/debug/usb/xhci/0000\:00\:14.0/ports/port02/portsc > Powered Connected Enabled Link:U0 PortSpeed:3 Change: Wake: > > -Mathias >
diff --git a/drivers/usb/host/xhci-tegra.c b/drivers/usb/host/xhci-tegra.c index 8163aefc6c6b..a6e36b3c968f 100644 --- a/drivers/usb/host/xhci-tegra.c +++ b/drivers/usb/host/xhci-tegra.c @@ -203,6 +203,7 @@ struct tegra_xusb_soc { bool scale_ss_clock; bool has_ipfs; + bool lpm_support; }; struct tegra_xusb_context { @@ -1779,6 +1780,7 @@ static const struct tegra_xusb_soc tegra186_soc = { .data_out = 0xec, .owner = 0xf0, }, + .lpm_support = true, }; static const char * const tegra194_supply_names[] = { @@ -1808,6 +1810,7 @@ static const struct tegra_xusb_soc tegra194_soc = { .data_out = 0x70, .owner = 0x74, }, + .lpm_support = true, }; MODULE_FIRMWARE("nvidia/tegra194/xusb.bin"); @@ -1832,7 +1835,11 @@ static struct platform_driver tegra_xusb_driver = { static void tegra_xhci_quirks(struct device *dev, struct xhci_hcd *xhci) { + struct tegra_xusb *tegra = dev_get_drvdata(dev); + xhci->quirks |= XHCI_PLAT; + if (tegra && tegra->soc->lpm_support) + xhci->quirks |= XHCI_LPM_SUPPORT; } static int tegra_xhci_setup(struct usb_hcd *hcd)
Tegra186 and Tegra194 xHC supports USB 3.0 LPM. This commit enables XHCI_LPM_SUPPORT quirk for Tegra186 and Tegra194. Signed-off-by: JC Kuo <jckuo@nvidia.com> --- drivers/usb/host/xhci-tegra.c | 7 +++++++ 1 file changed, 7 insertions(+)