Never seen XHCI auto handoff working on TI and RENESAS cards.
Eventually, we force handoff. This code forces the handoff
unconditionally. It saves 5 seconds boot time for each card.
Signed-off-by: Babu Moger <[email protected]>
---
v2:
Made changes per comments from Greg KH.
Extra space removal in assignment
Added both vendor and device id checks.
Resending the patch. Original discussion here.
https://marc.info/?t=145221162000007&r=1&w=4
drivers/usb/host/pci-quirks.c | 8 ++++++++
1 files changed, 8 insertions(+), 0 deletions(-)
diff --git a/drivers/usb/host/pci-quirks.c b/drivers/usb/host/pci-quirks.c
index 35af362..31c9502 100644
--- a/drivers/usb/host/pci-quirks.c
+++ b/drivers/usb/host/pci-quirks.c
@@ -996,6 +996,14 @@ static void quirk_usb_handoff_xhci(struct pci_dev *pdev)
}
val = readl(base + ext_cap_offset);
+ /* Auto handoff never worked for these devices. Force it and continue */
+ if ((pdev->vendor == PCI_VENDOR_ID_TI && pdev->device == 0x8241) ||
+ (pdev->vendor == PCI_VENDOR_ID_RENESAS
+ && pdev->device == 0x0014)) {
+ val = (val | XHCI_HC_OS_OWNED) & ~XHCI_HC_BIOS_OWNED;
+ writel(val, base + ext_cap_offset);
+ }
+
/* If the BIOS owns the HC, signal that the OS wants it, and wait */
if (val & XHCI_HC_BIOS_OWNED) {
writel(val | XHCI_HC_OS_OWNED, base + ext_cap_offset);
--
1.7.1
On 22.10.2016 01:25, Babu Moger wrote:
> Never seen XHCI auto handoff working on TI and RENESAS cards.
> Eventually, we force handoff. This code forces the handoff
> unconditionally. It saves 5 seconds boot time for each card.
>
> Signed-off-by: Babu Moger <[email protected]>
Do the Renesas and TI controllers still advertise the extended capability
for the handoff? (XHCI_EXT_CAPS_LEGACY)
Is this some known issue with these vendors controllers? Is there some documentation
about this, errata or anything?
Adding Yoshihiro Shimoda, he might know about the Renesas controller.
-Mathias
Hi,
> From: Mathias Nyman
> Sent: Monday, October 24, 2016 6:58 PM
>
> On 22.10.2016 01:25, Babu Moger wrote:
> > Never seen XHCI auto handoff working on TI and RENESAS cards.
> > Eventually, we force handoff. This code forces the handoff
> > unconditionally. It saves 5 seconds boot time for each card.
> >
> > Signed-off-by: Babu Moger <[email protected]>
>
> Do the Renesas and TI controllers still advertise the extended capability
> for the handoff? (XHCI_EXT_CAPS_LEGACY)
>
> Is this some known issue with these vendors controllers? Is there some documentation
> about this, errata or anything?
>
> Adding Yoshihiro Shimoda, he might know about the Renesas controller.
Thank you for adding me on this email.
However, unfortunately I don't know the detail about Renesas PCIe xHCI controllers.
(I know the xHCI controller of R-Car SoCs for now.)
Best regards,
Yoshihiro Shimoda
On 10/24/2016 5:54 AM, Yoshihiro Shimoda wrote:
> Hi,
>
>> From: Mathias Nyman
>> Sent: Monday, October 24, 2016 6:58 PM
>>
>> On 22.10.2016 01:25, Babu Moger wrote:
>>> Never seen XHCI auto handoff working on TI and RENESAS cards.
>>> Eventually, we force handoff. This code forces the handoff
>>> unconditionally. It saves 5 seconds boot time for each card.
>>>
>>> Signed-off-by: Babu Moger <[email protected]>
>> Do the Renesas and TI controllers still advertise the extended capability
>> for the handoff? (XHCI_EXT_CAPS_LEGACY)
I don't see this capability. Here is lspci output.
# lspci -s 0009:01:00.0 -vvv
0009:01:00.0 USB controller: Texas Instruments TUSB73x0 SuperSpeed USB
3.0 xHCI Host Controller (rev 02) (prog-if 30 [XHCI])
Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop-
ParErr- Stepping- SERR- FastB2B- DisINTx+
Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort-
<TAbort- <MAbort- >SERR- <PERR- INTx-
Latency: 0, Cache Line Size: 64 bytes
Interrupt: pin A routed to IRQ 0000000e
Region 0: Memory at 100000000 (64-bit, non-prefetchable) [size=64K]
Region 2: Memory at 100010000 (64-bit, non-prefetchable) [size=8K]
Region 4: [virtual] Memory at fffdfdc000000000 (32-bit,
non-prefetchable)
Region 5: [virtual] Memory at fffdfdc000000000 (32-bit,
non-prefetchable)
[virtual] Expansion ROM at fffdfdc000000000 [disabled]
Capabilities: [40] Power Management version 3
Flags: PMEClk- DSI- D1+ D2+ AuxCurrent=0mA
PME(D0+,D1+,D2+,D3hot+,D3cold-)
Status: D0 NoSoftRst+ PME-Enable- DSel=0 DScale=0 PME-
Capabilities: [48] MSI: Enable- Count=1/8 Maskable- 64bit+
Address: 0000000000000000 Data: 0000
Capabilities: [70] Express (v2) Endpoint, MSI 00
DevCap: MaxPayload 1024 bytes, PhantFunc 0, Latency L0s
unlimited, L1 unlimited
ExtTag- AttnBtn- AttnInd- PwrInd- RBE+ FLReset-
DevCtl: Report errors: Correctable- Non-Fatal- Fatal-
Unsupported-
RlxdOrd+ ExtTag- PhantFunc- AuxPwr- NoSnoop-
MaxPayload 256 bytes, MaxReadReq 512 bytes
DevSta: CorrErr+ UncorrErr- FatalErr- UnsuppReq+
AuxPwr- TransPend-
LnkCap: Port #0, Speed 5GT/s, Width x1, ASPM L0s L1,
Latency L0 <2us, L1 <64us
ClockPM+ Surprise- LLActRep- BwNot-
LnkCtl: ASPM Disabled; RCB 64 bytes Disabled- Retrain-
CommClk-
ExtSynch- ClockPM- AutWidDis- BWInt- AutBWInt-
LnkSta: Speed 5GT/s, Width x1, TrErr- Train- SlotClk+
DLActive- BWMgmt- ABWMgmt-
DevCap2: Completion Timeout: Not Supported,
TimeoutDis+, LTR-, OBFF Not Supported
DevCtl2: Completion Timeout: 50us to 50ms, TimeoutDis-,
LTR-, OBFF Disabled
LnkCtl2: Target Link Speed: 5GT/s, EnterCompliance-
SpeedDis-
Transmit Margin: Normal Operating Range,
EnterModifiedCompliance- ComplianceSOS-
Compliance De-emphasis: -6dB
LnkSta2: Current De-emphasis Level: -3.5dB,
EqualizationComplete-, EqualizationPhase1-
EqualizationPhase2-, EqualizationPhase3-,
LinkEqualizationRequest-
Capabilities: [c0] MSI-X: Enable+ Count=8 Masked-
Vector table: BAR=2 offset=00000000
PBA: BAR=2 offset=00001000
Capabilities: [100 v2] Advanced Error Reporting
UESta: DLP- SDES- TLP- FCP- CmpltTO- CmpltAbrt-
UnxCmplt- RxOF- MalfTLP- ECRC- UnsupReq- ACSViol-
UEMsk: DLP- SDES- TLP- FCP- CmpltTO- CmpltAbrt-
UnxCmplt- RxOF- MalfTLP- ECRC- UnsupReq- ACSViol-
UESvrt: DLP+ SDES+ TLP- FCP+ CmpltTO- CmpltAbrt-
UnxCmplt- RxOF+ MalfTLP+ ECRC- UnsupReq- ACSViol-
CESta: RxErr- BadTLP- BadDLLP- Rollover- Timeout-
NonFatalErr+
CEMsk: RxErr- BadTLP- BadDLLP- Rollover- Timeout-
NonFatalErr+
AERCap: First Error Pointer: 00, GenCap+ CGenEn-
ChkCap+ ChkEn-
Capabilities: [150 v1] Device Serial Number 08-00-28-00-00-20-00-00
Kernel driver in use: xhci_hcd
>>
>> Is this some known issue with these vendors controllers? Is there some documentation
>> about this, errata or anything?
>>
>> Adding Yoshihiro Shimoda, he might know about the Renesas controller.
> Thank you for adding me on this email.
> However, unfortunately I don't know the detail about Renesas PCIe xHCI controllers.
> (I know the xHCI controller of R-Car SoCs for now.)
>
> Best regards,
> Yoshihiro Shimoda
>
On 24.10.2016 17:52, Babu Moger wrote:
>
> On 10/24/2016 5:54 AM, Yoshihiro Shimoda wrote:
>> Hi,
>>
>>> From: Mathias Nyman
>>> Sent: Monday, October 24, 2016 6:58 PM
>>>
>>> On 22.10.2016 01:25, Babu Moger wrote:
>>>> Never seen XHCI auto handoff working on TI and RENESAS cards.
>>>> Eventually, we force handoff. This code forces the handoff
>>>> unconditionally. It saves 5 seconds boot time for each card.
>>>>
>>>> Signed-off-by: Babu Moger <[email protected]>
>>> Do the Renesas and TI controllers still advertise the extended capability
>>> for the handoff? (XHCI_EXT_CAPS_LEGACY)
> I don't see this capability. Here is lspci output.
>
It's not a PCI capability, it's a xhci Extended Capability.
If the capability is supported, and handoff fail, then quirk_usb_handoff_xhci() will print
"xHCI BIOS handoff failed (BIOS bug ?)"
Further on in the same function we wait for the "controller not ready" bit in the status
register to clear. If that times it prints out:
"xHCI HW not ready after 5 sec (HC bug?) status"
Do you see any of these two messages in the log?
-Mathias
On 10/25/2016 1:51 AM, Mathias Nyman wrote:
> On 24.10.2016 17:52, Babu Moger wrote:
>>
>> On 10/24/2016 5:54 AM, Yoshihiro Shimoda wrote:
>>> Hi,
>>>
>>>> From: Mathias Nyman
>>>> Sent: Monday, October 24, 2016 6:58 PM
>>>>
>>>> On 22.10.2016 01:25, Babu Moger wrote:
>>>>> Never seen XHCI auto handoff working on TI and RENESAS cards.
>>>>> Eventually, we force handoff. This code forces the handoff
>>>>> unconditionally. It saves 5 seconds boot time for each card.
>>>>>
>>>>> Signed-off-by: Babu Moger <[email protected]>
>>>> Do the Renesas and TI controllers still advertise the extended
>>>> capability
>>>> for the handoff? (XHCI_EXT_CAPS_LEGACY)
>> I don't see this capability. Here is lspci output.
>>
>
> It's not a PCI capability, it's a xhci Extended Capability.
>
> If the capability is supported, and handoff fail, then
> quirk_usb_handoff_xhci() will print
> "xHCI BIOS handoff failed (BIOS bug ?)"
>
Yes. I see these messages. After this patch I don't see these messages.
> Further on in the same function we wait for the "controller not ready"
> bit in the status
> register to clear. If that times it prints out:
>
> "xHCI HW not ready after 5 sec (HC bug?) status"
No. I didn't see this messages.
>
> Do you see any of these two messages in the log?
>
> -Mathias