[edk2-devel] [PATCH 1/1] MdeModulePkg\XhciDxe: Command Abort when times out
Wu, Hao A
hao.a.wu at intel.com
Wed Sep 20 08:19:39 UTC 2023
General level comments:
1. Some CI tests failed for the proposed patch: https://github.com/tianocore/edk2/pull/4844. Could you help to resolve them?
Please help to follow the step 11 in page: https://github.com/tianocore/tianocore.github.io/wiki/EDK-II-Development-Process and check the CI test results for the upcoming patch?
2. Could you help to update the title of this commit to "MdeModulePkg/XhciDxe: Abort the Address Device cmd when time out" to better reflect the purpose of the change?
3. Could you help to add a brief description as the commit log message?
Serveral inline comments below:
> -----Original Message-----
> From: Cai, Xianglei <xianglei.cai at intel.com>
> Sent: Monday, September 18, 2023 2:57 PM
> To: devel at edk2.groups.io
> Cc: Cai, Xianglei <xianglei.cai at intel.com>; Wu, Hao A <hao.a.wu at intel.com>;
> Ni, Ray <ray.ni at intel.com>; Wang, Jian J <jian.j.wang at intel.com>; Gao,
> Liming <gaoliming at byosoft.com.cn>; Shih, More <more.shih at intel.com>
> Subject: [PATCH 1/1] MdeModulePkg\XhciDxe: Command Abort when times
> out
>
> https://bugzilla.tianocore.org/show_bug.cgi?id=4552
>
> CC: Hao A Wu <hao.a.wu at intel.com>
> CC: Ray Ni <ray.ni at intel.com>
> CC: Jian J Wang <jian.j.wang at intel.com>
> CC: Liming Gao <gaoliming at byosoft.com.cn>
> CC: More Shih <more.shih at intel.com>
> Signed-off-by: Xianglei Cai <xianglei.cai at intel.com>
> ---
> MdeModulePkg/Bus/Pci/XhciDxe/XhciSched.c | 35
> ++++++++++++++++++++++++
> 1 file changed, 35 insertions(+)
>
> diff --git a/MdeModulePkg/Bus/Pci/XhciDxe/XhciSched.c
> b/MdeModulePkg/Bus/Pci/XhciDxe/XhciSched.c
> index 53421e64a850..5024d5e3acb9 100644
> --- a/MdeModulePkg/Bus/Pci/XhciDxe/XhciSched.c
> +++ b/MdeModulePkg/Bus/Pci/XhciDxe/XhciSched.c
> @@ -2121,6 +2121,29 @@ RingIntTransferDoorBell (
> return EFI_SUCCESS;
> }
>
> +/**
> + Set Command abort
> +
> + @param Xhc The XHCI Instance.
> + @param SlotId The slot id to be disabled.
> +
> + @retval EFI_SUCCESS Successfully disable the device slot.
This function does not return value. Please help to update the above function description.
> +
> +**/
> +VOID
> +XhcCmdRingCmdAbort (
> + IN USB_XHCI_INSTANCE *Xhc,
> + IN UINT8 SlotId
> + )
> +{
> + //
> + // Some 3rd party XHCI external cards don't support single 64-bytes
> +width register access,
> + // So divide it to two 32-bytes width register access.
> + //
The above comments does not apply here. Please help to remove it.
> + DEBUG ((DEBUG_INFO, "Command Ring Control set Command Abort,
> SlotId:
> +%d!\n", SlotId));
> + XhcSetOpRegBit (Xhc, XHC_CRCR_OFFSET, XHC_CRCR_CA); }
Sorry for a question, do we need to wait the CRR bit of the CRCR register to become 0 to ensure the command ring is stopped?
> +
> /**
> Assign and initialize the device slot for a new device.
>
> @@ -2331,6 +2354,12 @@ XhcInitializeDeviceSlot (
> Xhc->UsbDevContext[SlotId].XhciDevAddr = DeviceAddress;
> } else {
> DEBUG ((DEBUG_ERROR, " Slot %d address not assigned successfully.
> Status = %r\n", SlotId, Status));
> + //
> + // Software need abort the command if a command times out in xHCI spec
> 4.6.
> + //
Suggest to update "xHCI spec 4.6" to "xHCI spec Revision 1.2, section 4.6.1.2"
Best Regards,
Hao Wu
> + if (Status == EFI_TIMEOUT) {
> + XhcCmdRingCmdAbort (Xhc, SlotId);
> + }
> XhcDisableSlotCmd (Xhc, SlotId);
> }
>
> @@ -2547,6 +2576,12 @@ XhcInitializeDeviceSlot64 (
> Xhc->UsbDevContext[SlotId].XhciDevAddr = DeviceAddress;
> } else {
> DEBUG ((DEBUG_ERROR, " Slot %d address not assigned successfully.
> Status = %r\n", SlotId, Status));
> + //
> + // Software need abort the command if a command times out in xHCI spec
> 4.6.
> + //
> + if (Status == EFI_TIMEOUT) {
> + XhcCmdRingCmdAbort (Xhc, SlotId);
> + }
> XhcDisableSlotCmd64 (Xhc, SlotId);
> }
>
> --
> 2.42.0.windows.2
-=-=-=-=-=-=-=-=-=-=-=-
Groups.io Links: You receive all messages sent to this group.
View/Reply Online (#108900): https://edk2.groups.io/g/devel/message/108900
Mute This Topic: https://groups.io/mt/101435621/1813853
Group Owner: devel+owner at edk2.groups.io
Unsubscribe: https://edk2.groups.io/g/devel/unsub [edk2-devel-archive at redhat.com]
-=-=-=-=-=-=-=-=-=-=-=-
More information about the edk2-devel-archive
mailing list