public inbox for devel@edk2.groups.io
 help / color / mirror / Atom feed
From: Jon Hunter <jonathanh@nvidia.com>
To: "Luo, Heng" <heng.luo@intel.com>,
	"Wu, Hao A" <hao.a.wu@intel.com>,
	"devel@edk2.groups.io" <devel@edk2.groups.io>,
	"jbrasen@nvidia.com" <jbrasen@nvidia.com>
Cc: "Ni, Ray" <ray.ni@intel.com>
Subject: Re: [edk2-devel] [PATCH] MdeModulePkg/XhciDxe: Retry device slot init on failure
Date: Thu, 22 Oct 2020 14:57:04 +0100	[thread overview]
Message-ID: <206d9891-152b-3954-1b61-bf144cab4ed5@nvidia.com> (raw)
In-Reply-To: <040616b6-c7e9-d28d-69ac-bccb406d4730@nvidia.com>

Hi Heng,

On 22/10/2020 09:44, Jon Hunter wrote:
> Hi Heng,
> 
> On 22/10/2020 04:01, Luo, Heng wrote:
>> Hi Hao,
>> I applying this patch, it cannot fix BZ https://bugzilla.tianocore.org/show_bug.cgi?id=3007.
> 
> Can you elaborate more on why that is? Unfortunately, your patch does
> not work for the scenario we are testing.
> 
> In your case does the bad USB device never enumerate even on the 2nd
> attempt?
> 
> If so, then maybe we can apply your patch to do the disable slot and in
> our patch we just try the device init again once, without calling the
> slot disable because that is handled by your patch.

Applying the below change on top of your change, works for me.
One change in the below is not to return the status from the 
call to XhcDisableSlotCmd64() which is necessary to retry the
device initialisation.

Jon

diff --git a/MdeModulePkg/Bus/Pci/XhciDxe/XhciSched.c b/MdeModulePkg/Bus/Pci/XhciDxe/XhciSched.c
index 21bc9ce17556..82efd4a03f8e 100644
--- a/MdeModulePkg/Bus/Pci/XhciDxe/XhciSched.c
+++ b/MdeModulePkg/Bus/Pci/XhciDxe/XhciSched.c
@@ -1684,9 +1684,11 @@ XhcPollPortStatusChange (
   EFI_STATUS        Status;
   UINT8             Speed;
   UINT8             SlotId;
+  UINT8             Retries;
   USB_DEV_ROUTE     RouteChart;
   Status = EFI_SUCCESS;
+  Retries = XHC_INIT_DEVICE_SLOT_RETRIES;
   if ((PortState->PortChangeStatus & (USB_PORT_STAT_C_CONNECTION | USB_PORT_STAT_C_ENABLE | USB_PORT_STAT_C_OVERCURRENT | USB_PORT_STAT_C_RESET)) == 0) {
     return EFI_SUCCESS;
@@ -1728,17 +1730,29 @@ XhcPollPortStatusChange (
     } else if ((PortState->PortStatus & USB_PORT_STAT_SUPER_SPEED) != 0) {
       Speed = EFI_USB_SPEED_SUPER;
     }
-    //
-    // Execute Enable_Slot cmd for attached device, initialize device context and assign device address.
-    //
-    SlotId = XhcRouteStringToSlotId (Xhc, RouteChart);
-    if ((SlotId == 0) && ((PortState->PortChangeStatus & USB_PORT_STAT_C_RESET) != 0)) {
-      if (Xhc->HcCParams.Data.Csz == 0) {
-        Status = XhcInitializeDeviceSlot (Xhc, ParentRouteChart, Port, RouteChart, Speed);
-      } else {
-        Status = XhcInitializeDeviceSlot64 (Xhc, ParentRouteChart, Port, RouteChart, Speed);
+
+    do {
+      //
+      // Execute Enable_Slot cmd for attached device, initialize device context and assign device address.
+      //
+      SlotId = XhcRouteStringToSlotId (Xhc, RouteChart);
+      if ((SlotId == 0) && ((PortState->PortChangeStatus & USB_PORT_STAT_C_RESET) != 0)) {
+        if (Xhc->HcCParams.Data.Csz == 0) {
+          Status = XhcInitializeDeviceSlot (Xhc, ParentRouteChart, Port, RouteChart, Speed);
+        } else {
+          Status = XhcInitializeDeviceSlot64 (Xhc, ParentRouteChart, Port, RouteChart, Speed);
+        }
       }
-    }
+
+      //
+      // According to the xHCI specification (section 4.6.5), "a USB Transaction
+      // Error Completion Code for an Address Device Command may be due to a Stall
+      // response from a device. Software should issue a Disable Slot Command for
+      // the Device Slot then an Enable Slot Command to recover from this error."
+      // Therefore, retry the device slot initialization if it fails due to a
+      // device error.
+      //
+    } while ((Status == EFI_DEVICE_ERROR) && (Retries--));
   }
   return Status;
@@ -2248,7 +2262,7 @@ XhcInitializeDeviceSlot (
     Xhc->UsbDevContext[SlotId].XhciDevAddr = DeviceAddress;
   } else {
     DEBUG ((DEBUG_INFO, "    Address %d assigned unsuccessfully\n"));
-    Status = XhcDisableSlotCmd (Xhc, SlotId);
+    XhcDisableSlotCmd (Xhc, SlotId);
   }
   return Status;
@@ -2461,7 +2475,7 @@ XhcInitializeDeviceSlot64 (
     Xhc->UsbDevContext[SlotId].XhciDevAddr = DeviceAddress;
   } else {
     DEBUG ((DEBUG_INFO, "    Address %d assigned unsuccessfully\n"));
-    Status = XhcDisableSlotCmd64 (Xhc, SlotId);
+    XhcDisableSlotCmd64 (Xhc, SlotId);
   }
   return Status;
diff --git a/MdeModulePkg/Bus/Pci/XhciDxe/XhciSched.h b/MdeModulePkg/Bus/Pci/XhciDxe/XhciSched.h
index 2f1899502151..3f9cdb1c3609 100644
--- a/MdeModulePkg/Bus/Pci/XhciDxe/XhciSched.h
+++ b/MdeModulePkg/Bus/Pci/XhciDxe/XhciSched.h
@@ -11,6 +11,7 @@ SPDX-License-Identifier: BSD-2-Clause-Patent
 #define _EFI_XHCI_SCHED_H_
 #define XHC_URB_SIG      SIGNATURE_32 ('U', 'S', 'B', 'R')
+#define XHC_INIT_DEVICE_SLOT_RETRIES 1
 













-- 
nvpublic

  reply	other threads:[~2020-10-22 13:57 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-10-21 15:44 [PATCH] MdeModulePkg/XhciDxe: Retry device slot init on failure Jeff Brasen
2020-10-22  0:54 ` [edk2-devel] " Wu, Hao A
2020-10-22  3:01   ` Heng Luo
2020-10-22  8:44     ` Jon Hunter
2020-10-22 13:57       ` Jon Hunter [this message]
2020-10-23  0:53         ` Heng Luo
2020-10-23  1:09           ` Wu, Hao A
2020-10-23  1:27             ` Heng Luo
2020-10-22  6:40 ` Wu, Hao A
2020-10-22  8:39   ` Jon Hunter

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-list from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=206d9891-152b-3954-1b61-bf144cab4ed5@nvidia.com \
    --to=devel@edk2.groups.io \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox