1. Patchew
  2. EDK2
  3. View series

[edk2-devel] [PATCH v2] MdeModulePkg: Enabling OS boot from SD card through UEFI payload

Aiman Rosli posted 1 patch 2 years, 3 months ago
Failed in applying to current master (apply log)
There is a newer version of this series
MdeModulePkg/Bus/Pci/SdMmcPciHcDxe/SdDevice.c | 23 ++++++++++++++++---
1 file changed, 20 insertions(+), 3 deletions(-)
[edk2-devel] [PATCH v2] MdeModulePkg: Enabling OS boot from SD card through UEFI payload
Posted by Aiman Rosli 2 years, 3 months ago 
This changes is by adding 50ms delay during voltage switching from 3.3V to 1.8V,
plus adding a goto Voltage33Retry for 3.3V checking and retrying.

Signed-off-by: Aiman Rosli <muhammad.aiman.rosli@intel.com>
---
 MdeModulePkg/Bus/Pci/SdMmcPciHcDxe/SdDevice.c | 23 ++++++++++++++++---
 1 file changed, 20 insertions(+), 3 deletions(-)

diff --git a/MdeModulePkg/Bus/Pci/SdMmcPciHcDxe/SdDevice.c b/MdeModulePkg/Bus/Pci/SdMmcPciHcDxe/SdDevice.c
index 662f9f483c..527fe7dc20 100644
--- a/MdeModulePkg/Bus/Pci/SdMmcPciHcDxe/SdDevice.c
+++ b/MdeModulePkg/Bus/Pci/SdMmcPciHcDxe/SdDevice.c
@@ -1213,9 +1213,14 @@ SdCardIdentification (
   UINT32                         PresentState;
   UINT8                          HostCtrl2;
   UINTN                          Retry;
+  BOOLEAN                        ForceVoltage33;
+
+  ForceVoltage33 = FALSE;
 
   PciIo    = Private->PciIo;
   PassThru = &Private->PassThru;
+
+Voltage33Retry:
   //
   // 1. Send Cmd0 to the device
   //
@@ -1294,6 +1299,14 @@ SdCardIdentification (
     return EFI_UNSUPPORTED;
   }
 
+  //
+  // 1.8V had failed in the previous run, forcing a retry with 3.3V instead
+  //
+  if (ForceVoltage33 == TRUE) {
+    S18r           = FALSE;
+    ForceVoltage33 = FALSE;
+  }
+
   //
   // 5. Repeatly send Acmd41 with supply voltage window to the device.
   //    Note here we only support the cards complied with SD physical
@@ -1362,13 +1375,17 @@ SdCardIdentification (
         goto Error;
       }
 
-      gBS->Stall (1000);
+      // Workaround to add a delay of 50 ms in order for clock to stabilize before turning on the SD card again.
+      gBS->Stall (50000);
 
       SdMmcHcRwMmio (PciIo, Slot, SD_MMC_HC_PRESENT_STATE, TRUE, sizeof (PresentState), &PresentState);
       if (((PresentState >> 20) & 0xF) != 0xF) {
         DEBUG ((DEBUG_ERROR, "SdCardIdentification: SwitchVoltage fails with PresentState = 0x%x, It should be 0xF\n", PresentState));
-        Status = EFI_DEVICE_ERROR;
-        goto Error;
+        Status         = SdMmcHcReset (Private, Slot);
+        Status         = SdMmcHcInitHost (Private, Slot);
+        ForceVoltage33 = TRUE;
+        DEBUG ((DEBUG_ERROR, "SdCardIdentification: Switching to 1.8V had failed in the previous run, forcing a retry with 3.3V instead\n"));
+        goto Voltage33Retry;
       }
     }
 
-- 
2.34.1.windows.1



-=-=-=-=-=-=-=-=-=-=-=-
Groups.io Links: You receive all messages sent to this group.
View/Reply Online (#85756): https://edk2.groups.io/g/devel/message/85756
Mute This Topic: https://groups.io/mt/88482891/1787277
Group Owner: devel+owner@edk2.groups.io
Unsubscribe: https://edk2.groups.io/g/devel/unsub [importer@patchew.org]
-=-=-=-=-=-=-=-=-=-=-=-
Re: [edk2-devel] [PATCH v2] MdeModulePkg: Enabling OS boot from SD card through UEFI payload
Posted by Wu, Hao A 2 years, 3 months ago 
Please refer to the inline comments below:


> -----Original Message-----
> From: devel@edk2.groups.io <devel@edk2.groups.io> On Behalf Of Aiman
> Rosli
> Sent: Monday, January 17, 2022 8:56 PM
> To: devel@edk2.groups.io
> Cc: Rosli, Muhammad Aiman <muhammad.aiman.rosli@intel.com>
> Subject: [edk2-devel] [PATCH v2] MdeModulePkg: Enabling OS boot from SD
> card through UEFI payload
> 
> This changes is by adding 50ms delay during voltage switching from 3.3V to
> 1.8V, plus adding a goto Voltage33Retry for 3.3V checking and retrying.
> 
> Signed-off-by: Aiman Rosli <muhammad.aiman.rosli@intel.com>
> ---
>  MdeModulePkg/Bus/Pci/SdMmcPciHcDxe/SdDevice.c | 23
> ++++++++++++++++---
>  1 file changed, 20 insertions(+), 3 deletions(-)
> 
> diff --git a/MdeModulePkg/Bus/Pci/SdMmcPciHcDxe/SdDevice.c
> b/MdeModulePkg/Bus/Pci/SdMmcPciHcDxe/SdDevice.c
> index 662f9f483c..527fe7dc20 100644
> --- a/MdeModulePkg/Bus/Pci/SdMmcPciHcDxe/SdDevice.c
> +++ b/MdeModulePkg/Bus/Pci/SdMmcPciHcDxe/SdDevice.c
> @@ -1213,9 +1213,14 @@ SdCardIdentification (
>    UINT32                         PresentState;
>    UINT8                          HostCtrl2;
>    UINTN                          Retry;
> +  BOOLEAN                        ForceVoltage33;
> +
> +  ForceVoltage33 = FALSE;
> 
>    PciIo    = Private->PciIo;
>    PassThru = &Private->PassThru;
> +
> +Voltage33Retry:
>    //
>    // 1. Send Cmd0 to the device
>    //
> @@ -1294,6 +1299,14 @@ SdCardIdentification (
>      return EFI_UNSUPPORTED;
>    }
> 
> +  //
> +  // 1.8V had failed in the previous run, forcing a retry with 3.3V
> + instead  //  if (ForceVoltage33 == TRUE) {
> +    S18r           = FALSE;
> +    ForceVoltage33 = FALSE;


Like I mentioned in https://github.com/tianocore/edk2/pull/2413#discussion_r785611042
I do not think "ForceVoltage33 = FALSE" is necessary, could you help to double check?


> +  }
> +
>    //
>    // 5. Repeatly send Acmd41 with supply voltage window to the device.
>    //    Note here we only support the cards complied with SD physical
> @@ -1362,13 +1375,17 @@ SdCardIdentification (
>          goto Error;
>        }
> 
> -      gBS->Stall (1000);
> +      // Workaround to add a delay of 50 ms in order for clock to stabilize
> before turning on the SD card again.
> +      gBS->Stall (50000);


Sorry for a question.
As I mentioned in:
https://github.com/tianocore/edk2/pull/2413#discussion_r785610179
https://github.com/tianocore/edk2/pull/2413#discussion_r785610742

Can this 50 milliseconds delay be put in the below 1.8V switch failure handling logic?
Enlarging the delay (from 1ms to 50ms) in the normal initialization flow will bring performance impact.


> 
>        SdMmcHcRwMmio (PciIo, Slot, SD_MMC_HC_PRESENT_STATE, TRUE,
> sizeof (PresentState), &PresentState);
>        if (((PresentState >> 20) & 0xF) != 0xF) {
>          DEBUG ((DEBUG_ERROR, "SdCardIdentification: SwitchVoltage fails with
> PresentState = 0x%x, It should be 0xF\n", PresentState));
> -        Status = EFI_DEVICE_ERROR;
> -        goto Error;
> +        Status         = SdMmcHcReset (Private, Slot);
> +        Status         = SdMmcHcInitHost (Private, Slot);


Could you help to check my comment in:
https://github.com/tianocore/edk2/pull/2413#discussion_r785610742
to 1) add comments and 2) refine the error handling for the "reset and reinitialize the slot before the 3.3V retry"

Best Regards,
Hao Wu


> +        ForceVoltage33 = TRUE;
> +        DEBUG ((DEBUG_ERROR, "SdCardIdentification: Switching to 1.8V had
> failed in the previous run, forcing a retry with 3.3V instead\n"));
> +        goto Voltage33Retry;
>        }
>      }
> 
> --
> 2.34.1.windows.1
> 
> 
> 
> 
> 



-=-=-=-=-=-=-=-=-=-=-=-
Groups.io Links: You receive all messages sent to this group.
View/Reply Online (#85764): https://edk2.groups.io/g/devel/message/85764
Mute This Topic: https://groups.io/mt/88482891/1787277
Group Owner: devel+owner@edk2.groups.io
Unsubscribe: https://edk2.groups.io/g/devel/unsub [importer@patchew.org]
-=-=-=-=-=-=-=-=-=-=-=-

Patchew 2.1-devel-0870f84

© 2016 - 2024 Red Hat, Inc.