Reviewed-by: Daniel Schaefer
From: devel@edk2.groups.io on behalf of Chang, Abner via
groups.io
Sent: Tuesday, August 23, 2022 13:57
To: devel@edk2.groups.io
Cc: Leif Lindholm ; Michael D Kinney
; Daniel Schaefer ; Sunil V
L
Subject: [edk2-devel] [edk2
Reviewed-by: Daniel Schaefer
From: devel@edk2.groups.io on behalf of NickleLa
Sent: Tuesday, August 23, 2022 15:56
To: Sunil V L
Cc: abner.ch...@amd.com ; devel@edk2.groups.io
; Andrew Fish ; Leif Lindholm
; Michael D Kinney ;
Daniel Schaefer
Subject: Re
Thanks!
ghcr.io/tianocore/containers/fedora-35-test:cea91ec is the container it uses,
right?
I used this to build the RISC-V U540 platform in edk2-platforms and can confirm
it can still build and boot to shell fine.
Tested-by: Daniel Schaefer
On 8/11/22 18:49, Oliver Steffen wrote:
>
CpuExceptionHandlerLib to follow
the API changes, based on the ARM changes in e2ae0bed29ca0900bd35.
The functionality to RISC-V platforms should be none.
Signed-off-by: Daniel Schaefer
Cc: Sunil V L
---
Silicon/RISC-V/ProcessorPkg/Library/RiscVExceptionLib/CpuExceptionHandlerLib.c
| 82
Hi Oliver,
I assume you made sure that the build still works on at least one platform per
arch?
For example did you build a RISCV64 platform?
Thanks,
Daniel
On 8/10/22 20:03, Oliver Steffen wrote:
> Remove BaseTools/Bin/gcc*_linux_ext_dep.yaml to stop
> downloading gcc from external locations;
Hi Kun,
sorry for the late reply.
The change is okay - thanks for making sure to change this platform, too!
Reviewed-by: Daniel Schaefer
Thanks,
Daniel
On 8/11/22 06:52, Kun Qin wrote:
> Hi U5SeriesPkg maintainers,
>
> Can you please review the patch below and let me know if any
Yes, that's my GitHub ID.
Can add reviewed-by or acked-by. Whatever makes sense here.
Thanks,
Daniel
On 7/21/22 11:14, Michael D Kinney wrote:
> Cc: Andrew Fish
> Cc: Leif Lindholm
> Cc: Daniel Schaefer
> Cc: Sean Rhodes
> Cc: Debkumar De
> Cc: Erdem Aktas
> Cc:
Hi Steffen,
No, I don't think we've tried GCC12 yet.
Hm, I don't quite understand the bug report but it sounds like something
that'll be fixed by GCC?
Thanks,
Daniel
From: devel@edk2.groups.io on behalf of Oliver Steffen
Sent: Thursday, June 2, 2022 21:50
To:
Great! U500 confirmed working again and cleanup through generic code.
Reviewed-by: Daniel Schaefer
From: Chang, Abner (HPS SW/FW Technologist)
Sent: Monday, November 15, 2021 10:56
To: devel@edk2.groups.io
Cc: Chang, Abner (HPS SW/FW Technologist) ; Schaefer
Reviewed-by: Daniel Schaefer
From: Chang, Abner (HPS SW/FW Technologist)
Sent: Tuesday, October 19, 2021 16:09
To: devel@edk2.groups.io
Cc: Chang, Abner (HPS SW/FW Technologist) ; Sunil V L
; Schaefer, Daniel
Subject: [edk2-platforms][PATCH 00/30] EDK2 RISC-V
Reviewed-by: Daniel Schaefer
On 11/3/21 11:05, Michael D Kinney wrote:
> Fix build breaks for all architectures by adding RngLib mapping.
>
> Cc: Jian J Wang
> Cc: Ard Biesheuvel
> Cc: Abner Chang
> Cc: Daniel Schaefer
> Signed-off-by: Michael D Kinney
>
Reviewed-by: Daniel Schaefer
On 11/3/21 13:28, Yao, Jiewen wrote:
> Acked-by: Jiewen Yao
>
> Need ARM/RISC-V owner to confirm.
>
>> -Original Message-
>> From: Kinney, Michael D
>> Sent: Wednesday, November 3, 2021 11:06 AM
>> To: devel@edk2.group
Perfect, thanks.
I think BaseRngLib has an implementation for ARM or perhaps only AARCH64.
I'm working one for RISCV64: https://bugzilla.tianocore.org/show_bug.cgi?id=3676
But until then this is good.
Reviewed-by: Daniel Schaefer
On 11/3/21 11:05, Michael D Kinney wrote:
> Fix NetworkP
From: Abner Chang
Cc: Daniel Schaefer
Cc: Abner Chang
Cc: Sunil V L
Reviewed-by: Abner Chang
Signed-off-by: Daniel Schaefer
---
Platform/SiFive/U5SeriesPkg/FreedomU540HiFiveUnleashedBoard/U540.dsc | 5 ++
Platform/SiFive/U5SeriesPkg/FreedomU540HiFiveUnleashedBoard/U540.fdf | 1
Cc: Daniel Schaefer
Cc: Abner Chang
Cc: Sunil V L
Reviewed-by: Abner Chang
Signed-off-by: Daniel Schaefer
---
Platform/SiFive/U5SeriesPkg/FreedomU500VC707Board/U500.dsc| 12 +-
Platform/SiFive/U5SeriesPkg/FreedomU540HiFiveUnleashedBoard/U540.dsc | 12 +-
Platform/RISC-V
The Linux EFISTUB reads the FDT from the EFI system configuration
table. Before installing the FDT needs to be patched with the booting
hartid, because the kernel in S-Mode cannot determine it.
Cc: Daniel Schaefer
Cc: Abner Chang
Cc: Sunil V L
Reviewed-by: Abner Chang
Signed-off-by: Daniel
It's a generic platform file. Only the device tree decides what happens.
Cc: Daniel Schaefer
Cc: Abner Chang
Cc: Sunil V L
Reviewed-by: Abner Chang
Signed-off-by: Daniel Schaefer
---
Platform/SiFive/U5SeriesPkg/FreedomU540HiFiveUnleashedBoard/U54
Cc: Daniel Schaefer
Cc: Abner Chang
Cc: Sunil V L
Reviewed-by: Abner Chang
Signed-off-by: Daniel Schaefer
---
Platform/SiFive/U5SeriesPkg/FreedomU540HiFiveUnleashedBoard/U540.dsc
| 2 +
Platform/SiFive/U5SeriesPkg/FreedomU540HiFiveUnleashedBoard/U540.fdf
Cc: Daniel Schaefer
Cc: Abner Chang
Cc: Sunil V L
Reviewed-by: Abner Chang
Signed-off-by: Daniel Schaefer
---
Silicon/RISC-V/ProcessorPkg/Library/RiscVOpensbiLib/RiscVOpensbiLib.inf
| 13 +++-
Platform/SiFive/U5SeriesPkg/FreedomU540HiFiveUnleashedBoard
New platform files require more space in SEC.
Behavior is determined not by source code but by device tree.
Cc: Daniel Schaefer
Cc: Abner Chang
Cc: Sunil V L
Reviewed-by: Abner Chang
Signed-off-by: Daniel Schaefer
---
Platform/SiFive/U5SeriesPkg/FreedomU540HiFiveUnleashedBoard/Library
OpenSBI uses the device tree for platform specific initialization, so we
need to have it already in SEC.
Cc: Daniel Schaefer
Cc: Abner Chang
Cc: Sunil V L
Reviewed-by: Abner Chang
Signed-off-by: Daniel Schaefer
---
Platform/RISC-V/PlatformPkg/RiscVPlatformPkg.dec| 4
From: Abner Chang
Add RISC-V FirmwareContext library for different FirmwareContext
implementations. This instance uses SBI firmware extension to
get the pointer to FirmwareContext.
Cc: Sunil V L
Cc: Daniel Schaefer
Reviewed-by: Abner Chang
Reviewed-by: Daniel Schaefer
Signed-off-by: Abner
From: Abner Chang
Use RISC-V FirmwareContext library to set and get the
pointer of PeiService on SiFive U5 series platforms.
Cc: Sunil V L
Cc: Daniel Schaefer
Reviewed-by: Daniel Schaefer
Signed-off-by: Abner Chang
---
Platform/SiFive/U5SeriesPkg/FreedomU540HiFiveUnleashedBoard/U540.dsc
From: Abner Chang
Use RISC-V FirmwareContext library to set and get the
pointer of PeiService.
Cc: Sunil V L
Cc: Daniel Schaefer
Reviewed-by: Daniel Schaefer
Signed-off-by: Abner Chang
---
Silicon/RISC-V/ProcessorPkg/Library/PeiServicesTablePointerLibOpenSbi
Cc: Abner Chang
Cc: Sunil V L
Reviewed-by: Abner Chang
Signed-off-by: Daniel Schaefer
---
Platform/SiFive/U5SeriesPkg/FreedomU500VC707Board/U500.dsc
| 2 +-
Platform/SiFive/U5SeriesPkg/FreedomU540HiFiveUnleashedBoard/U540.dsc
| 2 +-
Platform/RISC-V
Signed-off-by: Daniel Schaefer
---
Platform/SiFive/U5SeriesPkg/FreedomU500VC707Board/U500.dsc
| 2 +-
Platform/SiFive/U5SeriesPkg/FreedomU540HiFiveUnleashedBoard/U540.dsc
Cc: Abner Chang
Cc: Sunil V L
Reviewed-by: Abner Chang
Signed-off-by: Daniel Schaefer
---
Platform/SiFive/U5SeriesPkg/Include/SiFiveU5MCCoreplex.h| 33
Platform/RISC-V/PlatformPkg/Universal/Pei/PlatformPei/Platform.c| 17
+-
Platform/SiFive
FirmwareContext library
RISC-V/PlatformPkg: Add FdtPeim to pass DTB from PEI to DXE via HOB
Daniel Schaefer (10):
U5SeriesPkg: Deduplicate PlatformPei
RISC-V: Split SMBIOS out of PlatformPei
RISC-V: Use U5 SMBIOS library only for those platforms
U540: Add and build device tree
RISC-V
Title says v2 but this is the first time I sent it. Sorry.
If there's a next one, it'll be v3, just to be consistent.
On 10/6/21 19:56, Daniel Schaefer wrote:
> OpenSBI can now use a generic platform implementation. The platform
> specific behavior is derived from the device
It's a generic platform file. Only the device tree decides what happens.
Cc: Daniel Schaefer
Cc: Abner Chang
Cc: Sunil V L
Signed-off-by: Daniel Schaefer
---
Platform/{SiFive/U5SeriesPkg/FreedomU540HiFiveUnleashedBoard =>
RISC-V/PlatformPkg}/Library/OpensbiPla
Cc: Daniel Schaefer
Cc: Abner Chang
Cc: Sunil V L
Signed-off-by: Daniel Schaefer
---
Platform/RISC-V/PlatformPkg/Library/OpensbiPlatformLibNull/Platform.c
| 27 ---
Platform/RISC-V/PlatformPkg/Universal/Sec/Riscv64/SecEntry.S
The Linux EFISTUB reads the FDT from the EFI system configuration
table. Before installing the FDT needs to be patched with the booting
hartid, because the kernel in S-Mode cannot determine it.
Cc: Daniel Schaefer
Cc: Abner Chang
Cc: Sunil V L
Signed-off-by: Daniel Schaefer
---
Platform
Cc: Daniel Schaefer
Cc: Abner Chang
Cc: Sunil V L
Signed-off-by: Daniel Schaefer
---
Platform/RISC-V/PlatformPkg/Library/ResetSystemLib/ResetSystemLib.c | 128
Platform/RISC-V/PlatformPkg/Library/ResetSystemLib/ResetSystemLib.inf | 33
+
Platform/SiFive
New platform files require more space in SEC.
Behavior is determined not by source code but by device tree.
Cc: Daniel Schaefer
Cc: Abner Chang
Cc: Sunil V L
Signed-off-by: Daniel Schaefer
---
Platform/SiFive/U5SeriesPkg/FreedomU500VC707Board/U500.fdf.inc
From: Abner Chang
Cc: Daniel Schaefer
Cc: Abner Chang
Cc: Sunil V L
Signed-off-by: Daniel Schaefer
---
Platform/RISC-V/PlatformPkg/Universal/FdtPeim/FdtPeim.c | 70
Platform/RISC-V/PlatformPkg/Universal/FdtPeim/FdtPeim.inf| 49
OpenSBI uses the device tree for platform specific initialization, so we
need to have it already in SEC.
Cc: Daniel Schaefer
Cc: Abner Chang
Cc: Sunil V L
Signed-off-by: Daniel Schaefer
---
Platform/RISC-V/PlatformPkg/RiscVPlatformPkg.dec| 4 ++
Platform/RISC-V
Device Tree is taken from Uboot.
Cc: Daniel Schaefer
Cc: Abner Chang
Cc: Sunil V L
Signed-off-by: Daniel Schaefer
---
Platform/SiFive/U5SeriesPkg/FreedomU540HiFiveUnleashedBoard/DeviceTree.fdf.inc
| 35 +++
Platform/SiFive/U5SeriesPkg/FreedomU540HiFiveUnleashedBoard
OpenSBI uses the device tree for platform specific initialization, so we
need to have it already in SEC.
Cc: Daniel Schaefer
Cc: Abner Chang
Cc: Sunil V L
Signed-off-by: Daniel Schaefer
---
Platform/RISC-V/PlatformPkg/RiscVPlatformPkg.dec| 4 ++
Platform/RISC-V
From: Abner Chang
Use RISC-V FirmwareContext library to set and get the
pointer of PeiService on SiFive U5 series platforms.
Cc: Sunil V L
Cc: Daniel Schaefer
Signed-off-by: Abner Chang
---
Platform/SiFive/U5SeriesPkg/FreedomU540HiFiveUnleashedBoard/U540.dsc | 2 ++
1 file changed, 2
patch.
Still way too much. Hopefully MinPlatform will help.
26 ./RiscvVirt.dec
13 ./RiscvVirt.uni
12 ./RiscvVirtPkgExtra.uni
78 ./VarStore.fdf.inc
66 ./RiscvVirt.fdf.inc
654 ./RiscvVirt.dsc
356 ./RiscvVirt.fdf
1205 total
Cc: Abner Chang
Cc: Sunil V L
Signed-off-by: Daniel
From: Abner Chang
Add RISC-V FirmwareContext library for different FirmwareContext
implementations. This instance uses SBI firmware extension to
get the pointer to FirmwareContext.
Cc: Sunil V L
Cc: Daniel Schaefer
Signed-off-by: Abner Chang
---
Silicon/RISC-V/ProcessorPkg/Include
From: Abner Chang
Use RISC-V FirmwareContext library to set and get the
pointer of PeiService.
Cc: Sunil V L
Cc: Daniel Schaefer
Signed-off-by: Abner Chang
---
Silicon/RISC-V/ProcessorPkg/Library/PeiServicesTablePointerLibOpenSbi/PeiServicesTablePointerLibOpenSbi.inf
| 4 ++--
Silicon
Cc: Abner Chang
Cc: Sunil V L
Signed-off-by: Daniel Schaefer
---
Platform/RISC-V/PlatformPkg/Library/PeiCoreInfoHobLibNull/CoreInfoHob.c
| 37 +
Platform/RISC-V/PlatformPkg/Library/PeiCoreInfoHobLibNull/PeiCoreInfoHobLib.inf
| 43
Platform/RISC
Cc: Abner Chang
Cc: Sunil V L
Signed-off-by: Daniel Schaefer
---
Platform/RISC-V/PlatformPkg/Universal/Pei/PlatformPei/Platform.c| 17
+-
Platform/SiFive/U5SeriesPkg/Include/SiFiveU5MCCoreplex.h| 33
Platform/SiFive/U5SeriesPkg/Library
FirmwareContext library
Silicon/RISC-V: PeiServiceTableLib uses RiscVFirmwareContextLib
SiFive/U5SeriesPkg: Use FirmwareContext library
RISC-V/PlatformPkg: Add FdtPeim to pass DTB from PEI to DXE via HOB
Daniel Schaefer (10):
U5SeriesPkg: Deduplicate PlatformPei
RISC-V: Split SMBIOS out of
Otherwise we can't load a 3rd party image because we're still in DXE.
MdeModulePkg/Universal/SecurityStubDxe/Defer3rdPartyImageLoad.c prevents
that.
Cc: Abner Chang
Cc: Sunil V L
Cc: Leif Lindholm
Reviewed-By: Sunil V L
Signed-off-by: Daniel Schaefer
---
Platform/RISC-V/P
While enumerating the PCIe devices, the driver tries to set some caching
attributes on the memory.
Cc: Abner Chang
Cc: Sunil V L
Cc: Leif Lindholm
Reviewed-By: Sunil V L
Signed-off-by: Daniel Schaefer
---
Silicon/RISC-V/ProcessorPkg/Universal/CpuDxe/CpuDxe.c | 4 ++--
1 file changed, 2
Cc: Abner Chang
Cc: Sunil V L
Cc: Leif Lindholm
Signed-off-by: Daniel Schaefer
---
Platform/SiFive/U5SeriesPkg/FreedomU500VC707Board/U500.dsc | 7
++-
Platform/SiFive/U5SeriesPkg/FreedomU500VC707Board/U500.fdf | 1 +
Platform/SiFive/U5SeriesPkg
SERT_EFI_ERROR (Status = Not Found)
> ASSERT [PciHostBridgeDxe]
> /edk2/MdeModulePkg/Bus/Pci/PciHostBridgeDxe/PciHostBridge.c(332): !EFI_ERROR
> (CheckStatus)
> QEMU: Terminated
Change works on QEMU 5.2 and 6.0.
Cc: Abner Chang
Cc: Sunil V L
Cc: Leif Lindholm
Review
Not DiskIoDxe because we don't have a disk, just for loading from
Ramdisks.
Cc: Abner Chang
Cc: Sunil V L
Cc: Leif Lindholm
Reviewed-By: Sunil V L
Signed-off-by: Daniel Schaefer
---
Platform/SiFive/U5SeriesPkg/FreedomU500VC707Board/U500.dsc | 7
++-
Platform/S
Changes from v1:
- Remove ShellLib override in 3/5
Tested on the unpublshed riscvvirt branch and it can still boot to
Linux.
Cc: Abner Chang
Cc: Sunil V L
Cc: Leif Lindholm
Daniel Schaefer (5):
U540, U500: Add filesystem drivers
RISC-V/CpuDxe: Ignore set memory attributes failure
U540
On 10/4/21 18:45, Sunil V L wrote:
> On Mon, Oct 04, 2021 at 01:23:57AM +0800, Daniel Schaefer wrote:
>> Cc: Abner Chang
>> Cc: Sunil V L
>> Cc: Leif Lindholm
>>
>> Signed-off-by: Daniel Schaefer
>> ---
>> Platform/SiFive/U5SeriesPkg
This series is for edk2-platforms of course. Excuse the incorrect subject
please.
On 10/4/21 01:23, Daniel Schaefer wrote:
> Some miscellaneous patches, that each compile and still boot to UEFI
> Shell, to allow for booting to Linux on HW and also prepare for Linux on
> RiscvVirtPkg (RI
SERT_EFI_ERROR (Status = Not Found)
> ASSERT [PciHostBridgeDxe]
> /edk2/MdeModulePkg/Bus/Pci/PciHostBridgeDxe/PciHostBridge.c(332): !EFI_ERROR
> (CheckStatus)
> QEMU: Terminated
Change works on QEMU 5.2 and 6.0.
Cc: Abner Chang
Cc: Sunil V L
Cc: Leif Lindholm
Signed-off-by: Daniel S
Otherwise we can't load a 3rd party image because we're still in DXE.
MdeModulePkg/Universal/SecurityStubDxe/Defer3rdPartyImageLoad.c prevents
that.
Cc: Abner Chang
Cc: Sunil V L
Cc: Leif Lindholm
Signed-off-by: Daniel Schaefer
---
Platform/RISC-V/PlatformPkg/Library/PlatformBoot
Cc: Leif Lindholm
Daniel Schaefer (5):
U540, U500: Add filesystem drivers
RISC-V/CpuDxe: Ignore set memory attributes failure
U540, U500: Add initrd command to boot Linux
U540: BuildCpuHob with 48 to indicate size memory space
Signal EndOfDxe in boot manager
Platform/RISC-V
While enumerating the PCIe devices, the driver tries to set some caching
attributes on the memory.
Cc: Abner Chang
Cc: Sunil V L
Cc: Leif Lindholm
Signed-off-by: Daniel Schaefer
---
Silicon/RISC-V/ProcessorPkg/Universal/CpuDxe/CpuDxe.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions
Not DiskIoDxe because we don't have a disk, just for loading from
Ramdisks.
Cc: Abner Chang
Cc: Sunil V L
Cc: Leif Lindholm
Signed-off-by: Daniel Schaefer
---
Platform/SiFive/U5SeriesPkg/FreedomU500VC707Board/U500.dsc | 7
++-
Platform/SiFive/U5SeriesPkg/FreedomU500VC707
Cc: Abner Chang
Cc: Sunil V L
Cc: Leif Lindholm
Signed-off-by: Daniel Schaefer
---
Platform/SiFive/U5SeriesPkg/FreedomU500VC707Board/U500.dsc | 7
+++
Platform/SiFive/U5SeriesPkg/FreedomU500VC707Board/U500.fdf | 1 +
Platform/SiFive/U5SeriesPkg
I'm happy. For the commits that I hadn't given it:
Reviewed-By: Daniel Schaefer
Thanks!
On 9/30/21 08:45, Abner Chang wrote:
> In V3: Address comments on V2.
> In V2: Remove HPE license on the files that just moved around or
>the changes in the file are just code re
Oh and this also needs to be followed up with a change to edk2-platforms.
On 9/29/21 07:16, Daniel Schaefer wrote:
> Please fix the issue in the maintainers file.
> Looks good otherwise:
>
> Reviewed-By: Daniel Schaefer
>
> On 9/28/21 16:31, Abner Chang wrote:
>> Reloc
Good so far.
Reviewed-By: Daniel Schaefer
But it needs an update to edk2-platforms, too. Lots of platform use this PCD.
On 9/28/21 16:31, Abner Chang wrote:
> PcdPciIoTranslation PCD is relocated to MdePkg and leveraged by
> both ARM and RISC-V arch. This patch removes the one from ArmPkg
> Signed-off-by: Abner Chang
> Cc: Michael D Kinney
> Cc: Liming Gao
> Cc: Zhiguang Liu
> Cc: Ard Biesheuvel
> Cc: Leif Lindholm
> Cc: Sami Mujawar
> Cc: Gerd Hoffmann
> Cc: Daniel Schaefer
> Cc: Sunil V L
> ---
> MdePkg/MdePkg.dec | 8
> 1
Reviewed-By: Daniel Schaefer
On 9/28/21 16:31, Abner Chang wrote:
> Relocate VirtioFdtDxe to OvmfPkg/Fdt, this driver is leverage by
> both ARM and RISC-V archs.
>
> Signed-off-by: Abner Chang
> Cc: Ard Biesheuvel
> Cc: Leif Lindholm
> Cc: Sami Mujawar
> Cc: Jiewen
Chang
> Cc: Ard Biesheuvel
> Cc: Leif Lindholm
> Cc: Sami Mujawar
> Cc: Jiewen Yao
> Cc: Jordan Justen
> Cc: Gerd Hoffmann
> Cc: Daniel Schaefer
> Cc: Sunil V L
> ---
> ArmVirtPkg/ArmVirtCloudHv.dsc | 2 +-
> ArmVirtPkg/ArmVirtKvmToo
Looks good. Just one minor typo in the comment.
Reviewed-By: Daniel Schaefer
On 9/28/21 16:31, Abner Chang wrote:
> Relocate FdtPciHostBridgeLib to OvmfPkg/Fdt, this library is
> leverage by both ARM and RISC-V archs. Also use
> PcdPciMmio32Translation and PcdPciMmio64Translati
Please fix the issue in the maintainers file.
Looks good otherwise:
Reviewed-By: Daniel Schaefer
On 9/28/21 16:31, Abner Chang wrote:
> Relocate PciPcdProducerLib to OvmfPkg/Fdt, this library is
> leverage by both ARM and RISC-V archs.
>
> Add OvmfPkg/Fdt maintainers in Mai
Translation
>
> Signed-off-by: Abner Chang
> Cc: Ard Biesheuvel
> Cc: Leif Lindholm
> Cc: Sami Mujawar
> Cc: Jiewen Yao
> Cc: Jordan Justen
> Cc: Gerd Hoffmann
> Cc: Daniel Schaefer
> Cc: Sunil V L
> Cc: Liming Gao
> Cc: Zhiguang Liu
> Cc: Michael D Ki
access. We relocate this PCD
> because this PCD is common to ARM and RSIC-V arch.
>
> Signed-off-by: Abner Chang
> Cc: Michael D Kinney
> Cc: Liming Gao
> Cc: Zhiguang Liu
> Cc: Ard Biesheuvel
> Cc: Leif Lindholm
> Cc: Sami Mujawar
> Cc: Gerd Hoffmann
> Cc: Danie
Looks good. I checked and you got all instances.
Reviewed-By: Daniel Schaefer
On 9/28/21 16:30, Abner Chang wrote:
> This is one of the series patches to restructure the location of modules under
> ArmVirtPkg for RiscVVirtPkg. RiscVVirtPkg leverage FDT Client protocol to
> parse
Hi,
we, as probably most users of EDK2, have a huge codebase with lots of
modules.
Autogen takes multiple minutes, which is unnecessary when doing a
rebuild. Luckily there is `--skip-autogen` but it only works if no INF,
DSC, ... file has been changed. The developer needs to know when they
PayloadPkg/Include/Linuxboot.h
create mode 100644 UefiPayloadPkg/Library/LbParseLib/LbParseLib.c
Cc: Cheng-Chieh Huang
Cc: Daniel Schaefer
Cc: Trammell Hudson
Cc: Maurice Ma
Cc: Guo Dong
Cc: Benjamin You
--
2.32.0.402.g57bb445576-goog
-=-=-=-=-=-=-=-=-=-=-=-
Groups.io Links: You r
On 7/16/21 11:56 PM, Ard Biesheuvel wrote:
On Fri, 16 Jul 2021 at 17:00, Kinney, Michael D
wrote:
Hi Ard,
I see you were involved in the OS side changes.
Can you explain what is required for the FW <-> OS interface with respect to
Load File Protocol and this media device path node.
What ha
Thanks!
Reviewed-By: Daniel Schaefer
On 7/13/21 11:24 PM, Takuto Naito wrote:
REF:https://bugzilla.tianocore.org/show_bug.cgi?id=3489
Fix build error of U540 and U500 caused by BdsDxe.
The edk2 patch
MdeModulePkg/BdsDxe: Update BdsEntry to use Variable Policy
requires
Second version because I forgot to update HPE copyright.
Signed-off-by: Daniel Schaefer
Cc: Jian J Wang
Cc: Hao A Wu
Cc: Dandan Bi
Cc: Liming Gao
Daniel Schaefer (1):
MdeModulePkg: Print which PCD was unable to be found
MdeModulePkg/Universal/PCD/Dxe/Service.c | 3 ++-
1 file changed, 2
When reaching the assert, it can't be debugged because the PCD is
unknown. Therefore the PCD's GUID and token number should be printed.
Signed-off-by: Daniel Schaefer
Cc: Jian J Wang
Cc: Hao A Wu
Cc: Dandan Bi
Cc: Liming Gao
---
MdeModulePkg/Universal/PCD/Dxe/Service.c | 3 +
When reaching the assert, it can't be debugged because the PCD is
unknown. Therefore the PCD's GUID and token number should be printed.
Signed-off-by: Daniel Schaefer
Cc: Jian J Wang
Cc: Hao A Wu
Cc: Dandan Bi
Cc: Liming Gao
---
MdeModulePkg/Universal/PCD/Dxe/Service.c | 1
On 7/13/21 11:44 AM, Wu, Hao A wrote:
-Original Message-
From: devel@edk2.groups.io On Behalf Of Daniel
Schaefer
Sent: Tuesday, July 13, 2021 9:10 AM
To: devel@edk2.groups.io
Cc: Trammell Hudson ; Daniel Schaefer
; Wang, Jian J ; Wu,
Hao A ; Ni, Ray ; Gao, Zhichao
Subject: [edk2-devel
Looks good. I compared it with existing R_RISCV_PCREL_LO12_I and looked
at the differences.
This one doesn't do use mRiscVPass1GotFixup.
I assume this is an optimization that's not possible here?
Haven't tested that it works but since it works for Pete:
Reviewed-by: Daniel Sc
protocol is registered.
Without this, calling RamDisk->Register() in the callback causes a crash:
ASSERT [RamDiskDxe] MdePkg/Library/BaseLib/LinkedList.c(75): List->ForwardLink
!= ((void *) 0)
Signed-off-by: Trammell Hudson
Cc: Daniel Schaefer
Cc: Jian J Wang
Cc: Hao A Wu
Cc: Ray
ersal/Disk/RamDiskDxe/RamDiskDriver.c | 11 ++-
1 file changed, 6 insertions(+), 5 deletions(-)
Cc: Daniel Schaefer
Cc: Trammell Hudson
Cc: Jian J Wang
Cc: Hao A Wu
Cc: Ray Ni
Cc: Zhichao Gao
--
2.31.1
-=-=-=-=-=-=-=-=-=-=-=-
Groups.io Links: You receive all messages sent to this
Reviewed-by: Daniel Schaefer
From: devel@edk2.groups.io on behalf of Abner Chang
Sent: Thursday, July 1, 2021 11:50
To: devel@edk2.groups.io ; g...@semihalf.com
Cc: l...@nuviainc.com ; ardb+tianoc...@kernel.org
; samer.el-haj-mahm...@arm.com
; sunny.w
v2:
> - Addressed Daniel's comment on formatting
>
> Testing:
> 1) Debian GCC 8.3.0 and booted sifive_u and QMEU virt models.
> 2) Debian 10.2.0 and booted QEMU virt model.
> 3) riscv-gnu-tool chain 9.2 and booted QEMU virt model.
>
> Signed-off-by: Sunil V L
edfishPkg/PrivateInclude/Library/RedfishCrtLib.h
called SIXTY_FOUR_BIT and THIRTY_TWO_BIT.
Not sure if they are for general consumption?
It seems that Mike suggests those to be defined in the platform's DSC?
See d0bf83e1cca
Thanks
Laszlo
Thanks
Liming
发件人: devel@edk2.groups.io 代表
can just use their distribution's riscv64
toolchain and compile edk2 with it. At least Debian and Fedora ship them.
Best regards
Heinrich
Signed-off-by: Sunil V L
Cc: Abner Chang
Cc: Daniel Schaefer
Cc: Heinrich Schuchardt
---
BaseTools/Source/C/GenFw/Elf64Convert.c | 45 ++
on my end.
https://github.com/vlsunil/riscv-edk2/commit/a742fab0d73ab568aa2d2578a5ccfc400ffd2fa5
Reviewed-by: Daniel Schaefer
Tested with GCC 8.3.0 on Debian Buster.
Can boot -machine sifive_u to UEFI Shell and -machine virt on QEMU to
Linux shell.
Sunil tested to compile on GCC 10.2.0 on Debi
ere a way to test compliance with
the spec?
From: devel@edk2.groups.io on behalf of Daniel Schaefer
Sent: Tuesday, May 25, 2021 02:59
To: devel@edk2.groups.io ; gaolim...@byosoft.com.cn
; kaaira7...@gmail.com ; Ni,
Ray ; mikub...@linux.microsoft.com
; isaa
and consistent quality. Similar
> approaches for the other elements of the spec should help to improve
> code sharing.
>
> Anyway, yes, it should be able to help you reduce the copies of “mostly
> common” code that you encountered and the code and spec are open to
> welcome the a
7; ;
'Leif Lindholm'
Subject: 回复: 回复: [edk2-devel] [PATCH v1 1/1] Add MemoryFence implementation for
RiscV64
Daniel:
Thanks for your information. Acked-by: Liming Gao
And, do you request to merge this patch for edk2 stable tag 202105?
Thanks
Liming
发件人: devel@edk2.groups
:
Now, it is clear to me. So, I suggest to merge this change when it is
verified on generic RISC-V QEMU virt machine. Is it OK?
Thanks
Liming
> -邮件原件-
> 发件人: devel@edk2.groups.io 代表 Daniel
> Schaefer
> 发送时间: 2021年5月18日 10:35
> 收件人: devel@edk2.groups.io; gaolim...@byosoft.c
you reduce the copies of “mostly
> common” code that you encountered and the code and spec are open to
> welcome the additional use and feedback from additional applications.
>
> Regards,
> Isaac
>
> *From:*devel@edk2.groups.io *On Behalf Of *Daniel
> Schaefer
> *Sent:* Wednesda
Hi,
that sounds like a great project!
I'm currently trying to create an equivalent of OvmfPkg for the RISCV64 generic
QEMU virt machine.
I don't like how much of my DSC and FDF file has modules that pretty much all
platforms should have.
MinPlatform would help reduce that, right?
Is MinPlatform
OpenSBI has started using those in v0.9. See:
https://github.com/riscv/opensbi/blob/v0.9/lib/utils/fdt/fdt_domain.c
Cc: Abner Chang
Cc: Leif Lindholm
Cc: Ard Biesheuvel
Cc: Daniel Schaefer
Cc: Anup Patel
Signed-off-by: Daniel Schaefer
---
EmbeddedPkg/Include/libfdt_env.h | 8
1
>> -----邮件原件-
>> 发件人: devel@edk2.groups.io 代表 Daniel
>> Schaefer
>> 发送时间: 2021年5月16日 2:13
>> 收件人: devel@edk2.groups.io
>> 抄送: Abner Chang ; Michael D Kinney
>> ; Liming Gao ;
>> Zhiguang Liu ; Leif Lindholm
>> 主题: [edk2-devel] [PATCH v1
Cc: Abner Chang
Cc: Michael D Kinney
Cc: Liming Gao
Cc: Zhiguang Liu
Cc: Leif Lindholm
Signed-off-by: Daniel Schaefer
---
MdePkg/Library/BaseLib/BaseLib.inf | 1 +
MdePkg/Library/BaseLib/RiscV64/MemoryFence.S | 33
2 files changed, 34 insertions(+)
diff
Some/all platforms are going to require EDK2 to build a device tree and
use it in the early stages of boot.
Cc: Bob Feng
Cc: Liming Gao
Cc: Yuwei Chen
Cc: Abner Chang
Signed-off-by: Daniel Schaefer
---
BaseTools/Conf/tools_def.template | 1 +
1 file changed, 1 insertion(+)
diff --git a
Ok sure, let's make only undef an error, not all other warnings. Then the
behaviour will also be the same as on MSVC.
From: devel@edk2.groups.io on behalf of Daniel Schaefer
Sent: Monday, March 8, 2021 11:44
To: Feng, Bob C ; devel@edk2.groups.io
Cc: L
, or even the EBC compiler
itself.
Can you please help me with this?
Thanks,
Daniel
On 3/4/21 10:12 AM, gaoliming wrote:
> Do you check whether EBC compiler supports this warning?
>
> And, do you evaluate CLANG compiler support for this warning?
>
> Thanks
> Liming
>>
Abner has reviewed and approved everything now.
Other people only reviewed v2 5/5.
Does anybody else have any comments?
From: devel@edk2.groups.io on behalf of Daniel Schaefer
Sent: Thursday, March 4, 2021 11:21
To: devel@edk2.groups.io
Cc: G Edhaya Chandran
--Original Message-
From: devel@edk2.groups.io On Behalf Of Daniel Schaefer
Sent: Tuesday, March 2, 2021 4:22 PM
To: devel@edk2.groups.io
Cc: Feng, Bob C ; Liming Gao ;
Chen, Christine ; Derek Lin
Subject: [edk2-devel] [PATCH v1 1/2] BaseTools: Make undefined VFR macro an
error (GCC)
VFR su
1 - 100 of 166 matches
Mail list logo
