+1

Hello Dov, Could we please use a better name here? 'QemuKernel' is confusing as it is used for initrd and cmdline only, in this case, right? 'QemuKernelLoader' is better in this regard, or other

FWIW, I strongly agree with Marvin on this: having ParseStatus in its current form is a bad idea since it adds no value but does incur a cost. Thanks, Michael

I don't need to be okay with anything, I'm not a maintainer nor an authority. But I gave my opinion, which is that it is dead code that makes the design/flow harder to understand for a third party, at

Liming, Bugzilla is created at https://bugzilla.tianocore.org/show_bug.cgi?id=3447 Thanks Zhiguang

I assume you are ok with the ParseStatus. I will send new version based on mail discussion. Thanks!

Dov: Can you submit one BZ for this new feature? I will add it into edk2 202108 stable tag planning. Thanks Liming

Zhiguang: Can you submit one BZ for this new feature? I will add it into edk2 202108 stable tag planning. Thanks Liming

REF: https://bugzilla.tianocore.org/show_bug.cgi?id=3246 MdeLibs.dsc.inc was added for some basic/default library instances provided by MdePkg, RegisterFilterLibNull which will be consumed by IoLib

If ParseElfImage() is not called, nothing will initialize ParseStatus and the load function will read random data. If AllocateZeroPool was used for the context, a common pattern throughout the

Memory buffer that is allocated by malloc() and realloc() will be shifted by 8 bytes because Oniguruma keeps its memory signature. This 8 bytes shift is not handled while calling free() to release

Hi Guo, Thanks for the comments. I described the reason in the V2 commint message: Explicitly define some PCDs as DynamicEx, or their default type will be Dynamic Thanks Zhiguang

Caller might call LoadElfImage() without firstly calling ParseElfImage() by mistake. ParseStatus is added to catch such mistake. I don't trust the caller would follow the contracts properly😊.

This patch 1) changed PCD type from Dynamic to DynamicEX 2) added 3 PCDs. It would be great if you could describe why 3 PCDs are added in the commit message. With that: Reviewed-by: Guo Dong

Reviewed-by: Guo Dong <guo.dong@...>

Reviewed-by: Guo Dong <guo.dong@...>

Hi Patrick Thanks for catching this issue. I updated the code, and you can find the code below https://github.com/LiuZhiguang001/edk2/tree/UniversalPayloadHeaders_v4 Please help

V1: Currently, BDS driver will link a PlatformBootManagerLib, which contains pl= atform sepcific logic. This patch get the platform specific logic from a protocol,= so that platform logic for Boot

From: Sean Brogan <spbrogan@...> REF: https://bugzilla.tianocore.org/show_bug.cgi?id=3443 Existing implementation could modify class global data that causes potential incorrect file mask to be