RE: How to pass the EFI_MEMORY_SP type to VM in QEMU

[Luo, Zhigang]

[AMD Official Use Only - AMD Internal Distribution Only]

Hi Gerd,

Thanks for your response. So, on X86, I need to add a new type in qemu and 
firmware. Is my understanding correct?

Regards,
Zhigang

-----Original Message-----
From: Gerd Hoffmann <kraxel@redhat.com>
Sent: Monday, November 4, 2024 4:53 AM
To: Luo, Zhigang <Zhigang.Luo@amd.com>
Cc: qemu-devel@nongnu.org; Philippe Mathieu-Daudé <philmd@linaro.org>
Subject: Re: How to pass the EFI_MEMORY_SP type to VM in QEMU

On Fri, Nov 01, 2024 at 02:45:44PM +0000, Luo, Zhigang wrote:
> Dear QEMU Community,
>
> I hope this message finds everyone well. My name is Zhigang, and I am 
> currently engaged in a project that requires virtualization capabilities 
> provided by QEMU. I am reaching out to the community for guidance on a 
> specific configuration that I need to implement.
>
> I am looking to pass the EFI_MEMORY_SP (Special Purpose Memory) type memory 
> from host to a virtual machine within QEMU. This memory needs to be 
> EFI_MEMORY_SP type in the virtual machine as well. This particular memory 
> type is essential for the functionality of my project, and I have been facing 
> some difficulties in setting it up correctly.
>
> I have reviewed the available documentation and resources, but I have not yet 
> found a clear method to achieve this. I would greatly appreciate it if anyone 
> could provide insights, share their experiences, or direct me to relevant 
> documentation or examples that pertain to configuring EFI_MEMORY_SP memory 
> type for a VM in QEMU.

The memory map is passed from qemu to firmware using the etc/e820 fw_cfg file 
(on x86).  That would be the natural way to this, for example by taking a 
approach simliar to the linux kernel and define a type for this 
(E820_TYPE_SOFT_RESERVED).

On arm / riscv etc. the memory map is passed via device tree from qemu to 
firmware, so on these platforms the device tree should be used for 
EFI_MEMORY_SP too.

take care,
  Gerd