When SaSbQemu platform is configured with multi memory nodes, like 
numa architecture, os will ignore any memory node in the device tree 
except the first one. The kernel reads UEFI memory map for memory 
information when booting via UEFI. 

However UEFI only allocates the first memory node memory space for 
SbsaQemu platform, in this scenario we can refer to the implement 
of "OvmfPkg/Fdt/HighMemDxe" and use the GCD services to add memory 
spaces for high memory node.

Instead of using FdtClientDxe driver to get the informatin of memory, 
we get the hardware information of memory via SMC calls, which is proposed
on get rid of DeviceTree from SbsaQemu:
https://openfw.io/edk2-devel/20240131132400.3022662-1-xiongyining1...@phytium.com.cn/

Changes in v4:
- Rename SbsaQemuHardwareInfoLib to HardwareInfoLib and use newer function 
names.
- Link to v3: 
https://openfw.io/edk2-devel/20240220073332.3249289-1-xiongyining1...@phytium.com.cn/

Changes in v3:
- when the memory node is the first one, there is no need to add it.

Changes in v2:
- get the information of memory via SMC rather than FdtClientDxe.
- add a new driver rather than use HighMemDxe. 

Xiong Yining (1):
  SbsaQemu: add memory space for the high memory nodes

 Platform/Qemu/SbsaQemu/SbsaQemu.dsc           |   1 +
 Platform/Qemu/SbsaQemu/SbsaQemu.fdf           |   1 +
 .../SbsaQemuHighMemDxe/SbsaQemuHighMemDxe.inf |  45 ++++++
 .../SbsaQemuHighMemDxe/SbsaQemuHighMemDxe.c   | 134 ++++++++++++++++++
 4 files changed, 181 insertions(+)
 create mode 100644 
Silicon/Qemu/SbsaQemu/Drivers/SbsaQemuHighMemDxe/SbsaQemuHighMemDxe.inf
 create mode 100644 
Silicon/Qemu/SbsaQemu/Drivers/SbsaQemuHighMemDxe/SbsaQemuHighMemDxe.c

-- 
2.34.1



-=-=-=-=-=-=-=-=-=-=-=-
Groups.io Links: You receive all messages sent to this group.
View/Reply Online (#117169): https://edk2.groups.io/g/devel/message/117169
Mute This Topic: https://groups.io/mt/105177587/21656
Group Owner: devel+ow...@edk2.groups.io
Unsubscribe: https://edk2.groups.io/g/devel/unsub [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-


Reply via email to