By default, highmem option is enabled for machine types later than
virt-2.12. This allows qemu to place devices and RAM in physical
address space above 32-bits. This can cause issues as according to the
documentation Xen supports up to 12GiB of physical address space.
Recently the issue was observed using runqemu, that was causing Linux
running on top of Xen to stall when trying to access ECAM space placed
by qemu at 256GiB mark. Even though this issue is most probably related
to QEMU and not Xen (the investigation showed that it can map ECAM
space correctly), it is best to avoid such issues by disabling highmem
on qemuarm.

Signed-off-by: Michal Orzel <[email protected]>
---
 classes/qemuboot-xen-defaults.bbclass | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/classes/qemuboot-xen-defaults.bbclass 
b/classes/qemuboot-xen-defaults.bbclass
index f1550cc..460cbe6 100644
--- a/classes/qemuboot-xen-defaults.bbclass
+++ b/classes/qemuboot-xen-defaults.bbclass
@@ -23,8 +23,10 @@ QB_DEFAULT_KERNEL:qemuarm = "xen-${MACHINE}"
 
 # 64-bit Arm: gic version 3
 QB_MACHINE:qemuarm64 = "-machine virt,gic-version=3 -machine 
virtualization=true"
-# 32-bit Arm
-QB_MACHINE:qemuarm = "-machine virt -machine virtualization=true"
+# 32-bit Arm: highmem=off
+# Disable highmem so that qemu does not use highmem IO regions that end up
+# being placed at the 256GiB mark (e.g. ECAM space) and can cause issues in 
Xen.
+QB_MACHINE:qemuarm = "-machine virt,highmem=off -machine virtualization=true"
 
 # Increase the default qemu memory allocation to allow for the hypervisor.
 # Use a weak assignment to allow for change of default and override elsewhere.
-- 
2.25.1

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#7352): 
https://lists.yoctoproject.org/g/meta-virtualization/message/7352
Mute This Topic: https://lists.yoctoproject.org/mt/91642064/21656
Group Owner: [email protected]
Unsubscribe: https://lists.yoctoproject.org/g/meta-virtualization/unsub 
[[email protected]]
-=-=-=-=-=-=-=-=-=-=-=-

Reply via email to