vmm: memory_manager: Apply NUMA policy to memory zones

Relying on the new option 'host_numa_node' from the 'memory-zone'
parameter, the user can now define which NUMA node from the host
should be used to back the current memory zone.

Signed-off-by: Sebastien Boeuf <sebastien.boeuf@intel.com>
This commit is contained in:
Sebastien Boeuf
2020-08-25 18:02:34 +02:00
committed by Bo Chen
parent e6f585a31c
commit 1b4591aecc
3 changed files with 91 additions and 1 deletions

View File

@@ -2217,7 +2217,12 @@ mod tests {
let mut cmd = GuestCommand::new(&guest);
cmd.args(&["--cpus", "boot=1"])
.args(&["--memory", "size=0"])
.args(&["--memory-zone", "size=1G", "size=3G,file=/dev/shm"])
.args(&[
"--memory-zone",
"size=1G",
"size=3G,file=/dev/shm",
"size=1G,host_numa_node=0",
])
.args(&["--kernel", guest.fw_path.as_str()])
.default_disks()
.default_net();