KVM: SVM: Consider NUMA affinity when allocating per-CPU save_area
authorLi RongQing <lirongqing@baidu.com>
Mon, 20 May 2024 12:08:58 +0000 (20:08 +0800)
committerSean Christopherson <seanjc@google.com>
Mon, 3 Jun 2024 23:14:11 +0000 (16:14 -0700)
commit99a49093ce92d6116a1635802f9c16d0db6e805a
tree9efb8da910363a8ec40e7ab9bece1eb6cf023571
parent9f44286d77ac72a15692c56d0fcbf7d2534e1f1e
KVM: SVM: Consider NUMA affinity when allocating per-CPU save_area

save_area of per-CPU svm_data are dominantly accessed from their
own local CPUs, so allocate them node-local for performance reason

so rename __snp_safe_alloc_page as snp_safe_alloc_page_node which
accepts numa node id as input parameter, svm_cpu_init call it with
node id switched from cpu id

Signed-off-by: Li RongQing <lirongqing@baidu.com>
Reviewed-by: Tom Lendacky <thomas.lendacky@amd.com>
Link: https://lore.kernel.org/r/20240520120858.13117-4-lirongqing@baidu.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
arch/x86/kvm/svm/sev.c
arch/x86/kvm/svm/svm.c
arch/x86/kvm/svm/svm.h