From: Li RongQing <lirongqing@baidu.com>
To: seanjc@google.com, pbonzini@redhat.com, kvm@vger.kernel.org,
thomas.lendacky@amd.com, yosryahmed@google.com,
pgonda@google.com
Cc: Li RongQing <lirongqing@baidu.com>
Subject: [PATCH 3/3] KVM: SVM: Consider NUMA affinity when allocating per-CPU save_area
Date: Mon, 6 May 2024 11:11:06 +0800 [thread overview]
Message-ID: <20240506031106.3894-4-lirongqing@baidu.com> (raw)
In-Reply-To: <20240506031106.3894-1-lirongqing@baidu.com>
save_area of per-CPU svm_data is dominantly accessed from their
own local CPUs, 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 calls it with
node id switched from cpu id
Signed-off-by: Li RongQing <lirongqing@baidu.com>
---
arch/x86/kvm/svm/sev.c | 6 +++---
arch/x86/kvm/svm/svm.c | 2 +-
arch/x86/kvm/svm/svm.h | 4 ++--
3 files changed, 6 insertions(+), 6 deletions(-)
diff --git a/arch/x86/kvm/svm/sev.c b/arch/x86/kvm/svm/sev.c
index 24608a2..8a19dd8 100644
--- a/arch/x86/kvm/svm/sev.c
+++ b/arch/x86/kvm/svm/sev.c
@@ -3179,13 +3179,13 @@ void sev_vcpu_deliver_sipi_vector(struct kvm_vcpu *vcpu, u8 vector)
ghcb_set_sw_exit_info_2(svm->sev_es.ghcb, 1);
}
-struct page *__snp_safe_alloc_page(gfp_t gfp)
+struct page *snp_safe_alloc_page_node(int node, gfp_t gfp)
{
unsigned long pfn;
struct page *p;
if (!cc_platform_has(CC_ATTR_HOST_SEV_SNP))
- return alloc_page(gfp | __GFP_ZERO);
+ return alloc_pages_node(node, gfp | __GFP_ZERO, 0);
/*
* Allocate an SNP-safe page to workaround the SNP erratum where
@@ -3196,7 +3196,7 @@ struct page *__snp_safe_alloc_page(gfp_t gfp)
* Allocate one extra page, choose a page which is not
* 2MB-aligned, and free the other.
*/
- p = alloc_pages(gfp | __GFP_ZERO, 1);
+ p = alloc_pages_node(node, gfp | __GFP_ZERO, 1);
if (!p)
return NULL;
diff --git a/arch/x86/kvm/svm/svm.c b/arch/x86/kvm/svm/svm.c
index c01dc9b..8df05f2 100644
--- a/arch/x86/kvm/svm/svm.c
+++ b/arch/x86/kvm/svm/svm.c
@@ -703,7 +703,7 @@ static int svm_cpu_init(int cpu)
int ret = -ENOMEM;
memset(sd, 0, sizeof(struct svm_cpu_data));
- sd->save_area = __snp_safe_alloc_page(GFP_KERNEL);
+ sd->save_area = snp_safe_alloc_page_node(cpu_to_node(cpu), GFP_KERNEL);
if (!sd->save_area)
return ret;
diff --git a/arch/x86/kvm/svm/svm.h b/arch/x86/kvm/svm/svm.h
index bbea3ee..982e42b 100644
--- a/arch/x86/kvm/svm/svm.h
+++ b/arch/x86/kvm/svm/svm.h
@@ -695,11 +695,11 @@ void sev_vcpu_deliver_sipi_vector(struct kvm_vcpu *vcpu, u8 vector);
void sev_es_prepare_switch_to_guest(struct sev_es_save_area *hostsa);
void sev_es_unmap_ghcb(struct vcpu_svm *svm);
-struct page *__snp_safe_alloc_page(gfp_t gfp);
+struct page *snp_safe_alloc_page_node(int node, gfp_t gfp);
static inline struct page *snp_safe_alloc_page(void)
{
- return __snp_safe_alloc_page(GFP_KERNEL_ACCOUNT);
+ return snp_safe_alloc_page_node(numa_node_id(), GFP_KERNEL_ACCOUNT);
}
/* vmenter.S */
--
2.9.4
prev parent reply other threads:[~2024-05-06 3:19 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <20240506031106.3894-1-lirongqing@baidu.com>
2024-05-06 3:11 ` [PATCH 1/3] KVM: SVM: remove useless input parameter in snp_safe_alloc_page Li RongQing
2024-05-06 3:11 ` Li RongQing [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20240506031106.3894-4-lirongqing@baidu.com \
--to=lirongqing@baidu.com \
--cc=kvm@vger.kernel.org \
--cc=pbonzini@redhat.com \
--cc=pgonda@google.com \
--cc=seanjc@google.com \
--cc=thomas.lendacky@amd.com \
--cc=yosryahmed@google.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).