From: Raghavendra Rao Ananta Date: Fri, 11 Aug 2023 04:51:26 +0000 (+0000) Subject: KVM: arm64: Invalidate the table entries upon a range X-Git-Tag: v6.6-rc1~36^2~9^2~2^2~1 X-Git-Url: http://git.ipfire.org/?a=commitdiff_plain;h=defc8cc7abf0fcee8d73e440ee02827348d060e0;p=thirdparty%2Flinux.git KVM: arm64: Invalidate the table entries upon a range Currently, during the operations such as a hugepage collapse, KVM would flush the entire VM's context using 'vmalls12e1is' TLBI operation. Specifically, if the VM is faulting on many hugepages (say after dirty-logging), it creates a performance penalty for the guest whose pages have already been faulted earlier as they would have to refill their TLBs again. Instead, leverage kvm_tlb_flush_vmid_range() for table entries. If the system supports it, only the required range will be flushed. Else, it'll fallback to the previous mechanism. Signed-off-by: Raghavendra Rao Ananta Reviewed-by: Gavin Shan Reviewed-by: Shaoqin Huang Signed-off-by: Marc Zyngier Link: https://lore.kernel.org/r/20230811045127.3308641-14-rananta@google.com --- diff --git a/arch/arm64/kvm/hyp/pgtable.c b/arch/arm64/kvm/hyp/pgtable.c index fb58a229b7440..1a4af37fd8c00 100644 --- a/arch/arm64/kvm/hyp/pgtable.c +++ b/arch/arm64/kvm/hyp/pgtable.c @@ -806,7 +806,8 @@ static bool stage2_try_break_pte(const struct kvm_pgtable_visit_ctx *ctx, * evicted pte value (if any). */ if (kvm_pte_table(ctx->old, ctx->level)) - kvm_call_hyp(__kvm_tlb_flush_vmid, mmu); + kvm_tlb_flush_vmid_range(mmu, ctx->addr, + kvm_granule_size(ctx->level)); else if (kvm_pte_valid(ctx->old)) kvm_call_hyp(__kvm_tlb_flush_vmid_ipa, mmu, ctx->addr, ctx->level);