kvm.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Paolo Bonzini <pbonzini@redhat.com>
To: linux-kernel@vger.kernel.org, kvm@vger.kernel.org
Cc: dmatlack@google.com, seanjc@google.com
Subject: [PATCH v3 28/31] KVM: MMU: pass struct kvm_page_fault to mmu_set_spte
Date: Fri, 24 Sep 2021 12:31:49 -0400	[thread overview]
Message-ID: <20210924163152.289027-29-pbonzini@redhat.com> (raw)
In-Reply-To: <20210924163152.289027-1-pbonzini@redhat.com>

mmu_set_spte is called for either PTE prefetching or page faults.  The
three boolean arguments write_fault, speculative and host_writable are
always respectively false/true/true for prefetching and coming from
a struct kvm_page_fault for page faults.

Let mmu_set_spte distinguish these two situation by accepting a
possibly NULL struct kvm_page_fault argument.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 arch/x86/kvm/mmu/mmu.c         | 17 ++++++++++-------
 arch/x86/kvm/mmu/paging_tmpl.h | 13 +++----------
 2 files changed, 13 insertions(+), 17 deletions(-)

diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c
index c208f001c302..4b304f60cf44 100644
--- a/arch/x86/kvm/mmu/mmu.c
+++ b/arch/x86/kvm/mmu/mmu.c
@@ -2675,9 +2675,8 @@ int mmu_try_to_unsync_pages(struct kvm_vcpu *vcpu, gfn_t gfn, bool can_unsync,
 }
 
 static int mmu_set_spte(struct kvm_vcpu *vcpu, u64 *sptep,
-			unsigned int pte_access, bool write_fault,
-			gfn_t gfn, kvm_pfn_t pfn, bool speculative,
-			bool host_writable)
+			unsigned int pte_access, gfn_t gfn,
+			kvm_pfn_t pfn, struct kvm_page_fault *fault)
 {
 	struct kvm_mmu_page *sp = sptep_to_sp(sptep);
 	int level = sp->role.level;
@@ -2687,6 +2686,11 @@ static int mmu_set_spte(struct kvm_vcpu *vcpu, u64 *sptep,
 	bool wrprot;
 	u64 spte;
 
+	/* Prefetching always gets a writable pfn.  */
+	bool host_writable = !fault || fault->map_writable;
+	bool speculative = !fault || fault->prefault;
+	bool write_fault = fault && fault->write;
+
 	pgprintk("%s: spte %llx write_fault %d gfn %llx\n", __func__,
 		 *sptep, write_fault, gfn);
 
@@ -2778,8 +2782,8 @@ static int direct_pte_prefetch_many(struct kvm_vcpu *vcpu,
 		return -1;
 
 	for (i = 0; i < ret; i++, gfn++, start++) {
-		mmu_set_spte(vcpu, start, access, false, gfn,
-			     page_to_pfn(pages[i]), true, true);
+		mmu_set_spte(vcpu, start, access, gfn,
+			     page_to_pfn(pages[i]), NULL);
 		put_page(pages[i]);
 	}
 
@@ -2981,8 +2985,7 @@ static int __direct_map(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault)
 		return -EFAULT;
 
 	ret = mmu_set_spte(vcpu, it.sptep, ACC_ALL,
-			   fault->write, base_gfn, fault->pfn,
-			   fault->prefault, fault->map_writable);
+			   base_gfn, fault->pfn, fault);
 	if (ret == RET_PF_SPURIOUS)
 		return ret;
 
diff --git a/arch/x86/kvm/mmu/paging_tmpl.h b/arch/x86/kvm/mmu/paging_tmpl.h
index fbbaa3f5fb4e..8c07c42a4d73 100644
--- a/arch/x86/kvm/mmu/paging_tmpl.h
+++ b/arch/x86/kvm/mmu/paging_tmpl.h
@@ -578,13 +578,7 @@ FNAME(prefetch_gpte)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp,
 	if (is_error_pfn(pfn))
 		return false;
 
-	/*
-	 * we call mmu_set_spte() with host_writable = true because
-	 * pte_prefetch_gfn_to_pfn always gets a writable pfn.
-	 */
-	mmu_set_spte(vcpu, spte, pte_access, false, gfn, pfn,
-		     true, true);
-
+	mmu_set_spte(vcpu, spte, pte_access, gfn, pfn, NULL);
 	kvm_release_pfn_clean(pfn);
 	return true;
 }
@@ -763,9 +757,8 @@ static int FNAME(fetch)(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault,
 	if (WARN_ON_ONCE(it.level != fault->goal_level))
 		return -EFAULT;
 
-	ret = mmu_set_spte(vcpu, it.sptep, gw->pte_access, fault->write,
-			   base_gfn, fault->pfn, fault->prefault,
-			   fault->map_writable);
+	ret = mmu_set_spte(vcpu, it.sptep, gw->pte_access,
+			   base_gfn, fault->pfn, fault);
 	if (ret == RET_PF_SPURIOUS)
 		return ret;
 
-- 
2.27.0



  parent reply	other threads:[~2021-09-24 16:33 UTC|newest]

Thread overview: 37+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-09-24 16:31 [PATCH v3 00/31] KVM: x86: pass arguments on the page fault path via struct kvm_page_fault Paolo Bonzini
2021-09-24 16:31 ` [PATCH v3 01/31] KVM: MMU: pass unadulterated gpa to direct_page_fault Paolo Bonzini
2021-09-24 16:31 ` [PATCH v3 02/31] KVM: MMU: Introduce struct kvm_page_fault Paolo Bonzini
2021-09-28 23:25   ` David Matlack
2021-09-29 11:13     ` Paolo Bonzini
2021-09-24 16:31 ` [PATCH v3 03/31] KVM: MMU: change mmu->page_fault() arguments to kvm_page_fault Paolo Bonzini
2021-09-24 16:31 ` [PATCH v3 04/31] KVM: MMU: change direct_page_fault() " Paolo Bonzini
2021-09-24 16:31 ` [PATCH v3 05/31] KVM: MMU: change page_fault_handle_page_track() " Paolo Bonzini
2021-09-24 16:31 ` [PATCH v3 06/31] KVM: MMU: change kvm_faultin_pfn() " Paolo Bonzini
2021-09-24 16:31 ` [PATCH v3 07/31] KVM: MMU: change handle_abnormal_pfn() " Paolo Bonzini
2021-09-24 16:31 ` [PATCH v3 08/31] KVM: MMU: change __direct_map() " Paolo Bonzini
2021-09-24 16:31 ` [PATCH v3 09/31] KVM: MMU: change FNAME(fetch)() " Paolo Bonzini
2021-09-24 16:31 ` [PATCH v3 10/31] KVM: MMU: change kvm_tdp_mmu_map() " Paolo Bonzini
2021-09-24 16:31 ` [PATCH v3 11/31] KVM: MMU: change tdp_mmu_map_handle_target_level() " Paolo Bonzini
2021-09-24 16:31 ` [PATCH v3 12/31] KVM: MMU: change fast_page_fault() " Paolo Bonzini
2021-09-24 16:31 ` [PATCH v3 13/31] KVM: MMU: change kvm_mmu_hugepage_adjust() " Paolo Bonzini
2021-09-24 16:31 ` [PATCH v3 14/31] KVM: MMU: change disallowed_hugepage_adjust() " Paolo Bonzini
2021-09-24 16:31 ` [PATCH v3 15/31] KVM: MMU: change tracepoints " Paolo Bonzini
2021-09-24 16:31 ` [PATCH v3 16/31] KVM: x86/mmu: Verify shadow walk doesn't terminate early in page faults Paolo Bonzini
2021-09-24 16:31 ` [PATCH v3 17/31] KVM: x86/mmu: Fold rmap_recycle into rmap_add Paolo Bonzini
2021-09-24 16:31 ` [PATCH v3 18/31] KVM: MMU: mark page dirty in make_spte Paolo Bonzini
2021-09-24 16:31 ` [PATCH v3 19/31] KVM: MMU: unify tdp_mmu_map_set_spte_atomic and tdp_mmu_set_spte_atomic_no_dirty_log Paolo Bonzini
2021-09-24 16:31 ` [PATCH v3 20/31] KVM: x86/mmu: Pass the memslot around via struct kvm_page_fault Paolo Bonzini
2021-09-24 16:31 ` [PATCH v3 21/31] KVM: x86/mmu: Avoid memslot lookup in page_fault_handle_page_track Paolo Bonzini
2021-09-24 16:31 ` [PATCH v3 22/31] KVM: MMU: inline set_spte in mmu_set_spte Paolo Bonzini
2021-09-24 16:31 ` [PATCH v3 23/31] KVM: MMU: inline set_spte in FNAME(sync_page) Paolo Bonzini
2021-09-24 16:31 ` [PATCH v3 24/31] KVM: MMU: clean up make_spte return value Paolo Bonzini
2021-09-24 16:31 ` [PATCH v3 25/31] KVM: MMU: remove unnecessary argument to mmu_set_spte Paolo Bonzini
2021-09-24 16:31 ` [PATCH v3 26/31] KVM: MMU: set ad_disabled in TDP MMU role Paolo Bonzini
2021-09-24 16:31 ` [PATCH v3 27/31] KVM: MMU: pass kvm_mmu_page struct to make_spte Paolo Bonzini
2021-09-24 16:31 ` Paolo Bonzini [this message]
2021-09-24 16:31 ` [PATCH v3 29/31] KVM: x86/mmu: Avoid memslot lookup in rmap_add Paolo Bonzini
2021-09-24 16:31 ` [PATCH v3 30/31] KVM: x86/mmu: Avoid memslot lookup in make_spte and mmu_try_to_unsync_pages Paolo Bonzini
2021-09-24 16:31 ` [PATCH v3 31/31] KVM: MMU: make spte an in-out argument in make_spte Paolo Bonzini
2021-09-28 23:20   ` David Matlack
2021-09-29 11:14     ` Paolo Bonzini
2021-09-28 23:27 ` [PATCH v3 00/31] KVM: x86: pass arguments on the page fault path via struct kvm_page_fault David Matlack

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=20210924163152.289027-29-pbonzini@redhat.com \
    --to=pbonzini@redhat.com \
    --cc=dmatlack@google.com \
    --cc=kvm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=seanjc@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 NNTP newsgroup(s).