linux-arm-kernel.lists.infradead.org archive mirror
 help / color / mirror / Atom feed
From: Dave.Martin@arm.com (Dave Martin)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH 4/4] KVM: arm64: Avoid mistaken attempts to save SVE state for vcpus
Date: Thu, 14 Jun 2018 12:42:26 +0100	[thread overview]
Message-ID: <1528976546-25999-4-git-send-email-Dave.Martin@arm.com> (raw)
In-Reply-To: <1528976546-25999-1-git-send-email-Dave.Martin@arm.com>

Commit e6b673b ("KVM: arm64: Optimise FPSIMD handling to reduce
guest/host thrashing") uses fpsimd_save() to save the FPSIMD state
for a vcpu when scheduling the vcpu out.  However, currently
current's value of TIF_SVE is restored before calling fpsimd_save()
which means that fpsimd_save() may erroneously attempt to save SVE
state from the vcpu.  This enables current's vector state to be
polluted with guest data.  current->thread.sve_state may be
unallocated or not large enough, so this can also trigger a NULL
dereference or buffer overrun.

Instead of this, TIF_SVE should be configured properly for the
guest when calling fpsimd_save() with the vcpu context loaded.

This patch ensures this my delaying restoration of current's
TIF_SVE until after the call to fpsimd_save().

Fixes: e6b673b741ea ("KVM: arm64: Optimise FPSIMD handling to reduce guest/host thrashing")
Signed-off-by: Dave Martin <Dave.Martin@arm.com>
---
 arch/arm64/kvm/fpsimd.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/arch/arm64/kvm/fpsimd.c b/arch/arm64/kvm/fpsimd.c
index 30c7a34..4aaf78e 100644
--- a/arch/arm64/kvm/fpsimd.c
+++ b/arch/arm64/kvm/fpsimd.c
@@ -99,9 +99,6 @@ void kvm_arch_vcpu_ctxsync_fp(struct kvm_vcpu *vcpu)
  */
 static void __kvm_arch_vcpu_put_fp(struct kvm_vcpu *vcpu)
 {
-	update_thread_flag(TIF_SVE,
-			   vcpu->arch.flags & KVM_ARM64_HOST_SVE_IN_USE);
-
 	if (vcpu->arch.flags & KVM_ARM64_FP_ENABLED) {
 		/* Clean guest FP state to memory and invalidate cpu view */
 		fpsimd_save();
@@ -119,6 +116,9 @@ static void __kvm_arch_vcpu_put_fp(struct kvm_vcpu *vcpu)
 		else
 			sysreg_clear_set(CPACR_EL1, CPACR_EL1_ZEN_EL0EN, 0);
 	}
+
+	update_thread_flag(TIF_SVE,
+			   vcpu->arch.flags & KVM_ARM64_HOST_SVE_IN_USE);
 }
 
 
-- 
2.1.4

  parent reply	other threads:[~2018-06-14 11:42 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-06-14 11:33 [PATCH 0/4] KVM: arm64: FPSIMD/SVE fixes for 4.17 Dave Martin
2018-06-14 11:42 ` [PATCH 1/4] arm64: introduce sysreg_clear_set() Dave Martin
2018-06-14 11:42   ` [PATCH 2/4] KVM: arm64: Don't mask softirq with IRQs disabled in vcpu_put() Dave Martin
2018-06-14 12:50     ` Marc Zyngier
2018-06-14 12:57       ` Dave Martin
2018-06-14 11:42   ` [PATCH 3/4] KVM: arm64/sve: Fix SVE trap restoration for non-current tasks Dave Martin
2018-06-14 11:42   ` Dave Martin [this message]
2018-06-15  8:50     ` [PATCH 4/4] KVM: arm64: Avoid mistaken attempts to save SVE state for vcpus Alex Bennée
2018-06-15  9:27       ` Dave Martin
2018-06-14 12:45 ` [PATCH 0/4] KVM: arm64: FPSIMD/SVE fixes for 4.17 Dave Martin

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=1528976546-25999-4-git-send-email-Dave.Martin@arm.com \
    --to=dave.martin@arm.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    /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).