Merge tag 'kvm-s390-next-4.10-1' of git://git.kernel.org/pub/scm/linux/kernel/git/kvms390/linux

KVM: s390: Changes for 4.10 (via kvm/next)

Two small optimizations to not do register reloading in
vcpu_put/get, instead do it in the ioctl path. This reduces
the overhead for schedule-intense workload that does not
exit to QEMU. (e.g. KVM guest with eventfd/irqfd that
does a lot of context switching with vhost or iothreads).

arch/s390/kvm/interrupt.c

@@ -415,7 +415,7 @@ static int __write_machine_check(struct kvm_vcpu *vcpu,
 	int rc;
 
 	mci.val = mchk->mcic;
-	/* take care of lazy register loading via vcpu load/put */
+	/* take care of lazy register loading */
 	save_fpu_regs();
 	save_access_regs(vcpu->run->s.regs.acrs);
 

arch/s390/kvm/kvm-s390.c

@@ -1812,22 +1812,7 @@ __u64 kvm_s390_get_cpu_timer(struct kvm_vcpu *vcpu)
 
 void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
 {
-	/* Save host register state */
-	save_fpu_regs();
-	vcpu->arch.host_fpregs.fpc = current->thread.fpu.fpc;
-	vcpu->arch.host_fpregs.regs = current->thread.fpu.regs;
-
-	if (MACHINE_HAS_VX)
-		current->thread.fpu.regs = vcpu->run->s.regs.vrs;
-	else
-		current->thread.fpu.regs = vcpu->run->s.regs.fprs;
-	current->thread.fpu.fpc = vcpu->run->s.regs.fpc;
-	if (test_fp_ctl(current->thread.fpu.fpc))
-		/* User space provided an invalid FPC, let's clear it */
-		current->thread.fpu.fpc = 0;
 
-	save_access_regs(vcpu->arch.host_acrs);
-	restore_access_regs(vcpu->run->s.regs.acrs);
 	gmap_enable(vcpu->arch.enabled_gmap);
 	atomic_or(CPUSTAT_RUNNING, &vcpu->arch.sie_block->cpuflags);
 	if (vcpu->arch.cputm_enabled && !is_vcpu_idle(vcpu))
@@ -1844,16 +1829,6 @@ void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
 	vcpu->arch.enabled_gmap = gmap_get_enabled();
 	gmap_disable(vcpu->arch.enabled_gmap);
 
-	/* Save guest register state */
-	save_fpu_regs();
-	vcpu->run->s.regs.fpc = current->thread.fpu.fpc;
-
-	/* Restore host register state */
-	current->thread.fpu.fpc = vcpu->arch.host_fpregs.fpc;
-	current->thread.fpu.regs = vcpu->arch.host_fpregs.regs;
-
-	save_access_regs(vcpu->run->s.regs.acrs);
-	restore_access_regs(vcpu->arch.host_acrs);
 }
 
 static void kvm_s390_vcpu_initial_reset(struct kvm_vcpu *vcpu)
@@ -2243,7 +2218,6 @@ int kvm_arch_vcpu_ioctl_set_sregs(struct kvm_vcpu *vcpu,
 {
 	memcpy(&vcpu->run->s.regs.acrs, &sregs->acrs, sizeof(sregs->acrs));
 	memcpy(&vcpu->arch.sie_block->gcr, &sregs->crs, sizeof(sregs->crs));
-	restore_access_regs(vcpu->run->s.regs.acrs);
 	return 0;
 }
 
@@ -2257,11 +2231,9 @@ int kvm_arch_vcpu_ioctl_get_sregs(struct kvm_vcpu *vcpu,
 
 int kvm_arch_vcpu_ioctl_set_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu)
 {
-	/* make sure the new values will be lazily loaded */
-	save_fpu_regs();
 	if (test_fp_ctl(fpu->fpc))
 		return -EINVAL;
-	current->thread.fpu.fpc = fpu->fpc;
+	vcpu->run->s.regs.fpc = fpu->fpc;
 	if (MACHINE_HAS_VX)
 		convert_fp_to_vx((__vector128 *) vcpu->run->s.regs.vrs,
 				 (freg_t *) fpu->fprs);
@@ -2279,7 +2251,7 @@ int kvm_arch_vcpu_ioctl_get_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu)
 				 (__vector128 *) vcpu->run->s.regs.vrs);
 	else
 		memcpy(fpu->fprs, vcpu->run->s.regs.fprs, sizeof(fpu->fprs));
-	fpu->fpc = current->thread.fpu.fpc;
+	fpu->fpc = vcpu->run->s.regs.fpc;
 	return 0;
 }
 
@@ -2740,6 +2712,20 @@ static void sync_regs(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 		if (riccb->valid)
 			vcpu->arch.sie_block->ecb3 |= 0x01;
 	}
+	save_access_regs(vcpu->arch.host_acrs);
+	restore_access_regs(vcpu->run->s.regs.acrs);
+	/* save host (userspace) fprs/vrs */
+	save_fpu_regs();
+	vcpu->arch.host_fpregs.fpc = current->thread.fpu.fpc;
+	vcpu->arch.host_fpregs.regs = current->thread.fpu.regs;
+	if (MACHINE_HAS_VX)
+		current->thread.fpu.regs = vcpu->run->s.regs.vrs;
+	else
+		current->thread.fpu.regs = vcpu->run->s.regs.fprs;
+	current->thread.fpu.fpc = vcpu->run->s.regs.fpc;
+	if (test_fp_ctl(current->thread.fpu.fpc))
+		/* User space provided an invalid FPC, let's clear it */
+		current->thread.fpu.fpc = 0;
 
 	kvm_run->kvm_dirty_regs = 0;
 }
@@ -2758,6 +2744,15 @@ static void store_regs(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 	kvm_run->s.regs.pft = vcpu->arch.pfault_token;
 	kvm_run->s.regs.pfs = vcpu->arch.pfault_select;
 	kvm_run->s.regs.pfc = vcpu->arch.pfault_compare;
+	save_access_regs(vcpu->run->s.regs.acrs);
+	restore_access_regs(vcpu->arch.host_acrs);
+	/* Save guest register state */
+	save_fpu_regs();
+	vcpu->run->s.regs.fpc = current->thread.fpu.fpc;
+	/* Restore will be done lazily at return */
+	current->thread.fpu.fpc = vcpu->arch.host_fpregs.fpc;
+	current->thread.fpu.regs = vcpu->arch.host_fpregs.regs;
+
 }
 
 int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
@@ -2874,7 +2869,7 @@ int kvm_s390_vcpu_store_status(struct kvm_vcpu *vcpu, unsigned long addr)
 {
 	/*
 	 * The guest FPRS and ACRS are in the host FPRS/ACRS due to the lazy
-	 * copying in vcpu load/put. Lets update our copies before we save
+	 * switch in the run ioctl. Let's update our copies before we save
 	 * it into the save area
 	 */
 	save_fpu_regs();