[PATCH 0/3] kvm-s390: revised version of kvm-s390 guest memory handling - rebased v2

[PATCH 0/3] kvm-s390: revised version of kvm-s390 guest memory handling - rebased v2

[ehrhardt]

From: Christian Ehrhardt <ehrhardt@de.ibm.com>

As requested this is a rebased patch on top of the already applied v3
of the patch series.

*updates to already applied version*
- remove dependency to KVM_REQ_MMU_RELOAD in generic code
- remove explicit barrier after test_and_clear_bit as it is implied
- ensure the wait_on_bit waiter is notified
- move the reset of requests to kvm_vcpu_release to drop them early
- ensure dropping all vcpu requests while freeing a vcpu
- ensure kick allocations (might_sleep) are out of atomic context
- update vcpu->cpu in kvm-s390 arch handler for load/put
- centralize consumption of vcpu->request bits
- updates on running vcpus can now be handled without need to rerun the vcpu
- kvm_arch_set_memory_region waits until the bit is consumed by the vcpu
- kickout only scheduled vcpus (wait might hang forever on non-scheduled vcpus)

Note: further unification of make_all_cpu_request and the kick mechanism is
planned, but it might be good to split it from this step towards commonality.

Patches included:
Subject: [PATCH 1/3] kvm-s390: infrastructure to kick vcpus out of guest state - rebased
Subject: [PATCH 2/3] kvm-s390: update vcpu->cpu - rebased
Subject: [PATCH 3/3] kvm-s390: streamline memslot handling - rebased v2

Overall-Diffstat:
 arch/s390/include/asm/kvm_host.h |    2 +-
 arch/s390/kvm/intercept.c        |   10 ++++++----
 arch/s390/kvm/kvm-s390.c         |   36 +++++++++++++++++++++++++-----------
 arch/s390/kvm/kvm-s390.h         |   22 +++++++++++++++++++++-
 arch/s390/kvm/sigp.c             |   31 +++++++++++++++++++++----------
 virt/kvm/kvm_main.c              |    6 ++++++
 6 files changed, 80 insertions(+), 27 deletions(-)

[PATCH 1/3] kvm-s390: infrastructure to kick vcpus out of guest state - rebased

[ehrhardt]

From: Christian Ehrhardt <ehrhardt@linux.vnet.ibm.com>

As requested this is a rebased patch on top of the already applied v3
of the patch series.

*updates to already applied version*
- ensure allocations (might_sleep) are out of atomic context
- centralize consumption of vcpu->request bits

To ensure vcpu's come out of guest context in certain cases this patch adds a
s390 specific way to kick them out of guest context. Currently it kicks them
out to rerun the vcpu_run path in the s390 code, but the mechanism itself is
expandable and with a new flag we could also add e.g. kicks to userspace etc.

Signed-off-by: Christian Ehrhardt <ehrhardt@linux.vnet.ibm.com>
---

[diffstat]
 include/asm/kvm_host.h |    2 +-
 kvm/intercept.c        |   10 ++++++----
 kvm/kvm-s390.c         |    7 +++----
 kvm/kvm-s390.h         |   16 +++++++++++++++-
 kvm/sigp.c             |   31 +++++++++++++++++++++----------
 5 files changed, 46 insertions(+), 20 deletions(-)

Index: kvm/arch/s390/kvm/intercept.c
===================================================================
--- kvm.orig/arch/s390/kvm/intercept.c
+++ kvm/arch/s390/kvm/intercept.c
@@ -141,10 +141,12 @@ static int handle_stop(struct kvm_vcpu *
 			rc = -ENOTSUPP;
 	}
 
-	if (vcpu->arch.local_int.action_bits & ACTION_RELOADVCPU_ON_STOP) {
-		vcpu->arch.local_int.action_bits &= ~ACTION_RELOADVCPU_ON_STOP;
-		rc = SIE_INTERCEPT_RERUNVCPU;
-		vcpu->run->exit_reason = KVM_EXIT_INTR;
+	if (vcpu->arch.local_int.action_bits & ACTION_VCPUREQUEST_ON_STOP) {
+		vcpu->arch.local_int.action_bits &= ~ACTION_VCPUREQUEST_ON_STOP;
+		if (kvm_s390_handle_vcpu_requests(vcpu, VCPUREQUESTLVL_SIGP)) {
+			rc = SIE_INTERCEPT_CHECKREQUESTS;
+			vcpu->run->exit_reason = KVM_EXIT_INTR;
+		}
 	}
 
 	if (vcpu->arch.local_int.action_bits & ACTION_STOP_ON_STOP) {
Index: kvm/arch/s390/include/asm/kvm_host.h
===================================================================
--- kvm.orig/arch/s390/include/asm/kvm_host.h
+++ kvm/arch/s390/include/asm/kvm_host.h
@@ -182,7 +182,7 @@ struct kvm_s390_interrupt_info {
 /* for local_interrupt.action_flags */
 #define ACTION_STORE_ON_STOP		(1<<0)
 #define ACTION_STOP_ON_STOP		(1<<1)
-#define ACTION_RELOADVCPU_ON_STOP	(1<<2)
+#define ACTION_VCPUREQUEST_ON_STOP	(1<<2)
 
 struct kvm_s390_local_interrupt {
 	spinlock_t lock;
Index: kvm/arch/s390/kvm/kvm-s390.c
===================================================================
--- kvm.orig/arch/s390/kvm/kvm-s390.c
+++ kvm/arch/s390/kvm/kvm-s390.c
@@ -484,8 +484,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_v
 
 rerun_vcpu:
 	if (vcpu->requests)
-		if (test_and_clear_bit(KVM_REQ_MMU_RELOAD, &vcpu->requests))
-			kvm_s390_vcpu_set_mem(vcpu);
+		kvm_s390_handle_vcpu_requests(vcpu, VCPUREQUESTLVL_VCPURUN);
 
 	/* verify, that memory has been registered */
 	if (!vcpu->arch.sie_block->gmslm) {
@@ -521,7 +520,7 @@ rerun_vcpu:
 		rc = kvm_handle_sie_intercept(vcpu);
 	} while (!signal_pending(current) && !rc);
 
-	if (rc == SIE_INTERCEPT_RERUNVCPU)
+	if (rc == SIE_INTERCEPT_CHECKREQUESTS)
 		goto rerun_vcpu;
 
 	if (signal_pending(current) && !rc) {
@@ -710,7 +709,7 @@ int kvm_arch_set_memory_region(struct kv
 						&kvm->vcpus[i]->requests))
 				continue;
 			kvm_s390_inject_sigp_stop(kvm->vcpus[i],
-						  ACTION_RELOADVCPU_ON_STOP);
+						  ACTION_VCPUREQUEST_ON_STOP);
 		}
 	}
 
Index: kvm/arch/s390/kvm/kvm-s390.h
===================================================================
--- kvm.orig/arch/s390/kvm/kvm-s390.h
+++ kvm/arch/s390/kvm/kvm-s390.h
@@ -25,7 +25,7 @@
 typedef int (*intercept_handler_t)(struct kvm_vcpu *vcpu);
 
 /* negativ values are error codes, positive values for internal conditions */
-#define SIE_INTERCEPT_RERUNVCPU		(1<<0)
+#define SIE_INTERCEPT_CHECKREQUESTS		(1<<0)
 int kvm_handle_sie_intercept(struct kvm_vcpu *vcpu);
 
 #define VM_EVENT(d_kvm, d_loglevel, d_string, d_args...)\
@@ -81,6 +81,20 @@ static inline void kvm_s390_vcpu_set_mem
 	up_read(&vcpu->kvm->slots_lock);
 }
 
+/* interception levels from which handle vcpu requests can be called */
+#define VCPUREQUESTLVL_SIGP		1
+#define VCPUREQUESTLVL_VCPURUN		2
+static inline unsigned long kvm_s390_handle_vcpu_requests(struct kvm_vcpu *vcpu,
+						   int level)
+{
+	BUG_ON(!level);
+
+	if (!vcpu->requests)
+		return 0;
+
+	return vcpu->requests;
+}
+
 /* implemented in priv.c */
 int kvm_s390_handle_b2(struct kvm_vcpu *vcpu);
 
Index: kvm/arch/s390/kvm/sigp.c
===================================================================
--- kvm.orig/arch/s390/kvm/sigp.c
+++ kvm/arch/s390/kvm/sigp.c
@@ -108,15 +108,9 @@ unlock:
 	return rc;
 }
 
-static int __inject_sigp_stop(struct kvm_s390_local_interrupt *li, int action)
+static int __inject_sigp_stop(struct kvm_s390_local_interrupt *li, int action,
+			      struct kvm_s390_interrupt_info *inti)
 {
-	struct kvm_s390_interrupt_info *inti;
-
-	inti = kzalloc(sizeof(*inti), GFP_KERNEL);
-	if (!inti)
-		return -ENOMEM;
-	inti->type = KVM_S390_SIGP_STOP;
-
 	spin_lock_bh(&li->lock);
 	list_add_tail(&inti->list, &li->list);
 	atomic_set(&li->active, 1);
@@ -133,11 +127,17 @@ static int __sigp_stop(struct kvm_vcpu *
 {
 	struct kvm_s390_float_interrupt *fi = &vcpu->kvm->arch.float_int;
 	struct kvm_s390_local_interrupt *li;
+	struct kvm_s390_interrupt_info *inti;
 	int rc;
 
 	if (cpu_addr >= KVM_MAX_VCPUS)
 		return 3; /* not operational */
 
+	inti = kzalloc(sizeof(*inti), GFP_KERNEL);
+	if (!inti)
+		return -ENOMEM;
+	inti->type = KVM_S390_SIGP_STOP;
+
 	spin_lock(&fi->lock);
 	li = fi->local_int[cpu_addr];
 	if (li == NULL) {
@@ -145,7 +145,7 @@ static int __sigp_stop(struct kvm_vcpu *
 		goto unlock;
 	}
 
-	rc = __inject_sigp_stop(li, action);
+	rc = __inject_sigp_stop(li, action, inti);
 
 unlock:
 	spin_unlock(&fi->lock);
@@ -156,7 +156,18 @@ unlock:
 int kvm_s390_inject_sigp_stop(struct kvm_vcpu *vcpu, int action)
 {
 	struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int;
-	return __inject_sigp_stop(li, action);
+	struct kvm_s390_interrupt_info *inti;
+	int rc;
+
+	inti = kzalloc(sizeof(*inti), GFP_KERNEL);
+	if (!inti)
+		return -ENOMEM;
+	inti->type = KVM_S390_SIGP_STOP;
+
+	rc = __inject_sigp_stop(li, action, inti);
+	if (rc)
+		kfree(inti);
+	return rc;
 }
 
 static int __sigp_set_arch(struct kvm_vcpu *vcpu, u32 parameter)

[PATCH 2/3] kvm-s390: update vcpu->cpu - rebased

[ehrhardt]

From: Christian Ehrhardt <ehrhardt@linux.vnet.ibm.com>

As requested this is a rebased patch on top of the already applied v3
of the patch series.

kvm on s390 formerly ignored vcpu->cpu.
This patch adds set/unset vcpu->cpu in kvm_arch_vcpu_load/put to allow
further architecture unification e.g. let generic code not find -1 on
currently scheduled vcpus.

Signed-off-by: Christian Ehrhardt <ehrhardt@linux.vnet.ibm.com>
---

[diffstat]
 kvm-s390.c |    2 ++
 1 file changed, 2 insertions(+)

[diff]
Index: kvm/arch/s390/kvm/kvm-s390.c
===================================================================
--- kvm.orig/arch/s390/kvm/kvm-s390.c
+++ kvm/arch/s390/kvm/kvm-s390.c
@@ -244,6 +244,7 @@ void kvm_arch_vcpu_uninit(struct kvm_vcp
 
 void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
 {
+	vcpu->cpu = cpu;
 	save_fp_regs(&vcpu->arch.host_fpregs);
 	save_access_regs(vcpu->arch.host_acrs);
 	vcpu->arch.guest_fpregs.fpc &= FPC_VALID_MASK;
@@ -253,6 +254,7 @@ void kvm_arch_vcpu_load(struct kvm_vcpu 
 
 void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
 {
+	vcpu->cpu = -1;
 	save_fp_regs(&vcpu->arch.guest_fpregs);
 	save_access_regs(vcpu->arch.guest_acrs);
 	restore_fp_regs(&vcpu->arch.host_fpregs);

[PATCH 3/3] kvm-s390: streamline memslot handling - rebased v2

[ehrhardt]

From: Christian Ehrhardt <ehrhardt@linux.vnet.ibm.com>

As requested this is a rebased patch on top of the already applied v3
of the patch series.

*updates to applied version*
- remove dependency to KVM_REQ_MMU_RELOAD in generic code
- remove explicit barrier after test_and_clear_bit as it is implied
- ensure the wait_on_bit waiter is notified
- ensure dropping vcpu all requests while freeing a vcpu
- kickout only scheduled vcpus (its superfluous and wait might hang forever on
  not running vcpus)
- kvm_arch_set_memory_region waits until the bit is consumed by the vcpu

This patch relocates the variables kvm-s390 uses to track guest mem addr/size.
As discussed dropping the variables at struct kvm_arch level allows to use the
common vcpu->request based mechanism to reload guest memory if e.g. changes
via set_memory_region.
The kick mechanism introduced in this series is used to ensure running vcpus
leave guest state to catch the update.

Signed-off-by: Christian Ehrhardt <ehrhardt@linux.vnet.ibm.com>
---

[diffstat]
 arch/s390/kvm/kvm-s390.c |   27 ++++++++++++++++++++-------
 arch/s390/kvm/kvm-s390.h |    6 ++++++
 virt/kvm/kvm_main.c      |    6 ++++++
 3 files changed, 32 insertions(+), 7 deletions(-)

Index: kvm/arch/s390/kvm/kvm-s390.c
===================================================================
--- kvm.orig/arch/s390/kvm/kvm-s390.c
+++ kvm/arch/s390/kvm/kvm-s390.c
@@ -674,6 +674,12 @@ long kvm_arch_vcpu_ioctl(struct file *fi
 	return -EINVAL;
 }
 
+static int wait_bit_schedule(void *word)
+{
+	schedule();
+	return 0;
+}
+
 /* Section: memory related */
 int kvm_arch_set_memory_region(struct kvm *kvm,
 				struct kvm_userspace_memory_region *mem,
@@ -681,6 +687,7 @@ int kvm_arch_set_memory_region(struct kv
 				int user_alloc)
 {
 	int i;
+	struct kvm_vcpu *vcpu;
 
 	/* A few sanity checks. We can have exactly one memory slot which has
 	   to start at guest virtual zero and which has to be located at a
@@ -706,13 +713,19 @@ int kvm_arch_set_memory_region(struct kv
 
 	/* request update of sie control block for all available vcpus */
 	for (i = 0; i < KVM_MAX_VCPUS; ++i) {
-		if (kvm->vcpus[i]) {
-			if (test_and_set_bit(KVM_REQ_MMU_RELOAD,
-						&kvm->vcpus[i]->requests))
-				continue;
-			kvm_s390_inject_sigp_stop(kvm->vcpus[i],
-						  ACTION_VCPUREQUEST_ON_STOP);
-		}
+		vcpu = kvm->vcpus[i];
+		if (!vcpu)
+			continue;
+
+		if (!test_and_set_bit(KVM_REQ_MMU_RELOAD, &vcpu->requests))
+			continue;
+
+		if (vcpu->cpu == -1)
+			continue;
+
+		kvm_s390_inject_sigp_stop(vcpu, ACTION_VCPUREQUEST_ON_STOP);
+		wait_on_bit(&vcpu->requests, KVM_REQ_MMU_RELOAD,
+			    wait_bit_schedule, TASK_UNINTERRUPTIBLE);
 	}
 
 	return 0;
Index: kvm/arch/s390/kvm/kvm-s390.h
===================================================================
--- kvm.orig/arch/s390/kvm/kvm-s390.h
+++ kvm/arch/s390/kvm/kvm-s390.h
@@ -92,6 +92,12 @@ static inline unsigned long kvm_s390_han
 	if (!vcpu->requests)
 		return 0;
 
+	/* requests that can be handled at all levels */
+	if (test_and_clear_bit(KVM_REQ_MMU_RELOAD, &vcpu->requests)) {
+		wake_up_bit(&vcpu->requests, KVM_REQ_MMU_RELOAD);
+		kvm_s390_vcpu_set_mem(vcpu);
+	}
+
 	return vcpu->requests;
 }
 
Index: kvm/virt/kvm/kvm_main.c
===================================================================
--- kvm.orig/virt/kvm/kvm_main.c
+++ kvm/virt/kvm/kvm_main.c
@@ -1681,6 +1681,12 @@ static int kvm_vcpu_mmap(struct file *fi
 static int kvm_vcpu_release(struct inode *inode, struct file *filp)
 {
 	struct kvm_vcpu *vcpu = filp->private_data;
+	int i;
+
+	vcpu->requests = 0;
+	smp_mb();
+	for (i = 0; i < sizeof(vcpu->requests); i++)
+		wake_up_bit(&vcpu->requests, i);
 
 	kvm_put_kvm(vcpu->kvm);
 	return 0;