[PATCH v5 0/6] kvm: Make ioeventfd usable on s390.

[PATCH v5 0/6] kvm: Make ioeventfd usable on s390.

[Cornelia Huck]

v5 of the ioeventfd patch set, this time with a proper return code
from __diag_virtio_hypercall(), otherwise unchanged.

v4 -> v5:
- Proper return code in __diag_virtio_hypercall()
v3 -> v4:
- Pass cookies in virtio-ccw notify hypercall
- Coding style
v2 -> v3:
- Added a patch exporting the virtio-ccw api and use it for the
  diagnose implementation.
- Better naming: We're dealing with virtio-ccw notifications only.
v1 -> v2:
- Move irqfd initialization from a module init function to kvm_init,
  eliminating the need for a second module for kvm/s390.
- Use kvm_io_device for s390 css devices.

Cornelia Huck (5):
  KVM: s390: Export virtio-ccw api.
  KVM: Initialize irqfd from kvm_init().
  KVM: Introduce KVM_VIRTIO_CCW_NOTIFY_BUS.
  KVM: ioeventfd for virtio-ccw devices.
  KVM: s390: Wire up ioeventfd.

Michael S. Tsirkin (1):
  virtio_ccw: pass a cookie value to kvm hypercall

 Documentation/virtual/kvm/api.txt       |  8 ++++++++
 arch/s390/include/uapi/asm/Kbuild       |  1 +
 arch/s390/include/uapi/asm/virtio-ccw.h | 21 +++++++++++++++++++++
 arch/s390/kvm/Kconfig                   |  1 +
 arch/s390/kvm/Makefile                  |  2 +-
 arch/s390/kvm/diag.c                    | 26 ++++++++++++++++++++++++++
 arch/s390/kvm/kvm-s390.c                |  1 +
 drivers/s390/kvm/virtio_ccw.c           | 16 +++++++++-------
 include/linux/kvm_host.h                | 14 ++++++++++++++
 include/uapi/linux/kvm.h                |  3 +++
 virt/kvm/eventfd.c                      | 24 +++++++++++++++---------
 virt/kvm/kvm_main.c                     |  6 ++++++
 12 files changed, 106 insertions(+), 17 deletions(-)
 create mode 100644 arch/s390/include/uapi/asm/virtio-ccw.h

-- 
1.7.12.4

[PATCH v5 1/6] virtio_ccw: pass a cookie value to kvm hypercall

[Cornelia Huck]

From: "Michael S. Tsirkin" <mst@redhat.com>

Lookups by channel/vq pair on host during virtio notifications might be
expensive.  Interpret hypercall return value as a cookie which host can
use to do device lookups for the next notification more efficiently.

[CH: Fix line > 80 chars]
Tested-by: Christian Borntraeger <borntraeger@de.ibm.com>
Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
---
 drivers/s390/kvm/virtio_ccw.c | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

diff --git a/drivers/s390/kvm/virtio_ccw.c b/drivers/s390/kvm/virtio_ccw.c
index 3217dfe..259a2fb 100644
--- a/drivers/s390/kvm/virtio_ccw.c
+++ b/drivers/s390/kvm/virtio_ccw.c
@@ -77,6 +77,7 @@ struct virtio_ccw_vq_info {
 	void *queue;
 	struct vq_info_block *info_block;
 	struct list_head node;
+	long cookie;
 };
 
 #define KVM_VIRTIO_CCW_RING_ALIGN 4096
@@ -145,15 +146,18 @@ static int ccw_io_helper(struct virtio_ccw_device *vcdev,
 }
 
 static inline long do_kvm_notify(struct subchannel_id schid,
-				 unsigned long queue_index)
+				 unsigned long queue_index,
+				 long cookie)
 {
 	register unsigned long __nr asm("1") = KVM_S390_VIRTIO_CCW_NOTIFY;
 	register struct subchannel_id __schid asm("2") = schid;
 	register unsigned long __index asm("3") = queue_index;
 	register long __rc asm("2");
+	register long __cookie asm("4") = cookie;
 
 	asm volatile ("diag 2,4,0x500\n"
-		      : "=d" (__rc) : "d" (__nr), "d" (__schid), "d" (__index)
+		      : "=d" (__rc) : "d" (__nr), "d" (__schid), "d" (__index),
+		      "d"(__cookie)
 		      : "memory", "cc");
 	return __rc;
 }
@@ -166,7 +170,8 @@ static void virtio_ccw_kvm_notify(struct virtqueue *vq)
 
 	vcdev = to_vc_device(info->vq->vdev);
 	ccw_device_get_schid(vcdev->cdev, &schid);
-	do_kvm_notify(schid, virtqueue_get_queue_index(vq));
+	info->cookie = do_kvm_notify(schid, virtqueue_get_queue_index(vq),
+				     info->cookie);
 }
 
 static int virtio_ccw_read_vq_conf(struct virtio_ccw_device *vcdev,
-- 
1.7.12.4

[PATCH v5 2/6] KVM: s390: Export virtio-ccw api.

[Cornelia Huck]

Export the virtio-ccw api in a header for usage by other code.

Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
---
 arch/s390/include/uapi/asm/Kbuild       |  1 +
 arch/s390/include/uapi/asm/virtio-ccw.h | 21 +++++++++++++++++++++
 drivers/s390/kvm/virtio_ccw.c           |  5 +----
 3 files changed, 23 insertions(+), 4 deletions(-)
 create mode 100644 arch/s390/include/uapi/asm/virtio-ccw.h

diff --git a/arch/s390/include/uapi/asm/Kbuild b/arch/s390/include/uapi/asm/Kbuild
index 7bf68ff..9ccd190 100644
--- a/arch/s390/include/uapi/asm/Kbuild
+++ b/arch/s390/include/uapi/asm/Kbuild
@@ -44,5 +44,6 @@ header-y += termios.h
 header-y += types.h
 header-y += ucontext.h
 header-y += unistd.h
+header-y += virtio-ccw.h
 header-y += vtoc.h
 header-y += zcrypt.h
diff --git a/arch/s390/include/uapi/asm/virtio-ccw.h b/arch/s390/include/uapi/asm/virtio-ccw.h
new file mode 100644
index 0000000..a9a4ebf
--- /dev/null
+++ b/arch/s390/include/uapi/asm/virtio-ccw.h
@@ -0,0 +1,21 @@
+/*
+ * Definitions for virtio-ccw devices.
+ *
+ * Copyright IBM Corp. 2013
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License (version 2 only)
+ * as published by the Free Software Foundation.
+ *
+ *  Author(s): Cornelia Huck <cornelia.huck@de.ibm.com>
+ */
+#ifndef __KVM_VIRTIO_CCW_H
+#define __KVM_VIRTIO_CCW_H
+
+/* Alignment of vring buffers. */
+#define KVM_VIRTIO_CCW_RING_ALIGN 4096
+
+/* Subcode for diagnose 500 (virtio hypercall). */
+#define KVM_S390_VIRTIO_CCW_NOTIFY 3
+
+#endif
diff --git a/drivers/s390/kvm/virtio_ccw.c b/drivers/s390/kvm/virtio_ccw.c
index 259a2fb..bf5ba3b 100644
--- a/drivers/s390/kvm/virtio_ccw.c
+++ b/drivers/s390/kvm/virtio_ccw.c
@@ -31,6 +31,7 @@
 #include <asm/irq.h>
 #include <asm/cio.h>
 #include <asm/ccwdev.h>
+#include <asm/virtio-ccw.h>
 
 /*
  * virtio related functions
@@ -80,10 +81,6 @@ struct virtio_ccw_vq_info {
 	long cookie;
 };
 
-#define KVM_VIRTIO_CCW_RING_ALIGN 4096
-
-#define KVM_S390_VIRTIO_CCW_NOTIFY 3
-
 #define CCW_CMD_SET_VQ 0x13
 #define CCW_CMD_VDEV_RESET 0x33
 #define CCW_CMD_SET_IND 0x43
-- 
1.7.12.4

[PATCH v5 3/6] KVM: Initialize irqfd from kvm_init().

[Cornelia Huck]

Currently, eventfd introduces module_init/module_exit functions
to initialize/cleanup the irqfd workqueue. This only works, however,
if no other module_init/module_exit functions are built into the
same module.

Let's just move the initialization and cleanup to kvm_init and kvm_exit.
This way, it is also clearer where kvm startup may fail.

Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
---
 include/linux/kvm_host.h | 13 +++++++++++++
 virt/kvm/eventfd.c       |  7 ++-----
 virt/kvm/kvm_main.c      |  6 ++++++
 3 files changed, 21 insertions(+), 5 deletions(-)

diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index 722cae7..3b768ef 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -423,6 +423,19 @@ void kvm_vcpu_uninit(struct kvm_vcpu *vcpu);
 int __must_check vcpu_load(struct kvm_vcpu *vcpu);
 void vcpu_put(struct kvm_vcpu *vcpu);
 
+#ifdef __KVM_HAVE_IOAPIC
+int kvm_irqfd_init(void);
+void kvm_irqfd_exit(void);
+#else
+static inline int kvm_irqfd_init(void)
+{
+	return 0;
+}
+
+static inline void kvm_irqfd_exit(void)
+{
+}
+#endif
 int kvm_init(void *opaque, unsigned vcpu_size, unsigned vcpu_align,
 		  struct module *module);
 void kvm_exit(void);
diff --git a/virt/kvm/eventfd.c b/virt/kvm/eventfd.c
index b6eea5c..f0ced1a 100644
--- a/virt/kvm/eventfd.c
+++ b/virt/kvm/eventfd.c
@@ -544,7 +544,7 @@ void kvm_irq_routing_update(struct kvm *kvm,
  * aggregated from all vm* instances. We need our own isolated single-thread
  * queue to prevent deadlock against flushing the normal work-queue.
  */
-static int __init irqfd_module_init(void)
+int kvm_irqfd_init(void)
 {
 	irqfd_cleanup_wq = create_singlethread_workqueue("kvm-irqfd-cleanup");
 	if (!irqfd_cleanup_wq)
@@ -553,13 +553,10 @@ static int __init irqfd_module_init(void)
 	return 0;
 }
 
-static void __exit irqfd_module_exit(void)
+void kvm_irqfd_exit(void)
 {
 	destroy_workqueue(irqfd_cleanup_wq);
 }
-
-module_init(irqfd_module_init);
-module_exit(irqfd_module_exit);
 #endif
 
 /*
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index adc68fe..7c188a3 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -2920,6 +2920,9 @@ int kvm_init(void *opaque, unsigned vcpu_size, unsigned vcpu_align,
 	int r;
 	int cpu;
 
+	r = kvm_irqfd_init();
+	if (r)
+		goto out_irqfd;
 	r = kvm_arch_init(opaque);
 	if (r)
 		goto out_fail;
@@ -3000,6 +3003,8 @@ out_free_0a:
 out_free_0:
 	kvm_arch_exit();
 out_fail:
+	kvm_irqfd_exit();
+out_irqfd:
 	return r;
 }
 EXPORT_SYMBOL_GPL(kvm_init);
@@ -3016,6 +3021,7 @@ void kvm_exit(void)
 	on_each_cpu(hardware_disable_nolock, NULL, 1);
 	kvm_arch_hardware_unsetup();
 	kvm_arch_exit();
+	kvm_irqfd_exit();
 	free_cpumask_var(cpus_hardware_enabled);
 }
 EXPORT_SYMBOL_GPL(kvm_exit);
-- 
1.7.12.4

[PATCH v5 4/6] KVM: Introduce KVM_VIRTIO_CCW_NOTIFY_BUS.

[Cornelia Huck]

Add a new bus type for virtio-ccw devices on s390.

Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
---
 include/linux/kvm_host.h | 1 +
 1 file changed, 1 insertion(+)

diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index 3b768ef..206247f 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -148,6 +148,7 @@ struct kvm_io_bus {
 enum kvm_bus {
 	KVM_MMIO_BUS,
 	KVM_PIO_BUS,
+	KVM_VIRTIO_CCW_NOTIFY_BUS,
 	KVM_NR_BUSES
 };
 
-- 
1.7.12.4

[PATCH v5 5/6] KVM: ioeventfd for virtio-ccw devices.

[Cornelia Huck]

Enhance KVM_IOEVENTFD with a new flag that allows to attach to virtio-ccw
devices on s390 via the KVM_VIRTIO_CCW_NOTIFY_BUS.

Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
---
 Documentation/virtual/kvm/api.txt |  8 ++++++++
 include/uapi/linux/kvm.h          |  3 +++
 virt/kvm/eventfd.c                | 17 +++++++++++++----
 3 files changed, 24 insertions(+), 4 deletions(-)

diff --git a/Documentation/virtual/kvm/api.txt b/Documentation/virtual/kvm/api.txt
index c2534c3..86232d6 100644
--- a/Documentation/virtual/kvm/api.txt
+++ b/Documentation/virtual/kvm/api.txt
@@ -1468,15 +1468,23 @@ struct kvm_ioeventfd {
 	__u8  pad[36];
 };
 
+For the special case of virtio-ccw devices on s390, the ioevent is matched
+to a subchannel/virtqueue tuple instead.
+
 The following flags are defined:
 
 #define KVM_IOEVENTFD_FLAG_DATAMATCH (1 << kvm_ioeventfd_flag_nr_datamatch)
 #define KVM_IOEVENTFD_FLAG_PIO       (1 << kvm_ioeventfd_flag_nr_pio)
 #define KVM_IOEVENTFD_FLAG_DEASSIGN  (1 << kvm_ioeventfd_flag_nr_deassign)
+#define KVM_IOEVENTFD_FLAG_VIRTIO_CCW_NOTIFY \
+	(1 << kvm_ioeventfd_flag_nr_virtio_ccw_notify)
 
 If datamatch flag is set, the event will be signaled only if the written value
 to the registered address is equal to datamatch in struct kvm_ioeventfd.
 
+For virtio-ccw devices, addr contains the subchannel id and datamatch the
+virtqueue index.
+
 
 4.60 KVM_DIRTY_TLB
 
diff --git a/include/uapi/linux/kvm.h b/include/uapi/linux/kvm.h
index 9a2db57..8f3e5ae 100644
--- a/include/uapi/linux/kvm.h
+++ b/include/uapi/linux/kvm.h
@@ -448,12 +448,15 @@ enum {
 	kvm_ioeventfd_flag_nr_datamatch,
 	kvm_ioeventfd_flag_nr_pio,
 	kvm_ioeventfd_flag_nr_deassign,
+	kvm_ioeventfd_flag_nr_virtio_ccw_notify,
 	kvm_ioeventfd_flag_nr_max,
 };
 
 #define KVM_IOEVENTFD_FLAG_DATAMATCH (1 << kvm_ioeventfd_flag_nr_datamatch)
 #define KVM_IOEVENTFD_FLAG_PIO       (1 << kvm_ioeventfd_flag_nr_pio)
 #define KVM_IOEVENTFD_FLAG_DEASSIGN  (1 << kvm_ioeventfd_flag_nr_deassign)
+#define KVM_IOEVENTFD_FLAG_VIRTIO_CCW_NOTIFY \
+	(1 << kvm_ioeventfd_flag_nr_virtio_ccw_notify)
 
 #define KVM_IOEVENTFD_VALID_FLAG_MASK  ((1 << kvm_ioeventfd_flag_nr_max) - 1)
 
diff --git a/virt/kvm/eventfd.c b/virt/kvm/eventfd.c
index f0ced1a..c9b6494 100644
--- a/virt/kvm/eventfd.c
+++ b/virt/kvm/eventfd.c
@@ -675,15 +675,24 @@ ioeventfd_check_collision(struct kvm *kvm, struct _ioeventfd *p)
 	return false;
 }
 
+static enum kvm_bus ioeventfd_bus_from_flags(__u32 flags)
+{
+	if (flags & KVM_IOEVENTFD_FLAG_PIO)
+		return KVM_PIO_BUS;
+	if (flags & KVM_IOEVENTFD_FLAG_VIRTIO_CCW_NOTIFY)
+		return KVM_VIRTIO_CCW_NOTIFY_BUS;
+	return KVM_MMIO_BUS;
+}
+
 static int
 kvm_assign_ioeventfd(struct kvm *kvm, struct kvm_ioeventfd *args)
 {
-	int                       pio = args->flags & KVM_IOEVENTFD_FLAG_PIO;
-	enum kvm_bus              bus_idx = pio ? KVM_PIO_BUS : KVM_MMIO_BUS;
+	enum kvm_bus              bus_idx;
 	struct _ioeventfd        *p;
 	struct eventfd_ctx       *eventfd;
 	int                       ret;
 
+	bus_idx = ioeventfd_bus_from_flags(args->flags);
 	/* must be natural-word sized */
 	switch (args->len) {
 	case 1:
@@ -758,12 +767,12 @@ fail:
 static int
 kvm_deassign_ioeventfd(struct kvm *kvm, struct kvm_ioeventfd *args)
 {
-	int                       pio = args->flags & KVM_IOEVENTFD_FLAG_PIO;
-	enum kvm_bus              bus_idx = pio ? KVM_PIO_BUS : KVM_MMIO_BUS;
+	enum kvm_bus              bus_idx;
 	struct _ioeventfd        *p, *tmp;
 	struct eventfd_ctx       *eventfd;
 	int                       ret = -ENOENT;
 
+	bus_idx = ioeventfd_bus_from_flags(args->flags);
 	eventfd = eventfd_ctx_fdget(args->fd);
 	if (IS_ERR(eventfd))
 		return PTR_ERR(eventfd);
-- 
1.7.12.4

[PATCH v5 6/6] KVM: s390: Wire up ioeventfd.

[Cornelia Huck]

Enable ioeventfd support on s390 and hook up diagnose 500 virtio-ccw
notifications.

Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
---
 arch/s390/kvm/Kconfig    |  1 +
 arch/s390/kvm/Makefile   |  2 +-
 arch/s390/kvm/diag.c     | 26 ++++++++++++++++++++++++++
 arch/s390/kvm/kvm-s390.c |  1 +
 4 files changed, 29 insertions(+), 1 deletion(-)

diff --git a/arch/s390/kvm/Kconfig b/arch/s390/kvm/Kconfig
index b58dd86..3c43e30 100644
--- a/arch/s390/kvm/Kconfig
+++ b/arch/s390/kvm/Kconfig
@@ -22,6 +22,7 @@ config KVM
 	select PREEMPT_NOTIFIERS
 	select ANON_INODES
 	select HAVE_KVM_CPU_RELAX_INTERCEPT
+	select HAVE_KVM_EVENTFD
 	---help---
 	  Support hosting paravirtualized guest machines using the SIE
 	  virtualization capability on the mainframe. This should work
diff --git a/arch/s390/kvm/Makefile b/arch/s390/kvm/Makefile
index 3975722..8fe9d65 100644
--- a/arch/s390/kvm/Makefile
+++ b/arch/s390/kvm/Makefile
@@ -6,7 +6,7 @@
 # it under the terms of the GNU General Public License (version 2 only)
 # as published by the Free Software Foundation.
 
-common-objs = $(addprefix ../../../virt/kvm/, kvm_main.o)
+common-objs = $(addprefix ../../../virt/kvm/, kvm_main.o eventfd.o)
 
 ccflags-y := -Ivirt/kvm -Iarch/s390/kvm
 
diff --git a/arch/s390/kvm/diag.c b/arch/s390/kvm/diag.c
index a390687..1c01a99 100644
--- a/arch/s390/kvm/diag.c
+++ b/arch/s390/kvm/diag.c
@@ -13,6 +13,7 @@
 
 #include <linux/kvm.h>
 #include <linux/kvm_host.h>
+#include <asm/virtio-ccw.h>
 #include "kvm-s390.h"
 #include "trace.h"
 #include "trace-s390.h"
@@ -104,6 +105,29 @@ static int __diag_ipl_functions(struct kvm_vcpu *vcpu)
 	return -EREMOTE;
 }
 
+static int __diag_virtio_hypercall(struct kvm_vcpu *vcpu)
+{
+	int ret, idx;
+
+	/* No virtio-ccw notification? Get out quickly. */
+	if (!vcpu->kvm->arch.css_support ||
+	    (vcpu->run->s.regs.gprs[1] != KVM_S390_VIRTIO_CCW_NOTIFY))
+		return -EOPNOTSUPP;
+
+	idx = srcu_read_lock(&vcpu->kvm->srcu);
+	/*
+	 * The layout is as follows:
+	 * - gpr 2 contains the subchannel id (passed as addr)
+	 * - gpr 3 contains the virtqueue index (passed as datamatch)
+	 */
+	ret = kvm_io_bus_write(vcpu->kvm, KVM_VIRTIO_CCW_NOTIFY_BUS,
+				vcpu->run->s.regs.gprs[2],
+				8, &vcpu->run->s.regs.gprs[3]);
+	srcu_read_unlock(&vcpu->kvm->srcu, idx);
+	/* kvm_io_bus_write returns -EOPNOTSUPP if it found no match. */
+	return ret < 0 ? ret : 0;
+}
+
 int kvm_s390_handle_diag(struct kvm_vcpu *vcpu)
 {
 	int code = (vcpu->arch.sie_block->ipb & 0xfff0000) >> 16;
@@ -118,6 +142,8 @@ int kvm_s390_handle_diag(struct kvm_vcpu *vcpu)
 		return __diag_time_slice_end_directed(vcpu);
 	case 0x308:
 		return __diag_ipl_functions(vcpu);
+	case 0x500:
+		return __diag_virtio_hypercall(vcpu);
 	default:
 		return -EOPNOTSUPP;
 	}
diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c
index 4377d18..c7671e5 100644
--- a/arch/s390/kvm/kvm-s390.c
+++ b/arch/s390/kvm/kvm-s390.c
@@ -142,6 +142,7 @@ int kvm_dev_ioctl_check_extension(long ext)
 	case KVM_CAP_ONE_REG:
 	case KVM_CAP_ENABLE_CAP:
 	case KVM_CAP_S390_CSS_SUPPORT:
+	case KVM_CAP_IOEVENTFD:
 		r = 1;
 		break;
 	case KVM_CAP_NR_VCPUS:
-- 
1.7.12.4

Re: [PATCH v5 0/6] kvm: Make ioeventfd usable on s390.

[Michael S. Tsirkin]

On Thu, Feb 28, 2013 at 12:33:15PM +0100, Cornelia Huck wrote:
> v5 of the ioeventfd patch set, this time with a proper return code
> from __diag_virtio_hypercall(), otherwise unchanged.
> 
> v4 -> v5:
> - Proper return code in __diag_virtio_hypercall()
> v3 -> v4:
> - Pass cookies in virtio-ccw notify hypercall
> - Coding style
> v2 -> v3:
> - Added a patch exporting the virtio-ccw api and use it for the
>   diagnose implementation.
> - Better naming: We're dealing with virtio-ccw notifications only.
> v1 -> v2:
> - Move irqfd initialization from a module init function to kvm_init,
>   eliminating the need for a second module for kvm/s390.
> - Use kvm_io_device for s390 css devices.

The patches look obviously correct to me.

I take no strong position on whether the userspace interface
and the addition of virtio-specific bits in s390 kvm are
acceptable, but looks ok to me too.

Acked-by: Michael S. Tsirkin <mst@redhat.com>

> Cornelia Huck (5):
>   KVM: s390: Export virtio-ccw api.
>   KVM: Initialize irqfd from kvm_init().
>   KVM: Introduce KVM_VIRTIO_CCW_NOTIFY_BUS.
>   KVM: ioeventfd for virtio-ccw devices.
>   KVM: s390: Wire up ioeventfd.
> 
> Michael S. Tsirkin (1):
>   virtio_ccw: pass a cookie value to kvm hypercall
> 
>  Documentation/virtual/kvm/api.txt       |  8 ++++++++
>  arch/s390/include/uapi/asm/Kbuild       |  1 +
>  arch/s390/include/uapi/asm/virtio-ccw.h | 21 +++++++++++++++++++++
>  arch/s390/kvm/Kconfig                   |  1 +
>  arch/s390/kvm/Makefile                  |  2 +-
>  arch/s390/kvm/diag.c                    | 26 ++++++++++++++++++++++++++
>  arch/s390/kvm/kvm-s390.c                |  1 +
>  drivers/s390/kvm/virtio_ccw.c           | 16 +++++++++-------
>  include/linux/kvm_host.h                | 14 ++++++++++++++
>  include/uapi/linux/kvm.h                |  3 +++
>  virt/kvm/eventfd.c                      | 24 +++++++++++++++---------
>  virt/kvm/kvm_main.c                     |  6 ++++++
>  12 files changed, 106 insertions(+), 17 deletions(-)
>  create mode 100644 arch/s390/include/uapi/asm/virtio-ccw.h
> 
> -- 
> 1.7.12.4

Re: [PATCH v5 6/6] KVM: s390: Wire up ioeventfd.

[Marcelo Tosatti]

On Thu, Feb 28, 2013 at 12:33:21PM +0100, Cornelia Huck wrote:
> Enable ioeventfd support on s390 and hook up diagnose 500 virtio-ccw
> notifications.
> 
> Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
> ---
>  arch/s390/kvm/Kconfig    |  1 +
>  arch/s390/kvm/Makefile   |  2 +-
>  arch/s390/kvm/diag.c     | 26 ++++++++++++++++++++++++++
>  arch/s390/kvm/kvm-s390.c |  1 +
>  4 files changed, 29 insertions(+), 1 deletion(-)
> 
> diff --git a/arch/s390/kvm/Kconfig b/arch/s390/kvm/Kconfig
> index b58dd86..3c43e30 100644
> --- a/arch/s390/kvm/Kconfig
> +++ b/arch/s390/kvm/Kconfig
> @@ -22,6 +22,7 @@ config KVM
>  	select PREEMPT_NOTIFIERS
>  	select ANON_INODES
>  	select HAVE_KVM_CPU_RELAX_INTERCEPT
> +	select HAVE_KVM_EVENTFD
>  	---help---
>  	  Support hosting paravirtualized guest machines using the SIE
>  	  virtualization capability on the mainframe. This should work
> diff --git a/arch/s390/kvm/Makefile b/arch/s390/kvm/Makefile
> index 3975722..8fe9d65 100644
> --- a/arch/s390/kvm/Makefile
> +++ b/arch/s390/kvm/Makefile
> @@ -6,7 +6,7 @@
>  # it under the terms of the GNU General Public License (version 2 only)
>  # as published by the Free Software Foundation.
>  
> -common-objs = $(addprefix ../../../virt/kvm/, kvm_main.o)
> +common-objs = $(addprefix ../../../virt/kvm/, kvm_main.o eventfd.o)
>  
>  ccflags-y := -Ivirt/kvm -Iarch/s390/kvm
>  
> diff --git a/arch/s390/kvm/diag.c b/arch/s390/kvm/diag.c
> index a390687..1c01a99 100644
> --- a/arch/s390/kvm/diag.c
> +++ b/arch/s390/kvm/diag.c
> @@ -13,6 +13,7 @@
>  
>  #include <linux/kvm.h>
>  #include <linux/kvm_host.h>
> +#include <asm/virtio-ccw.h>
>  #include "kvm-s390.h"
>  #include "trace.h"
>  #include "trace-s390.h"
> @@ -104,6 +105,29 @@ static int __diag_ipl_functions(struct kvm_vcpu *vcpu)
>  	return -EREMOTE;
>  }
>  
> +static int __diag_virtio_hypercall(struct kvm_vcpu *vcpu)
> +{
> +	int ret, idx;
> +
> +	/* No virtio-ccw notification? Get out quickly. */
> +	if (!vcpu->kvm->arch.css_support ||
> +	    (vcpu->run->s.regs.gprs[1] != KVM_S390_VIRTIO_CCW_NOTIFY))
> +		return -EOPNOTSUPP;
> +
> +	idx = srcu_read_lock(&vcpu->kvm->srcu);
> +	/*
> +	 * The layout is as follows:
> +	 * - gpr 2 contains the subchannel id (passed as addr)
> +	 * - gpr 3 contains the virtqueue index (passed as datamatch)
> +	 */
> +	ret = kvm_io_bus_write(vcpu->kvm, KVM_VIRTIO_CCW_NOTIFY_BUS,
> +				vcpu->run->s.regs.gprs[2],
> +				8, &vcpu->run->s.regs.gprs[3]);
> +	srcu_read_unlock(&vcpu->kvm->srcu, idx);
> +	/* kvm_io_bus_write returns -EOPNOTSUPP if it found no match. */
> +	return ret < 0 ? ret : 0;
> +}
> +

What about the cookie?

Re: [PATCH v5 6/6] KVM: s390: Wire up ioeventfd.

[Cornelia Huck]

On Mon, 4 Mar 2013 21:07:41 -0300
Marcelo Tosatti <mtosatti@redhat.com> wrote:

> On Thu, Feb 28, 2013 at 12:33:21PM +0100, Cornelia Huck wrote:
> > Enable ioeventfd support on s390 and hook up diagnose 500 virtio-ccw
> > notifications.
> > 
> > Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
> > ---
> >  arch/s390/kvm/Kconfig    |  1 +
> >  arch/s390/kvm/Makefile   |  2 +-
> >  arch/s390/kvm/diag.c     | 26 ++++++++++++++++++++++++++
> >  arch/s390/kvm/kvm-s390.c |  1 +
> >  4 files changed, 29 insertions(+), 1 deletion(-)
> > 
> > diff --git a/arch/s390/kvm/Kconfig b/arch/s390/kvm/Kconfig
> > index b58dd86..3c43e30 100644
> > --- a/arch/s390/kvm/Kconfig
> > +++ b/arch/s390/kvm/Kconfig
> > @@ -22,6 +22,7 @@ config KVM
> >  	select PREEMPT_NOTIFIERS
> >  	select ANON_INODES
> >  	select HAVE_KVM_CPU_RELAX_INTERCEPT
> > +	select HAVE_KVM_EVENTFD
> >  	---help---
> >  	  Support hosting paravirtualized guest machines using the SIE
> >  	  virtualization capability on the mainframe. This should work
> > diff --git a/arch/s390/kvm/Makefile b/arch/s390/kvm/Makefile
> > index 3975722..8fe9d65 100644
> > --- a/arch/s390/kvm/Makefile
> > +++ b/arch/s390/kvm/Makefile
> > @@ -6,7 +6,7 @@
> >  # it under the terms of the GNU General Public License (version 2 only)
> >  # as published by the Free Software Foundation.
> >  
> > -common-objs = $(addprefix ../../../virt/kvm/, kvm_main.o)
> > +common-objs = $(addprefix ../../../virt/kvm/, kvm_main.o eventfd.o)
> >  
> >  ccflags-y := -Ivirt/kvm -Iarch/s390/kvm
> >  
> > diff --git a/arch/s390/kvm/diag.c b/arch/s390/kvm/diag.c
> > index a390687..1c01a99 100644
> > --- a/arch/s390/kvm/diag.c
> > +++ b/arch/s390/kvm/diag.c
> > @@ -13,6 +13,7 @@
> >  
> >  #include <linux/kvm.h>
> >  #include <linux/kvm_host.h>
> > +#include <asm/virtio-ccw.h>
> >  #include "kvm-s390.h"
> >  #include "trace.h"
> >  #include "trace-s390.h"
> > @@ -104,6 +105,29 @@ static int __diag_ipl_functions(struct kvm_vcpu *vcpu)
> >  	return -EREMOTE;
> >  }
> >  
> > +static int __diag_virtio_hypercall(struct kvm_vcpu *vcpu)
> > +{
> > +	int ret, idx;
> > +
> > +	/* No virtio-ccw notification? Get out quickly. */
> > +	if (!vcpu->kvm->arch.css_support ||
> > +	    (vcpu->run->s.regs.gprs[1] != KVM_S390_VIRTIO_CCW_NOTIFY))
> > +		return -EOPNOTSUPP;
> > +
> > +	idx = srcu_read_lock(&vcpu->kvm->srcu);
> > +	/*
> > +	 * The layout is as follows:
> > +	 * - gpr 2 contains the subchannel id (passed as addr)
> > +	 * - gpr 3 contains the virtqueue index (passed as datamatch)
> > +	 */
> > +	ret = kvm_io_bus_write(vcpu->kvm, KVM_VIRTIO_CCW_NOTIFY_BUS,
> > +				vcpu->run->s.regs.gprs[2],
> > +				8, &vcpu->run->s.regs.gprs[3]);
> > +	srcu_read_unlock(&vcpu->kvm->srcu, idx);
> > +	/* kvm_io_bus_write returns -EOPNOTSUPP if it found no match. */
> > +	return ret < 0 ? ret : 0;
> > +}
> > +
> 
> What about the cookie?

This is the not-performance-optimized version which went through our
testing and which I'd like to see applied. I also have a not-yet-tested
patch that adds cookie support (see below), but I'd like to postpone
this to 3.10.

commit 8fce4a8a3478252f7ee2eb74d2732673f5911120
Author: Cornelia Huck <cornelia.huck@de.ibm.com>
Date:   Tue Feb 26 17:05:01 2013 +0100

    virtio-ccw: ioeventfd: use cookies
    
    Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>

diff --git a/arch/s390/kvm/diag.c b/arch/s390/kvm/diag.c
index 1c01a99..bd2a013 100644
--- a/arch/s390/kvm/diag.c
+++ b/arch/s390/kvm/diag.c
@@ -108,22 +108,29 @@ static int __diag_ipl_functions(struct kvm_vcpu *vcpu)
 static int __diag_virtio_hypercall(struct kvm_vcpu *vcpu)
 {
 	int ret, idx;
+	long cookie;
 
 	/* No virtio-ccw notification? Get out quickly. */
 	if (!vcpu->kvm->arch.css_support ||
 	    (vcpu->run->s.regs.gprs[1] != KVM_S390_VIRTIO_CCW_NOTIFY))
 		return -EOPNOTSUPP;
 
+	cookie = vcpu->run->s.regs.gprs[4];
 	idx = srcu_read_lock(&vcpu->kvm->srcu);
 	/*
 	 * The layout is as follows:
 	 * - gpr 2 contains the subchannel id (passed as addr)
 	 * - gpr 3 contains the virtqueue index (passed as datamatch)
+	 * - gpr 4 contains the index on the bus (optionally)
 	 */
-	ret = kvm_io_bus_write(vcpu->kvm, KVM_VIRTIO_CCW_NOTIFY_BUS,
-				vcpu->run->s.regs.gprs[2],
-				8, &vcpu->run->s.regs.gprs[3]);
+	ret = kvm_io_bus_write_cookie(vcpu->kvm, KVM_VIRTIO_CCW_NOTIFY_BUS,
+				      vcpu->run->s.regs.gprs[2],
+				      8, &vcpu->run->s.regs.gprs[3],
+				      &cookie);
 	srcu_read_unlock(&vcpu->kvm->srcu, idx);
+	if (!ret)
+		/* Return cookie in gpr 2. */
+		vcpu->run->s.regs.gprs[2] = cookie;
 	/* kvm_io_bus_write returns -EOPNOTSUPP if it found no match. */
 	return ret < 0 ? ret : 0;
 }
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index 206247f..2f65a3a 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -154,6 +154,8 @@ enum kvm_bus {
 
 int kvm_io_bus_write(struct kvm *kvm, enum kvm_bus bus_idx, gpa_t addr,
 		     int len, const void *val);
+int kvm_io_bus_write_cookie(struct kvm *kvm, enum kvm_bus bus_idx, gpa_t addr,
+			    int len, const void *val, long *cookie);
 int kvm_io_bus_read(struct kvm *kvm, enum kvm_bus bus_idx, gpa_t addr, int len,
 		    void *val);
 int kvm_io_bus_register_dev(struct kvm *kvm, enum kvm_bus bus_idx, gpa_t addr,
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index 7c188a3..b77ea5f 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -2710,6 +2710,51 @@ int kvm_io_bus_write(struct kvm *kvm, enum kvm_bus bus_idx, gpa_t addr,
 	return -EOPNOTSUPP;
 }
 
+/* kvm_io_bus_write_cookie - called under kvm->slots_lock */
+int kvm_io_bus_write_cookie(struct kvm *kvm, enum kvm_bus bus_idx, gpa_t addr,
+			    int len, const void *val, long *cookie)
+{
+	int idx, ret = -EOPNOTSUPP;
+	struct kvm_io_bus *bus;
+	struct kvm_io_range range;
+
+	range = (struct kvm_io_range) {
+		.addr = addr,
+		.len = len,
+	};
+
+	if (!cookie)
+		return -EINVAL;
+
+	bus = srcu_dereference(kvm->buses[bus_idx], &kvm->srcu);
+
+	/* First try the device referenced by *cookie. */
+	if (kvm_io_bus_sort_cmp(&range, &bus->range[*cookie]) == 0)
+		if (!kvm_iodevice_write(bus->range[*cookie].dev, addr, len,
+					val))
+			return 0;
+
+	/*
+	 * *cookie contained garbage; fall back to search and return the
+	 * correct value in *cookie.
+	 */
+	idx = kvm_io_bus_get_first_dev(bus, addr, len);
+	if (idx < 0)
+		goto out;
+
+	while (idx < bus->dev_count &&
+		kvm_io_bus_sort_cmp(&range, &bus->range[idx]) == 0) {
+		if (!kvm_iodevice_write(bus->range[idx].dev, addr, len, val)) {
+			ret = 0;
+			goto out;
+		}
+		idx++;
+	}
+out:
+	*cookie = idx;
+	return ret;
+}
+
 /* kvm_io_bus_read - called under kvm->slots_lock */
 int kvm_io_bus_read(struct kvm *kvm, enum kvm_bus bus_idx, gpa_t addr,
 		    int len, void *val)

Re: [PATCH v5 6/6] KVM: s390: Wire up ioeventfd.

[Michael S. Tsirkin]

On Tue, Mar 05, 2013 at 09:13:08AM +0100, Cornelia Huck wrote:
> On Mon, 4 Mar 2013 21:07:41 -0300
> Marcelo Tosatti <mtosatti@redhat.com> wrote:
> 
> > On Thu, Feb 28, 2013 at 12:33:21PM +0100, Cornelia Huck wrote:
> > > Enable ioeventfd support on s390 and hook up diagnose 500 virtio-ccw
> > > notifications.
> > > 
> > > Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
> > > ---
> > >  arch/s390/kvm/Kconfig    |  1 +
> > >  arch/s390/kvm/Makefile   |  2 +-
> > >  arch/s390/kvm/diag.c     | 26 ++++++++++++++++++++++++++
> > >  arch/s390/kvm/kvm-s390.c |  1 +
> > >  4 files changed, 29 insertions(+), 1 deletion(-)
> > > 
> > > diff --git a/arch/s390/kvm/Kconfig b/arch/s390/kvm/Kconfig
> > > index b58dd86..3c43e30 100644
> > > --- a/arch/s390/kvm/Kconfig
> > > +++ b/arch/s390/kvm/Kconfig
> > > @@ -22,6 +22,7 @@ config KVM
> > >  	select PREEMPT_NOTIFIERS
> > >  	select ANON_INODES
> > >  	select HAVE_KVM_CPU_RELAX_INTERCEPT
> > > +	select HAVE_KVM_EVENTFD
> > >  	---help---
> > >  	  Support hosting paravirtualized guest machines using the SIE
> > >  	  virtualization capability on the mainframe. This should work
> > > diff --git a/arch/s390/kvm/Makefile b/arch/s390/kvm/Makefile
> > > index 3975722..8fe9d65 100644
> > > --- a/arch/s390/kvm/Makefile
> > > +++ b/arch/s390/kvm/Makefile
> > > @@ -6,7 +6,7 @@
> > >  # it under the terms of the GNU General Public License (version 2 only)
> > >  # as published by the Free Software Foundation.
> > >  
> > > -common-objs = $(addprefix ../../../virt/kvm/, kvm_main.o)
> > > +common-objs = $(addprefix ../../../virt/kvm/, kvm_main.o eventfd.o)
> > >  
> > >  ccflags-y := -Ivirt/kvm -Iarch/s390/kvm
> > >  
> > > diff --git a/arch/s390/kvm/diag.c b/arch/s390/kvm/diag.c
> > > index a390687..1c01a99 100644
> > > --- a/arch/s390/kvm/diag.c
> > > +++ b/arch/s390/kvm/diag.c
> > > @@ -13,6 +13,7 @@
> > >  
> > >  #include <linux/kvm.h>
> > >  #include <linux/kvm_host.h>
> > > +#include <asm/virtio-ccw.h>
> > >  #include "kvm-s390.h"
> > >  #include "trace.h"
> > >  #include "trace-s390.h"
> > > @@ -104,6 +105,29 @@ static int __diag_ipl_functions(struct kvm_vcpu *vcpu)
> > >  	return -EREMOTE;
> > >  }
> > >  
> > > +static int __diag_virtio_hypercall(struct kvm_vcpu *vcpu)
> > > +{
> > > +	int ret, idx;
> > > +
> > > +	/* No virtio-ccw notification? Get out quickly. */
> > > +	if (!vcpu->kvm->arch.css_support ||
> > > +	    (vcpu->run->s.regs.gprs[1] != KVM_S390_VIRTIO_CCW_NOTIFY))
> > > +		return -EOPNOTSUPP;
> > > +
> > > +	idx = srcu_read_lock(&vcpu->kvm->srcu);
> > > +	/*
> > > +	 * The layout is as follows:
> > > +	 * - gpr 2 contains the subchannel id (passed as addr)
> > > +	 * - gpr 3 contains the virtqueue index (passed as datamatch)
> > > +	 */
> > > +	ret = kvm_io_bus_write(vcpu->kvm, KVM_VIRTIO_CCW_NOTIFY_BUS,
> > > +				vcpu->run->s.regs.gprs[2],
> > > +				8, &vcpu->run->s.regs.gprs[3]);
> > > +	srcu_read_unlock(&vcpu->kvm->srcu, idx);
> > > +	/* kvm_io_bus_write returns -EOPNOTSUPP if it found no match. */
> > > +	return ret < 0 ? ret : 0;
> > > +}
> > > +
> > 
> > What about the cookie?
> 
> This is the not-performance-optimized version which went through our
> testing and which I'd like to see applied. I also have a not-yet-tested
> patch that adds cookie support (see below), but I'd like to postpone
> this to 3.10.

I agree it's best to delay this optimization, focus on basic
functionality now. As guest with cookie support is
compatible with host without said support (all we do
really is reserve a gpr), there won't be issues adding it later.
Some comments on the draft below.


> commit 8fce4a8a3478252f7ee2eb74d2732673f5911120
> Author: Cornelia Huck <cornelia.huck@de.ibm.com>
> Date:   Tue Feb 26 17:05:01 2013 +0100
> 
>     virtio-ccw: ioeventfd: use cookies
>     
>     Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
> 
> diff --git a/arch/s390/kvm/diag.c b/arch/s390/kvm/diag.c
> index 1c01a99..bd2a013 100644
> --- a/arch/s390/kvm/diag.c
> +++ b/arch/s390/kvm/diag.c
> @@ -108,22 +108,29 @@ static int __diag_ipl_functions(struct kvm_vcpu *vcpu)
>  static int __diag_virtio_hypercall(struct kvm_vcpu *vcpu)
>  {
>  	int ret, idx;
> +	long cookie;
>  
>  	/* No virtio-ccw notification? Get out quickly. */
>  	if (!vcpu->kvm->arch.css_support ||
>  	    (vcpu->run->s.regs.gprs[1] != KVM_S390_VIRTIO_CCW_NOTIFY))
>  		return -EOPNOTSUPP;
>  
> +	cookie = vcpu->run->s.regs.gprs[4];

Hmm signed so this can become negative.

>  	idx = srcu_read_lock(&vcpu->kvm->srcu);
>  	/*
>  	 * The layout is as follows:
>  	 * - gpr 2 contains the subchannel id (passed as addr)
>  	 * - gpr 3 contains the virtqueue index (passed as datamatch)
> +	 * - gpr 4 contains the index on the bus (optionally)
>  	 */
> -	ret = kvm_io_bus_write(vcpu->kvm, KVM_VIRTIO_CCW_NOTIFY_BUS,
> -				vcpu->run->s.regs.gprs[2],
> -				8, &vcpu->run->s.regs.gprs[3]);
> +	ret = kvm_io_bus_write_cookie(vcpu->kvm, KVM_VIRTIO_CCW_NOTIFY_BUS,
> +				      vcpu->run->s.regs.gprs[2],
> +				      8, &vcpu->run->s.regs.gprs[3],
> +				      &cookie);
>  	srcu_read_unlock(&vcpu->kvm->srcu, idx);
> +	if (!ret)
> +		/* Return cookie in gpr 2. */
> +		vcpu->run->s.regs.gprs[2] = cookie;

We should also set gprs[2] to something sane if exiting to userspace
so avoid adding overhead for that case.
Probably # of devices + 1 or a negative value.
Are there 32 bit guests? If yes can they pass a negative value?

>  	/* kvm_io_bus_write returns -EOPNOTSUPP if it found no match. */
>  	return ret < 0 ? ret : 0;
>  }
> diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
> index 206247f..2f65a3a 100644
> --- a/include/linux/kvm_host.h
> +++ b/include/linux/kvm_host.h
> @@ -154,6 +154,8 @@ enum kvm_bus {
>  
>  int kvm_io_bus_write(struct kvm *kvm, enum kvm_bus bus_idx, gpa_t addr,
>  		     int len, const void *val);
> +int kvm_io_bus_write_cookie(struct kvm *kvm, enum kvm_bus bus_idx, gpa_t addr,
> +			    int len, const void *val, long *cookie);
>  int kvm_io_bus_read(struct kvm *kvm, enum kvm_bus bus_idx, gpa_t addr, int len,
>  		    void *val);
>  int kvm_io_bus_register_dev(struct kvm *kvm, enum kvm_bus bus_idx, gpa_t addr,
> diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
> index 7c188a3..b77ea5f 100644
> --- a/virt/kvm/kvm_main.c
> +++ b/virt/kvm/kvm_main.c
> @@ -2710,6 +2710,51 @@ int kvm_io_bus_write(struct kvm *kvm, enum kvm_bus bus_idx, gpa_t addr,
>  	return -EOPNOTSUPP;
>  }
>  
> +/* kvm_io_bus_write_cookie - called under kvm->slots_lock */
> +int kvm_io_bus_write_cookie(struct kvm *kvm, enum kvm_bus bus_idx, gpa_t addr,
> +			    int len, const void *val, long *cookie)
> +{
> +	int idx, ret = -EOPNOTSUPP;
> +	struct kvm_io_bus *bus;
> +	struct kvm_io_range range;
> +
> +	range = (struct kvm_io_range) {
> +		.addr = addr,
> +		.len = len,
> +	};
> +
> +	if (!cookie)
> +		return -EINVAL;
> +

Hmm does this ever trigger?

> +	bus = srcu_dereference(kvm->buses[bus_idx], &kvm->srcu);
> +
> +	/* First try the device referenced by *cookie. */
> +	if (kvm_io_bus_sort_cmp(&range, &bus->range[*cookie]) == 0)

We need to range-check the cookie.

> +		if (!kvm_iodevice_write(bus->range[*cookie].dev, addr, len,
> +					val))
> +			return 0;
> +
> +	/*
> +	 * *cookie contained garbage; fall back to search and return the
> +	 * correct value in *cookie.
> +	 */
> +	idx = kvm_io_bus_get_first_dev(bus, addr, len);
> +	if (idx < 0)
> +		goto out;
> +
> +	while (idx < bus->dev_count &&
> +		kvm_io_bus_sort_cmp(&range, &bus->range[idx]) == 0) {
> +		if (!kvm_iodevice_write(bus->range[idx].dev, addr, len, val)) {
> +			ret = 0;
> +			goto out;
> +		}
> +		idx++;
> +	}
> +out:
> +	*cookie = idx;
> +	return ret;
> +}
> +
>  /* kvm_io_bus_read - called under kvm->slots_lock */
>  int kvm_io_bus_read(struct kvm *kvm, enum kvm_bus bus_idx, gpa_t addr,
>  		    int len, void *val)

Re: [PATCH v5 6/6] KVM: s390: Wire up ioeventfd.

[Cornelia Huck]

On Tue, 5 Mar 2013 15:19:43 +0200
"Michael S. Tsirkin" <mst@redhat.com> wrote:

> On Tue, Mar 05, 2013 at 09:13:08AM +0100, Cornelia Huck wrote:
> > On Mon, 4 Mar 2013 21:07:41 -0300
> > Marcelo Tosatti <mtosatti@redhat.com> wrote:
> > 
> > > On Thu, Feb 28, 2013 at 12:33:21PM +0100, Cornelia Huck wrote:
> > > > Enable ioeventfd support on s390 and hook up diagnose 500 virtio-ccw
> > > > notifications.
> > > > 
> > > > Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
> > > > ---
> > > >  arch/s390/kvm/Kconfig    |  1 +
> > > >  arch/s390/kvm/Makefile   |  2 +-
> > > >  arch/s390/kvm/diag.c     | 26 ++++++++++++++++++++++++++
> > > >  arch/s390/kvm/kvm-s390.c |  1 +
> > > >  4 files changed, 29 insertions(+), 1 deletion(-)
> > > > 
> > > > diff --git a/arch/s390/kvm/Kconfig b/arch/s390/kvm/Kconfig
> > > > index b58dd86..3c43e30 100644
> > > > --- a/arch/s390/kvm/Kconfig
> > > > +++ b/arch/s390/kvm/Kconfig
> > > > @@ -22,6 +22,7 @@ config KVM
> > > >  	select PREEMPT_NOTIFIERS
> > > >  	select ANON_INODES
> > > >  	select HAVE_KVM_CPU_RELAX_INTERCEPT
> > > > +	select HAVE_KVM_EVENTFD
> > > >  	---help---
> > > >  	  Support hosting paravirtualized guest machines using the SIE
> > > >  	  virtualization capability on the mainframe. This should work
> > > > diff --git a/arch/s390/kvm/Makefile b/arch/s390/kvm/Makefile
> > > > index 3975722..8fe9d65 100644
> > > > --- a/arch/s390/kvm/Makefile
> > > > +++ b/arch/s390/kvm/Makefile
> > > > @@ -6,7 +6,7 @@
> > > >  # it under the terms of the GNU General Public License (version 2 only)
> > > >  # as published by the Free Software Foundation.
> > > >  
> > > > -common-objs = $(addprefix ../../../virt/kvm/, kvm_main.o)
> > > > +common-objs = $(addprefix ../../../virt/kvm/, kvm_main.o eventfd.o)
> > > >  
> > > >  ccflags-y := -Ivirt/kvm -Iarch/s390/kvm
> > > >  
> > > > diff --git a/arch/s390/kvm/diag.c b/arch/s390/kvm/diag.c
> > > > index a390687..1c01a99 100644
> > > > --- a/arch/s390/kvm/diag.c
> > > > +++ b/arch/s390/kvm/diag.c
> > > > @@ -13,6 +13,7 @@
> > > >  
> > > >  #include <linux/kvm.h>
> > > >  #include <linux/kvm_host.h>
> > > > +#include <asm/virtio-ccw.h>
> > > >  #include "kvm-s390.h"
> > > >  #include "trace.h"
> > > >  #include "trace-s390.h"
> > > > @@ -104,6 +105,29 @@ static int __diag_ipl_functions(struct kvm_vcpu *vcpu)
> > > >  	return -EREMOTE;
> > > >  }
> > > >  
> > > > +static int __diag_virtio_hypercall(struct kvm_vcpu *vcpu)
> > > > +{
> > > > +	int ret, idx;
> > > > +
> > > > +	/* No virtio-ccw notification? Get out quickly. */
> > > > +	if (!vcpu->kvm->arch.css_support ||
> > > > +	    (vcpu->run->s.regs.gprs[1] != KVM_S390_VIRTIO_CCW_NOTIFY))
> > > > +		return -EOPNOTSUPP;
> > > > +
> > > > +	idx = srcu_read_lock(&vcpu->kvm->srcu);
> > > > +	/*
> > > > +	 * The layout is as follows:
> > > > +	 * - gpr 2 contains the subchannel id (passed as addr)
> > > > +	 * - gpr 3 contains the virtqueue index (passed as datamatch)
> > > > +	 */
> > > > +	ret = kvm_io_bus_write(vcpu->kvm, KVM_VIRTIO_CCW_NOTIFY_BUS,
> > > > +				vcpu->run->s.regs.gprs[2],
> > > > +				8, &vcpu->run->s.regs.gprs[3]);
> > > > +	srcu_read_unlock(&vcpu->kvm->srcu, idx);
> > > > +	/* kvm_io_bus_write returns -EOPNOTSUPP if it found no match. */
> > > > +	return ret < 0 ? ret : 0;
> > > > +}
> > > > +
> > > 
> > > What about the cookie?
> > 
> > This is the not-performance-optimized version which went through our
> > testing and which I'd like to see applied. I also have a not-yet-tested
> > patch that adds cookie support (see below), but I'd like to postpone
> > this to 3.10.
> 
> I agree it's best to delay this optimization, focus on basic
> functionality now. As guest with cookie support is
> compatible with host without said support (all we do
> really is reserve a gpr), there won't be issues adding it later.
> Some comments on the draft below.
> 
> 
> > commit 8fce4a8a3478252f7ee2eb74d2732673f5911120
> > Author: Cornelia Huck <cornelia.huck@de.ibm.com>
> > Date:   Tue Feb 26 17:05:01 2013 +0100
> > 
> >     virtio-ccw: ioeventfd: use cookies
> >     
> >     Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
> > 
> > diff --git a/arch/s390/kvm/diag.c b/arch/s390/kvm/diag.c
> > index 1c01a99..bd2a013 100644
> > --- a/arch/s390/kvm/diag.c
> > +++ b/arch/s390/kvm/diag.c
> > @@ -108,22 +108,29 @@ static int __diag_ipl_functions(struct kvm_vcpu *vcpu)
> >  static int __diag_virtio_hypercall(struct kvm_vcpu *vcpu)
> >  {
> >  	int ret, idx;
> > +	long cookie;
> >  
> >  	/* No virtio-ccw notification? Get out quickly. */
> >  	if (!vcpu->kvm->arch.css_support ||
> >  	    (vcpu->run->s.regs.gprs[1] != KVM_S390_VIRTIO_CCW_NOTIFY))
> >  		return -EOPNOTSUPP;
> >  
> > +	cookie = vcpu->run->s.regs.gprs[4];
> 
> Hmm signed so this can become negative.
> 
> >  	idx = srcu_read_lock(&vcpu->kvm->srcu);
> >  	/*
> >  	 * The layout is as follows:
> >  	 * - gpr 2 contains the subchannel id (passed as addr)
> >  	 * - gpr 3 contains the virtqueue index (passed as datamatch)
> > +	 * - gpr 4 contains the index on the bus (optionally)
> >  	 */
> > -	ret = kvm_io_bus_write(vcpu->kvm, KVM_VIRTIO_CCW_NOTIFY_BUS,
> > -				vcpu->run->s.regs.gprs[2],
> > -				8, &vcpu->run->s.regs.gprs[3]);
> > +	ret = kvm_io_bus_write_cookie(vcpu->kvm, KVM_VIRTIO_CCW_NOTIFY_BUS,
> > +				      vcpu->run->s.regs.gprs[2],
> > +				      8, &vcpu->run->s.regs.gprs[3],
> > +				      &cookie);
> >  	srcu_read_unlock(&vcpu->kvm->srcu, idx);
> > +	if (!ret)
> > +		/* Return cookie in gpr 2. */
> > +		vcpu->run->s.regs.gprs[2] = cookie;
> 
> We should also set gprs[2] to something sane if exiting to userspace
> so avoid adding overhead for that case.
> Probably # of devices + 1 or a negative value.

Negative value would probably be the most obvious.

> Are there 32 bit guests? If yes can they pass a negative value?

No, we only support 64 bit guests.

> 
> >  	/* kvm_io_bus_write returns -EOPNOTSUPP if it found no match. */
> >  	return ret < 0 ? ret : 0;
> >  }
> > diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
> > index 206247f..2f65a3a 100644
> > --- a/include/linux/kvm_host.h
> > +++ b/include/linux/kvm_host.h
> > @@ -154,6 +154,8 @@ enum kvm_bus {
> >  
> >  int kvm_io_bus_write(struct kvm *kvm, enum kvm_bus bus_idx, gpa_t addr,
> >  		     int len, const void *val);
> > +int kvm_io_bus_write_cookie(struct kvm *kvm, enum kvm_bus bus_idx, gpa_t addr,
> > +			    int len, const void *val, long *cookie);
> >  int kvm_io_bus_read(struct kvm *kvm, enum kvm_bus bus_idx, gpa_t addr, int len,
> >  		    void *val);
> >  int kvm_io_bus_register_dev(struct kvm *kvm, enum kvm_bus bus_idx, gpa_t addr,
> > diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
> > index 7c188a3..b77ea5f 100644
> > --- a/virt/kvm/kvm_main.c
> > +++ b/virt/kvm/kvm_main.c
> > @@ -2710,6 +2710,51 @@ int kvm_io_bus_write(struct kvm *kvm, enum kvm_bus bus_idx, gpa_t addr,
> >  	return -EOPNOTSUPP;
> >  }
> >  
> > +/* kvm_io_bus_write_cookie - called under kvm->slots_lock */
> > +int kvm_io_bus_write_cookie(struct kvm *kvm, enum kvm_bus bus_idx, gpa_t addr,
> > +			    int len, const void *val, long *cookie)
> > +{
> > +	int idx, ret = -EOPNOTSUPP;
> > +	struct kvm_io_bus *bus;
> > +	struct kvm_io_range range;
> > +
> > +	range = (struct kvm_io_range) {
> > +		.addr = addr,
> > +		.len = len,
> > +	};
> > +
> > +	if (!cookie)
> > +		return -EINVAL;
> > +
> 
> Hmm does this ever trigger?

Probably not.

> 
> > +	bus = srcu_dereference(kvm->buses[bus_idx], &kvm->srcu);
> > +
> > +	/* First try the device referenced by *cookie. */
> > +	if (kvm_io_bus_sort_cmp(&range, &bus->range[*cookie]) == 0)
> 
> We need to range-check the cookie.

Yup.

> 
> > +		if (!kvm_iodevice_write(bus->range[*cookie].dev, addr, len,
> > +					val))
> > +			return 0;
> > +
> > +	/*
> > +	 * *cookie contained garbage; fall back to search and return the
> > +	 * correct value in *cookie.
> > +	 */
> > +	idx = kvm_io_bus_get_first_dev(bus, addr, len);
> > +	if (idx < 0)
> > +		goto out;
> > +
> > +	while (idx < bus->dev_count &&
> > +		kvm_io_bus_sort_cmp(&range, &bus->range[idx]) == 0) {
> > +		if (!kvm_iodevice_write(bus->range[idx].dev, addr, len, val)) {
> > +			ret = 0;
> > +			goto out;
> > +		}
> > +		idx++;
> > +	}
> > +out:
> > +	*cookie = idx;
> > +	return ret;
> > +}
> > +
> >  /* kvm_io_bus_read - called under kvm->slots_lock */
> >  int kvm_io_bus_read(struct kvm *kvm, enum kvm_bus bus_idx, gpa_t addr,
> >  		    int len, void *val)
> 

Re: [PATCH v5 0/6] kvm: Make ioeventfd usable on s390.

[Marcelo Tosatti]

On Thu, Feb 28, 2013 at 12:33:15PM +0100, Cornelia Huck wrote:
> v5 of the ioeventfd patch set, this time with a proper return code
> from __diag_virtio_hypercall(), otherwise unchanged.
> 
> v4 -> v5:
> - Proper return code in __diag_virtio_hypercall()
> v3 -> v4:
> - Pass cookies in virtio-ccw notify hypercall
> - Coding style
> v2 -> v3:
> - Added a patch exporting the virtio-ccw api and use it for the
>   diagnose implementation.
> - Better naming: We're dealing with virtio-ccw notifications only.
> v1 -> v2:
> - Move irqfd initialization from a module init function to kvm_init,
>   eliminating the need for a second module for kvm/s390.
> - Use kvm_io_device for s390 css devices.
> 
> Cornelia Huck (5):
>   KVM: s390: Export virtio-ccw api.
>   KVM: Initialize irqfd from kvm_init().
>   KVM: Introduce KVM_VIRTIO_CCW_NOTIFY_BUS.
>   KVM: ioeventfd for virtio-ccw devices.
>   KVM: s390: Wire up ioeventfd.
> 
> Michael S. Tsirkin (1):
>   virtio_ccw: pass a cookie value to kvm hypercall

Applied, thanks.