From 98fda169290b3b28c0f2db2b8f02290c13da50ef Mon Sep 17 00:00:00 2001
From: Radim Krčmář <rkrcmar@redhat.com>
Date: Wed, 4 Sep 2013 22:32:24 +0200
Subject: kvm: remove .done from struct kvm_async_pf
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

'.done' is used to mark the completion of 'async_pf_execute()', but
'cancel_work_sync()' returns true when the work was canceled, so we
use it instead.

Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 include/linux/kvm_host.h | 1 -
 1 file changed, 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index 0fbbc7aa02cb..749bdb12cd15 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -190,7 +190,6 @@ struct kvm_async_pf {
 	unsigned long addr;
 	struct kvm_arch_async_pf arch;
 	struct page *page;
-	bool done;
 };
 
 void kvm_clear_async_pf_completion_queue(struct kvm_vcpu *vcpu);
-- 
cgit v1.2.3-58-ga151


From 2f303b74a62fb74983c0a66e2df353be963c527c Mon Sep 17 00:00:00 2001
From: Paolo Bonzini <pbonzini@redhat.com>
Date: Wed, 25 Sep 2013 13:53:07 +0200
Subject: KVM: Convert kvm_lock back to non-raw spinlock

In commit e935b8372cf8 ("KVM: Convert kvm_lock to raw_spinlock"),
the kvm_lock was made a raw lock.  However, the kvm mmu_shrink()
function tries to grab the (non-raw) mmu_lock within the scope of
the raw locked kvm_lock being held.  This leads to the following:

BUG: sleeping function called from invalid context at kernel/rtmutex.c:659
in_atomic(): 1, irqs_disabled(): 0, pid: 55, name: kswapd0
Preemption disabled at:[<ffffffffa0376eac>] mmu_shrink+0x5c/0x1b0 [kvm]

Pid: 55, comm: kswapd0 Not tainted 3.4.34_preempt-rt
Call Trace:
 [<ffffffff8106f2ad>] __might_sleep+0xfd/0x160
 [<ffffffff817d8d64>] rt_spin_lock+0x24/0x50
 [<ffffffffa0376f3c>] mmu_shrink+0xec/0x1b0 [kvm]
 [<ffffffff8111455d>] shrink_slab+0x17d/0x3a0
 [<ffffffff81151f00>] ? mem_cgroup_iter+0x130/0x260
 [<ffffffff8111824a>] balance_pgdat+0x54a/0x730
 [<ffffffff8111fe47>] ? set_pgdat_percpu_threshold+0xa7/0xd0
 [<ffffffff811185bf>] kswapd+0x18f/0x490
 [<ffffffff81070961>] ? get_parent_ip+0x11/0x50
 [<ffffffff81061970>] ? __init_waitqueue_head+0x50/0x50
 [<ffffffff81118430>] ? balance_pgdat+0x730/0x730
 [<ffffffff81060d2b>] kthread+0xdb/0xe0
 [<ffffffff8106e122>] ? finish_task_switch+0x52/0x100
 [<ffffffff817e1e94>] kernel_thread_helper+0x4/0x10
 [<ffffffff81060c50>] ? __init_kthread_worker+0x

After the previous patch, kvm_lock need not be a raw spinlock anymore,
so change it back.

Reported-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Cc: kvm@vger.kernel.org
Cc: gleb@redhat.com
Cc: jan.kiszka@siemens.com
Reviewed-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 Documentation/virtual/kvm/locking.txt |  2 +-
 arch/x86/kvm/mmu.c                    |  5 ++---
 arch/x86/kvm/x86.c                    |  8 ++++----
 include/linux/kvm_host.h              |  2 +-
 virt/kvm/kvm_main.c                   | 18 +++++++++---------
 5 files changed, 17 insertions(+), 18 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/virtual/kvm/locking.txt b/Documentation/virtual/kvm/locking.txt
index ba9e1c2150c2..f8869410d40c 100644
--- a/Documentation/virtual/kvm/locking.txt
+++ b/Documentation/virtual/kvm/locking.txt
@@ -132,7 +132,7 @@ See the comments in spte_has_volatile_bits() and mmu_spte_update().
 ------------
 
 Name:		kvm_lock
-Type:		raw_spinlock
+Type:		spinlock_t
 Arch:		any
 Protects:	- vm_list
 
diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index dce0df8150df..cf95cfe050a6 100644
--- a/arch/x86/kvm/mmu.c
+++ b/arch/x86/kvm/mmu.c
@@ -4428,7 +4428,7 @@ mmu_shrink_scan(struct shrinker *shrink, struct shrink_control *sc)
 	int nr_to_scan = sc->nr_to_scan;
 	unsigned long freed = 0;
 
-	raw_spin_lock(&kvm_lock);
+	spin_lock(&kvm_lock);
 
 	list_for_each_entry(kvm, &vm_list, vm_list) {
 		int idx;
@@ -4478,9 +4478,8 @@ unlock:
 		break;
 	}
 
-	raw_spin_unlock(&kvm_lock);
+	spin_unlock(&kvm_lock);
 	return freed;
-
 }
 
 static unsigned long
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index e5ca72a5cdb6..187f824b1454 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -5263,7 +5263,7 @@ static int kvmclock_cpufreq_notifier(struct notifier_block *nb, unsigned long va
 
 	smp_call_function_single(freq->cpu, tsc_khz_changed, freq, 1);
 
-	raw_spin_lock(&kvm_lock);
+	spin_lock(&kvm_lock);
 	list_for_each_entry(kvm, &vm_list, vm_list) {
 		kvm_for_each_vcpu(i, vcpu, kvm) {
 			if (vcpu->cpu != freq->cpu)
@@ -5273,7 +5273,7 @@ static int kvmclock_cpufreq_notifier(struct notifier_block *nb, unsigned long va
 				send_ipi = 1;
 		}
 	}
-	raw_spin_unlock(&kvm_lock);
+	spin_unlock(&kvm_lock);
 
 	if (freq->old < freq->new && send_ipi) {
 		/*
@@ -5426,12 +5426,12 @@ static void pvclock_gtod_update_fn(struct work_struct *work)
 	struct kvm_vcpu *vcpu;
 	int i;
 
-	raw_spin_lock(&kvm_lock);
+	spin_lock(&kvm_lock);
 	list_for_each_entry(kvm, &vm_list, vm_list)
 		kvm_for_each_vcpu(i, vcpu, kvm)
 			set_bit(KVM_REQ_MASTERCLOCK_UPDATE, &vcpu->requests);
 	atomic_set(&kvm_guest_has_master_clock, 0);
-	raw_spin_unlock(&kvm_lock);
+	spin_unlock(&kvm_lock);
 }
 
 static DECLARE_WORK(pvclock_gtod_work, pvclock_gtod_update_fn);
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index 749bdb12cd15..7c961e1e9270 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -142,7 +142,7 @@ struct kvm;
 struct kvm_vcpu;
 extern struct kmem_cache *kvm_vcpu_cache;
 
-extern raw_spinlock_t kvm_lock;
+extern spinlock_t kvm_lock;
 extern struct list_head vm_list;
 
 struct kvm_io_range {
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index eb94343c2ed2..d469114aff09 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -70,7 +70,7 @@ MODULE_LICENSE("GPL");
  * 		kvm->lock --> kvm->slots_lock --> kvm->irq_lock
  */
 
-DEFINE_RAW_SPINLOCK(kvm_lock);
+DEFINE_SPINLOCK(kvm_lock);
 static DEFINE_RAW_SPINLOCK(kvm_count_lock);
 LIST_HEAD(vm_list);
 
@@ -491,9 +491,9 @@ static struct kvm *kvm_create_vm(unsigned long type)
 	if (r)
 		goto out_err;
 
-	raw_spin_lock(&kvm_lock);
+	spin_lock(&kvm_lock);
 	list_add(&kvm->vm_list, &vm_list);
-	raw_spin_unlock(&kvm_lock);
+	spin_unlock(&kvm_lock);
 
 	return kvm;
 
@@ -582,9 +582,9 @@ static void kvm_destroy_vm(struct kvm *kvm)
 	struct mm_struct *mm = kvm->mm;
 
 	kvm_arch_sync_events(kvm);
-	raw_spin_lock(&kvm_lock);
+	spin_lock(&kvm_lock);
 	list_del(&kvm->vm_list);
-	raw_spin_unlock(&kvm_lock);
+	spin_unlock(&kvm_lock);
 	kvm_free_irq_routing(kvm);
 	for (i = 0; i < KVM_NR_BUSES; i++)
 		kvm_io_bus_destroy(kvm->buses[i]);
@@ -3054,10 +3054,10 @@ static int vm_stat_get(void *_offset, u64 *val)
 	struct kvm *kvm;
 
 	*val = 0;
-	raw_spin_lock(&kvm_lock);
+	spin_lock(&kvm_lock);
 	list_for_each_entry(kvm, &vm_list, vm_list)
 		*val += *(u32 *)((void *)kvm + offset);
-	raw_spin_unlock(&kvm_lock);
+	spin_unlock(&kvm_lock);
 	return 0;
 }
 
@@ -3071,12 +3071,12 @@ static int vcpu_stat_get(void *_offset, u64 *val)
 	int i;
 
 	*val = 0;
-	raw_spin_lock(&kvm_lock);
+	spin_lock(&kvm_lock);
 	list_for_each_entry(kvm, &vm_list, vm_list)
 		kvm_for_each_vcpu(i, vcpu, kvm)
 			*val += *(u32 *)((void *)vcpu + offset);
 
-	raw_spin_unlock(&kvm_lock);
+	spin_unlock(&kvm_lock);
 	return 0;
 }
 
-- 
cgit v1.2.3-58-ga151


From 6d9d41e57440e32a3400f37aa05ef7a1a09ced64 Mon Sep 17 00:00:00 2001
From: Christoffer Dall <christoffer.dall@linaro.org>
Date: Wed, 2 Oct 2013 14:22:28 -0700
Subject: KVM: Move gfn_to_index to x86 specific code

The gfn_to_index function relies on huge page defines which either may
not make sense on systems that don't support huge pages or are defined
in an unconvenient way for other architectures.  Since this is
x86-specific, move the function to arch/x86/include/asm/kvm_host.h.

Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
Signed-off-by: Gleb Natapov <gleb@redhat.com>
---
 arch/x86/include/asm/kvm_host.h | 7 +++++++
 include/linux/kvm_host.h        | 7 -------
 2 files changed, 7 insertions(+), 7 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index 8dd143a65d60..5cbf3166257c 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -79,6 +79,13 @@
 #define KVM_HPAGE_MASK(x)	(~(KVM_HPAGE_SIZE(x) - 1))
 #define KVM_PAGES_PER_HPAGE(x)	(KVM_HPAGE_SIZE(x) / PAGE_SIZE)
 
+static inline gfn_t gfn_to_index(gfn_t gfn, gfn_t base_gfn, int level)
+{
+	/* KVM_HPAGE_GFN_SHIFT(PT_PAGE_TABLE_LEVEL) must be 0. */
+	return (gfn >> KVM_HPAGE_GFN_SHIFT(level)) -
+		(base_gfn >> KVM_HPAGE_GFN_SHIFT(level));
+}
+
 #define SELECTOR_TI_MASK (1 << 2)
 #define SELECTOR_RPL_MASK 0x03
 
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index 7c961e1e9270..f6dccde755f6 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -841,13 +841,6 @@ static inline int memslot_id(struct kvm *kvm, gfn_t gfn)
 	return gfn_to_memslot(kvm, gfn)->id;
 }
 
-static inline gfn_t gfn_to_index(gfn_t gfn, gfn_t base_gfn, int level)
-{
-	/* KVM_HPAGE_GFN_SHIFT(PT_PAGE_TABLE_LEVEL) must be 0. */
-	return (gfn >> KVM_HPAGE_GFN_SHIFT(level)) -
-		(base_gfn >> KVM_HPAGE_GFN_SHIFT(level));
-}
-
 static inline gfn_t
 hva_to_gfn_memslot(unsigned long hva, struct kvm_memory_slot *slot)
 {
-- 
cgit v1.2.3-58-ga151


From f2e106692d5189303997ad7b96de8d8123aa5613 Mon Sep 17 00:00:00 2001
From: chai wen <chaiw.fnst@cn.fujitsu.com>
Date: Mon, 14 Oct 2013 22:22:33 +0800
Subject: KVM: Drop FOLL_GET in GUP when doing async page fault

Page pinning is not mandatory in kvm async page fault processing since
after async page fault event is delivered to a guest it accesses page once
again and does its own GUP.  Drop the FOLL_GET flag in GUP in async_pf
code, and do some simplifying in check/clear processing.

Suggested-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Gu zheng <guz.fnst@cn.fujitsu.com>
Signed-off-by: chai wen <chaiw.fnst@cn.fujitsu.com>
Signed-off-by: Gleb Natapov <gleb@redhat.com>
---
 arch/x86/kvm/x86.c         |  4 ++--
 include/linux/kvm_host.h   |  2 +-
 include/trace/events/kvm.h | 10 ++++------
 virt/kvm/async_pf.c        | 17 +++++------------
 4 files changed, 12 insertions(+), 21 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index c951c71dc80b..edf2a07df3a3 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -7298,7 +7298,7 @@ void kvm_arch_async_page_ready(struct kvm_vcpu *vcpu, struct kvm_async_pf *work)
 	int r;
 
 	if ((vcpu->arch.mmu.direct_map != work->arch.direct_map) ||
-	      is_error_page(work->page))
+	      work->wakeup_all)
 		return;
 
 	r = kvm_mmu_reload(vcpu);
@@ -7408,7 +7408,7 @@ void kvm_arch_async_page_present(struct kvm_vcpu *vcpu,
 	struct x86_exception fault;
 
 	trace_kvm_async_pf_ready(work->arch.token, work->gva);
-	if (is_error_page(work->page))
+	if (work->wakeup_all)
 		work->arch.token = ~0; /* broadcast wakeup */
 	else
 		kvm_del_async_pf_gfn(vcpu, work->arch.gfn);
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index f6dccde755f6..c9d4236ab442 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -189,7 +189,7 @@ struct kvm_async_pf {
 	gva_t gva;
 	unsigned long addr;
 	struct kvm_arch_async_pf arch;
-	struct page *page;
+	bool   wakeup_all;
 };
 
 void kvm_clear_async_pf_completion_queue(struct kvm_vcpu *vcpu);
diff --git a/include/trace/events/kvm.h b/include/trace/events/kvm.h
index 7005d1109ec9..131a0bda7aec 100644
--- a/include/trace/events/kvm.h
+++ b/include/trace/events/kvm.h
@@ -296,23 +296,21 @@ DEFINE_EVENT(kvm_async_pf_nopresent_ready, kvm_async_pf_ready,
 
 TRACE_EVENT(
 	kvm_async_pf_completed,
-	TP_PROTO(unsigned long address, struct page *page, u64 gva),
-	TP_ARGS(address, page, gva),
+	TP_PROTO(unsigned long address, u64 gva),
+	TP_ARGS(address, gva),
 
 	TP_STRUCT__entry(
 		__field(unsigned long, address)
-		__field(pfn_t, pfn)
 		__field(u64, gva)
 		),
 
 	TP_fast_assign(
 		__entry->address = address;
-		__entry->pfn = page ? page_to_pfn(page) : 0;
 		__entry->gva = gva;
 		),
 
-	TP_printk("gva %#llx address %#lx pfn %#llx",  __entry->gva,
-		  __entry->address, __entry->pfn)
+	TP_printk("gva %#llx address %#lx",  __entry->gva,
+		  __entry->address)
 );
 
 #endif
diff --git a/virt/kvm/async_pf.c b/virt/kvm/async_pf.c
index b197950ac4d5..8631d9c14320 100644
--- a/virt/kvm/async_pf.c
+++ b/virt/kvm/async_pf.c
@@ -56,7 +56,6 @@ void kvm_async_pf_vcpu_init(struct kvm_vcpu *vcpu)
 
 static void async_pf_execute(struct work_struct *work)
 {
-	struct page *page = NULL;
 	struct kvm_async_pf *apf =
 		container_of(work, struct kvm_async_pf, work);
 	struct mm_struct *mm = apf->mm;
@@ -68,13 +67,12 @@ static void async_pf_execute(struct work_struct *work)
 
 	use_mm(mm);
 	down_read(&mm->mmap_sem);
-	get_user_pages(current, mm, addr, 1, 1, 0, &page, NULL);
+	get_user_pages(current, mm, addr, 1, 1, 0, NULL, NULL);
 	up_read(&mm->mmap_sem);
 	unuse_mm(mm);
 
 	spin_lock(&vcpu->async_pf.lock);
 	list_add_tail(&apf->link, &vcpu->async_pf.done);
-	apf->page = page;
 	spin_unlock(&vcpu->async_pf.lock);
 
 	/*
@@ -82,7 +80,7 @@ static void async_pf_execute(struct work_struct *work)
 	 * this point
 	 */
 
-	trace_kvm_async_pf_completed(addr, page, gva);
+	trace_kvm_async_pf_completed(addr, gva);
 
 	if (waitqueue_active(&vcpu->wq))
 		wake_up_interruptible(&vcpu->wq);
@@ -112,8 +110,6 @@ void kvm_clear_async_pf_completion_queue(struct kvm_vcpu *vcpu)
 			list_entry(vcpu->async_pf.done.next,
 				   typeof(*work), link);
 		list_del(&work->link);
-		if (!is_error_page(work->page))
-			kvm_release_page_clean(work->page);
 		kmem_cache_free(async_pf_cache, work);
 	}
 	spin_unlock(&vcpu->async_pf.lock);
@@ -133,14 +129,11 @@ void kvm_check_async_pf_completion(struct kvm_vcpu *vcpu)
 		list_del(&work->link);
 		spin_unlock(&vcpu->async_pf.lock);
 
-		if (work->page)
-			kvm_arch_async_page_ready(vcpu, work);
+		kvm_arch_async_page_ready(vcpu, work);
 		kvm_arch_async_page_present(vcpu, work);
 
 		list_del(&work->queue);
 		vcpu->async_pf.queued--;
-		if (!is_error_page(work->page))
-			kvm_release_page_clean(work->page);
 		kmem_cache_free(async_pf_cache, work);
 	}
 }
@@ -163,7 +156,7 @@ int kvm_setup_async_pf(struct kvm_vcpu *vcpu, gva_t gva, gfn_t gfn,
 	if (!work)
 		return 0;
 
-	work->page = NULL;
+	work->wakeup_all = false;
 	work->vcpu = vcpu;
 	work->gva = gva;
 	work->addr = gfn_to_hva(vcpu->kvm, gfn);
@@ -203,7 +196,7 @@ int kvm_async_pf_wakeup_all(struct kvm_vcpu *vcpu)
 	if (!work)
 		return -ENOMEM;
 
-	work->page = KVM_ERR_PTR_BAD_PAGE;
+	work->wakeup_all = true;
 	INIT_LIST_HEAD(&work->queue); /* for list_del to work */
 
 	spin_lock(&vcpu->async_pf.lock);
-- 
cgit v1.2.3-58-ga151


From 5587027ce9d59a57aecaa190be1c8e560aaff45d Mon Sep 17 00:00:00 2001
From: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Date: Mon, 7 Oct 2013 22:18:00 +0530
Subject: kvm: Add struct kvm arg to memslot APIs

We will use that in the later patch to find the kvm ops handler

Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Alexander Graf <agraf@suse.de>
---
 arch/arm/kvm/arm.c                 |  5 +++--
 arch/ia64/kvm/kvm-ia64.c           |  5 +++--
 arch/mips/kvm/kvm_mips.c           |  5 +++--
 arch/powerpc/include/asm/kvm_ppc.h |  6 ++++--
 arch/powerpc/kvm/book3s.c          |  4 ++--
 arch/powerpc/kvm/booke.c           |  4 ++--
 arch/powerpc/kvm/powerpc.c         |  9 +++++----
 arch/s390/kvm/kvm-s390.c           |  5 +++--
 arch/x86/kvm/x86.c                 |  5 +++--
 include/linux/kvm_host.h           |  5 +++--
 virt/kvm/kvm_main.c                | 12 ++++++------
 11 files changed, 37 insertions(+), 28 deletions(-)

(limited to 'include/linux')

diff --git a/arch/arm/kvm/arm.c b/arch/arm/kvm/arm.c
index cc5adb9349ef..e312e4a53f8d 100644
--- a/arch/arm/kvm/arm.c
+++ b/arch/arm/kvm/arm.c
@@ -152,12 +152,13 @@ int kvm_arch_vcpu_fault(struct kvm_vcpu *vcpu, struct vm_fault *vmf)
 	return VM_FAULT_SIGBUS;
 }
 
-void kvm_arch_free_memslot(struct kvm_memory_slot *free,
+void kvm_arch_free_memslot(struct kvm *kvm, struct kvm_memory_slot *free,
 			   struct kvm_memory_slot *dont)
 {
 }
 
-int kvm_arch_create_memslot(struct kvm_memory_slot *slot, unsigned long npages)
+int kvm_arch_create_memslot(struct kvm *kvm, struct kvm_memory_slot *slot,
+			    unsigned long npages)
 {
 	return 0;
 }
diff --git a/arch/ia64/kvm/kvm-ia64.c b/arch/ia64/kvm/kvm-ia64.c
index bdfd8789b376..985bf80c622e 100644
--- a/arch/ia64/kvm/kvm-ia64.c
+++ b/arch/ia64/kvm/kvm-ia64.c
@@ -1550,12 +1550,13 @@ int kvm_arch_vcpu_fault(struct kvm_vcpu *vcpu, struct vm_fault *vmf)
 	return VM_FAULT_SIGBUS;
 }
 
-void kvm_arch_free_memslot(struct kvm_memory_slot *free,
+void kvm_arch_free_memslot(struct kvm *kvm, struct kvm_memory_slot *free,
 			   struct kvm_memory_slot *dont)
 {
 }
 
-int kvm_arch_create_memslot(struct kvm_memory_slot *slot, unsigned long npages)
+int kvm_arch_create_memslot(struct kvm *kvm, struct kvm_memory_slot *slot,
+			    unsigned long npages)
 {
 	return 0;
 }
diff --git a/arch/mips/kvm/kvm_mips.c b/arch/mips/kvm/kvm_mips.c
index a7b044536de4..73b34827826c 100644
--- a/arch/mips/kvm/kvm_mips.c
+++ b/arch/mips/kvm/kvm_mips.c
@@ -198,12 +198,13 @@ kvm_arch_dev_ioctl(struct file *filp, unsigned int ioctl, unsigned long arg)
 	return -ENOIOCTLCMD;
 }
 
-void kvm_arch_free_memslot(struct kvm_memory_slot *free,
+void kvm_arch_free_memslot(struct kvm *kvm, struct kvm_memory_slot *free,
 			   struct kvm_memory_slot *dont)
 {
 }
 
-int kvm_arch_create_memslot(struct kvm_memory_slot *slot, unsigned long npages)
+int kvm_arch_create_memslot(struct kvm *kvm, struct kvm_memory_slot *slot,
+			    unsigned long npages)
 {
 	return 0;
 }
diff --git a/arch/powerpc/include/asm/kvm_ppc.h b/arch/powerpc/include/asm/kvm_ppc.h
index c13f15db476c..20f461637090 100644
--- a/arch/powerpc/include/asm/kvm_ppc.h
+++ b/arch/powerpc/include/asm/kvm_ppc.h
@@ -134,9 +134,11 @@ extern struct page *kvm_alloc_hpt(unsigned long nr_pages);
 extern void kvm_release_hpt(struct page *page, unsigned long nr_pages);
 extern int kvmppc_core_init_vm(struct kvm *kvm);
 extern void kvmppc_core_destroy_vm(struct kvm *kvm);
-extern void kvmppc_core_free_memslot(struct kvm_memory_slot *free,
+extern void kvmppc_core_free_memslot(struct kvm *kvm,
+				     struct kvm_memory_slot *free,
 				     struct kvm_memory_slot *dont);
-extern int kvmppc_core_create_memslot(struct kvm_memory_slot *slot,
+extern int kvmppc_core_create_memslot(struct kvm *kvm,
+				      struct kvm_memory_slot *slot,
 				      unsigned long npages);
 extern int kvmppc_core_prepare_memory_region(struct kvm *kvm,
 				struct kvm_memory_slot *memslot,
diff --git a/arch/powerpc/kvm/book3s.c b/arch/powerpc/kvm/book3s.c
index 39d2994f9d27..130fe1d75bac 100644
--- a/arch/powerpc/kvm/book3s.c
+++ b/arch/powerpc/kvm/book3s.c
@@ -761,13 +761,13 @@ int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm, struct kvm_dirty_log *log)
 	return kvmppc_ops->get_dirty_log(kvm, log);
 }
 
-void kvmppc_core_free_memslot(struct kvm_memory_slot *free,
+void kvmppc_core_free_memslot(struct kvm *kvm, struct kvm_memory_slot *free,
 			      struct kvm_memory_slot *dont)
 {
 	kvmppc_ops->free_memslot(free, dont);
 }
 
-int kvmppc_core_create_memslot(struct kvm_memory_slot *slot,
+int kvmppc_core_create_memslot(struct kvm *kvm, struct kvm_memory_slot *slot,
 			       unsigned long npages)
 {
 	return kvmppc_ops->create_memslot(slot, npages);
diff --git a/arch/powerpc/kvm/booke.c b/arch/powerpc/kvm/booke.c
index 1769354d0bf8..cb2d986a3382 100644
--- a/arch/powerpc/kvm/booke.c
+++ b/arch/powerpc/kvm/booke.c
@@ -1662,12 +1662,12 @@ int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm, struct kvm_dirty_log *log)
 	return -ENOTSUPP;
 }
 
-void kvmppc_core_free_memslot(struct kvm_memory_slot *free,
+void kvmppc_core_free_memslot(struct kvm *kvm, struct kvm_memory_slot *free,
 			      struct kvm_memory_slot *dont)
 {
 }
 
-int kvmppc_core_create_memslot(struct kvm_memory_slot *slot,
+int kvmppc_core_create_memslot(struct kvm *kvm, struct kvm_memory_slot *slot,
 			       unsigned long npages)
 {
 	return 0;
diff --git a/arch/powerpc/kvm/powerpc.c b/arch/powerpc/kvm/powerpc.c
index 9aaa07efa4b6..b103d747934a 100644
--- a/arch/powerpc/kvm/powerpc.c
+++ b/arch/powerpc/kvm/powerpc.c
@@ -417,15 +417,16 @@ long kvm_arch_dev_ioctl(struct file *filp,
 	return -EINVAL;
 }
 
-void kvm_arch_free_memslot(struct kvm_memory_slot *free,
+void kvm_arch_free_memslot(struct kvm *kvm, struct kvm_memory_slot *free,
 			   struct kvm_memory_slot *dont)
 {
-	kvmppc_core_free_memslot(free, dont);
+	kvmppc_core_free_memslot(kvm, free, dont);
 }
 
-int kvm_arch_create_memslot(struct kvm_memory_slot *slot, unsigned long npages)
+int kvm_arch_create_memslot(struct kvm *kvm, struct kvm_memory_slot *slot,
+			    unsigned long npages)
 {
-	return kvmppc_core_create_memslot(slot, npages);
+	return kvmppc_core_create_memslot(kvm, slot, npages);
 }
 
 void kvm_arch_memslots_updated(struct kvm *kvm)
diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c
index 1e4e7b97337a..bedda67cc222 100644
--- a/arch/s390/kvm/kvm-s390.c
+++ b/arch/s390/kvm/kvm-s390.c
@@ -1089,12 +1089,13 @@ int kvm_arch_vcpu_fault(struct kvm_vcpu *vcpu, struct vm_fault *vmf)
 	return VM_FAULT_SIGBUS;
 }
 
-void kvm_arch_free_memslot(struct kvm_memory_slot *free,
+void kvm_arch_free_memslot(struct kvm *kvm, struct kvm_memory_slot *free,
 			   struct kvm_memory_slot *dont)
 {
 }
 
-int kvm_arch_create_memslot(struct kvm_memory_slot *slot, unsigned long npages)
+int kvm_arch_create_memslot(struct kvm *kvm, struct kvm_memory_slot *slot,
+			    unsigned long npages)
 {
 	return 0;
 }
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index edf2a07df3a3..666526a55c46 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -7080,7 +7080,7 @@ void kvm_arch_destroy_vm(struct kvm *kvm)
 	kfree(rcu_dereference_check(kvm->arch.apic_map, 1));
 }
 
-void kvm_arch_free_memslot(struct kvm_memory_slot *free,
+void kvm_arch_free_memslot(struct kvm *kvm, struct kvm_memory_slot *free,
 			   struct kvm_memory_slot *dont)
 {
 	int i;
@@ -7101,7 +7101,8 @@ void kvm_arch_free_memslot(struct kvm_memory_slot *free,
 	}
 }
 
-int kvm_arch_create_memslot(struct kvm_memory_slot *slot, unsigned long npages)
+int kvm_arch_create_memslot(struct kvm *kvm, struct kvm_memory_slot *slot,
+			    unsigned long npages)
 {
 	int i;
 
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index c9d4236ab442..8b0107dc2067 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -507,9 +507,10 @@ int kvm_set_memory_region(struct kvm *kvm,
 			  struct kvm_userspace_memory_region *mem);
 int __kvm_set_memory_region(struct kvm *kvm,
 			    struct kvm_userspace_memory_region *mem);
-void kvm_arch_free_memslot(struct kvm_memory_slot *free,
+void kvm_arch_free_memslot(struct kvm *kvm, struct kvm_memory_slot *free,
 			   struct kvm_memory_slot *dont);
-int kvm_arch_create_memslot(struct kvm_memory_slot *slot, unsigned long npages);
+int kvm_arch_create_memslot(struct kvm *kvm, struct kvm_memory_slot *slot,
+			    unsigned long npages);
 void kvm_arch_memslots_updated(struct kvm *kvm);
 int kvm_arch_prepare_memory_region(struct kvm *kvm,
 				struct kvm_memory_slot *memslot,
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index 94c6e3f6f244..0932c3b64155 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -542,13 +542,13 @@ static void kvm_destroy_dirty_bitmap(struct kvm_memory_slot *memslot)
 /*
  * Free any memory in @free but not in @dont.
  */
-static void kvm_free_physmem_slot(struct kvm_memory_slot *free,
+static void kvm_free_physmem_slot(struct kvm *kvm, struct kvm_memory_slot *free,
 				  struct kvm_memory_slot *dont)
 {
 	if (!dont || free->dirty_bitmap != dont->dirty_bitmap)
 		kvm_destroy_dirty_bitmap(free);
 
-	kvm_arch_free_memslot(free, dont);
+	kvm_arch_free_memslot(kvm, free, dont);
 
 	free->npages = 0;
 }
@@ -559,7 +559,7 @@ void kvm_free_physmem(struct kvm *kvm)
 	struct kvm_memory_slot *memslot;
 
 	kvm_for_each_memslot(memslot, slots)
-		kvm_free_physmem_slot(memslot, NULL);
+		kvm_free_physmem_slot(kvm, memslot, NULL);
 
 	kfree(kvm->memslots);
 }
@@ -823,7 +823,7 @@ int __kvm_set_memory_region(struct kvm *kvm,
 	if (change == KVM_MR_CREATE) {
 		new.userspace_addr = mem->userspace_addr;
 
-		if (kvm_arch_create_memslot(&new, npages))
+		if (kvm_arch_create_memslot(kvm, &new, npages))
 			goto out_free;
 	}
 
@@ -899,7 +899,7 @@ int __kvm_set_memory_region(struct kvm *kvm,
 
 	kvm_arch_commit_memory_region(kvm, mem, &old, change);
 
-	kvm_free_physmem_slot(&old, &new);
+	kvm_free_physmem_slot(kvm, &old, &new);
 	kfree(old_memslots);
 
 	return 0;
@@ -907,7 +907,7 @@ int __kvm_set_memory_region(struct kvm *kvm,
 out_slots:
 	kfree(slots);
 out_free:
-	kvm_free_physmem_slot(&new, &old);
+	kvm_free_physmem_slot(kvm, &new, &old);
 out:
 	return r;
 }
-- 
cgit v1.2.3-58-ga151


From ec53500fae421e07c5d035918ca454a429732ef4 Mon Sep 17 00:00:00 2001
From: Alex Williamson <alex.williamson@redhat.com>
Date: Wed, 30 Oct 2013 11:02:17 -0600
Subject: kvm: Add VFIO device

So far we've succeeded at making KVM and VFIO mostly unaware of each
other, but areas are cropping up where a connection beyond eventfds
and irqfds needs to be made.  This patch introduces a KVM-VFIO device
that is meant to be a gateway for such interaction.  The user creates
the device and can add and remove VFIO groups to it via file
descriptors.  When a group is added, KVM verifies the group is valid
and gets a reference to it via the VFIO external user interface.

Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 Documentation/virtual/kvm/devices/vfio.txt |  22 +++
 arch/x86/kvm/Kconfig                       |   1 +
 arch/x86/kvm/Makefile                      |   2 +-
 include/linux/kvm_host.h                   |   1 +
 include/uapi/linux/kvm.h                   |   4 +
 virt/kvm/Kconfig                           |   3 +
 virt/kvm/kvm_main.c                        |   5 +
 virt/kvm/vfio.c                            | 220 +++++++++++++++++++++++++++++
 8 files changed, 257 insertions(+), 1 deletion(-)
 create mode 100644 Documentation/virtual/kvm/devices/vfio.txt
 create mode 100644 virt/kvm/vfio.c

(limited to 'include/linux')

diff --git a/Documentation/virtual/kvm/devices/vfio.txt b/Documentation/virtual/kvm/devices/vfio.txt
new file mode 100644
index 000000000000..ef51740c67ca
--- /dev/null
+++ b/Documentation/virtual/kvm/devices/vfio.txt
@@ -0,0 +1,22 @@
+VFIO virtual device
+===================
+
+Device types supported:
+  KVM_DEV_TYPE_VFIO
+
+Only one VFIO instance may be created per VM.  The created device
+tracks VFIO groups in use by the VM and features of those groups
+important to the correctness and acceleration of the VM.  As groups
+are enabled and disabled for use by the VM, KVM should be updated
+about their presence.  When registered with KVM, a reference to the
+VFIO-group is held by KVM.
+
+Groups:
+  KVM_DEV_VFIO_GROUP
+
+KVM_DEV_VFIO_GROUP attributes:
+  KVM_DEV_VFIO_GROUP_ADD: Add a VFIO group to VFIO-KVM device tracking
+  KVM_DEV_VFIO_GROUP_DEL: Remove a VFIO group from VFIO-KVM device tracking
+
+For each, kvm_device_attr.addr points to an int32_t file descriptor
+for the VFIO group.
diff --git a/arch/x86/kvm/Kconfig b/arch/x86/kvm/Kconfig
index a47a3e54b964..b89c5db2b832 100644
--- a/arch/x86/kvm/Kconfig
+++ b/arch/x86/kvm/Kconfig
@@ -38,6 +38,7 @@ config KVM
 	select PERF_EVENTS
 	select HAVE_KVM_MSI
 	select HAVE_KVM_CPU_RELAX_INTERCEPT
+	select KVM_VFIO
 	---help---
 	  Support hosting fully virtualized guest machines using hardware
 	  virtualization extensions.  You will need a fairly recent
diff --git a/arch/x86/kvm/Makefile b/arch/x86/kvm/Makefile
index bf4fb04d0112..25d22b2d6509 100644
--- a/arch/x86/kvm/Makefile
+++ b/arch/x86/kvm/Makefile
@@ -9,7 +9,7 @@ KVM := ../../../virt/kvm
 
 kvm-y			+= $(KVM)/kvm_main.o $(KVM)/ioapic.o \
 				$(KVM)/coalesced_mmio.o $(KVM)/irq_comm.o \
-				$(KVM)/eventfd.o $(KVM)/irqchip.o
+				$(KVM)/eventfd.o $(KVM)/irqchip.o $(KVM)/vfio.o
 kvm-$(CONFIG_KVM_DEVICE_ASSIGNMENT)	+= $(KVM)/assigned-dev.o $(KVM)/iommu.o
 kvm-$(CONFIG_KVM_ASYNC_PF)	+= $(KVM)/async_pf.o
 
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index c9d4236ab442..7beddbd38ac7 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -1058,6 +1058,7 @@ struct kvm_device *kvm_device_from_filp(struct file *filp);
 
 extern struct kvm_device_ops kvm_mpic_ops;
 extern struct kvm_device_ops kvm_xics_ops;
+extern struct kvm_device_ops kvm_vfio_ops;
 
 #ifdef CONFIG_HAVE_KVM_CPU_RELAX_INTERCEPT
 
diff --git a/include/uapi/linux/kvm.h b/include/uapi/linux/kvm.h
index 32c60b98ddf8..509cfbfe9658 100644
--- a/include/uapi/linux/kvm.h
+++ b/include/uapi/linux/kvm.h
@@ -845,6 +845,10 @@ struct kvm_device_attr {
 #define KVM_DEV_TYPE_FSL_MPIC_20	1
 #define KVM_DEV_TYPE_FSL_MPIC_42	2
 #define KVM_DEV_TYPE_XICS		3
+#define KVM_DEV_TYPE_VFIO		4
+#define  KVM_DEV_VFIO_GROUP			1
+#define   KVM_DEV_VFIO_GROUP_ADD			1
+#define   KVM_DEV_VFIO_GROUP_DEL			2
 
 /*
  * ioctls for VM fds
diff --git a/virt/kvm/Kconfig b/virt/kvm/Kconfig
index 779262f59e25..fbe1a48bd629 100644
--- a/virt/kvm/Kconfig
+++ b/virt/kvm/Kconfig
@@ -27,3 +27,6 @@ config HAVE_KVM_MSI
 
 config HAVE_KVM_CPU_RELAX_INTERCEPT
        bool
+
+config KVM_VFIO
+       bool
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index 9ca014da134d..82c4047aa0e3 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -2270,6 +2270,11 @@ static int kvm_ioctl_create_device(struct kvm *kvm,
 	case KVM_DEV_TYPE_XICS:
 		ops = &kvm_xics_ops;
 		break;
+#endif
+#ifdef CONFIG_KVM_VFIO
+	case KVM_DEV_TYPE_VFIO:
+		ops = &kvm_vfio_ops;
+		break;
 #endif
 	default:
 		return -ENODEV;
diff --git a/virt/kvm/vfio.c b/virt/kvm/vfio.c
new file mode 100644
index 000000000000..597c258245ea
--- /dev/null
+++ b/virt/kvm/vfio.c
@@ -0,0 +1,220 @@
+/*
+ * VFIO-KVM bridge pseudo device
+ *
+ * Copyright (C) 2013 Red Hat, Inc.  All rights reserved.
+ *     Author: Alex Williamson <alex.williamson@redhat.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/errno.h>
+#include <linux/file.h>
+#include <linux/kvm_host.h>
+#include <linux/list.h>
+#include <linux/module.h>
+#include <linux/mutex.h>
+#include <linux/slab.h>
+#include <linux/uaccess.h>
+#include <linux/vfio.h>
+
+struct kvm_vfio_group {
+	struct list_head node;
+	struct vfio_group *vfio_group;
+};
+
+struct kvm_vfio {
+	struct list_head group_list;
+	struct mutex lock;
+};
+
+static struct vfio_group *kvm_vfio_group_get_external_user(struct file *filep)
+{
+	struct vfio_group *vfio_group;
+	struct vfio_group *(*fn)(struct file *);
+
+	fn = symbol_get(vfio_group_get_external_user);
+	if (!fn)
+		return ERR_PTR(-EINVAL);
+
+	vfio_group = fn(filep);
+
+	symbol_put(vfio_group_get_external_user);
+
+	return vfio_group;
+}
+
+static void kvm_vfio_group_put_external_user(struct vfio_group *vfio_group)
+{
+	void (*fn)(struct vfio_group *);
+
+	fn = symbol_get(vfio_group_put_external_user);
+	if (!fn)
+		return;
+
+	fn(vfio_group);
+
+	symbol_put(vfio_group_put_external_user);
+}
+
+static int kvm_vfio_set_group(struct kvm_device *dev, long attr, u64 arg)
+{
+	struct kvm_vfio *kv = dev->private;
+	struct vfio_group *vfio_group;
+	struct kvm_vfio_group *kvg;
+	void __user *argp = (void __user *)arg;
+	struct fd f;
+	int32_t fd;
+	int ret;
+
+	switch (attr) {
+	case KVM_DEV_VFIO_GROUP_ADD:
+		if (get_user(fd, (int32_t __user *)argp))
+			return -EFAULT;
+
+		f = fdget(fd);
+		if (!f.file)
+			return -EBADF;
+
+		vfio_group = kvm_vfio_group_get_external_user(f.file);
+		fdput(f);
+
+		if (IS_ERR(vfio_group))
+			return PTR_ERR(vfio_group);
+
+		mutex_lock(&kv->lock);
+
+		list_for_each_entry(kvg, &kv->group_list, node) {
+			if (kvg->vfio_group == vfio_group) {
+				mutex_unlock(&kv->lock);
+				kvm_vfio_group_put_external_user(vfio_group);
+				return -EEXIST;
+			}
+		}
+
+		kvg = kzalloc(sizeof(*kvg), GFP_KERNEL);
+		if (!kvg) {
+			mutex_unlock(&kv->lock);
+			kvm_vfio_group_put_external_user(vfio_group);
+			return -ENOMEM;
+		}
+
+		list_add_tail(&kvg->node, &kv->group_list);
+		kvg->vfio_group = vfio_group;
+
+		mutex_unlock(&kv->lock);
+
+		return 0;
+
+	case KVM_DEV_VFIO_GROUP_DEL:
+		if (get_user(fd, (int32_t __user *)argp))
+			return -EFAULT;
+
+		f = fdget(fd);
+		if (!f.file)
+			return -EBADF;
+
+		vfio_group = kvm_vfio_group_get_external_user(f.file);
+		fdput(f);
+
+		if (IS_ERR(vfio_group))
+			return PTR_ERR(vfio_group);
+
+		ret = -ENOENT;
+
+		mutex_lock(&kv->lock);
+
+		list_for_each_entry(kvg, &kv->group_list, node) {
+			if (kvg->vfio_group != vfio_group)
+				continue;
+
+			list_del(&kvg->node);
+			kvm_vfio_group_put_external_user(kvg->vfio_group);
+			kfree(kvg);
+			ret = 0;
+			break;
+		}
+
+		mutex_unlock(&kv->lock);
+
+		kvm_vfio_group_put_external_user(vfio_group);
+
+		return ret;
+	}
+
+	return -ENXIO;
+}
+
+static int kvm_vfio_set_attr(struct kvm_device *dev,
+			     struct kvm_device_attr *attr)
+{
+	switch (attr->group) {
+	case KVM_DEV_VFIO_GROUP:
+		return kvm_vfio_set_group(dev, attr->attr, attr->addr);
+	}
+
+	return -ENXIO;
+}
+
+static int kvm_vfio_has_attr(struct kvm_device *dev,
+			     struct kvm_device_attr *attr)
+{
+	switch (attr->group) {
+	case KVM_DEV_VFIO_GROUP:
+		switch (attr->attr) {
+		case KVM_DEV_VFIO_GROUP_ADD:
+		case KVM_DEV_VFIO_GROUP_DEL:
+			return 0;
+		}
+
+		break;
+	}
+
+	return -ENXIO;
+}
+
+static void kvm_vfio_destroy(struct kvm_device *dev)
+{
+	struct kvm_vfio *kv = dev->private;
+	struct kvm_vfio_group *kvg, *tmp;
+
+	list_for_each_entry_safe(kvg, tmp, &kv->group_list, node) {
+		kvm_vfio_group_put_external_user(kvg->vfio_group);
+		list_del(&kvg->node);
+		kfree(kvg);
+	}
+
+	kfree(kv);
+	kfree(dev); /* alloc by kvm_ioctl_create_device, free by .destroy */
+}
+
+static int kvm_vfio_create(struct kvm_device *dev, u32 type)
+{
+	struct kvm_device *tmp;
+	struct kvm_vfio *kv;
+
+	/* Only one VFIO "device" per VM */
+	list_for_each_entry(tmp, &dev->kvm->devices, vm_node)
+		if (tmp->ops == &kvm_vfio_ops)
+			return -EBUSY;
+
+	kv = kzalloc(sizeof(*kv), GFP_KERNEL);
+	if (!kv)
+		return -ENOMEM;
+
+	INIT_LIST_HEAD(&kv->group_list);
+	mutex_init(&kv->lock);
+
+	dev->private = kv;
+
+	return 0;
+}
+
+struct kvm_device_ops kvm_vfio_ops = {
+	.name = "kvm-vfio",
+	.create = kvm_vfio_create,
+	.destroy = kvm_vfio_destroy,
+	.set_attr = kvm_vfio_set_attr,
+	.has_attr = kvm_vfio_has_attr,
+};
-- 
cgit v1.2.3-58-ga151


From d96eb2c6f480769bff32054e78b964860dae4d56 Mon Sep 17 00:00:00 2001
From: Alex Williamson <alex.williamson@redhat.com>
Date: Wed, 30 Oct 2013 11:02:23 -0600
Subject: kvm/x86: Convert iommu_flags to iommu_noncoherent

Default to operating in coherent mode.  This simplifies the logic when
we switch to a model of registering and unregistering noncoherent I/O
with KVM.

Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 arch/ia64/include/asm/kvm_host.h |  2 +-
 arch/x86/include/asm/kvm_host.h  |  2 +-
 arch/x86/kvm/vmx.c               |  2 +-
 arch/x86/kvm/x86.c               |  2 +-
 include/linux/kvm_host.h         |  3 ---
 virt/kvm/iommu.c                 | 16 ++++++++--------
 6 files changed, 12 insertions(+), 15 deletions(-)

(limited to 'include/linux')

diff --git a/arch/ia64/include/asm/kvm_host.h b/arch/ia64/include/asm/kvm_host.h
index 95a3ff93777c..db95f570705f 100644
--- a/arch/ia64/include/asm/kvm_host.h
+++ b/arch/ia64/include/asm/kvm_host.h
@@ -476,7 +476,7 @@ struct kvm_arch {
 
 	struct list_head assigned_dev_head;
 	struct iommu_domain *iommu_domain;
-	int iommu_flags;
+	bool iommu_noncoherent;
 
 	unsigned long irq_sources_bitmap;
 	unsigned long irq_states[KVM_IOAPIC_NUM_PINS];
diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index 5cbf3166257c..91b35e4005d3 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -564,7 +564,7 @@ struct kvm_arch {
 
 	struct list_head assigned_dev_head;
 	struct iommu_domain *iommu_domain;
-	int iommu_flags;
+	bool iommu_noncoherent;
 	struct kvm_pic *vpic;
 	struct kvm_ioapic *vioapic;
 	struct kvm_pit *vpit;
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 6850b0f1d52b..727a5e980c43 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -7446,7 +7446,7 @@ static u64 vmx_get_mt_mask(struct kvm_vcpu *vcpu, gfn_t gfn, bool is_mmio)
 	if (is_mmio)
 		ret = MTRR_TYPE_UNCACHABLE << VMX_EPT_MT_EPTE_SHIFT;
 	else if (vcpu->kvm->arch.iommu_domain &&
-		!(vcpu->kvm->arch.iommu_flags & KVM_IOMMU_CACHE_COHERENCY))
+		 vcpu->kvm->arch.iommu_noncoherent)
 		ret = kvm_get_guest_memory_type(vcpu, gfn) <<
 		      VMX_EPT_MT_EPTE_SHIFT;
 	else
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 15f9540a2b1f..92ad83e5b132 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -2719,7 +2719,7 @@ static void wbinvd_ipi(void *garbage)
 static bool need_emulate_wbinvd(struct kvm_vcpu *vcpu)
 {
 	return vcpu->kvm->arch.iommu_domain &&
-		!(vcpu->kvm->arch.iommu_flags & KVM_IOMMU_CACHE_COHERENCY);
+	       vcpu->kvm->arch.iommu_noncoherent;
 }
 
 void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index 7beddbd38ac7..ed64880e4915 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -746,9 +746,6 @@ void kvm_unregister_irq_ack_notifier(struct kvm *kvm,
 int kvm_request_irq_source_id(struct kvm *kvm);
 void kvm_free_irq_source_id(struct kvm *kvm, int irq_source_id);
 
-/* For vcpu->arch.iommu_flags */
-#define KVM_IOMMU_CACHE_COHERENCY	0x1
-
 #ifdef CONFIG_KVM_DEVICE_ASSIGNMENT
 int kvm_iommu_map_pages(struct kvm *kvm, struct kvm_memory_slot *slot);
 void kvm_iommu_unmap_pages(struct kvm *kvm, struct kvm_memory_slot *slot);
diff --git a/virt/kvm/iommu.c b/virt/kvm/iommu.c
index a3b14109049b..d32d156a423a 100644
--- a/virt/kvm/iommu.c
+++ b/virt/kvm/iommu.c
@@ -79,7 +79,7 @@ int kvm_iommu_map_pages(struct kvm *kvm, struct kvm_memory_slot *slot)
 	flags = IOMMU_READ;
 	if (!(slot->flags & KVM_MEM_READONLY))
 		flags |= IOMMU_WRITE;
-	if (kvm->arch.iommu_flags & KVM_IOMMU_CACHE_COHERENCY)
+	if (!kvm->arch.iommu_noncoherent)
 		flags |= IOMMU_CACHE;
 
 
@@ -158,7 +158,8 @@ int kvm_assign_device(struct kvm *kvm,
 {
 	struct pci_dev *pdev = NULL;
 	struct iommu_domain *domain = kvm->arch.iommu_domain;
-	int r, last_flags;
+	int r;
+	bool noncoherent;
 
 	/* check if iommu exists and in use */
 	if (!domain)
@@ -174,15 +175,13 @@ int kvm_assign_device(struct kvm *kvm,
 		return r;
 	}
 
-	last_flags = kvm->arch.iommu_flags;
-	if (iommu_domain_has_cap(kvm->arch.iommu_domain,
-				 IOMMU_CAP_CACHE_COHERENCY))
-		kvm->arch.iommu_flags |= KVM_IOMMU_CACHE_COHERENCY;
+	noncoherent = !iommu_domain_has_cap(kvm->arch.iommu_domain,
+					    IOMMU_CAP_CACHE_COHERENCY);
 
 	/* Check if need to update IOMMU page table for guest memory */
-	if ((last_flags ^ kvm->arch.iommu_flags) ==
-			KVM_IOMMU_CACHE_COHERENCY) {
+	if (noncoherent != kvm->arch.iommu_noncoherent) {
 		kvm_iommu_unmap_memslots(kvm);
+		kvm->arch.iommu_noncoherent = noncoherent;
 		r = kvm_iommu_map_memslots(kvm);
 		if (r)
 			goto out_unmap;
@@ -342,6 +341,7 @@ int kvm_iommu_unmap_guest(struct kvm *kvm)
 	mutex_lock(&kvm->slots_lock);
 	kvm_iommu_unmap_memslots(kvm);
 	kvm->arch.iommu_domain = NULL;
+	kvm->arch.iommu_noncoherent = false;
 	mutex_unlock(&kvm->slots_lock);
 
 	iommu_domain_free(domain);
-- 
cgit v1.2.3-58-ga151


From e0f0bbc527f6e9c0261f1d16b2a0b47612b7f235 Mon Sep 17 00:00:00 2001
From: Alex Williamson <alex.williamson@redhat.com>
Date: Wed, 30 Oct 2013 11:02:30 -0600
Subject: kvm: Create non-coherent DMA registeration

We currently use some ad-hoc arch variables tied to legacy KVM device
assignment to manage emulation of instructions that depend on whether
non-coherent DMA is present.  Create an interface for this, adapting
legacy KVM device assignment and adding VFIO via the KVM-VFIO device.
For now we assume that non-coherent DMA is possible any time we have a
VFIO group.  Eventually an interface can be developed as part of the
VFIO external user interface to query the coherency of a group.

Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 arch/x86/include/asm/kvm_host.h |  2 ++
 arch/x86/kvm/vmx.c              |  3 +--
 arch/x86/kvm/x86.c              | 22 +++++++++++++++++++--
 include/linux/kvm_host.h        | 19 ++++++++++++++++++
 virt/kvm/iommu.c                |  6 ++++++
 virt/kvm/vfio.c                 | 44 +++++++++++++++++++++++++++++++++++++++++
 6 files changed, 92 insertions(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index 91b35e4005d3..de388c55e7ec 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -565,6 +565,8 @@ struct kvm_arch {
 	struct list_head assigned_dev_head;
 	struct iommu_domain *iommu_domain;
 	bool iommu_noncoherent;
+#define __KVM_HAVE_ARCH_NONCOHERENT_DMA
+	atomic_t noncoherent_dma_count;
 	struct kvm_pic *vpic;
 	struct kvm_ioapic *vioapic;
 	struct kvm_pit *vpit;
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 727a5e980c43..fabf7421ec18 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -7445,8 +7445,7 @@ static u64 vmx_get_mt_mask(struct kvm_vcpu *vcpu, gfn_t gfn, bool is_mmio)
 	 */
 	if (is_mmio)
 		ret = MTRR_TYPE_UNCACHABLE << VMX_EPT_MT_EPTE_SHIFT;
-	else if (vcpu->kvm->arch.iommu_domain &&
-		 vcpu->kvm->arch.iommu_noncoherent)
+	else if (kvm_arch_has_noncoherent_dma(vcpu->kvm))
 		ret = kvm_get_guest_memory_type(vcpu, gfn) <<
 		      VMX_EPT_MT_EPTE_SHIFT;
 	else
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 92ad83e5b132..ec35d09937da 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -2718,8 +2718,7 @@ static void wbinvd_ipi(void *garbage)
 
 static bool need_emulate_wbinvd(struct kvm_vcpu *vcpu)
 {
-	return vcpu->kvm->arch.iommu_domain &&
-	       vcpu->kvm->arch.iommu_noncoherent;
+	return kvm_arch_has_noncoherent_dma(vcpu->kvm);
 }
 
 void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
@@ -6998,6 +6997,7 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 	INIT_LIST_HEAD(&kvm->arch.active_mmu_pages);
 	INIT_LIST_HEAD(&kvm->arch.zapped_obsolete_pages);
 	INIT_LIST_HEAD(&kvm->arch.assigned_dev_head);
+	atomic_set(&kvm->arch.noncoherent_dma_count, 0);
 
 	/* Reserve bit 0 of irq_sources_bitmap for userspace irq source */
 	set_bit(KVM_USERSPACE_IRQ_SOURCE_ID, &kvm->arch.irq_sources_bitmap);
@@ -7437,6 +7437,24 @@ bool kvm_arch_can_inject_async_page_present(struct kvm_vcpu *vcpu)
 			kvm_x86_ops->interrupt_allowed(vcpu);
 }
 
+void kvm_arch_register_noncoherent_dma(struct kvm *kvm)
+{
+	atomic_inc(&kvm->arch.noncoherent_dma_count);
+}
+EXPORT_SYMBOL_GPL(kvm_arch_register_noncoherent_dma);
+
+void kvm_arch_unregister_noncoherent_dma(struct kvm *kvm)
+{
+	atomic_dec(&kvm->arch.noncoherent_dma_count);
+}
+EXPORT_SYMBOL_GPL(kvm_arch_unregister_noncoherent_dma);
+
+bool kvm_arch_has_noncoherent_dma(struct kvm *kvm)
+{
+	return atomic_read(&kvm->arch.noncoherent_dma_count);
+}
+EXPORT_SYMBOL_GPL(kvm_arch_has_noncoherent_dma);
+
 EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_exit);
 EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_inj_virq);
 EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_page_fault);
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index ed64880e4915..92aae88756db 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -670,6 +670,25 @@ static inline void kvm_arch_free_vm(struct kvm *kvm)
 }
 #endif
 
+#ifdef __KVM_HAVE_ARCH_NONCOHERENT_DMA
+void kvm_arch_register_noncoherent_dma(struct kvm *kvm);
+void kvm_arch_unregister_noncoherent_dma(struct kvm *kvm);
+bool kvm_arch_has_noncoherent_dma(struct kvm *kvm);
+#else
+static inline void kvm_arch_register_noncoherent_dma(struct kvm *kvm)
+{
+}
+
+static inline void kvm_arch_unregister_noncoherent_dma(struct kvm *kvm)
+{
+}
+
+static inline bool kvm_arch_has_noncoherent_dma(struct kvm *kvm)
+{
+	return false;
+}
+#endif
+
 static inline wait_queue_head_t *kvm_arch_vcpu_wq(struct kvm_vcpu *vcpu)
 {
 #ifdef __KVM_HAVE_ARCH_WQP
diff --git a/virt/kvm/iommu.c b/virt/kvm/iommu.c
index d32d156a423a..c7d9ce122529 100644
--- a/virt/kvm/iommu.c
+++ b/virt/kvm/iommu.c
@@ -140,6 +140,9 @@ static int kvm_iommu_map_memslots(struct kvm *kvm)
 	struct kvm_memslots *slots;
 	struct kvm_memory_slot *memslot;
 
+	if (kvm->arch.iommu_noncoherent)
+		kvm_arch_register_noncoherent_dma(kvm);
+
 	idx = srcu_read_lock(&kvm->srcu);
 	slots = kvm_memslots(kvm);
 
@@ -327,6 +330,9 @@ static int kvm_iommu_unmap_memslots(struct kvm *kvm)
 
 	srcu_read_unlock(&kvm->srcu, idx);
 
+	if (kvm->arch.iommu_noncoherent)
+		kvm_arch_unregister_noncoherent_dma(kvm);
+
 	return 0;
 }
 
diff --git a/virt/kvm/vfio.c b/virt/kvm/vfio.c
index 597c258245ea..ca4260e35037 100644
--- a/virt/kvm/vfio.c
+++ b/virt/kvm/vfio.c
@@ -27,6 +27,7 @@ struct kvm_vfio_group {
 struct kvm_vfio {
 	struct list_head group_list;
 	struct mutex lock;
+	bool noncoherent;
 };
 
 static struct vfio_group *kvm_vfio_group_get_external_user(struct file *filep)
@@ -58,6 +59,43 @@ static void kvm_vfio_group_put_external_user(struct vfio_group *vfio_group)
 	symbol_put(vfio_group_put_external_user);
 }
 
+/*
+ * Groups can use the same or different IOMMU domains.  If the same then
+ * adding a new group may change the coherency of groups we've previously
+ * been told about.  We don't want to care about any of that so we retest
+ * each group and bail as soon as we find one that's noncoherent.  This
+ * means we only ever [un]register_noncoherent_dma once for the whole device.
+ */
+static void kvm_vfio_update_coherency(struct kvm_device *dev)
+{
+	struct kvm_vfio *kv = dev->private;
+	bool noncoherent = false;
+	struct kvm_vfio_group *kvg;
+
+	mutex_lock(&kv->lock);
+
+	list_for_each_entry(kvg, &kv->group_list, node) {
+		/*
+		 * TODO: We need an interface to check the coherency of
+		 * the IOMMU domain this group is using.  For now, assume
+		 * it's always noncoherent.
+		 */
+		noncoherent = true;
+		break;
+	}
+
+	if (noncoherent != kv->noncoherent) {
+		kv->noncoherent = noncoherent;
+
+		if (kv->noncoherent)
+			kvm_arch_register_noncoherent_dma(dev->kvm);
+		else
+			kvm_arch_unregister_noncoherent_dma(dev->kvm);
+	}
+
+	mutex_unlock(&kv->lock);
+}
+
 static int kvm_vfio_set_group(struct kvm_device *dev, long attr, u64 arg)
 {
 	struct kvm_vfio *kv = dev->private;
@@ -105,6 +143,8 @@ static int kvm_vfio_set_group(struct kvm_device *dev, long attr, u64 arg)
 
 		mutex_unlock(&kv->lock);
 
+		kvm_vfio_update_coherency(dev);
+
 		return 0;
 
 	case KVM_DEV_VFIO_GROUP_DEL:
@@ -140,6 +180,8 @@ static int kvm_vfio_set_group(struct kvm_device *dev, long attr, u64 arg)
 
 		kvm_vfio_group_put_external_user(vfio_group);
 
+		kvm_vfio_update_coherency(dev);
+
 		return ret;
 	}
 
@@ -185,6 +227,8 @@ static void kvm_vfio_destroy(struct kvm_device *dev)
 		kfree(kvg);
 	}
 
+	kvm_vfio_update_coherency(dev);
+
 	kfree(kv);
 	kfree(dev); /* alloc by kvm_ioctl_create_device, free by .destroy */
 }
-- 
cgit v1.2.3-58-ga151


From 81e87e26796782e014fd1f2bb9cd8fb6ce4021a8 Mon Sep 17 00:00:00 2001
From: "Michael S. Tsirkin" <mst@redhat.com>
Date: Wed, 30 Oct 2013 21:43:01 +0200
Subject: kvm_host: typo fix

fix up typo in comment.

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 include/linux/kvm_host.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index 92aae88756db..9bb1048d010e 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -804,7 +804,7 @@ static inline void kvm_guest_enter(void)
 
 	/* KVM does not hold any references to rcu protected data when it
 	 * switches CPU into a guest mode. In fact switching to a guest mode
-	 * is very similar to exiting to userspase from rcu point of view. In
+	 * is very similar to exiting to userspace from rcu point of view. In
 	 * addition CPU may stay in a guest mode for quite a long time (up to
 	 * one time slice). Lets treat guest mode as quiescent state, just like
 	 * we do with user-mode execution.
-- 
cgit v1.2.3-58-ga151


From ce332f662deb545c8a4f3f58debcca26bb2e44b0 Mon Sep 17 00:00:00 2001
From: "Michael S. Tsirkin" <mst@redhat.com>
Date: Mon, 4 Nov 2013 22:36:17 +0200
Subject: srcu: API for barrier after srcu read unlock

srcu read lock/unlock include a full memory barrier
but that's an implementation detail.
Add an API for make memory fencing explicit for
users that need this barrier, to make sure we
can change it as needed without breaking all users.

Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Reviewed-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Gleb Natapov <gleb@redhat.com>
---
 include/linux/srcu.h | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/srcu.h b/include/linux/srcu.h
index c114614ed172..9b058eecd403 100644
--- a/include/linux/srcu.h
+++ b/include/linux/srcu.h
@@ -237,4 +237,18 @@ static inline void srcu_read_unlock(struct srcu_struct *sp, int idx)
 	__srcu_read_unlock(sp, idx);
 }
 
+/**
+ * smp_mb__after_srcu_read_unlock - ensure full ordering after srcu_read_unlock
+ *
+ * Converts the preceding srcu_read_unlock into a two-way memory barrier.
+ *
+ * Call this after srcu_read_unlock, to guarantee that all memory operations
+ * that occur after smp_mb__after_srcu_read_unlock will appear to happen after
+ * the preceding srcu_read_unlock.
+ */
+static inline void smp_mb__after_srcu_read_unlock(void)
+{
+	/* __srcu_read_unlock has smp_mb() internally so nothing to do here. */
+}
+
 #endif
-- 
cgit v1.2.3-58-ga151


From 8b414521bc5375ae8ba18c083af95d44b8da0d04 Mon Sep 17 00:00:00 2001
From: Marcelo Tosatti <mtosatti@redhat.com>
Date: Fri, 11 Oct 2013 21:39:26 -0300
Subject: hung_task: add method to reset detector

In certain occasions it is possible for a hung task detector
positive to be false: continuation from a paused VM, for example.

Add a method to reset detection, similar as is done
with other kernel watchdogs.

Acked-by: Don Zickus <dzickus@redhat.com>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Gleb Natapov <gleb@redhat.com>
---
 arch/x86/kernel/pvclock.c |  1 +
 include/linux/sched.h     |  8 ++++++++
 kernel/hung_task.c        | 11 +++++++++++
 3 files changed, 20 insertions(+)

(limited to 'include/linux')

diff --git a/arch/x86/kernel/pvclock.c b/arch/x86/kernel/pvclock.c
index 6279928c0a71..2f355d229a58 100644
--- a/arch/x86/kernel/pvclock.c
+++ b/arch/x86/kernel/pvclock.c
@@ -48,6 +48,7 @@ void pvclock_touch_watchdogs(void)
 	touch_softlockup_watchdog_sync();
 	clocksource_touch_watchdog();
 	rcu_cpu_stall_reset();
+	reset_hung_task_detector();
 }
 
 static atomic64_t last_value = ATOMIC64_INIT(0);
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 6682da36b293..7bb4b4a2a101 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -285,6 +285,14 @@ static inline void lockup_detector_init(void)
 }
 #endif
 
+#ifdef CONFIG_DETECT_HUNG_TASK
+void reset_hung_task_detector(void);
+#else
+static inline void reset_hung_task_detector(void)
+{
+}
+#endif
+
 /* Attach to any functions which should be ignored in wchan output. */
 #define __sched		__attribute__((__section__(".sched.text")))
 
diff --git a/kernel/hung_task.c b/kernel/hung_task.c
index 3e97fb126e6b..dfdf51534b3e 100644
--- a/kernel/hung_task.c
+++ b/kernel/hung_task.c
@@ -203,6 +203,14 @@ int proc_dohung_task_timeout_secs(struct ctl_table *table, int write,
 	return ret;
 }
 
+static atomic_t reset_hung_task = ATOMIC_INIT(0);
+
+void reset_hung_task_detector(void)
+{
+	atomic_set(&reset_hung_task, 1);
+}
+EXPORT_SYMBOL_GPL(reset_hung_task_detector);
+
 /*
  * kthread which checks for tasks stuck in D state
  */
@@ -216,6 +224,9 @@ static int watchdog(void *dummy)
 		while (schedule_timeout_interruptible(timeout_jiffies(timeout)))
 			timeout = sysctl_hung_task_timeout_secs;
 
+		if (atomic_xchg(&reset_hung_task, 0))
+			continue;
+
 		check_hung_uninterruptible_tasks(timeout);
 	}
 
-- 
cgit v1.2.3-58-ga151