[GIT PULL 0/4] KVM: s390: features and test for 5.11

[GIT PULL 0/4] KVM: s390: features and test for 5.11

[Christian Borntraeger]

Paolo,

the following changes since commit f8394f232b1eab649ce2df5c5f15b0e528c92091:

  Linux 5.10-rc3 (2020-11-08 16:10:16 -0800)

are available in the Git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/kvms390/linux.git  tags/kvm-s390-next-5.11-1

for you to fetch changes up to 50a05be484cb70d9dfb55fa5a6ed57eab193901f:

  KVM: s390: track synchronous pfault events in kvm_stat (2020-12-10 14:20:26 +0100)

----------------------------------------------------------------
KVM: s390: Features and Test for 5.11

- memcg accouting for s390 specific parts of kvm and gmap
- selftest for diag318
- new kvm_stat for when async_pf falls back to sync

The selftest even triggers a non-critical bug that is unrelated
to diag318, fix will follow later.

----------------------------------------------------------------
Christian Borntraeger (3):
      KVM: s390: Add memcg accounting to KVM allocations
      s390/gmap: make gmap memcg aware
      KVM: s390: track synchronous pfault events in kvm_stat

Collin Walling (1):
      KVM: selftests: sync_regs test for diag318

 arch/s390/include/asm/kvm_host.h                   |  1 +
 arch/s390/kvm/guestdbg.c                           |  8 +--
 arch/s390/kvm/intercept.c                          |  2 +-
 arch/s390/kvm/interrupt.c                          | 10 +--
 arch/s390/kvm/kvm-s390.c                           | 22 +++---
 arch/s390/kvm/priv.c                               |  4 +-
 arch/s390/kvm/pv.c                                 |  6 +-
 arch/s390/kvm/vsie.c                               |  4 +-
 arch/s390/mm/gmap.c                                | 30 ++++----
 tools/testing/selftests/kvm/Makefile               |  2 +-
 .../kvm/include/s390x/diag318_test_handler.h       | 13 ++++
 .../selftests/kvm/lib/s390x/diag318_test_handler.c | 82 ++++++++++++++++++++++
 tools/testing/selftests/kvm/s390x/sync_regs_test.c | 16 ++++-
 13 files changed, 156 insertions(+), 44 deletions(-)
 create mode 100644 tools/testing/selftests/kvm/include/s390x/diag318_test_handler.h
 create mode 100644 tools/testing/selftests/kvm/lib/s390x/diag318_test_handler.c

[GIT PULL 1/4] KVM: s390: Add memcg accounting to KVM allocations

[Christian Borntraeger]

Almost all kvm allocations in the s390x KVM code can be attributed to
the process that triggers the allocation (in other words, no global
allocation for other guests). This will help the memcg controller to
make the right decisions.

Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Acked-by: Janosch Frank <frankja@linux.ibm.com>
Acked-by: Cornelia Huck <cohuck@redhat.com>
---
 arch/s390/kvm/guestdbg.c  |  8 ++++----
 arch/s390/kvm/intercept.c |  2 +-
 arch/s390/kvm/interrupt.c | 10 +++++-----
 arch/s390/kvm/kvm-s390.c  | 20 ++++++++++----------
 arch/s390/kvm/priv.c      |  4 ++--
 arch/s390/kvm/pv.c        |  6 +++---
 arch/s390/kvm/vsie.c      |  4 ++--
 7 files changed, 27 insertions(+), 27 deletions(-)

diff --git a/arch/s390/kvm/guestdbg.c b/arch/s390/kvm/guestdbg.c
index 394a5f53805b..3765c4223bf9 100644
--- a/arch/s390/kvm/guestdbg.c
+++ b/arch/s390/kvm/guestdbg.c
@@ -184,7 +184,7 @@ static int __import_wp_info(struct kvm_vcpu *vcpu,
 	if (wp_info->len < 0 || wp_info->len > MAX_WP_SIZE)
 		return -EINVAL;
 
-	wp_info->old_data = kmalloc(bp_data->len, GFP_KERNEL);
+	wp_info->old_data = kmalloc(bp_data->len, GFP_KERNEL_ACCOUNT);
 	if (!wp_info->old_data)
 		return -ENOMEM;
 	/* try to backup the original value */
@@ -234,7 +234,7 @@ int kvm_s390_import_bp_data(struct kvm_vcpu *vcpu,
 	if (nr_wp > 0) {
 		wp_info = kmalloc_array(nr_wp,
 					sizeof(*wp_info),
-					GFP_KERNEL);
+					GFP_KERNEL_ACCOUNT);
 		if (!wp_info) {
 			ret = -ENOMEM;
 			goto error;
@@ -243,7 +243,7 @@ int kvm_s390_import_bp_data(struct kvm_vcpu *vcpu,
 	if (nr_bp > 0) {
 		bp_info = kmalloc_array(nr_bp,
 					sizeof(*bp_info),
-					GFP_KERNEL);
+					GFP_KERNEL_ACCOUNT);
 		if (!bp_info) {
 			ret = -ENOMEM;
 			goto error;
@@ -349,7 +349,7 @@ static struct kvm_hw_wp_info_arch *any_wp_changed(struct kvm_vcpu *vcpu)
 		if (!wp_info || !wp_info->old_data || wp_info->len <= 0)
 			continue;
 
-		temp = kmalloc(wp_info->len, GFP_KERNEL);
+		temp = kmalloc(wp_info->len, GFP_KERNEL_ACCOUNT);
 		if (!temp)
 			continue;
 
diff --git a/arch/s390/kvm/intercept.c b/arch/s390/kvm/intercept.c
index e7a7c499a73f..72b25b7cc6ae 100644
--- a/arch/s390/kvm/intercept.c
+++ b/arch/s390/kvm/intercept.c
@@ -398,7 +398,7 @@ int handle_sthyi(struct kvm_vcpu *vcpu)
 	if (!kvm_s390_pv_cpu_is_protected(vcpu) && (addr & ~PAGE_MASK))
 		return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION);
 
-	sctns = (void *)get_zeroed_page(GFP_KERNEL);
+	sctns = (void *)get_zeroed_page(GFP_KERNEL_ACCOUNT);
 	if (!sctns)
 		return -ENOMEM;
 
diff --git a/arch/s390/kvm/interrupt.c b/arch/s390/kvm/interrupt.c
index 2f177298c663..e3183bd05910 100644
--- a/arch/s390/kvm/interrupt.c
+++ b/arch/s390/kvm/interrupt.c
@@ -1792,7 +1792,7 @@ struct kvm_s390_interrupt_info *kvm_s390_get_io_int(struct kvm *kvm,
 		goto out;
 	}
 gisa_out:
-	tmp_inti = kzalloc(sizeof(*inti), GFP_KERNEL);
+	tmp_inti = kzalloc(sizeof(*inti), GFP_KERNEL_ACCOUNT);
 	if (tmp_inti) {
 		tmp_inti->type = KVM_S390_INT_IO(1, 0, 0, 0);
 		tmp_inti->io.io_int_word = isc_to_int_word(isc);
@@ -2015,7 +2015,7 @@ int kvm_s390_inject_vm(struct kvm *kvm,
 	struct kvm_s390_interrupt_info *inti;
 	int rc;
 
-	inti = kzalloc(sizeof(*inti), GFP_KERNEL);
+	inti = kzalloc(sizeof(*inti), GFP_KERNEL_ACCOUNT);
 	if (!inti)
 		return -ENOMEM;
 
@@ -2414,7 +2414,7 @@ static int enqueue_floating_irq(struct kvm_device *dev,
 		return -EINVAL;
 
 	while (len >= sizeof(struct kvm_s390_irq)) {
-		inti = kzalloc(sizeof(*inti), GFP_KERNEL);
+		inti = kzalloc(sizeof(*inti), GFP_KERNEL_ACCOUNT);
 		if (!inti)
 			return -ENOMEM;
 
@@ -2462,7 +2462,7 @@ static int register_io_adapter(struct kvm_device *dev,
 	if (dev->kvm->arch.adapters[adapter_info.id] != NULL)
 		return -EINVAL;
 
-	adapter = kzalloc(sizeof(*adapter), GFP_KERNEL);
+	adapter = kzalloc(sizeof(*adapter), GFP_KERNEL_ACCOUNT);
 	if (!adapter)
 		return -ENOMEM;
 
@@ -3290,7 +3290,7 @@ int kvm_s390_gib_init(u8 nisc)
 		goto out;
 	}
 
-	gib = (struct kvm_s390_gib *)get_zeroed_page(GFP_KERNEL | GFP_DMA);
+	gib = (struct kvm_s390_gib *)get_zeroed_page(GFP_KERNEL_ACCOUNT | GFP_DMA);
 	if (!gib) {
 		rc = -ENOMEM;
 		goto out;
diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c
index 6b74b92c1a58..19804c388d61 100644
--- a/arch/s390/kvm/kvm-s390.c
+++ b/arch/s390/kvm/kvm-s390.c
@@ -1254,7 +1254,7 @@ static int kvm_s390_set_processor(struct kvm *kvm, struct kvm_device_attr *attr)
 		ret = -EBUSY;
 		goto out;
 	}
-	proc = kzalloc(sizeof(*proc), GFP_KERNEL);
+	proc = kzalloc(sizeof(*proc), GFP_KERNEL_ACCOUNT);
 	if (!proc) {
 		ret = -ENOMEM;
 		goto out;
@@ -1416,7 +1416,7 @@ static int kvm_s390_get_processor(struct kvm *kvm, struct kvm_device_attr *attr)
 	struct kvm_s390_vm_cpu_processor *proc;
 	int ret = 0;
 
-	proc = kzalloc(sizeof(*proc), GFP_KERNEL);
+	proc = kzalloc(sizeof(*proc), GFP_KERNEL_ACCOUNT);
 	if (!proc) {
 		ret = -ENOMEM;
 		goto out;
@@ -1444,7 +1444,7 @@ static int kvm_s390_get_machine(struct kvm *kvm, struct kvm_device_attr *attr)
 	struct kvm_s390_vm_cpu_machine *mach;
 	int ret = 0;
 
-	mach = kzalloc(sizeof(*mach), GFP_KERNEL);
+	mach = kzalloc(sizeof(*mach), GFP_KERNEL_ACCOUNT);
 	if (!mach) {
 		ret = -ENOMEM;
 		goto out;
@@ -1812,7 +1812,7 @@ static long kvm_s390_get_skeys(struct kvm *kvm, struct kvm_s390_skeys *args)
 	if (args->count < 1 || args->count > KVM_S390_SKEYS_MAX)
 		return -EINVAL;
 
-	keys = kvmalloc_array(args->count, sizeof(uint8_t), GFP_KERNEL);
+	keys = kvmalloc_array(args->count, sizeof(uint8_t), GFP_KERNEL_ACCOUNT);
 	if (!keys)
 		return -ENOMEM;
 
@@ -1857,7 +1857,7 @@ static long kvm_s390_set_skeys(struct kvm *kvm, struct kvm_s390_skeys *args)
 	if (args->count < 1 || args->count > KVM_S390_SKEYS_MAX)
 		return -EINVAL;
 
-	keys = kvmalloc_array(args->count, sizeof(uint8_t), GFP_KERNEL);
+	keys = kvmalloc_array(args->count, sizeof(uint8_t), GFP_KERNEL_ACCOUNT);
 	if (!keys)
 		return -ENOMEM;
 
@@ -2625,7 +2625,7 @@ static void sca_dispose(struct kvm *kvm)
 
 int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 {
-	gfp_t alloc_flags = GFP_KERNEL;
+	gfp_t alloc_flags = GFP_KERNEL_ACCOUNT;
 	int i, rc;
 	char debug_name[16];
 	static unsigned long sca_offset;
@@ -2670,7 +2670,7 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 
 	BUILD_BUG_ON(sizeof(struct sie_page2) != 4096);
 	kvm->arch.sie_page2 =
-	     (struct sie_page2 *) get_zeroed_page(GFP_KERNEL | GFP_DMA);
+	     (struct sie_page2 *) get_zeroed_page(GFP_KERNEL_ACCOUNT | GFP_DMA);
 	if (!kvm->arch.sie_page2)
 		goto out_err;
 
@@ -2900,7 +2900,7 @@ static int sca_switch_to_extended(struct kvm *kvm)
 	if (kvm->arch.use_esca)
 		return 0;
 
-	new_sca = alloc_pages_exact(sizeof(*new_sca), GFP_KERNEL|__GFP_ZERO);
+	new_sca = alloc_pages_exact(sizeof(*new_sca), GFP_KERNEL_ACCOUNT | __GFP_ZERO);
 	if (!new_sca)
 		return -ENOMEM;
 
@@ -3133,7 +3133,7 @@ void kvm_s390_vcpu_unsetup_cmma(struct kvm_vcpu *vcpu)
 
 int kvm_s390_vcpu_setup_cmma(struct kvm_vcpu *vcpu)
 {
-	vcpu->arch.sie_block->cbrlo = get_zeroed_page(GFP_KERNEL);
+	vcpu->arch.sie_block->cbrlo = get_zeroed_page(GFP_KERNEL_ACCOUNT);
 	if (!vcpu->arch.sie_block->cbrlo)
 		return -ENOMEM;
 	return 0;
@@ -3243,7 +3243,7 @@ int kvm_arch_vcpu_create(struct kvm_vcpu *vcpu)
 	int rc;
 
 	BUILD_BUG_ON(sizeof(struct sie_page) != 4096);
-	sie_page = (struct sie_page *) get_zeroed_page(GFP_KERNEL);
+	sie_page = (struct sie_page *) get_zeroed_page(GFP_KERNEL_ACCOUNT);
 	if (!sie_page)
 		return -ENOMEM;
 
diff --git a/arch/s390/kvm/priv.c b/arch/s390/kvm/priv.c
index cd74989ce0b0..9928f785c677 100644
--- a/arch/s390/kvm/priv.c
+++ b/arch/s390/kvm/priv.c
@@ -879,7 +879,7 @@ static int handle_stsi(struct kvm_vcpu *vcpu)
 	switch (fc) {
 	case 1: /* same handling for 1 and 2 */
 	case 2:
-		mem = get_zeroed_page(GFP_KERNEL);
+		mem = get_zeroed_page(GFP_KERNEL_ACCOUNT);
 		if (!mem)
 			goto out_no_data;
 		if (stsi((void *) mem, fc, sel1, sel2))
@@ -888,7 +888,7 @@ static int handle_stsi(struct kvm_vcpu *vcpu)
 	case 3:
 		if (sel1 != 2 || sel2 != 2)
 			goto out_no_data;
-		mem = get_zeroed_page(GFP_KERNEL);
+		mem = get_zeroed_page(GFP_KERNEL_ACCOUNT);
 		if (!mem)
 			goto out_no_data;
 		handle_stsi_3_2_2(vcpu, (void *) mem);
diff --git a/arch/s390/kvm/pv.c b/arch/s390/kvm/pv.c
index eb99e2f95ebe..373b654c84bd 100644
--- a/arch/s390/kvm/pv.c
+++ b/arch/s390/kvm/pv.c
@@ -60,7 +60,7 @@ int kvm_s390_pv_create_cpu(struct kvm_vcpu *vcpu, u16 *rc, u16 *rrc)
 	if (kvm_s390_pv_cpu_get_handle(vcpu))
 		return -EINVAL;
 
-	vcpu->arch.pv.stor_base = __get_free_pages(GFP_KERNEL,
+	vcpu->arch.pv.stor_base = __get_free_pages(GFP_KERNEL_ACCOUNT,
 						   get_order(uv_info.guest_cpu_stor_len));
 	if (!vcpu->arch.pv.stor_base)
 		return -ENOMEM;
@@ -72,7 +72,7 @@ int kvm_s390_pv_create_cpu(struct kvm_vcpu *vcpu, u16 *rc, u16 *rrc)
 	uvcb.stor_origin = (u64)vcpu->arch.pv.stor_base;
 
 	/* Alloc Secure Instruction Data Area Designation */
-	vcpu->arch.sie_block->sidad = __get_free_page(GFP_KERNEL | __GFP_ZERO);
+	vcpu->arch.sie_block->sidad = __get_free_page(GFP_KERNEL_ACCOUNT | __GFP_ZERO);
 	if (!vcpu->arch.sie_block->sidad) {
 		free_pages(vcpu->arch.pv.stor_base,
 			   get_order(uv_info.guest_cpu_stor_len));
@@ -120,7 +120,7 @@ static int kvm_s390_pv_alloc_vm(struct kvm *kvm)
 	struct kvm_memory_slot *memslot;
 
 	kvm->arch.pv.stor_var = NULL;
-	kvm->arch.pv.stor_base = __get_free_pages(GFP_KERNEL, get_order(base));
+	kvm->arch.pv.stor_base = __get_free_pages(GFP_KERNEL_ACCOUNT, get_order(base));
 	if (!kvm->arch.pv.stor_base)
 		return -ENOMEM;
 
diff --git a/arch/s390/kvm/vsie.c b/arch/s390/kvm/vsie.c
index 4f3cbf6003a9..c5d0a58b2c29 100644
--- a/arch/s390/kvm/vsie.c
+++ b/arch/s390/kvm/vsie.c
@@ -1234,7 +1234,7 @@ static struct vsie_page *get_vsie_page(struct kvm *kvm, unsigned long addr)
 
 	mutex_lock(&kvm->arch.vsie.mutex);
 	if (kvm->arch.vsie.page_count < nr_vcpus) {
-		page = alloc_page(GFP_KERNEL | __GFP_ZERO | GFP_DMA);
+		page = alloc_page(GFP_KERNEL_ACCOUNT | __GFP_ZERO | GFP_DMA);
 		if (!page) {
 			mutex_unlock(&kvm->arch.vsie.mutex);
 			return ERR_PTR(-ENOMEM);
@@ -1336,7 +1336,7 @@ int kvm_s390_handle_vsie(struct kvm_vcpu *vcpu)
 void kvm_s390_vsie_init(struct kvm *kvm)
 {
 	mutex_init(&kvm->arch.vsie.mutex);
-	INIT_RADIX_TREE(&kvm->arch.vsie.addr_to_page, GFP_KERNEL);
+	INIT_RADIX_TREE(&kvm->arch.vsie.addr_to_page, GFP_KERNEL_ACCOUNT);
 }
 
 /* Destroy the vsie data structures. To be called when a vm is destroyed. */
-- 
2.28.0

[GIT PULL 2/4] s390/gmap: make gmap memcg aware

[Christian Borntraeger]

gmap allocations can be attributed to a process.

Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Acked-by: Heiko Carstens <hca@linux.ibm.com>
Acked-by: Janosch Frank <frankja@linux.ibm.com>
Acked-by: Cornelia Huck <cohuck@redhat.com>
---
 arch/s390/mm/gmap.c | 30 +++++++++++++++---------------
 1 file changed, 15 insertions(+), 15 deletions(-)

diff --git a/arch/s390/mm/gmap.c b/arch/s390/mm/gmap.c
index cfb0017f33a7..0160ac97a27d 100644
--- a/arch/s390/mm/gmap.c
+++ b/arch/s390/mm/gmap.c
@@ -2,7 +2,7 @@
 /*
  *  KVM guest address space mapping code
  *
- *    Copyright IBM Corp. 2007, 2016, 2018
+ *    Copyright IBM Corp. 2007, 2020
  *    Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com>
  *		 David Hildenbrand <david@redhat.com>
  *		 Janosch Frank <frankja@linux.vnet.ibm.com>
@@ -56,19 +56,19 @@ static struct gmap *gmap_alloc(unsigned long limit)
 		atype = _ASCE_TYPE_REGION1;
 		etype = _REGION1_ENTRY_EMPTY;
 	}
-	gmap = kzalloc(sizeof(struct gmap), GFP_KERNEL);
+	gmap = kzalloc(sizeof(struct gmap), GFP_KERNEL_ACCOUNT);
 	if (!gmap)
 		goto out;
 	INIT_LIST_HEAD(&gmap->crst_list);
 	INIT_LIST_HEAD(&gmap->children);
 	INIT_LIST_HEAD(&gmap->pt_list);
-	INIT_RADIX_TREE(&gmap->guest_to_host, GFP_KERNEL);
-	INIT_RADIX_TREE(&gmap->host_to_guest, GFP_ATOMIC);
-	INIT_RADIX_TREE(&gmap->host_to_rmap, GFP_ATOMIC);
+	INIT_RADIX_TREE(&gmap->guest_to_host, GFP_KERNEL_ACCOUNT);
+	INIT_RADIX_TREE(&gmap->host_to_guest, GFP_ATOMIC | __GFP_ACCOUNT);
+	INIT_RADIX_TREE(&gmap->host_to_rmap, GFP_ATOMIC | __GFP_ACCOUNT);
 	spin_lock_init(&gmap->guest_table_lock);
 	spin_lock_init(&gmap->shadow_lock);
 	refcount_set(&gmap->ref_count, 1);
-	page = alloc_pages(GFP_KERNEL, CRST_ALLOC_ORDER);
+	page = alloc_pages(GFP_KERNEL_ACCOUNT, CRST_ALLOC_ORDER);
 	if (!page)
 		goto out_free;
 	page->index = 0;
@@ -309,7 +309,7 @@ static int gmap_alloc_table(struct gmap *gmap, unsigned long *table,
 	unsigned long *new;
 
 	/* since we dont free the gmap table until gmap_free we can unlock */
-	page = alloc_pages(GFP_KERNEL, CRST_ALLOC_ORDER);
+	page = alloc_pages(GFP_KERNEL_ACCOUNT, CRST_ALLOC_ORDER);
 	if (!page)
 		return -ENOMEM;
 	new = (unsigned long *) page_to_phys(page);
@@ -594,7 +594,7 @@ int __gmap_link(struct gmap *gmap, unsigned long gaddr, unsigned long vmaddr)
 	if (pmd_large(*pmd) && !gmap->mm->context.allow_gmap_hpage_1m)
 		return -EFAULT;
 	/* Link gmap segment table entry location to page table. */
-	rc = radix_tree_preload(GFP_KERNEL);
+	rc = radix_tree_preload(GFP_KERNEL_ACCOUNT);
 	if (rc)
 		return rc;
 	ptl = pmd_lock(mm, pmd);
@@ -1218,11 +1218,11 @@ static int gmap_protect_rmap(struct gmap *sg, unsigned long raddr,
 		vmaddr = __gmap_translate(parent, paddr);
 		if (IS_ERR_VALUE(vmaddr))
 			return vmaddr;
-		rmap = kzalloc(sizeof(*rmap), GFP_KERNEL);
+		rmap = kzalloc(sizeof(*rmap), GFP_KERNEL_ACCOUNT);
 		if (!rmap)
 			return -ENOMEM;
 		rmap->raddr = raddr;
-		rc = radix_tree_preload(GFP_KERNEL);
+		rc = radix_tree_preload(GFP_KERNEL_ACCOUNT);
 		if (rc) {
 			kfree(rmap);
 			return rc;
@@ -1741,7 +1741,7 @@ int gmap_shadow_r2t(struct gmap *sg, unsigned long saddr, unsigned long r2t,
 
 	BUG_ON(!gmap_is_shadow(sg));
 	/* Allocate a shadow region second table */
-	page = alloc_pages(GFP_KERNEL, CRST_ALLOC_ORDER);
+	page = alloc_pages(GFP_KERNEL_ACCOUNT, CRST_ALLOC_ORDER);
 	if (!page)
 		return -ENOMEM;
 	page->index = r2t & _REGION_ENTRY_ORIGIN;
@@ -1825,7 +1825,7 @@ int gmap_shadow_r3t(struct gmap *sg, unsigned long saddr, unsigned long r3t,
 
 	BUG_ON(!gmap_is_shadow(sg));
 	/* Allocate a shadow region second table */
-	page = alloc_pages(GFP_KERNEL, CRST_ALLOC_ORDER);
+	page = alloc_pages(GFP_KERNEL_ACCOUNT, CRST_ALLOC_ORDER);
 	if (!page)
 		return -ENOMEM;
 	page->index = r3t & _REGION_ENTRY_ORIGIN;
@@ -1909,7 +1909,7 @@ int gmap_shadow_sgt(struct gmap *sg, unsigned long saddr, unsigned long sgt,
 
 	BUG_ON(!gmap_is_shadow(sg) || (sgt & _REGION3_ENTRY_LARGE));
 	/* Allocate a shadow segment table */
-	page = alloc_pages(GFP_KERNEL, CRST_ALLOC_ORDER);
+	page = alloc_pages(GFP_KERNEL_ACCOUNT, CRST_ALLOC_ORDER);
 	if (!page)
 		return -ENOMEM;
 	page->index = sgt & _REGION_ENTRY_ORIGIN;
@@ -2116,7 +2116,7 @@ int gmap_shadow_page(struct gmap *sg, unsigned long saddr, pte_t pte)
 	parent = sg->parent;
 	prot = (pte_val(pte) & _PAGE_PROTECT) ? PROT_READ : PROT_WRITE;
 
-	rmap = kzalloc(sizeof(*rmap), GFP_KERNEL);
+	rmap = kzalloc(sizeof(*rmap), GFP_KERNEL_ACCOUNT);
 	if (!rmap)
 		return -ENOMEM;
 	rmap->raddr = (saddr & PAGE_MASK) | _SHADOW_RMAP_PGTABLE;
@@ -2128,7 +2128,7 @@ int gmap_shadow_page(struct gmap *sg, unsigned long saddr, pte_t pte)
 			rc = vmaddr;
 			break;
 		}
-		rc = radix_tree_preload(GFP_KERNEL);
+		rc = radix_tree_preload(GFP_KERNEL_ACCOUNT);
 		if (rc)
 			break;
 		rc = -EAGAIN;
-- 
2.28.0

[GIT PULL 3/4] KVM: selftests: sync_regs test for diag318

[Christian Borntraeger]

From: Collin Walling <walling@linux.ibm.com>

The DIAGNOSE 0x0318 instruction, unique to s390x, is a privileged call
that must be intercepted via SIE, handled in userspace, and the
information set by the instruction is communicated back to KVM.

To test the instruction interception, an ad-hoc handler is defined which
simply has a VM execute the instruction and then userspace will extract
the necessary info. The handler is defined such that the instruction
invocation occurs only once. It is up to the caller to determine how the
info returned by this handler should be used.

The diag318 info is communicated from userspace to KVM via a sync_regs
call. This is tested during a sync_regs test, where the diag318 info is
requested via the handler, then the info is stored in the appropriate
register in KVM via a sync registers call.

If KVM does not support diag318, then the tests will print a message
stating that diag318 was skipped, and the asserts will simply test
against a value of 0.

Signed-off-by: Collin Walling <walling@linux.ibm.com>
Link: https://lore.kernel.org/r/20201207154125.10322-1-walling@linux.ibm.com
Acked-by: Janosch Frank <frankja@linux.ibm.com>
Acked-by: Cornelia Huck <cohuck@redhat.com>
Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
---
 tools/testing/selftests/kvm/Makefile          |  2 +-
 .../kvm/include/s390x/diag318_test_handler.h  | 13 +++
 .../kvm/lib/s390x/diag318_test_handler.c      | 82 +++++++++++++++++++
 .../selftests/kvm/s390x/sync_regs_test.c      | 16 +++-
 4 files changed, 111 insertions(+), 2 deletions(-)
 create mode 100644 tools/testing/selftests/kvm/include/s390x/diag318_test_handler.h
 create mode 100644 tools/testing/selftests/kvm/lib/s390x/diag318_test_handler.c

diff --git a/tools/testing/selftests/kvm/Makefile b/tools/testing/selftests/kvm/Makefile
index 30afbad36cd5..1b9c257fca5e 100644
--- a/tools/testing/selftests/kvm/Makefile
+++ b/tools/testing/selftests/kvm/Makefile
@@ -36,7 +36,7 @@ endif
 LIBKVM = lib/assert.c lib/elf.c lib/io.c lib/kvm_util.c lib/sparsebit.c lib/test_util.c
 LIBKVM_x86_64 = lib/x86_64/processor.c lib/x86_64/vmx.c lib/x86_64/svm.c lib/x86_64/ucall.c
 LIBKVM_aarch64 = lib/aarch64/processor.c lib/aarch64/ucall.c
-LIBKVM_s390x = lib/s390x/processor.c lib/s390x/ucall.c
+LIBKVM_s390x = lib/s390x/processor.c lib/s390x/ucall.c lib/s390x/diag318_test_handler.c
 
 TEST_GEN_PROGS_x86_64 = x86_64/cr4_cpuid_sync_test
 TEST_GEN_PROGS_x86_64 += x86_64/evmcs_test
diff --git a/tools/testing/selftests/kvm/include/s390x/diag318_test_handler.h b/tools/testing/selftests/kvm/include/s390x/diag318_test_handler.h
new file mode 100644
index 000000000000..b0ed71302722
--- /dev/null
+++ b/tools/testing/selftests/kvm/include/s390x/diag318_test_handler.h
@@ -0,0 +1,13 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later
+ *
+ * Test handler for the s390x DIAGNOSE 0x0318 instruction.
+ *
+ * Copyright (C) 2020, IBM
+ */
+
+#ifndef SELFTEST_KVM_DIAG318_TEST_HANDLER
+#define SELFTEST_KVM_DIAG318_TEST_HANDLER
+
+uint64_t get_diag318_info(void);
+
+#endif
diff --git a/tools/testing/selftests/kvm/lib/s390x/diag318_test_handler.c b/tools/testing/selftests/kvm/lib/s390x/diag318_test_handler.c
new file mode 100644
index 000000000000..86b9e611ad87
--- /dev/null
+++ b/tools/testing/selftests/kvm/lib/s390x/diag318_test_handler.c
@@ -0,0 +1,82 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Test handler for the s390x DIAGNOSE 0x0318 instruction.
+ *
+ * Copyright (C) 2020, IBM
+ */
+
+#include "test_util.h"
+#include "kvm_util.h"
+
+#define VCPU_ID	6
+
+#define ICPT_INSTRUCTION	0x04
+#define IPA0_DIAG		0x8300
+
+static void guest_code(void)
+{
+	uint64_t diag318_info = 0x12345678;
+
+	asm volatile ("diag %0,0,0x318\n" : : "d" (diag318_info));
+}
+
+/*
+ * The DIAGNOSE 0x0318 instruction call must be handled via userspace. As such,
+ * we create an ad-hoc VM here to handle the instruction then extract the
+ * necessary data. It is up to the caller to decide what to do with that data.
+ */
+static uint64_t diag318_handler(void)
+{
+	struct kvm_vm *vm;
+	struct kvm_run *run;
+	uint64_t reg;
+	uint64_t diag318_info;
+
+	vm = vm_create_default(VCPU_ID, 0, guest_code);
+	vcpu_run(vm, VCPU_ID);
+	run = vcpu_state(vm, VCPU_ID);
+
+	TEST_ASSERT(run->exit_reason == KVM_EXIT_S390_SIEIC,
+		    "DIAGNOSE 0x0318 instruction was not intercepted");
+	TEST_ASSERT(run->s390_sieic.icptcode == ICPT_INSTRUCTION,
+		    "Unexpected intercept code: 0x%x", run->s390_sieic.icptcode);
+	TEST_ASSERT((run->s390_sieic.ipa & 0xff00) == IPA0_DIAG,
+		    "Unexpected IPA0 code: 0x%x", (run->s390_sieic.ipa & 0xff00));
+
+	reg = (run->s390_sieic.ipa & 0x00f0) >> 4;
+	diag318_info = run->s.regs.gprs[reg];
+
+	TEST_ASSERT(diag318_info != 0, "DIAGNOSE 0x0318 info not set");
+
+	kvm_vm_free(vm);
+
+	return diag318_info;
+}
+
+uint64_t get_diag318_info(void)
+{
+	static uint64_t diag318_info;
+	static bool printed_skip;
+
+	/*
+	 * If KVM does not support diag318, then return 0 to
+	 * ensure tests do not break.
+	 */
+	if (!kvm_check_cap(KVM_CAP_S390_DIAG318)) {
+		if (!printed_skip) {
+			fprintf(stdout, "KVM_CAP_S390_DIAG318 not supported. "
+				"Skipping diag318 test.\n");
+			printed_skip = true;
+		}
+		return 0;
+	}
+
+	/*
+	 * If a test has previously requested the diag318 info,
+	 * then don't bother spinning up a temporary VM again.
+	 */
+	if (!diag318_info)
+		diag318_info = diag318_handler();
+
+	return diag318_info;
+}
diff --git a/tools/testing/selftests/kvm/s390x/sync_regs_test.c b/tools/testing/selftests/kvm/s390x/sync_regs_test.c
index 5731ccf34917..caf7b8859a94 100644
--- a/tools/testing/selftests/kvm/s390x/sync_regs_test.c
+++ b/tools/testing/selftests/kvm/s390x/sync_regs_test.c
@@ -20,6 +20,7 @@
 
 #include "test_util.h"
 #include "kvm_util.h"
+#include "diag318_test_handler.h"
 
 #define VCPU_ID 5
 
@@ -70,7 +71,7 @@ static void compare_sregs(struct kvm_sregs *left, struct kvm_sync_regs *right)
 
 #undef REG_COMPARE
 
-#define TEST_SYNC_FIELDS   (KVM_SYNC_GPRS|KVM_SYNC_ACRS|KVM_SYNC_CRS)
+#define TEST_SYNC_FIELDS   (KVM_SYNC_GPRS|KVM_SYNC_ACRS|KVM_SYNC_CRS|KVM_SYNC_DIAG318)
 #define INVALID_SYNC_FIELD 0x80000000
 
 int main(int argc, char *argv[])
@@ -152,6 +153,12 @@ int main(int argc, char *argv[])
 
 	run->kvm_valid_regs = TEST_SYNC_FIELDS;
 	run->kvm_dirty_regs = KVM_SYNC_GPRS | KVM_SYNC_ACRS;
+
+	if (get_diag318_info() > 0) {
+		run->s.regs.diag318 = get_diag318_info();
+		run->kvm_dirty_regs |= KVM_SYNC_DIAG318;
+	}
+
 	rv = _vcpu_run(vm, VCPU_ID);
 	TEST_ASSERT(rv == 0, "vcpu_run failed: %d\n", rv);
 	TEST_ASSERT(run->exit_reason == KVM_EXIT_S390_SIEIC,
@@ -164,6 +171,9 @@ int main(int argc, char *argv[])
 	TEST_ASSERT(run->s.regs.acrs[0]  == 1 << 11,
 		    "acr0 sync regs value incorrect 0x%x.",
 		    run->s.regs.acrs[0]);
+	TEST_ASSERT(run->s.regs.diag318 == get_diag318_info(),
+		    "diag318 sync regs value incorrect 0x%llx.",
+		    run->s.regs.diag318);
 
 	vcpu_regs_get(vm, VCPU_ID, &regs);
 	compare_regs(&regs, &run->s.regs);
@@ -177,6 +187,7 @@ int main(int argc, char *argv[])
 	run->kvm_valid_regs = TEST_SYNC_FIELDS;
 	run->kvm_dirty_regs = 0;
 	run->s.regs.gprs[11] = 0xDEADBEEF;
+	run->s.regs.diag318 = 0x4B1D;
 	rv = _vcpu_run(vm, VCPU_ID);
 	TEST_ASSERT(rv == 0, "vcpu_run failed: %d\n", rv);
 	TEST_ASSERT(run->exit_reason == KVM_EXIT_S390_SIEIC,
@@ -186,6 +197,9 @@ int main(int argc, char *argv[])
 	TEST_ASSERT(run->s.regs.gprs[11] != 0xDEADBEEF,
 		    "r11 sync regs value incorrect 0x%llx.",
 		    run->s.regs.gprs[11]);
+	TEST_ASSERT(run->s.regs.diag318 != 0x4B1D,
+		    "diag318 sync regs value incorrect 0x%llx.",
+		    run->s.regs.diag318);
 
 	kvm_vm_free(vm);
 
-- 
2.28.0

[GIT PULL 4/4] KVM: s390: track synchronous pfault events in kvm_stat

[Christian Borntraeger]

Right now we do count pfault (pseudo page faults aka async page faults
start and completion events). What we do not count is, if an async page
fault would have been possible by the host, but it was disabled by the
guest (e.g. interrupts off, pfault disabled, secure execution....).  Let
us count those as well in the pfault_sync counter.

Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Link: https://lore.kernel.org/r/20201125090658.38463-1-borntraeger@de.ibm.com
---
 arch/s390/include/asm/kvm_host.h | 1 +
 arch/s390/kvm/kvm-s390.c         | 2 ++
 2 files changed, 3 insertions(+)

diff --git a/arch/s390/include/asm/kvm_host.h b/arch/s390/include/asm/kvm_host.h
index 463c24e26000..74f9a036bab2 100644
--- a/arch/s390/include/asm/kvm_host.h
+++ b/arch/s390/include/asm/kvm_host.h
@@ -459,6 +459,7 @@ struct kvm_vcpu_stat {
 	u64 diagnose_308;
 	u64 diagnose_500;
 	u64 diagnose_other;
+	u64 pfault_sync;
 };
 
 #define PGM_OPERATION			0x01
diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c
index 19804c388d61..065f94f22fd3 100644
--- a/arch/s390/kvm/kvm-s390.c
+++ b/arch/s390/kvm/kvm-s390.c
@@ -60,6 +60,7 @@
 struct kvm_stats_debugfs_item debugfs_entries[] = {
 	VCPU_STAT("userspace_handled", exit_userspace),
 	VCPU_STAT("exit_null", exit_null),
+	VCPU_STAT("pfault_sync", pfault_sync),
 	VCPU_STAT("exit_validity", exit_validity),
 	VCPU_STAT("exit_stop_request", exit_stop_request),
 	VCPU_STAT("exit_external_request", exit_external_request),
@@ -4111,6 +4112,7 @@ static int vcpu_post_run(struct kvm_vcpu *vcpu, int exit_reason)
 		current->thread.gmap_pfault = 0;
 		if (kvm_arch_setup_async_pf(vcpu))
 			return 0;
+		vcpu->stat.pfault_sync++;
 		return kvm_arch_fault_in_page(vcpu, current->thread.gmap_addr, 1);
 	}
 	return vcpu_post_run_fault_in_sie(vcpu);
-- 
2.28.0

Re: [GIT PULL 0/4] KVM: s390: features and test for 5.11

[Paolo Bonzini]

On 10/12/20 15:25, Christian Borntraeger wrote:
> Paolo,
> 
> the following changes since commit f8394f232b1eab649ce2df5c5f15b0e528c92091:
> 
>    Linux 5.10-rc3 (2020-11-08 16:10:16 -0800)
> 
> are available in the Git repository at:
> 
>    git://git.kernel.org/pub/scm/linux/kernel/git/kvms390/linux.git  tags/kvm-s390-next-5.11-1
> 
> for you to fetch changes up to 50a05be484cb70d9dfb55fa5a6ed57eab193901f:
> 
>    KVM: s390: track synchronous pfault events in kvm_stat (2020-12-10 14:20:26 +0100)
> 
> ----------------------------------------------------------------
> KVM: s390: Features and Test for 5.11
> 
> - memcg accouting for s390 specific parts of kvm and gmap
> - selftest for diag318
> - new kvm_stat for when async_pf falls back to sync
> 
> The selftest even triggers a non-critical bug that is unrelated
> to diag318, fix will follow later.
> 
> ----------------------------------------------------------------
> Christian Borntraeger (3):
>        KVM: s390: Add memcg accounting to KVM allocations
>        s390/gmap: make gmap memcg aware
>        KVM: s390: track synchronous pfault events in kvm_stat
> 
> Collin Walling (1):
>        KVM: selftests: sync_regs test for diag318
> 
>   arch/s390/include/asm/kvm_host.h                   |  1 +
>   arch/s390/kvm/guestdbg.c                           |  8 +--
>   arch/s390/kvm/intercept.c                          |  2 +-
>   arch/s390/kvm/interrupt.c                          | 10 +--
>   arch/s390/kvm/kvm-s390.c                           | 22 +++---
>   arch/s390/kvm/priv.c                               |  4 +-
>   arch/s390/kvm/pv.c                                 |  6 +-
>   arch/s390/kvm/vsie.c                               |  4 +-
>   arch/s390/mm/gmap.c                                | 30 ++++----
>   tools/testing/selftests/kvm/Makefile               |  2 +-
>   .../kvm/include/s390x/diag318_test_handler.h       | 13 ++++
>   .../selftests/kvm/lib/s390x/diag318_test_handler.c | 82 ++++++++++++++++++++++
>   tools/testing/selftests/kvm/s390x/sync_regs_test.c | 16 ++++-
>   13 files changed, 156 insertions(+), 44 deletions(-)
>   create mode 100644 tools/testing/selftests/kvm/include/s390x/diag318_test_handler.h
>   create mode 100644 tools/testing/selftests/kvm/lib/s390x/diag318_test_handler.c
> 

Pulled, thanks.

Paolo