Merge tag 'arm-perf-4.1' of git://git.kernel.org/pub/scm/linux/kernel/git/will/linux into next/drivers

Merge "arm-cci PMU updates for 4.1" from Will Deacon:

CCI-400 PMU updates

This series reworks some of the CCI-400 PMU code so that it can be used
on both ARM and ARM64-based systems, without the need to boot in secure
mode on the latter. This paves the way for CCI-500 support in future.

* tag 'arm-perf-4.1' of git://git.kernel.org/pub/scm/linux/kernel/git/will/linux:
  arm-cci: Fix CCI PMU event validation
  arm-cci: Split the code for PMU vs driver support
  arm-cci: Get rid of secure transactions for PMU driver
  arm-cci: Abstract the CCI400 PMU specific definitions
  arm-cci: Rearrange code for splitting PMU vs driver code
  drivers: cci: reject groups spanning multiple HW PMUs
  + Linux 4.0-rc4

Signed-off-by: Olof Johansson <olof@lixom.net>

11 files changed, 76 insertions, 82 deletions

diff --git a/arch/s390/include/asm/kvm_host.h b/arch/s390/include/asm/kvm_host.h
index d84559e31f32..f407bbf5ee94 100644
--- a/arch/s390/include/asm/kvm_host.h
+++ b/arch/s390/include/asm/kvm_host.h
@@ -515,15 +515,15 @@ struct s390_io_adapter {
 #define S390_ARCH_FAC_MASK_SIZE_U64 \
 	(S390_ARCH_FAC_MASK_SIZE_BYTE / sizeof(u64))
 
-struct s390_model_fac {
-	/* facilities used in SIE context */
-	__u64 sie[S390_ARCH_FAC_LIST_SIZE_U64];
-	/* subset enabled by kvm */
-	__u64 kvm[S390_ARCH_FAC_LIST_SIZE_U64];
+struct kvm_s390_fac {
+	/* facility list requested by guest */
+	__u64 list[S390_ARCH_FAC_LIST_SIZE_U64];
+	/* facility mask supported by kvm & hosting machine */
+	__u64 mask[S390_ARCH_FAC_LIST_SIZE_U64];
 };
 
 struct kvm_s390_cpu_model {
-	struct s390_model_fac *fac;
+	struct kvm_s390_fac *fac;
 	struct cpuid cpu_id;
 	unsigned short ibc;
 };
diff --git a/arch/s390/include/asm/mmu_context.h b/arch/s390/include/asm/mmu_context.h
index f49b71954654..8fb3802f8fad 100644
--- a/arch/s390/include/asm/mmu_context.h
+++ b/arch/s390/include/asm/mmu_context.h
@@ -62,6 +62,7 @@ static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next,
 {
 	int cpu = smp_processor_id();
 
+	S390_lowcore.user_asce = next->context.asce_bits | __pa(next->pgd);
 	if (prev == next)
 		return;
 	if (MACHINE_HAS_TLB_LC)
@@ -73,7 +74,6 @@ static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next,
 	atomic_dec(&prev->context.attach_count);
 	if (MACHINE_HAS_TLB_LC)
 		cpumask_clear_cpu(cpu, &prev->context.cpu_attach_mask);
-	S390_lowcore.user_asce = next->context.asce_bits | __pa(next->pgd);
 }
 
 #define finish_arch_post_lock_switch finish_arch_post_lock_switch
diff --git a/arch/s390/include/asm/page.h b/arch/s390/include/asm/page.h
index 7b2ac6e44166..53eacbd4f09b 100644
--- a/arch/s390/include/asm/page.h
+++ b/arch/s390/include/asm/page.h
@@ -37,16 +37,7 @@ static inline void storage_key_init_range(unsigned long start, unsigned long end
 #endif
 }
 
-static inline void clear_page(void *page)
-{
-	register unsigned long reg1 asm ("1") = 0;
-	register void *reg2 asm ("2") = page;
-	register unsigned long reg3 asm ("3") = 4096;
-	asm volatile(
-		"	mvcl	2,0"
-		: "+d" (reg2), "+d" (reg3) : "d" (reg1)
-		: "memory", "cc");
-}
+#define clear_page(page)	memset((page), 0, PAGE_SIZE)
 
 /*
  * copy_page uses the mvcl instruction with 0xb0 padding byte in order to
diff --git a/arch/s390/kernel/jump_label.c b/arch/s390/kernel/jump_label.c
index cb2d51e779df..830066f936c8 100644
--- a/arch/s390/kernel/jump_label.c
+++ b/arch/s390/kernel/jump_label.c
@@ -36,16 +36,20 @@ static void jump_label_make_branch(struct jump_entry *entry, struct insn *insn)
 	insn->offset = (entry->target - entry->code) >> 1;
 }
 
-static void jump_label_bug(struct jump_entry *entry, struct insn *insn)
+static void jump_label_bug(struct jump_entry *entry, struct insn *expected,
+			   struct insn *new)
 {
 	unsigned char *ipc = (unsigned char *)entry->code;
-	unsigned char *ipe = (unsigned char *)insn;
+	unsigned char *ipe = (unsigned char *)expected;
+	unsigned char *ipn = (unsigned char *)new;
 
 	pr_emerg("Jump label code mismatch at %pS [%p]\n", ipc, ipc);
 	pr_emerg("Found:    %02x %02x %02x %02x %02x %02x\n",
 		 ipc[0], ipc[1], ipc[2], ipc[3], ipc[4], ipc[5]);
 	pr_emerg("Expected: %02x %02x %02x %02x %02x %02x\n",
 		 ipe[0], ipe[1], ipe[2], ipe[3], ipe[4], ipe[5]);
+	pr_emerg("New:      %02x %02x %02x %02x %02x %02x\n",
+		 ipn[0], ipn[1], ipn[2], ipn[3], ipn[4], ipn[5]);
 	panic("Corrupted kernel text");
 }
 
@@ -69,10 +73,10 @@ static void __jump_label_transform(struct jump_entry *entry,
 	}
 	if (init) {
 		if (memcmp((void *)entry->code, &orignop, sizeof(orignop)))
-			jump_label_bug(entry, &old);
+			jump_label_bug(entry, &orignop, &new);
 	} else {
 		if (memcmp((void *)entry->code, &old, sizeof(old)))
-			jump_label_bug(entry, &old);
+			jump_label_bug(entry, &old, &new);
 	}
 	probe_kernel_write((void *)entry->code, &new, sizeof(new));
 }
diff --git a/arch/s390/kernel/module.c b/arch/s390/kernel/module.c
index 36154a2f1814..2ca95862e336 100644
--- a/arch/s390/kernel/module.c
+++ b/arch/s390/kernel/module.c
@@ -436,6 +436,7 @@ int module_finalize(const Elf_Ehdr *hdr,
 		    const Elf_Shdr *sechdrs,
 		    struct module *me)
 {
+	jump_label_apply_nops(me);
 	vfree(me->arch.syminfo);
 	me->arch.syminfo = NULL;
 	return 0;
diff --git a/arch/s390/kernel/processor.c b/arch/s390/kernel/processor.c
index 26108232fcaa..dc488e13b7e3 100644
--- a/arch/s390/kernel/processor.c
+++ b/arch/s390/kernel/processor.c
@@ -18,7 +18,7 @@
 
 static DEFINE_PER_CPU(struct cpuid, cpu_id);
 
-void cpu_relax(void)
+void notrace cpu_relax(void)
 {
 	if (!smp_cpu_mtid && MACHINE_HAS_DIAG44)
 		asm volatile("diag 0,0,0x44");
diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c
index 0c3623927563..f6579cfde2df 100644
--- a/arch/s390/kvm/kvm-s390.c
+++ b/arch/s390/kvm/kvm-s390.c
@@ -522,7 +522,7 @@ static int kvm_s390_set_processor(struct kvm *kvm, struct kvm_device_attr *attr)
 		memcpy(&kvm->arch.model.cpu_id, &proc->cpuid,
 		       sizeof(struct cpuid));
 		kvm->arch.model.ibc = proc->ibc;
-		memcpy(kvm->arch.model.fac->kvm, proc->fac_list,
+		memcpy(kvm->arch.model.fac->list, proc->fac_list,
 		       S390_ARCH_FAC_LIST_SIZE_BYTE);
 	} else
 		ret = -EFAULT;
@@ -556,7 +556,7 @@ static int kvm_s390_get_processor(struct kvm *kvm, struct kvm_device_attr *attr)
 	}
 	memcpy(&proc->cpuid, &kvm->arch.model.cpu_id, sizeof(struct cpuid));
 	proc->ibc = kvm->arch.model.ibc;
-	memcpy(&proc->fac_list, kvm->arch.model.fac->kvm, S390_ARCH_FAC_LIST_SIZE_BYTE);
+	memcpy(&proc->fac_list, kvm->arch.model.fac->list, S390_ARCH_FAC_LIST_SIZE_BYTE);
 	if (copy_to_user((void __user *)attr->addr, proc, sizeof(*proc)))
 		ret = -EFAULT;
 	kfree(proc);
@@ -576,10 +576,10 @@ static int kvm_s390_get_machine(struct kvm *kvm, struct kvm_device_attr *attr)
 	}
 	get_cpu_id((struct cpuid *) &mach->cpuid);
 	mach->ibc = sclp_get_ibc();
-	memcpy(&mach->fac_mask, kvm_s390_fac_list_mask,
-	       kvm_s390_fac_list_mask_size() * sizeof(u64));
+	memcpy(&mach->fac_mask, kvm->arch.model.fac->mask,
+	       S390_ARCH_FAC_LIST_SIZE_BYTE);
 	memcpy((unsigned long *)&mach->fac_list, S390_lowcore.stfle_fac_list,
-	       S390_ARCH_FAC_LIST_SIZE_U64);
+	       S390_ARCH_FAC_LIST_SIZE_BYTE);
 	if (copy_to_user((void __user *)attr->addr, mach, sizeof(*mach)))
 		ret = -EFAULT;
 	kfree(mach);
@@ -778,15 +778,18 @@ long kvm_arch_vm_ioctl(struct file *filp,
 static int kvm_s390_query_ap_config(u8 *config)
 {
 	u32 fcn_code = 0x04000000UL;
-	u32 cc;
+	u32 cc = 0;
 
+	memset(config, 0, 128);
 	asm volatile(
 		"lgr 0,%1\n"
 		"lgr 2,%2\n"
 		".long 0xb2af0000\n"		/* PQAP(QCI) */
-		"ipm %0\n"
+		"0: ipm %0\n"
 		"srl %0,28\n"
-		: "=r" (cc)
+		"1:\n"
+		EX_TABLE(0b, 1b)
+		: "+r" (cc)
 		: "r" (fcn_code), "r" (config)
 		: "cc", "0", "2", "memory"
 	);
@@ -839,9 +842,13 @@ static int kvm_s390_crypto_init(struct kvm *kvm)
 
 	kvm_s390_set_crycb_format(kvm);
 
-	/* Disable AES/DEA protected key functions by default */
-	kvm->arch.crypto.aes_kw = 0;
-	kvm->arch.crypto.dea_kw = 0;
+	/* Enable AES/DEA protected key functions by default */
+	kvm->arch.crypto.aes_kw = 1;
+	kvm->arch.crypto.dea_kw = 1;
+	get_random_bytes(kvm->arch.crypto.crycb->aes_wrapping_key_mask,
+			 sizeof(kvm->arch.crypto.crycb->aes_wrapping_key_mask));
+	get_random_bytes(kvm->arch.crypto.crycb->dea_wrapping_key_mask,
+			 sizeof(kvm->arch.crypto.crycb->dea_wrapping_key_mask));
 
 	return 0;
 }
@@ -886,40 +893,29 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 	/*
 	 * The architectural maximum amount of facilities is 16 kbit. To store
 	 * this amount, 2 kbyte of memory is required. Thus we need a full
-	 * page to hold the active copy (arch.model.fac->sie) and the current
-	 * facilities set (arch.model.fac->kvm). Its address size has to be
+	 * page to hold the guest facility list (arch.model.fac->list) and the
+	 * facility mask (arch.model.fac->mask). Its address size has to be
 	 * 31 bits and word aligned.
 	 */
 	kvm->arch.model.fac =
-		(struct s390_model_fac *) get_zeroed_page(GFP_KERNEL | GFP_DMA);
+		(struct kvm_s390_fac *) get_zeroed_page(GFP_KERNEL | GFP_DMA);
 	if (!kvm->arch.model.fac)
 		goto out_nofac;
 
-	memcpy(kvm->arch.model.fac->kvm, S390_lowcore.stfle_fac_list,
-	       S390_ARCH_FAC_LIST_SIZE_U64);
-
-	/*
-	 * If this KVM host runs *not* in a LPAR, relax the facility bits
-	 * of the kvm facility mask by all missing facilities. This will allow
-	 * to determine the right CPU model by means of the remaining facilities.
-	 * Live guest migration must prohibit the migration of KVMs running in
-	 * a LPAR to non LPAR hosts.
-	 */
-	if (!MACHINE_IS_LPAR)
-		for (i = 0; i < kvm_s390_fac_list_mask_size(); i++)
-			kvm_s390_fac_list_mask[i] &= kvm->arch.model.fac->kvm[i];
-
-	/*
-	 * Apply the kvm facility mask to limit the kvm supported/tolerated
-	 * facility list.
-	 */
+	/* Populate the facility mask initially. */
+	memcpy(kvm->arch.model.fac->mask, S390_lowcore.stfle_fac_list,
+	       S390_ARCH_FAC_LIST_SIZE_BYTE);
 	for (i = 0; i < S390_ARCH_FAC_LIST_SIZE_U64; i++) {
 		if (i < kvm_s390_fac_list_mask_size())
-			kvm->arch.model.fac->kvm[i] &= kvm_s390_fac_list_mask[i];
+			kvm->arch.model.fac->mask[i] &= kvm_s390_fac_list_mask[i];
 		else
-			kvm->arch.model.fac->kvm[i] = 0UL;
+			kvm->arch.model.fac->mask[i] = 0UL;
 	}
 
+	/* Populate the facility list initially. */
+	memcpy(kvm->arch.model.fac->list, kvm->arch.model.fac->mask,
+	       S390_ARCH_FAC_LIST_SIZE_BYTE);
+
 	kvm_s390_get_cpu_id(&kvm->arch.model.cpu_id);
 	kvm->arch.model.ibc = sclp_get_ibc() & 0x0fff;
 
@@ -1165,8 +1161,6 @@ int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu)
 
 	mutex_lock(&vcpu->kvm->lock);
 	vcpu->arch.cpu_id = vcpu->kvm->arch.model.cpu_id;
-	memcpy(vcpu->kvm->arch.model.fac->sie, vcpu->kvm->arch.model.fac->kvm,
-	       S390_ARCH_FAC_LIST_SIZE_BYTE);
 	vcpu->arch.sie_block->ibc = vcpu->kvm->arch.model.ibc;
 	mutex_unlock(&vcpu->kvm->lock);
 
@@ -1212,7 +1206,7 @@ struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm,
 		vcpu->arch.sie_block->scaol = (__u32)(__u64)kvm->arch.sca;
 		set_bit(63 - id, (unsigned long *) &kvm->arch.sca->mcn);
 	}
-	vcpu->arch.sie_block->fac = (int) (long) kvm->arch.model.fac->sie;
+	vcpu->arch.sie_block->fac = (int) (long) kvm->arch.model.fac->list;
 
 	spin_lock_init(&vcpu->arch.local_int.lock);
 	vcpu->arch.local_int.float_int = &kvm->arch.float_int;
diff --git a/arch/s390/kvm/kvm-s390.h b/arch/s390/kvm/kvm-s390.h
index 985c2114d7ef..c34109aa552d 100644
--- a/arch/s390/kvm/kvm-s390.h
+++ b/arch/s390/kvm/kvm-s390.h
@@ -128,7 +128,8 @@ static inline void kvm_s390_set_psw_cc(struct kvm_vcpu *vcpu, unsigned long cc)
 /* test availability of facility in a kvm intance */
 static inline int test_kvm_facility(struct kvm *kvm, unsigned long nr)
 {
-	return __test_facility(nr, kvm->arch.model.fac->kvm);
+	return __test_facility(nr, kvm->arch.model.fac->mask) &&
+		__test_facility(nr, kvm->arch.model.fac->list);
 }
 
 /* are cpu states controlled by user space */
diff --git a/arch/s390/kvm/priv.c b/arch/s390/kvm/priv.c
index bdd9b5b17e03..351116939ea2 100644
--- a/arch/s390/kvm/priv.c
+++ b/arch/s390/kvm/priv.c
@@ -348,7 +348,7 @@ static int handle_stfl(struct kvm_vcpu *vcpu)
 	 * We need to shift the lower 32 facility bits (bit 0-31) from a u64
 	 * into a u32 memory representation. They will remain bits 0-31.
 	 */
-	fac = *vcpu->kvm->arch.model.fac->sie >> 32;
+	fac = *vcpu->kvm->arch.model.fac->list >> 32;
 	rc = write_guest_lc(vcpu, offsetof(struct _lowcore, stfl_fac_list),
 			    &fac, sizeof(fac));
 	if (rc)
diff --git a/arch/s390/pci/pci.c b/arch/s390/pci/pci.c
index 753a56731951..f0b85443e060 100644
--- a/arch/s390/pci/pci.c
+++ b/arch/s390/pci/pci.c
@@ -287,7 +287,7 @@ void __iomem *pci_iomap_range(struct pci_dev *pdev,
 	addr = ZPCI_IOMAP_ADDR_BASE | ((u64) idx << 48);
 	return (void __iomem *) addr + offset;
 }
-EXPORT_SYMBOL_GPL(pci_iomap_range);
+EXPORT_SYMBOL(pci_iomap_range);
 
 void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long maxlen)
 {
@@ -309,7 +309,7 @@ void pci_iounmap(struct pci_dev *pdev, void __iomem *addr)
 	}
 	spin_unlock(&zpci_iomap_lock);
 }
-EXPORT_SYMBOL_GPL(pci_iounmap);
+EXPORT_SYMBOL(pci_iounmap);
 
 static int pci_read(struct pci_bus *bus, unsigned int devfn, int where,
 		    int size, u32 *val)
@@ -483,9 +483,8 @@ void arch_teardown_msi_irqs(struct pci_dev *pdev)
 	airq_iv_free_bit(zpci_aisb_iv, zdev->aisb);
 }
 
-static void zpci_map_resources(struct zpci_dev *zdev)
+static void zpci_map_resources(struct pci_dev *pdev)
 {
-	struct pci_dev *pdev = zdev->pdev;
 	resource_size_t len;
 	int i;
 
@@ -499,9 +498,8 @@ static void zpci_map_resources(struct zpci_dev *zdev)
 	}
 }
 
-static void zpci_unmap_resources(struct zpci_dev *zdev)
+static void zpci_unmap_resources(struct pci_dev *pdev)
 {
-	struct pci_dev *pdev = zdev->pdev;
 	resource_size_t len;
 	int i;
 
@@ -651,7 +649,7 @@ int pcibios_add_device(struct pci_dev *pdev)
 
 	zdev->pdev = pdev;
 	pdev->dev.groups = zpci_attr_groups;
-	zpci_map_resources(zdev);
+	zpci_map_resources(pdev);
 
 	for (i = 0; i < PCI_BAR_COUNT; i++) {
 		res = &pdev->resource[i];
@@ -663,6 +661,11 @@ int pcibios_add_device(struct pci_dev *pdev)
 	return 0;
 }
 
+void pcibios_release_device(struct pci_dev *pdev)
+{
+	zpci_unmap_resources(pdev);
+}
+
 int pcibios_enable_device(struct pci_dev *pdev, int mask)
 {
 	struct zpci_dev *zdev = get_zdev(pdev);
@@ -670,7 +673,6 @@ int pcibios_enable_device(struct pci_dev *pdev, int mask)
 	zdev->pdev = pdev;
 	zpci_debug_init_device(zdev);
 	zpci_fmb_enable_device(zdev);
-	zpci_map_resources(zdev);
 
 	return pci_enable_resources(pdev, mask);
 }
@@ -679,7 +681,6 @@ void pcibios_disable_device(struct pci_dev *pdev)
 {
 	struct zpci_dev *zdev = get_zdev(pdev);
 
-	zpci_unmap_resources(zdev);
 	zpci_fmb_disable_device(zdev);
 	zpci_debug_exit_device(zdev);
 	zdev->pdev = NULL;
@@ -688,7 +689,8 @@ void pcibios_disable_device(struct pci_dev *pdev)
 #ifdef CONFIG_HIBERNATE_CALLBACKS
 static int zpci_restore(struct device *dev)
 {
-	struct zpci_dev *zdev = get_zdev(to_pci_dev(dev));
+	struct pci_dev *pdev = to_pci_dev(dev);
+	struct zpci_dev *zdev = get_zdev(pdev);
 	int ret = 0;
 
 	if (zdev->state != ZPCI_FN_STATE_ONLINE)
@@ -698,7 +700,7 @@ static int zpci_restore(struct device *dev)
 	if (ret)
 		goto out;
 
-	zpci_map_resources(zdev);
+	zpci_map_resources(pdev);
 	zpci_register_ioat(zdev, 0, zdev->start_dma + PAGE_OFFSET,
 			   zdev->start_dma + zdev->iommu_size - 1,
 			   (u64) zdev->dma_table);
@@ -709,12 +711,14 @@ out:
 
 static int zpci_freeze(struct device *dev)
 {
-	struct zpci_dev *zdev = get_zdev(to_pci_dev(dev));
+	struct pci_dev *pdev = to_pci_dev(dev);
+	struct zpci_dev *zdev = get_zdev(pdev);
 
 	if (zdev->state != ZPCI_FN_STATE_ONLINE)
 		return 0;
 
 	zpci_unregister_ioat(zdev, 0);
+	zpci_unmap_resources(pdev);
 	return clp_disable_fh(zdev);
 }
 
diff --git a/arch/s390/pci/pci_mmio.c b/arch/s390/pci/pci_mmio.c
index 8aa271b3d1ad..b1bb2b72302c 100644
--- a/arch/s390/pci/pci_mmio.c
+++ b/arch/s390/pci/pci_mmio.c
@@ -64,8 +64,7 @@ SYSCALL_DEFINE3(s390_pci_mmio_write, unsigned long, mmio_addr,
 	if (copy_from_user(buf, user_buffer, length))
 		goto out;
 
-	memcpy_toio(io_addr, buf, length);
-	ret = 0;
+	ret = zpci_memcpy_toio(io_addr, buf, length);
 out:
 	if (buf != local_buf)
 		kfree(buf);
@@ -98,16 +97,16 @@ SYSCALL_DEFINE3(s390_pci_mmio_read, unsigned long, mmio_addr,
 		goto out;
 	io_addr = (void __iomem *)((pfn << PAGE_SHIFT) | (mmio_addr & ~PAGE_MASK));
 
-	ret = -EFAULT;
-	if ((unsigned long) io_addr < ZPCI_IOMAP_ADDR_BASE)
+	if ((unsigned long) io_addr < ZPCI_IOMAP_ADDR_BASE) {
+		ret = -EFAULT;
 		goto out;
-
-	memcpy_fromio(buf, io_addr, length);
-
-	if (copy_to_user(user_buffer, buf, length))
+	}
+	ret = zpci_memcpy_fromio(buf, io_addr, length);
+	if (ret)
 		goto out;
+	if (copy_to_user(user_buffer, buf, length))
+		ret = -EFAULT;
 
-	ret = 0;
 out:
 	if (buf != local_buf)
 		kfree(buf);