01-patch-3.10.103-104

Signed-off-by: Simon Eisenmann <simon@longsleep.org>

Makefile

@@ -1,6 +1,6 @@
 VERSION = 3
 PATCHLEVEL = 10
-SUBLEVEL = 103
+SUBLEVEL = 104
 EXTRAVERSION =
 NAME = TOSSUG Baby Fish
 

arch/mips/kvm/kvm_tlb.c

@@ -182,7 +182,7 @@ static int kvm_mips_map_page(struct kvm *kvm, gfn_t gfn)
         srcu_idx = srcu_read_lock(&kvm->srcu);
 	pfn = kvm_mips_gfn_to_pfn(kvm, gfn);
 
-	if (kvm_mips_is_error_pfn(pfn)) {
+	if (is_error_noslot_pfn(pfn)) {
 		kvm_err("Couldn't get pfn for gfn %#" PRIx64 "!\n", gfn);
 		err = -EFAULT;
 		goto out;

arch/powerpc/kernel/process.c

@@ -1088,16 +1088,6 @@ void start_thread(struct pt_regs *regs, unsigned long start, unsigned long sp)
 		current->thread.regs = regs - 1;
 	}
 
-#ifdef CONFIG_PPC_TRANSACTIONAL_MEM
-	/*
-	 * Clear any transactional state, we're exec()ing. The cause is
-	 * not important as there will never be a recheckpoint so it's not
-	 * user visible.
-	 */
-	if (MSR_TM_SUSPENDED(mfmsr()))
-		tm_reclaim_current(0);
-#endif
-
 	memset(regs->gpr, 0, sizeof(regs->gpr));
 	regs->ctr = 0;
 	regs->link = 0;

drivers/acpi/sysfs.c

@@ -492,23 +492,22 @@ static void acpi_global_event_handler(u32 event_type, acpi_handle device,
 static int get_status(u32 index, acpi_event_status *status,
 		      acpi_handle *handle)
 {
-	int result = 0;
+	int result;
 
 	if (index >= num_gpes + ACPI_NUM_FIXED_EVENTS)
-		goto end;
+		return -EINVAL;
 
 	if (index < num_gpes) {
 		result = acpi_get_gpe_device(index, handle);
 		if (result) {
 			ACPI_EXCEPTION((AE_INFO, AE_NOT_FOUND,
 					"Invalid GPE 0x%x", index));
-			goto end;
+			return result;
 		}
 		result = acpi_get_gpe_status(*handle, index, status);
 	} else if (index < (num_gpes + ACPI_NUM_FIXED_EVENTS))
 		result = acpi_get_event_status(index - num_gpes, status);
 
-end:
 	return result;
 }
 

drivers/crypto/nx/nx.c

@@ -309,7 +309,7 @@ static void nx_of_update_msc(struct device   *dev,
 		     ((bytes_so_far + sizeof(struct msc_triplet)) <= lenp) &&
 		     i < msc->triplets;
 		     i++) {
-			if (msc->fc > NX_MAX_FC || msc->mode > NX_MAX_MODE) {
+			if (msc->fc >= NX_MAX_FC || msc->mode >= NX_MAX_MODE) {
 				dev_err(dev, "unknown function code/mode "
 					"combo: %d/%d (ignored)\n", msc->fc,
 					msc->mode);

drivers/net/xen-netback/common.h

@@ -66,6 +66,8 @@ struct xenvif {
 	/* The shared rings and indexes. */
 	struct xen_netif_tx_back_ring tx;
 	struct xen_netif_rx_back_ring rx;
+	atomic_t ring_refcnt;
+	wait_queue_head_t waiting_to_unmap;
 
 	/* Frontend feature information. */
 	u8 can_sg:1;
@@ -120,6 +122,8 @@ void xenvif_free(struct xenvif *vif);
 
 void xenvif_get(struct xenvif *vif);
 void xenvif_put(struct xenvif *vif);
+void xenvif_get_rings(struct xenvif *vif);
+void xenvif_put_rings(struct xenvif *vif);
 
 int xenvif_xenbus_init(void);
 

drivers/net/xen-netback/interface.c

@@ -44,12 +44,23 @@ void xenvif_get(struct xenvif *vif)
 	atomic_inc(&vif->refcnt);
 }
 
+void xenvif_get_rings(struct xenvif *vif)
+{
+	atomic_inc(&vif->ring_refcnt);
+}
+
 void xenvif_put(struct xenvif *vif)
 {
 	if (atomic_dec_and_test(&vif->refcnt))
 		wake_up(&vif->waiting_to_free);
 }
 
+void xenvif_put_rings(struct xenvif *vif)
+{
+	if (atomic_dec_and_test(&vif->ring_refcnt))
+		wake_up(&vif->waiting_to_unmap);
+}
+
 int xenvif_schedulable(struct xenvif *vif)
 {
 	return netif_running(vif->dev) && netif_carrier_ok(vif->dev);
@@ -91,6 +102,7 @@ static int xenvif_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	/* Reserve ring slots for the worst-case number of fragments. */
 	vif->rx_req_cons_peek += xen_netbk_count_skb_slots(vif, skb);
 	xenvif_get(vif);
+	xenvif_get_rings(vif);
 
 	if (vif->can_queue && xen_netbk_must_stop_queue(vif))
 		netif_stop_queue(dev);
@@ -271,6 +283,7 @@ struct xenvif *xenvif_alloc(struct device *parent, domid_t domid,
 	vif->dev = dev;
 	INIT_LIST_HEAD(&vif->schedule_list);
 	INIT_LIST_HEAD(&vif->notify_list);
+	init_waitqueue_head(&vif->waiting_to_unmap);
 
 	vif->credit_bytes = vif->remaining_credit = ~0UL;
 	vif->credit_usec  = 0UL;
@@ -365,12 +378,12 @@ void xenvif_disconnect(struct xenvif *vif)
 	if (netif_carrier_ok(vif->dev))
 		xenvif_carrier_off(vif);
 
+	disable_irq(vif->irq);
+	xen_netbk_unmap_frontend_rings(vif);
 	if (vif->irq) {
 		unbind_from_irqhandler(vif->irq, vif);
 		vif->irq = 0;
 	}
-
-	xen_netbk_unmap_frontend_rings(vif);
 }
 
 void xenvif_free(struct xenvif *vif)

drivers/net/xen-netback/netback.c

@@ -814,6 +814,7 @@ static void xen_netbk_rx_action(struct xen_netbk *netbk)
 			xenvif_put(vif);
 		npo.meta_cons += sco->meta_slots_used;
 		dev_kfree_skb(skb);
+		xenvif_put_rings(vif);
 	}
 
 	list_for_each_entry_safe(vif, tmp, &notify, notify_list) {
@@ -1864,6 +1865,9 @@ static int xen_netbk_kthread(void *data)
 
 void xen_netbk_unmap_frontend_rings(struct xenvif *vif)
 {
+	atomic_dec(&vif->ring_refcnt);
+	wait_event(vif->waiting_to_unmap, atomic_read(&vif->ring_refcnt) == 0);
+
 	if (vif->tx.sring)
 		xenbus_unmap_ring_vfree(xenvif_to_xenbus_device(vif),
 					vif->tx.sring);
@@ -1882,6 +1886,8 @@ int xen_netbk_map_frontend_rings(struct xenvif *vif,
 
 	int err = -ENOMEM;
 
+	atomic_set(&vif->ring_refcnt, 1);
+
 	err = xenbus_map_ring_valloc(xenvif_to_xenbus_device(vif),
 				     tx_ring_ref, &addr);
 	if (err)

drivers/pci/pci-sysfs.c

@@ -1308,10 +1308,10 @@ int __must_check pci_create_sysfs_dev_files (struct pci_dev *pdev)
 	if (!sysfs_initialized)
 		return -EACCES;
 
-	if (pdev->cfg_size < PCI_CFG_SPACE_EXP_SIZE)
-		retval = sysfs_create_bin_file(&pdev->dev.kobj, &pci_config_attr);
-	else
+	if (pdev->cfg_size > PCI_CFG_SPACE_SIZE)
 		retval = sysfs_create_bin_file(&pdev->dev.kobj, &pcie_config_attr);
+	else
+		retval = sysfs_create_bin_file(&pdev->dev.kobj, &pci_config_attr);
 	if (retval)
 		goto err;
 
@@ -1368,10 +1368,10 @@ int __must_check pci_create_sysfs_dev_files (struct pci_dev *pdev)
 err_resource_files:
 	pci_remove_resource_files(pdev);
 err_config_file:
-	if (pdev->cfg_size < PCI_CFG_SPACE_EXP_SIZE)
-		sysfs_remove_bin_file(&pdev->dev.kobj, &pci_config_attr);
-	else
+	if (pdev->cfg_size > PCI_CFG_SPACE_SIZE)
 		sysfs_remove_bin_file(&pdev->dev.kobj, &pcie_config_attr);
+	else
+		sysfs_remove_bin_file(&pdev->dev.kobj, &pci_config_attr);
 err:
 	return retval;
 }
@@ -1405,10 +1405,10 @@ void pci_remove_sysfs_dev_files(struct pci_dev *pdev)
 
 	pci_remove_capabilities_sysfs(pdev);
 
-	if (pdev->cfg_size < PCI_CFG_SPACE_EXP_SIZE)
-		sysfs_remove_bin_file(&pdev->dev.kobj, &pci_config_attr);
-	else
+	if (pdev->cfg_size > PCI_CFG_SPACE_SIZE)
 		sysfs_remove_bin_file(&pdev->dev.kobj, &pcie_config_attr);
+	else
+		sysfs_remove_bin_file(&pdev->dev.kobj, &pci_config_attr);
 
 	pci_remove_resource_files(pdev);
 

drivers/pci/quirks.c

@@ -292,6 +292,18 @@ static void quirk_citrine(struct pci_dev *dev)
 }
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_IBM,	PCI_DEVICE_ID_IBM_CITRINE,	quirk_citrine);
 
+/*
+ * This chip can cause bus lockups if config addresses above 0x600
+ * are read or written.
+ */
+static void quirk_nfp6000(struct pci_dev *dev)
+{
+	dev->cfg_size = 0x600;
+}
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_NETRONOME,	PCI_DEVICE_ID_NETRONOME_NFP4000,	quirk_nfp6000);
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_NETRONOME,	PCI_DEVICE_ID_NETRONOME_NFP6000,	quirk_nfp6000);
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_NETRONOME,	PCI_DEVICE_ID_NETRONOME_NFP6000_VF,	quirk_nfp6000);
+
 /*  On IBM Crocodile ipr SAS adapters, expand BAR to system page size */
 static void quirk_extend_bar_to_page(struct pci_dev *dev)
 {

drivers/scsi/aacraid/commctrl.c

@@ -63,7 +63,7 @@ static int ioctl_send_fib(struct aac_dev * dev, void __user *arg)
 	struct fib *fibptr;
 	struct hw_fib * hw_fib = (struct hw_fib *)0;
 	dma_addr_t hw_fib_pa = (dma_addr_t)0LL;
-	unsigned size;
+	unsigned int size, osize;
 	int retval;
 
 	if (dev->in_reset) {
@@ -87,7 +87,8 @@ static int ioctl_send_fib(struct aac_dev * dev, void __user *arg)
 	 *	will not overrun the buffer when we copy the memory. Return
 	 *	an error if we would.
 	 */
-	size = le16_to_cpu(kfib->header.Size) + sizeof(struct aac_fibhdr);
+	osize = size = le16_to_cpu(kfib->header.Size) +
+		sizeof(struct aac_fibhdr);
 	if (size < le16_to_cpu(kfib->header.SenderSize))
 		size = le16_to_cpu(kfib->header.SenderSize);
 	if (size > dev->max_fib_size) {
@@ -118,6 +119,14 @@ static int ioctl_send_fib(struct aac_dev * dev, void __user *arg)
 		goto cleanup;
 	}
 
+	/* Sanity check the second copy */
+	if ((osize != le16_to_cpu(kfib->header.Size) +
+		sizeof(struct aac_fibhdr))
+		|| (size < le16_to_cpu(kfib->header.SenderSize))) {
+		retval = -EINVAL;
+		goto cleanup;
+	}
+
 	if (kfib->header.Command == cpu_to_le16(TakeABreakPt)) {
 		aac_adapter_interrupt(dev);
 		/*

drivers/scsi/megaraid/megaraid_sas_base.c

@@ -3470,7 +3470,7 @@ static int megasas_init_fw(struct megasas_instance *instance)
 	/* Find first memory bar */
 	bar_list = pci_select_bars(instance->pdev, IORESOURCE_MEM);
 	instance->bar = find_first_bit(&bar_list, sizeof(unsigned long));
-	if (pci_request_selected_regions(instance->pdev, instance->bar,
+	if (pci_request_selected_regions(instance->pdev, 1<<instance->bar,
 					 "megasas: LSI")) {
 		printk(KERN_DEBUG "megasas: IO memory region busy!\n");
 		return -EBUSY;
@@ -3640,7 +3640,7 @@ static int megasas_init_fw(struct megasas_instance *instance)
 	iounmap(instance->reg_set);
 
       fail_ioremap:
-	pci_release_selected_regions(instance->pdev, instance->bar);
+	pci_release_selected_regions(instance->pdev, 1<<instance->bar);
 
 	return -EINVAL;
 }
@@ -3661,7 +3661,7 @@ static void megasas_release_mfi(struct megasas_instance *instance)
 
 	iounmap(instance->reg_set);
 
-	pci_release_selected_regions(instance->pdev, instance->bar);
+	pci_release_selected_regions(instance->pdev, 1<<instance->bar);
 }
 
 /**

drivers/scsi/megaraid/megaraid_sas_fusion.c

@@ -2020,7 +2020,7 @@ megasas_release_fusion(struct megasas_instance *instance)
 
 	iounmap(instance->reg_set);
 
-	pci_release_selected_regions(instance->pdev, instance->bar);
+	pci_release_selected_regions(instance->pdev, 1<<instance->bar);
 }
 
 /**

drivers/staging/comedi/drivers/daqboard2000.c

@@ -678,7 +678,7 @@ static const void *daqboard2000_find_boardinfo(struct comedi_device *dev,
 	const struct daq200_boardtype *board;
 	int i;
 
-	if (pcidev->subsystem_device != PCI_VENDOR_ID_IOTECH)
+	if (pcidev->subsystem_vendor != PCI_VENDOR_ID_IOTECH)
 		return NULL;
 
 	for (i = 0; i < ARRAY_SIZE(boardtypes); i++) {

include/linux/mm.h

@@ -1743,6 +1743,7 @@ static inline struct page *follow_page(struct vm_area_struct *vma,
 #define FOLL_HWPOISON	0x100	/* check page is hwpoisoned */
 #define FOLL_NUMA	0x200	/* force NUMA hinting page fault */
 #define FOLL_MIGRATION	0x400	/* wait for page to replace migration entry */
+#define FOLL_COW	0x4000	/* internal GUP flag */
 
 typedef int (*pte_fn_t)(pte_t *pte, pgtable_t token, unsigned long addr,
 			void *data);

include/linux/pci_ids.h

@@ -2472,6 +2472,13 @@
 #define PCI_DEVICE_ID_KORENIX_JETCARDF2	0x1700
 #define PCI_DEVICE_ID_KORENIX_JETCARDF3	0x17ff
 
+#define PCI_VENDOR_ID_NETRONOME		0x19ee
+#define PCI_DEVICE_ID_NETRONOME_NFP3200	0x3200
+#define PCI_DEVICE_ID_NETRONOME_NFP3240	0x3240
+#define PCI_DEVICE_ID_NETRONOME_NFP4000	0x4000
+#define PCI_DEVICE_ID_NETRONOME_NFP6000	0x6000
+#define PCI_DEVICE_ID_NETRONOME_NFP6000_VF	0x6003
+
 #define PCI_VENDOR_ID_QMI		0x1a32
 
 #define PCI_VENDOR_ID_AZWAVE		0x1a3b

mm/memory.c

@@ -1464,6 +1464,16 @@ int zap_vma_ptes(struct vm_area_struct *vma, unsigned long address,
 }
 EXPORT_SYMBOL_GPL(zap_vma_ptes);
 
+/*
+ * FOLL_FORCE can write to even unwritable pte's, but only
+ * after we've gone through a COW cycle and they are dirty.
+ */
+static inline bool can_follow_write_pte(pte_t pte, unsigned int flags)
+{
+	return pte_write(pte) ||
+		((flags & FOLL_FORCE) && (flags & FOLL_COW) && pte_dirty(pte));
+}
+
 /**
  * follow_page_mask - look up a page descriptor from a user-virtual address
  * @vma: vm_area_struct mapping @address
@@ -1571,7 +1581,7 @@ struct page *follow_page_mask(struct vm_area_struct *vma,
 	}
 	if ((flags & FOLL_NUMA) && pte_numa(pte))
 		goto no_page;
-	if ((flags & FOLL_WRITE) && !pte_write(pte))
+	if ((flags & FOLL_WRITE) && !can_follow_write_pte(pte, flags))
 		goto unlock;
 
 	page = vm_normal_page(vma, address, pte);
@@ -1879,7 +1889,7 @@ long __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
 				 */
 				if ((ret & VM_FAULT_WRITE) &&
 				    !(vma->vm_flags & VM_WRITE))
-					foll_flags &= ~FOLL_WRITE;
+					foll_flags |= FOLL_COW;
 
 				cond_resched();
 			}
@@ -3836,8 +3846,18 @@ static int __handle_mm_fault(struct mm_struct *mm, struct vm_area_struct *vma,
 	if (unlikely(pmd_none(*pmd)) &&
 	    unlikely(__pte_alloc(mm, vma, pmd, address)))
 		return VM_FAULT_OOM;
-	/* if an huge pmd materialized from under us just retry later */
-	if (unlikely(pmd_trans_huge(*pmd)))
+	/*
+	 * If a huge pmd materialized under us just retry later.  Use
+	 * pmd_trans_unstable() instead of pmd_trans_huge() to ensure the pmd
+	 * didn't become pmd_trans_huge under us and then back to pmd_none, as
+	 * a result of MADV_DONTNEED running immediately after a huge pmd fault
+	 * in a different thread of this mm, in turn leading to a misleading
+	 * pmd_trans_huge() retval.  All we have to ensure is that it is a
+	 * regular pmd that we can walk with pte_offset_map() and we can do that
+	 * through an atomic read in C, which is what pmd_trans_unstable()
+	 * provides.
+	 */
+	if (unlikely(pmd_trans_unstable(pmd)))
 		return 0;
 	/*
 	 * A regular pmd is established and it can't morph into a huge pmd