More Xen updates

@@ -4,6 +4,7 @@
 #include <linux/threads.h>
 #include <linux/mm.h>		/* for struct page */
 #include <linux/pagemap.h>
+#include <linux/sched.h>
 
 #ifdef CONFIG_PARAVIRT
 #include <asm/paravirt.h>

712 lines not on this page · raw message

61 lines not on this page · raw message

@@ -29,7 +29,7 @@
      *   OUT: GMFN bases of extents that were allocated
      *   (NB. This command also updates the mach_to_phys translation table)
      */
-    GUEST_HANDLE(ulong) extent_start;
+    ulong extent_start;
 
     /* Number of extents, and size/alignment of each (2^extent_order pages). */
     unsigned long  nr_extents;

@@ -50,7 +50,6 @@
     domid_t        domid;
 
 };
-DEFINE_GUEST_HANDLE_STRUCT(xen_memory_reservation);
 
 /*
  * Returns the maximum machine frame number of mapped RAM in this system.

@@ -86,7 +85,7 @@
      * any large discontiguities in the machine address space, 2MB gaps in
      * the machphys table will be represented by an MFN base of zero.
      */
-    GUEST_HANDLE(ulong) extent_start;
+    ulong extent_start;
 
     /*
      * Number of extents written to the above array. This will be smaller

@@ -94,7 +93,6 @@
      */
     unsigned int nr_extents;
 };
-DEFINE_GUEST_HANDLE_STRUCT(xen_machphys_mfn_list);
 
 /*
  * Sets the GPFN at which a particular page appears in the specified guest's

@@ -117,7 +115,6 @@
     /* GPFN where the source mapping page should appear. */
     unsigned long gpfn;
 };
-DEFINE_GUEST_HANDLE_STRUCT(xen_add_to_physmap);
 
 /*
  * Translates a list of domain-specific GPFNs into MFNs. Returns a -ve error

@@ -132,14 +129,13 @@
     unsigned long nr_gpfns;
 
     /* List of GPFNs to translate. */
-    GUEST_HANDLE(ulong) gpfn_list;
+    ulong gpfn_list;
 
     /*
      * Output list to contain MFN translations. May be the same as the input
      * list (in which case each input GPFN is overwritten with the output MFN).
      */
-    GUEST_HANDLE(ulong) mfn_list;
+    ulong mfn_list;
 };
-DEFINE_GUEST_HANDLE_STRUCT(xen_translate_gpfn_list);
 
 #endif /* __XEN_PUBLIC_MEMORY_H__ */

91 lines not on this page · raw message

@@ -435,24 +435,6 @@
 	return xen_max;
 }
 
-static int map_pte_fn(pte_t *pte, struct page *pmd_page,
-		      unsigned long addr, void *data)
-{
-	unsigned long **frames = (unsigned long **)data;
-
-	set_pte_at(&init_mm, addr, pte, mfn_pte((*frames)[0], PAGE_KERNEL));
-	(*frames)++;
-	return 0;
-}
-
-static int unmap_pte_fn(pte_t *pte, struct page *pmd_page,
-			unsigned long addr, void *data)
-{
-
-	set_pte_at(&init_mm, addr, pte, __pte(0));
-	return 0;
-}
-
 static int gnttab_map(unsigned int start_idx, unsigned int end_idx)
 {
 	struct gnttab_setup_table setup;

@@ -476,17 +458,9 @@
 
 	BUG_ON(rc || setup.status);
 
-	if (shared == NULL) {
-		struct vm_struct *area;
-		area = xen_alloc_vm_area(PAGE_SIZE * max_nr_grant_frames());
-		BUG_ON(area == NULL);
-		shared = area->addr;
-	}
-	rc = apply_to_page_range(&init_mm, (unsigned long)shared,
-				 PAGE_SIZE * nr_gframes,
-				 map_pte_fn, &frames);
+	rc = arch_gnttab_map_shared(frames, nr_gframes, max_nr_grant_frames(),
+				    &shared);
 	BUG_ON(rc);
-	frames -= nr_gframes; /* adjust after map_pte_fn() */
 
 	kfree(frames);
 

@@ -502,10 +476,7 @@
 
 static int gnttab_suspend(void)
 {
-	apply_to_page_range(&init_mm, (unsigned long)shared,
-			    PAGE_SIZE * nr_grant_frames,
-			    unmap_pte_fn, NULL);
-
+	arch_gnttab_unmap_shared(shared, nr_grant_frames);
 	return 0;
 }
 

@@ -103,6 +103,12 @@
 void gnttab_grant_foreign_transfer_ref(grant_ref_t, domid_t domid,
 				       unsigned long pfn);
 
+int arch_gnttab_map_shared(unsigned long *frames, unsigned long nr_gframes,
+			   unsigned long max_nr_gframes,
+			   struct grant_entry **__shared);
+void arch_gnttab_unmap_shared(struct grant_entry *shared,
+			      unsigned long nr_gframes);
+
 #define gnttab_map_vaddr(map) ((void *)(map.host_virt_addr))
 
 #endif /* __ASM_GNTTAB_H__ */

@@ -669,7 +669,9 @@
    everything is pinned. */
 static __init void xen_alloc_pte_init(struct mm_struct *mm, u32 pfn)
 {
+#ifdef CONFIG_FLATMEM
 	BUG_ON(mem_map);	/* should only be used early */
+#endif
 	make_lowmem_page_readonly(__va(PFN_PHYS(pfn)));
 }
 

@@ -88,6 +88,7 @@
 	struct blk_shadow shadow[BLK_RING_SIZE];
 	unsigned long shadow_free;
 	int feature_barrier;
+	int is_ready;
 
 	/**
 	 * The number of people holding this device open.  We won't allow a

@@ -839,6 +840,8 @@
 	spin_unlock_irq(&blkif_io_lock);
 
 	add_disk(info->gd);
+
+	info->is_ready = 1;
 }
 
 /**

@@ -931,6 +934,13 @@
 	return 0;
 }
 
+static int blkfront_is_ready(struct xenbus_device *dev)
+{
+	struct blkfront_info *info = dev->dev.driver_data;
+
+	return info->is_ready;
+}
+
 static int blkif_open(struct inode *inode, struct file *filep)
 {
 	struct blkfront_info *info = inode->i_bdev->bd_disk->private_data;

@@ -977,6 +987,7 @@
 	.remove = blkfront_remove,
 	.resume = blkfront_resume,
 	.otherend_changed = backend_changed,
+	.is_ready = blkfront_is_ready,
 };
 
 static int __init xlblk_init(void)

@@ -846,6 +846,7 @@
 {
 	struct xenbus_device *xendev = to_xenbus_device(dev);
 	struct device_driver *drv = data;
+	struct xenbus_driver *xendrv;
 
 	/*
 	 * A device with no driver will never connect. We care only about

@@ -858,7 +859,9 @@
 	if (drv && (dev->driver != drv))
 		return 0;
 
-	return (xendev->state != XenbusStateConnected);
+	xendrv = to_xenbus_driver(dev->driver);
+	return (xendev->state != XenbusStateConnected ||
+		(xendrv->is_ready && !xendrv->is_ready(xendev)));
 }
 
 static int exists_disconnected_device(struct device_driver *drv)

@@ -97,6 +97,7 @@
 	int (*uevent)(struct xenbus_device *, char **, int, char *, int);
 	struct device_driver driver;
 	int (*read_otherend_details)(struct xenbus_device *dev);
+	int (*is_ready)(struct xenbus_device *dev);
 };
 
 static inline struct xenbus_driver *to_xenbus_driver(struct device_driver *drv)

@@ -1015,3 +1015,4 @@
 MODULE_DESCRIPTION("Xen virtual block device frontend");
 MODULE_LICENSE("GPL");
 MODULE_ALIAS_BLOCKDEV_MAJOR(XENVBD_MAJOR);
+MODULE_ALIAS("xen:vbd");

@@ -1809,3 +1809,4 @@
 
 MODULE_DESCRIPTION("Xen virtual network device frontend");
 MODULE_LICENSE("GPL");
+MODULE_ALIAS("xen:vif");

@@ -88,6 +88,16 @@
 	return match_device(drv->ids, to_xenbus_device(_dev)) != NULL;
 }
 
+static int xenbus_uevent(struct device *_dev, struct kobj_uevent_env *env)
+{
+	struct xenbus_device *dev = to_xenbus_device(_dev);
+
+	if (add_uevent_var(env, "MODALIAS=xen:%s", dev->devicetype))
+		return -ENOMEM;
+
+	return 0;
+}
+
 /* device/<type>/<id> => <type>-<id> */
 static int frontend_bus_id(char bus_id[BUS_ID_SIZE], const char *nodename)
 {

@@ -166,6 +176,7 @@
 	.bus = {
 		.name     = "xen",
 		.match    = xenbus_match,
+		.uevent   = xenbus_uevent,
 		.probe    = xenbus_dev_probe,
 		.remove   = xenbus_dev_remove,
 		.shutdown = xenbus_dev_shutdown,

@@ -438,6 +449,12 @@
 }
 DEVICE_ATTR(devtype, S_IRUSR | S_IRGRP | S_IROTH, xendev_show_devtype, NULL);
 
+static ssize_t xendev_show_modalias(struct device *dev,
+				    struct device_attribute *attr, char *buf)
+{
+	return sprintf(buf, "xen:%s\n", to_xenbus_device(dev)->devicetype);
+}
+DEVICE_ATTR(modalias, S_IRUSR | S_IRGRP | S_IROTH, xendev_show_modalias, NULL);
 
 int xenbus_probe_node(struct xen_bus_type *bus,
 		      const char *type,

@@ -492,10 +509,16 @@
 
 	err = device_create_file(&xendev->dev, &dev_attr_devtype);
 	if (err)
-		goto fail_remove_file;
+		goto fail_remove_nodename;
+
+	err = device_create_file(&xendev->dev, &dev_attr_modalias);
+	if (err)
+		goto fail_remove_devtype;
 
 	return 0;
-fail_remove_file:
+fail_remove_devtype:
+	device_remove_file(&xendev->dev, &dev_attr_devtype);
+fail_remove_nodename:
 	device_remove_file(&xendev->dev, &dev_attr_nodename);
 fail_unregister:
 	device_unregister(&xendev->dev);

@@ -466,7 +466,7 @@
 
 	setup.dom        = DOMID_SELF;
 	setup.nr_frames  = nr_gframes;
-	setup.frame_list = frames;
+	set_xen_guest_handle(setup.frame_list, frames);
 
 	rc = HYPERVISOR_grant_table_op(GNTTABOP_setup_table, &setup, 1);
 	if (rc == -ENOSYS) {

@@ -21,6 +21,30 @@
 	__DEFINE_GUEST_HANDLE(name, struct name)
 #define DEFINE_GUEST_HANDLE(name) __DEFINE_GUEST_HANDLE(name, name)
 #define GUEST_HANDLE(name)        __guest_handle_ ## name
+
+#ifdef __XEN__
+#if defined(__i386__)
+#define set_xen_guest_handle(hnd, val)			\
+	do {						\
+		if (sizeof(hnd) == 8)			\
+			*(uint64_t *)&(hnd) = 0;	\
+		(hnd).p = val;				\
+	} while (0)
+#elif defined(__x86_64__)
+#define set_xen_guest_handle(hnd, val)	do { (hnd).p = val; } while (0)
+#endif
+#else
+#if defined(__i386__)
+#define set_xen_guest_handle(hnd, val)			\
+	do {						\
+		if (sizeof(hnd) == 8)			\
+			*(uint64_t *)&(hnd) = 0;	\
+		(hnd) = val;				\
+	} while (0)
+#elif defined(__x86_64__)
+#define set_xen_guest_handle(hnd, val)	do { (hnd) = val; } while (0)
+#endif
+#endif
 
 #ifndef __ASSEMBLY__
 /* Guest handles for primitive C types. */

@@ -185,6 +185,7 @@
     grant_handle_t handle;
     uint64_t dev_bus_addr;
 };
+DEFINE_GUEST_HANDLE_STRUCT(gnttab_map_grant_ref);
 
 /*
  * GNTTABOP_unmap_grant_ref: Destroy one or more grant-reference mappings

@@ -206,6 +207,7 @@
     /* OUT parameters. */
     int16_t  status;              /* GNTST_* */
 };
+DEFINE_GUEST_HANDLE_STRUCT(gnttab_unmap_grant_ref);
 
 /*
  * GNTTABOP_setup_table: Set up a grant table for <dom> comprising at least

@@ -223,8 +225,9 @@
     uint32_t nr_frames;
     /* OUT parameters. */
     int16_t  status;              /* GNTST_* */
-    ulong *frame_list;
+    GUEST_HANDLE(ulong) frame_list;
 };
+DEFINE_GUEST_HANDLE_STRUCT(gnttab_setup_table);
 
 /*
  * GNTTABOP_dump_table: Dump the contents of the grant table to the

@@ -237,6 +240,7 @@
     /* OUT parameters. */
     int16_t status;               /* GNTST_* */
 };
+DEFINE_GUEST_HANDLE_STRUCT(gnttab_dump_table);
 
 /*
  * GNTTABOP_transfer_grant_ref: Transfer <frame> to a foreign domain. The

@@ -255,7 +259,7 @@
     /* OUT parameters. */
     int16_t       status;
 };
-
+DEFINE_GUEST_HANDLE_STRUCT(gnttab_transfer);
 
 /*
  * GNTTABOP_copy: Hypervisor based copy

@@ -296,6 +300,7 @@
 	/* OUT parameters. */
 	int16_t       status;
 };
+DEFINE_GUEST_HANDLE_STRUCT(gnttab_copy);
 
 /*
  * GNTTABOP_query_size: Query the current and maximum sizes of the shared

@@ -313,7 +318,7 @@
     uint32_t max_nr_frames;
     int16_t  status;              /* GNTST_* */
 };
-
+DEFINE_GUEST_HANDLE_STRUCT(gnttab_query_size);
 
 /*
  * Bitfield values for update_pin_status.flags.

@@ -57,6 +57,16 @@
 #define __HYPERVISOR_event_channel_op     32
 #define __HYPERVISOR_physdev_op           33
 #define __HYPERVISOR_hvm_op               34
+
+/* Architecture-specific hypercall definitions. */
+#define __HYPERVISOR_arch_0               48
+#define __HYPERVISOR_arch_1               49
+#define __HYPERVISOR_arch_2               50
+#define __HYPERVISOR_arch_3               51
+#define __HYPERVISOR_arch_4               52
+#define __HYPERVISOR_arch_5               53
+#define __HYPERVISOR_arch_6               54
+#define __HYPERVISOR_arch_7               55
 
 /*
  * VIRTUAL INTERRUPTS

@@ -78,8 +78,18 @@
 #define VIRQ_CONSOLE    2  /* (DOM0) Bytes received on emergency console. */
 #define VIRQ_DOM_EXC    3  /* (DOM0) Exceptional event for some domain.   */
 #define VIRQ_DEBUGGER   6  /* (DOM0) A domain has paused for debugging.   */
-#define NR_VIRQS        8
 
+/* Architecture-specific VIRQ definitions. */
+#define VIRQ_ARCH_0    16
+#define VIRQ_ARCH_1    17
+#define VIRQ_ARCH_2    18
+#define VIRQ_ARCH_3    19
+#define VIRQ_ARCH_4    20
+#define VIRQ_ARCH_5    21
+#define VIRQ_ARCH_6    22
+#define VIRQ_ARCH_7    23
+
+#define NR_VIRQS       24
 /*
  * MMU-UPDATE REQUESTS
  *

rename from arch/x86/xen/features.c
rename to drivers/xen/features.c

@@ -3,6 +3,7 @@
 
 #include <linux/init.h>
 #include <linux/irqreturn.h>
+#include <xen/xen-ops.h>
 
 /* These are code, but not functions.  Defined in entry.S */
 extern const char xen_hypervisor_callback[];

@@ -10,7 +11,6 @@
 
 void xen_copy_trap_info(struct trap_info *traps);
 
-DECLARE_PER_CPU(struct vcpu_info *, xen_vcpu);
 DECLARE_PER_CPU(unsigned long, xen_cr3);
 DECLARE_PER_CPU(unsigned long, xen_current_cr3);
 

rename from arch/x86/xen/events.c
rename to drivers/xen/events.c
@@ -33,11 +33,10 @@
 #include <asm/xen/hypercall.h>
 #include <asm/xen/hypervisor.h>
 
+#include <xen/xen-ops.h>
 #include <xen/events.h>
 #include <xen/interface/xen.h>
 #include <xen/interface/event_channel.h>
-
-#include "xen-ops.h"
 
 /*
  * This lock protects updates to the following mapping and reference-count

new file mode 100644
@@ -0,0 +1,8 @@
+#ifndef INCLUDE_XEN_OPS_H
+#define INCLUDE_XEN_OPS_H
+
+#include <linux/percpu.h>
+
+DECLARE_PER_CPU(struct vcpu_info *, xen_vcpu);
+
+#endif /* INCLUDE_XEN_OPS_H */

232 lines not on this page · raw message

41 lines not on this page · raw message

77 lines not on this page · raw message

@@ -156,6 +156,10 @@
 void xen_set_pte_at(struct mm_struct *mm, unsigned long addr,
 		    pte_t *ptep, pte_t pteval)
 {
+	/* updates to init_mm may be done without lock */
+	if (mm == &init_mm)
+		preempt_disable();
+
 	if (mm == current->mm || mm == &init_mm) {
 		if (paravirt_get_lazy_mode() == PARAVIRT_LAZY_MMU) {
 			struct multicall_space mcs;

@@ -163,12 +167,16 @@
 
 			MULTI_update_va_mapping(mcs.mc, addr, pteval, 0);
 			xen_mc_issue(PARAVIRT_LAZY_MMU);
-			return;
+			goto out;
 		} else
 			if (HYPERVISOR_update_va_mapping(addr, pteval, 0) == 0)
-				return;
+				goto out;
 	}
 	xen_set_pte(ptep, pteval);
+
+out:
+	if (mm == &init_mm)
+		preempt_enable();
 }
 
 pteval_t xen_pte_val(pte_t pte)

@@ -532,26 +532,37 @@
 static void xen_flush_tlb(void)
 {
 	struct mmuext_op *op;
-	struct multicall_space mcs = xen_mc_entry(sizeof(*op));
+	struct multicall_space mcs;
+
+	preempt_disable();
+
+	mcs = xen_mc_entry(sizeof(*op));
 
 	op = mcs.args;
 	op->cmd = MMUEXT_TLB_FLUSH_LOCAL;
 	MULTI_mmuext_op(mcs.mc, op, 1, NULL, DOMID_SELF);
 
 	xen_mc_issue(PARAVIRT_LAZY_MMU);
+
+	preempt_enable();
 }
 
 static void xen_flush_tlb_single(unsigned long addr)
 {
 	struct mmuext_op *op;
-	struct multicall_space mcs = xen_mc_entry(sizeof(*op));
+	struct multicall_space mcs;
 
+	preempt_disable();
+
+	mcs = xen_mc_entry(sizeof(*op));
 	op = mcs.args;
 	op->cmd = MMUEXT_INVLPG_LOCAL;
 	op->arg1.linear_addr = addr & PAGE_MASK;
 	MULTI_mmuext_op(mcs.mc, op, 1, NULL, DOMID_SELF);
 
 	xen_mc_issue(PARAVIRT_LAZY_MMU);
+
+	preempt_enable();
 }
 
 static void xen_flush_tlb_others(const cpumask_t *cpus, struct mm_struct *mm,

@@ -85,6 +85,7 @@
 		 */
 		uint64_t time[4];
 };
+DEFINE_GUEST_HANDLE_STRUCT(vcpu_runstate_info);
 
 /* VCPU is currently running on a physical CPU. */
 #define RUNSTATE_running  0

@@ -119,6 +120,7 @@
 #define VCPUOP_register_runstate_memory_area 5
 struct vcpu_register_runstate_memory_area {
 		union {
+				GUEST_HANDLE(vcpu_runstate_info) h;
 				struct vcpu_runstate_info *v;
 				uint64_t p;
 		} addr;

@@ -134,6 +136,7 @@
 struct vcpu_set_periodic_timer {
 		uint64_t period_ns;
 };
+DEFINE_GUEST_HANDLE_STRUCT(vcpu_set_periodic_timer);
 
 /*
  * Set or stop a VCPU's single-shot timer. Every VCPU has one single-shot

@@ -145,6 +148,7 @@
 		uint64_t timeout_abs_ns;
 		uint32_t flags;			   /* VCPU_SSHOTTMR_??? */
 };
+DEFINE_GUEST_HANDLE_STRUCT(vcpu_set_singleshot_timer);
 
 /* Flags to VCPUOP_set_singleshot_timer. */
  /* Require the timeout to be in the future (return -ETIME if it's passed). */

@@ -164,5 +168,6 @@
     uint32_t offset; /* offset within page */
     uint32_t rsvd;   /* unused */
 };
+DEFINE_GUEST_HANDLE_STRUCT(vcpu_register_vcpu_info);
 
 #endif /* __XEN_PUBLIC_VCPU_H__ */

@@ -75,7 +75,6 @@
 struct blkfront_info
 {
 	struct xenbus_device *xbdev;
-	dev_t dev;
 	struct gendisk *gd;
 	int vdevice;
 	blkif_vdev_t handle;

@@ -903,7 +902,7 @@
 		break;
 
 	case XenbusStateClosing:
-		bd = bdget(info->dev);
+		bd = bdget_disk(info->gd, 0);
 		if (bd == NULL)
 			xenbus_dev_fatal(dev, -ENODEV, "bdget failed");
 

new file mode 100644
@@ -0,0 +1,168 @@
+#ifndef __XEN_PAGE_H
+#define __XEN_PAGE_H
+
+#include <linux/pfn.h>
+
+#include <asm/uaccess.h>
+#include <asm/pgtable.h>
+
+#include <xen/features.h>
+
+/* Xen machine address */
+typedef struct xmaddr {
+	phys_addr_t maddr;
+} xmaddr_t;
+
+/* Xen pseudo-physical address */
+typedef struct xpaddr {
+	phys_addr_t paddr;
+} xpaddr_t;
+
+#define XMADDR(x)	((xmaddr_t) { .maddr = (x) })
+#define XPADDR(x)	((xpaddr_t) { .paddr = (x) })
+
+/**** MACHINE <-> PHYSICAL CONVERSION MACROS ****/
+#define INVALID_P2M_ENTRY	(~0UL)
+#define FOREIGN_FRAME_BIT	(1UL<<31)
+#define FOREIGN_FRAME(m)	((m) | FOREIGN_FRAME_BIT)
+
+extern unsigned long *phys_to_machine_mapping;
+
+static inline unsigned long pfn_to_mfn(unsigned long pfn)
+{
+	if (xen_feature(XENFEAT_auto_translated_physmap))
+		return pfn;
+
+	return phys_to_machine_mapping[(unsigned int)(pfn)] &
+		~FOREIGN_FRAME_BIT;
+}
+
+static inline int phys_to_machine_mapping_valid(unsigned long pfn)
+{
+	if (xen_feature(XENFEAT_auto_translated_physmap))
+		return 1;
+
+	return (phys_to_machine_mapping[pfn] != INVALID_P2M_ENTRY);
+}
+
+static inline unsigned long mfn_to_pfn(unsigned long mfn)
+{
+	unsigned long pfn;
+
+	if (xen_feature(XENFEAT_auto_translated_physmap))
+		return mfn;
+
+#if 0
+	if (unlikely((mfn >> machine_to_phys_order) != 0))
+		return max_mapnr;
+#endif
+
+	pfn = 0;
+	/*
+	 * The array access can fail (e.g., device space beyond end of RAM).
+	 * In such cases it doesn't matter what we return (we return garbage),
+	 * but we must handle the fault without crashing!
+	 */
+	__get_user(pfn, &machine_to_phys_mapping[mfn]);
+
+	return pfn;
+}
+
+static inline xmaddr_t phys_to_machine(xpaddr_t phys)
+{
+	unsigned offset = phys.paddr & ~PAGE_MASK;
+	return XMADDR(PFN_PHYS((u64)pfn_to_mfn(PFN_DOWN(phys.paddr))) | offset);
+}
+
+static inline xpaddr_t machine_to_phys(xmaddr_t machine)
+{
+	unsigned offset = machine.maddr & ~PAGE_MASK;
+	return XPADDR(PFN_PHYS((u64)mfn_to_pfn(PFN_DOWN(machine.maddr))) | offset);
+}
+
+/*
+ * We detect special mappings in one of two ways:
+ *  1. If the MFN is an I/O page then Xen will set the m2p entry
+ *     to be outside our maximum possible pseudophys range.
+ *  2. If the MFN belongs to a different domain then we will certainly
+ *     not have MFN in our p2m table. Conversely, if the page is ours,
+ *     then we'll have p2m(m2p(MFN))==MFN.
+ * If we detect a special mapping then it doesn't have a 'struct page'.
+ * We force !pfn_valid() by returning an out-of-range pointer.
+ *
+ * NB. These checks require that, for any MFN that is not in our reservation,
+ * there is no PFN such that p2m(PFN) == MFN. Otherwise we can get confused if
+ * we are foreign-mapping the MFN, and the other domain as m2p(MFN) == PFN.
+ * Yikes! Various places must poke in INVALID_P2M_ENTRY for safety.
+ *
+ * NB2. When deliberately mapping foreign pages into the p2m table, you *must*
+ *      use FOREIGN_FRAME(). This will cause pte_pfn() to choke on it, as we
+ *      require. In all the cases we care about, the FOREIGN_FRAME bit is
+ *      masked (e.g., pfn_to_mfn()) so behaviour there is correct.
+ */
+static inline unsigned long mfn_to_local_pfn(unsigned long mfn)
+{
+	extern unsigned long max_mapnr;
+	unsigned long pfn = mfn_to_pfn(mfn);
+	if ((pfn < max_mapnr)
+	    && !xen_feature(XENFEAT_auto_translated_physmap)
+	    && (phys_to_machine_mapping[pfn] != mfn))
+		return max_mapnr; /* force !pfn_valid() */
+	return pfn;
+}
+
+static inline void set_phys_to_machine(unsigned long pfn, unsigned long mfn)
+{
+	if (xen_feature(XENFEAT_auto_translated_physmap)) {
+		BUG_ON(pfn != mfn && mfn != INVALID_P2M_ENTRY);
+		return;
+	}
+	phys_to_machine_mapping[pfn] = mfn;
+}
+
+/* VIRT <-> MACHINE conversion */
+#define virt_to_machine(v)	(phys_to_machine(XPADDR(__pa(v))))
+#define virt_to_mfn(v)		(pfn_to_mfn(PFN_DOWN(__pa(v))))
+#define mfn_to_virt(m)		(__va(mfn_to_pfn(m) << PAGE_SHIFT))
+
+static inline unsigned long pte_mfn(pte_t pte)
+{
+	return (pte.pte & ~_PAGE_NX) >> PAGE_SHIFT;
+}
+
+static inline pte_t mfn_pte(unsigned long page_nr, pgprot_t pgprot)
+{
+	pte_t pte;
+
+	pte.pte = ((phys_addr_t)page_nr << PAGE_SHIFT) |
+		(pgprot_val(pgprot) & __supported_pte_mask);
+
+	return pte;
+}
+
+static inline pteval_t pte_val_ma(pte_t pte)
+{
+	return pte.pte;
+}
+
+static inline pte_t __pte_ma(pteval_t x)
+{
+	return (pte_t) { .pte = x };
+}
+
+#ifdef CONFIG_X86_PAE
+#define pmd_val_ma(v) ((v).pmd)
+#define pud_val_ma(v) ((v).pgd.pgd)
+#define __pmd_ma(x)	((pmd_t) { (x) } )
+#else  /* !X86_PAE */
+#define pmd_val_ma(v)	((v).pud.pgd.pgd)
+#endif	/* CONFIG_X86_PAE */
+
+#define pgd_val_ma(x)	((x).pgd)
+
+
+xmaddr_t arbitrary_virt_to_machine(unsigned long address);
+void make_lowmem_page_readonly(void *vaddr);
+void make_lowmem_page_readwrite(void *vaddr);
+
+#endif /* __XEN_PAGE_H */

169 lines not on this page · raw message

@@ -469,7 +469,7 @@
 	for_each_online_cpu(i) {
 		struct vcpu_info *v = per_cpu(xen_vcpu, i);
 		printk("%d: masked=%d pending=%d event_sel %08lx\n  ", i,
-			(get_irq_regs() && i == cpu) ? !(get_irq_regs()->flags & X86_EFLAGS_IF) : v->evtchn_upcall_mask,
+			(get_irq_regs() && i == cpu) ? xen_irqs_disabled(get_irq_regs()) : v->evtchn_upcall_mask,
 			v->evtchn_upcall_pending,
 			v->evtchn_pending_sel);
 	}

@@ -527,7 +527,10 @@
 		if (__get_cpu_var(nesting_count)++)
 			goto out;
 
-		/* NB. No need for a barrier here -- XCHG is a barrier on x86. */
+#ifndef CONFIG_X86 /* No need for a barrier -- XCHG is a barrier on x86. */
+		/* Clear master flag /before/ clearing selector flag. */
+		rmb();
+#endif
 		pending_words = xchg(&vcpu_info->evtchn_pending_sel, 0);
 		while (pending_words != 0) {
 			unsigned long pending_bits;

@@ -539,10 +542,8 @@
 				int port = (word_idx * BITS_PER_LONG) + bit_idx;
 				int irq = evtchn_to_irq[port];
 
-				if (irq != -1) {
-					regs->orig_ax = ~irq;
-					do_IRQ(regs);
-				}
+				if (irq != -1)
+					xen_do_IRQ(irq, regs);
 			}
 		}
 

new file mode 100644
@@ -0,0 +1,22 @@
+#ifndef __XEN_EVENTS_H
+#define __XEN_EVENTS_H
+
+enum ipi_vector {
+	XEN_RESCHEDULE_VECTOR,
+	XEN_CALL_FUNCTION_VECTOR,
+
+	XEN_NR_IPIS,
+};
+
+static inline int xen_irqs_disabled(struct pt_regs *regs)
+{
+	return raw_irqs_disabled_flags(regs->flags);
+}
+
+static inline void xen_do_IRQ(int irq, struct pt_regs *regs)
+{
+	regs->orig_ax = ~irq;
+	do_IRQ(regs);
+}
+
+#endif /* __XEN_EVENTS_H */

@@ -5,13 +5,7 @@
 
 #include <xen/interface/event_channel.h>
 #include <asm/xen/hypercall.h>
-
-enum ipi_vector {
-	XEN_RESCHEDULE_VECTOR,
-	XEN_CALL_FUNCTION_VECTOR,
-
-	XEN_NR_IPIS,
-};
+#include <asm/xen/events.h>
 
 int bind_evtchn_to_irq(unsigned int evtchn);
 int bind_evtchn_to_irqhandler(unsigned int evtchn,

@@ -586,6 +586,22 @@
 	rebind_irq_to_cpu(irq, tcpu);
 }
 
+int resend_irq_on_evtchn(unsigned int irq)
+{
+	int masked, evtchn = evtchn_from_irq(irq);
+	struct shared_info *s = HYPERVISOR_shared_info;
+
+	if (!VALID_EVTCHN(evtchn))
+		return 1;
+
+	masked = sync_test_and_set_bit(evtchn, s->evtchn_mask);
+	sync_set_bit(evtchn, s->evtchn_pending);
+	if (!masked)
+		unmask_evtchn(evtchn);
+
+	return 1;
+}
+
 static void enable_dynirq(unsigned int irq)
 {
 	int evtchn = evtchn_from_irq(irq);

@@ -31,6 +31,7 @@
 void unbind_from_irqhandler(unsigned int irq, void *dev_id);
 
 void xen_send_IPI_one(unsigned int cpu, enum ipi_vector vector);
+int resend_irq_on_evtchn(unsigned int irq);
 
 static inline void notify_remote_via_evtchn(int port)
 {

@@ -478,7 +478,7 @@
 
 	if (shared == NULL) {
 		struct vm_struct *area;
-		area = alloc_vm_area(PAGE_SIZE * max_nr_grant_frames());
+		area = xen_alloc_vm_area(PAGE_SIZE * max_nr_grant_frames());
 		BUG_ON(area == NULL);
 		shared = area->addr;
 	}

@@ -399,7 +399,7 @@
 
 	*vaddr = NULL;
 
-	area = alloc_vm_area(PAGE_SIZE);
+	area = xen_alloc_vm_area(PAGE_SIZE);
 	if (!area)
 		return -ENOMEM;
 

@@ -409,7 +409,7 @@
 		BUG();
 
 	if (op.status != GNTST_okay) {
-		free_vm_area(area);
+		xen_free_vm_area(area);
 		xenbus_dev_fatal(dev, op.status,
 				 "mapping in shared page %d from domain %d",
 				 gnt_ref, dev->otherend_id);

@@ -508,7 +508,7 @@
 		BUG();
 
 	if (op.status == GNTST_okay)
-		free_vm_area(area);
+		xen_free_vm_area(area);
 	else
 		xenbus_dev_error(dev, op.status,
 				 "unmapping page at handle %d error %d",

new file mode 100644
@@ -0,0 +1,7 @@
+#ifndef __XEN_GRANT_TABLE_H
+#define __XEN_GRANT_TABLE_H
+
+#define xen_alloc_vm_area(size)	alloc_vm_area(size)
+#define xen_free_vm_area(area)	free_vm_area(area)
+
+#endif /* __XEN_GRANT_TABLE_H */

@@ -39,6 +39,7 @@
 
 #include <asm/xen/hypervisor.h>
 #include <xen/interface/grant_table.h>
+#include <asm/xen/grant_table.h>
 
 /* NR_GRANT_FRAMES must be less than or equal to that configured in Xen */
 #define NR_GRANT_FRAMES 4

@@ -47,6 +47,7 @@
 
 #include <xen/interface/grant_table.h>
 #include <xen/interface/io/blkif.h>
+#include <xen/interface/io/protocols.h>
 
 #include <asm/xen/hypervisor.h>
 

@@ -614,6 +615,12 @@
 		message = "writing event-channel";
 		goto abort_transaction;
 	}
+	err = xenbus_printf(xbt, dev->nodename, "protocol", "%s",
+			    XEN_IO_PROTO_ABI_NATIVE);
+	if (err) {
+		message = "writing protocol";
+		goto abort_transaction;
+	}
 
 	err = xenbus_transaction_end(xbt, 0);
 	if (err) {

new file mode 100644
@@ -0,0 +1,21 @@
+#ifndef __XEN_PROTOCOLS_H__
+#define __XEN_PROTOCOLS_H__
+
+#define XEN_IO_PROTO_ABI_X86_32     "x86_32-abi"
+#define XEN_IO_PROTO_ABI_X86_64     "x86_64-abi"
+#define XEN_IO_PROTO_ABI_IA64       "ia64-abi"
+#define XEN_IO_PROTO_ABI_POWERPC64  "powerpc64-abi"
+
+#if defined(__i386__)
+# define XEN_IO_PROTO_ABI_NATIVE XEN_IO_PROTO_ABI_X86_32
+#elif defined(__x86_64__)
+# define XEN_IO_PROTO_ABI_NATIVE XEN_IO_PROTO_ABI_X86_64
+#elif defined(__ia64__)
+# define XEN_IO_PROTO_ABI_NATIVE XEN_IO_PROTO_ABI_IA64
+#elif defined(__powerpc64__)
+# define XEN_IO_PROTO_ABI_NATIVE XEN_IO_PROTO_ABI_POWERPC64
+#else
+# error arch fixup needed here
+#endif
+
+#endif

@@ -1016,3 +1016,4 @@
 MODULE_LICENSE("GPL");
 MODULE_ALIAS_BLOCKDEV_MAJOR(XENVBD_MAJOR);
 MODULE_ALIAS("xen:vbd");
+MODULE_ALIAS("xenblk");

@@ -1810,3 +1810,4 @@
 MODULE_DESCRIPTION("Xen virtual network device frontend");
 MODULE_LICENSE("GPL");
 MODULE_ALIAS("xen:vif");
+MODULE_ALIAS("xennet");

340 lines not on this page · raw message

550 lines not on this page · raw message

124 lines not on this page · raw message

114 lines not on this page · raw message

@@ -381,11 +381,15 @@ EXPORT_SYMBOL_GPL(gnttab_cancel_free_callback);
 static int grow_gnttab_list(unsigned int more_frames)
 {
 	unsigned int new_nr_grant_frames, extra_entries, i;
+	unsigned int nr_glist_frames, new_nr_glist_frames;
 
 	new_nr_grant_frames = nr_grant_frames + more_frames;
 	extra_entries       = more_frames * GREFS_PER_GRANT_FRAME;
 
-	for (i = nr_grant_frames; i < new_nr_grant_frames; i++) {
+	nr_glist_frames = (nr_grant_frames * GREFS_PER_GRANT_FRAME + RPP - 1) / RPP;
+	new_nr_glist_frames =
+		(new_nr_grant_frames * GREFS_PER_GRANT_FRAME + RPP - 1) / RPP;
+	for (i = nr_glist_frames; i < new_nr_glist_frames; i++) {
 		gnttab_list[i] = (grant_ref_t *)__get_free_page(GFP_ATOMIC);
 		if (!gnttab_list[i])
 			goto grow_nomem;

@@ -407,7 +411,7 @@ static int grow_gnttab_list(unsigned int more_frames)
 	return 0;
 
 grow_nomem:
-	for ( ; i >= nr_grant_frames; i--)
+	for ( ; i >= nr_glist_frames; i--)
 		free_page((unsigned long) gnttab_list[i]);
 	return -ENOMEM;
 }

@@ -530,7 +534,7 @@ static int gnttab_expand(unsigned int req_entries)
 static int __devinit gnttab_init(void)
 {
 	int i;
-	unsigned int max_nr_glist_frames;
+	unsigned int max_nr_glist_frames, nr_glist_frames;
 	unsigned int nr_init_grefs;
 
 	if (!is_running_on_xen())

@@ -543,15 +547,15 @@ static int __devinit gnttab_init(void)
 	 * grant reference free list on the current hypervisor.
 	 */
 	max_nr_glist_frames = (boot_max_nr_grant_frames *
-			       GREFS_PER_GRANT_FRAME /
-			       (PAGE_SIZE / sizeof(grant_ref_t)));
+			       GREFS_PER_GRANT_FRAME / RPP);
 
 	gnttab_list = kmalloc(max_nr_glist_frames * sizeof(grant_ref_t *),
 			      GFP_KERNEL);
 	if (gnttab_list == NULL)
 		return -ENOMEM;
 
-	for (i = 0; i < nr_grant_frames; i++) {
+	nr_glist_frames = (nr_grant_frames * GREFS_PER_GRANT_FRAME + RPP - 1) / RPP;
+	for (i = 0; i < nr_glist_frames; i++) {
 		gnttab_list[i] = (grant_ref_t *)__get_free_page(GFP_KERNEL);
 		if (gnttab_list[i] == NULL)
 			goto ini_nomem;