Support PCI devices with multiple MSIs, performance improvement for

kernel-based backends (by not populated m2p overrides when mapping),
 and assorted minor bug fixes and cleanups.
 -----BEGIN PGP SIGNATURE-----
 Version: GnuPG v1.4.10 (GNU/Linux)
 
 iQEcBAABAgAGBQJTPTGyAAoJEFxbo/MsZsTRnjgH/10j5CbOK1RFvIyCSslGTf4G
 slhK8P8dhhplGAxwXXji322lWNYEx9Jd+V0Bhxnvr4drSlsP/qkWuBWf+u1LBvRq
 AVPM99tk0XHCVAuvMMNo/lc62dTIR9IpQvnY6WhHSHnSlfqyVcdnbaGk8/LRuxWJ
 u2F0MXzDNH00b/kt6hDBt3F7CkHfjwsEn43LCkkxyHPp5MJGD7bGDIe+bKtnjv9u
 D9VJtCWQkrjWQ6jNpjdP833JCNCGQrXtVO3DeTAGs3T1tGmiEsqp6kT6Gp5zCFnh
 oaQk9jfQL2S+IVnVhHVMW9nTwNPPrnIrD69FlgTrK301mcYW1mKoFotTogzHu+0=
 =2IG+
 -----END PGP SIGNATURE-----

Merge tag 'stable/for-linus-3.15-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip

Pull Xen features and fixes from David Vrabel:
 "Support PCI devices with multiple MSIs, performance improvement for
  kernel-based backends (by not populated m2p overrides when mapping),
  and assorted minor bug fixes and cleanups"

* tag 'stable/for-linus-3.15-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip:
  xen/acpi-processor: fix enabling interrupts on syscore_resume
  xen/grant-table: Refactor gnttab_[un]map_refs to avoid m2p_override
  xen: remove XEN_PRIVILEGED_GUEST
  xen: add support for MSI message groups
  xen-pciback: Use pci_enable_msix_exact() instead of pci_enable_msix()
  xen/xenbus: remove unused xenbus_bind_evtchn()
  xen/events: remove unnecessary call to bind_evtchn_to_cpu()
  xen/events: remove the unused resend_irq_on_evtchn()
  drivers:xen-selfballoon:reset 'frontswap_inertia_counter' after frontswap_shrink
  drivers: xen: Include appropriate header file in pcpu.c
  drivers: xen: Mark function as static in platform-pci.c

drivers/xen/events/events_base.c

@@ -388,10 +388,10 @@ static void xen_irq_init(unsigned irq)
 	list_add_tail(&info->list, &xen_irq_list_head);
 }
 
-static int __must_check xen_allocate_irq_dynamic(void)
+static int __must_check xen_allocate_irqs_dynamic(int nvec)
 {
 	int first = 0;
-	int irq;
+	int i, irq;
 
 #ifdef CONFIG_X86_IO_APIC
 	/*
@@ -405,14 +405,22 @@ static int __must_check xen_allocate_irq_dynamic(void)
 		first = get_nr_irqs_gsi();
 #endif
 
-	irq = irq_alloc_desc_from(first, -1);
+	irq = irq_alloc_descs_from(first, nvec, -1);
 
-	if (irq >= 0)
-		xen_irq_init(irq);
+	if (irq >= 0) {
+		for (i = 0; i < nvec; i++)
+			xen_irq_init(irq + i);
+	}
 
 	return irq;
 }
 
+static inline int __must_check xen_allocate_irq_dynamic(void)
+{
+
+	return xen_allocate_irqs_dynamic(1);
+}
+
 static int __must_check xen_allocate_irq_gsi(unsigned gsi)
 {
 	int irq;
@@ -466,9 +474,6 @@ static void xen_evtchn_close(unsigned int port)
 	close.port = port;
 	if (HYPERVISOR_event_channel_op(EVTCHNOP_close, &close) != 0)
 		BUG();
-
-	/* Closed ports are implicitly re-bound to VCPU0. */
-	bind_evtchn_to_cpu(port, 0);
 }
 
 static void pirq_query_unmask(int irq)
@@ -730,22 +735,25 @@ int xen_allocate_pirq_msi(struct pci_dev *dev, struct msi_desc *msidesc)
 }
 
 int xen_bind_pirq_msi_to_irq(struct pci_dev *dev, struct msi_desc *msidesc,
-			     int pirq, const char *name, domid_t domid)
+			     int pirq, int nvec, const char *name, domid_t domid)
 {
-	int irq, ret;
+	int i, irq, ret;
 
 	mutex_lock(&irq_mapping_update_lock);
 
-	irq = xen_allocate_irq_dynamic();
+	irq = xen_allocate_irqs_dynamic(nvec);
 	if (irq < 0)
 		goto out;
 
-	irq_set_chip_and_handler_name(irq, &xen_pirq_chip, handle_edge_irq,
-			name);
+	for (i = 0; i < nvec; i++) {
+		irq_set_chip_and_handler_name(irq + i, &xen_pirq_chip, handle_edge_irq, name);
+
+		ret = xen_irq_info_pirq_setup(irq + i, 0, pirq + i, 0, domid,
+					      i == 0 ? 0 : PIRQ_MSI_GROUP);
+		if (ret < 0)
+			goto error_irq;
+	}
 
-	ret = xen_irq_info_pirq_setup(irq, 0, pirq, 0, domid, 0);
-	if (ret < 0)
-		goto error_irq;
 	ret = irq_set_msi_desc(irq, msidesc);
 	if (ret < 0)
 		goto error_irq;
@@ -753,7 +761,8 @@ out:
 	mutex_unlock(&irq_mapping_update_lock);
 	return irq;
 error_irq:
-	__unbind_from_irq(irq);
+	for (; i >= 0; i--)
+		__unbind_from_irq(irq + i);
 	mutex_unlock(&irq_mapping_update_lock);
 	return ret;
 }
@@ -767,7 +776,12 @@ int xen_destroy_irq(int irq)
 
 	mutex_lock(&irq_mapping_update_lock);
 
-	if (xen_initial_domain()) {
+	/*
+	 * If trying to remove a vector in a MSI group different
+	 * than the first one skip the PIRQ unmap unless this vector
+	 * is the first one in the group.
+	 */
+	if (xen_initial_domain() && !(info->u.pirq.flags & PIRQ_MSI_GROUP)) {
 		unmap_irq.pirq = info->u.pirq.pirq;
 		unmap_irq.domid = info->u.pirq.domid;
 		rc = HYPERVISOR_physdev_op(PHYSDEVOP_unmap_pirq, &unmap_irq);
@@ -1329,26 +1343,6 @@ static int set_affinity_irq(struct irq_data *data, const struct cpumask *dest,
 	return rebind_irq_to_cpu(data->irq, tcpu);
 }
 
-static int retrigger_evtchn(int evtchn)
-{
-	int masked;
-
-	if (!VALID_EVTCHN(evtchn))
-		return 0;
-
-	masked = test_and_set_mask(evtchn);
-	set_evtchn(evtchn);
-	if (!masked)
-		unmask_evtchn(evtchn);
-
-	return 1;
-}
-
-int resend_irq_on_evtchn(unsigned int irq)
-{
-	return retrigger_evtchn(evtchn_from_irq(irq));
-}
-
 static void enable_dynirq(struct irq_data *data)
 {
 	int evtchn = evtchn_from_irq(data->irq);
@@ -1383,7 +1377,18 @@ static void mask_ack_dynirq(struct irq_data *data)
 
 static int retrigger_dynirq(struct irq_data *data)
 {
-	return retrigger_evtchn(evtchn_from_irq(data->irq));
+	unsigned int evtchn = evtchn_from_irq(data->irq);
+	int masked;
+
+	if (!VALID_EVTCHN(evtchn))
+		return 0;
+
+	masked = test_and_set_mask(evtchn);
+	set_evtchn(evtchn);
+	if (!masked)
+		unmask_evtchn(evtchn);
+
+	return 1;
 }
 
 static void restore_pirqs(void)

drivers/xen/events/events_internal.h

@@ -53,6 +53,7 @@ struct irq_info {
 
 #define PIRQ_NEEDS_EOI	(1 << 0)
 #define PIRQ_SHAREABLE	(1 << 1)
+#define PIRQ_MSI_GROUP	(1 << 2)
 
 struct evtchn_ops {
 	unsigned (*max_channels)(void);

drivers/xen/grant-table.c

@@ -933,9 +933,6 @@ int gnttab_map_refs(struct gnttab_map_grant_ref *map_ops,
 		    struct page **pages, unsigned int count)
 {
 	int i, ret;
-	bool lazy = false;
-	pte_t *pte;
-	unsigned long mfn;
 
 	ret = HYPERVISOR_grant_table_op(GNTTABOP_map_grant_ref, map_ops, count);
 	if (ret)
@@ -947,45 +944,7 @@ int gnttab_map_refs(struct gnttab_map_grant_ref *map_ops,
 			gnttab_retry_eagain_gop(GNTTABOP_map_grant_ref, map_ops + i,
 						&map_ops[i].status, __func__);
 
-	/* this is basically a nop on x86 */
-	if (xen_feature(XENFEAT_auto_translated_physmap)) {
-		for (i = 0; i < count; i++) {
-			if (map_ops[i].status)
-				continue;
-			set_phys_to_machine(map_ops[i].host_addr >> PAGE_SHIFT,
-					map_ops[i].dev_bus_addr >> PAGE_SHIFT);
-		}
-		return ret;
-	}
-
-	if (!in_interrupt() && paravirt_get_lazy_mode() == PARAVIRT_LAZY_NONE) {
-		arch_enter_lazy_mmu_mode();
-		lazy = true;
-	}
-
-	for (i = 0; i < count; i++) {
-		/* Do not add to override if the map failed. */
-		if (map_ops[i].status)
-			continue;
-
-		if (map_ops[i].flags & GNTMAP_contains_pte) {
-			pte = (pte_t *) (mfn_to_virt(PFN_DOWN(map_ops[i].host_addr)) +
-				(map_ops[i].host_addr & ~PAGE_MASK));
-			mfn = pte_mfn(*pte);
-		} else {
-			mfn = PFN_DOWN(map_ops[i].dev_bus_addr);
-		}
-		ret = m2p_add_override(mfn, pages[i], kmap_ops ?
-				       &kmap_ops[i] : NULL);
-		if (ret)
-			goto out;
-	}
-
- out:
-	if (lazy)
-		arch_leave_lazy_mmu_mode();
-
-	return ret;
+	return set_foreign_p2m_mapping(map_ops, kmap_ops, pages, count);
 }
 EXPORT_SYMBOL_GPL(gnttab_map_refs);
 
@@ -993,39 +952,13 @@ int gnttab_unmap_refs(struct gnttab_unmap_grant_ref *unmap_ops,
 		      struct gnttab_map_grant_ref *kmap_ops,
 		      struct page **pages, unsigned int count)
 {
-	int i, ret;
-	bool lazy = false;
+	int ret;
 
 	ret = HYPERVISOR_grant_table_op(GNTTABOP_unmap_grant_ref, unmap_ops, count);
 	if (ret)
 		return ret;
 
-	/* this is basically a nop on x86 */
-	if (xen_feature(XENFEAT_auto_translated_physmap)) {
-		for (i = 0; i < count; i++) {
-			set_phys_to_machine(unmap_ops[i].host_addr >> PAGE_SHIFT,
-					INVALID_P2M_ENTRY);
-		}
-		return ret;
-	}
-
-	if (!in_interrupt() && paravirt_get_lazy_mode() == PARAVIRT_LAZY_NONE) {
-		arch_enter_lazy_mmu_mode();
-		lazy = true;
-	}
-
-	for (i = 0; i < count; i++) {
-		ret = m2p_remove_override(pages[i], kmap_ops ?
-				       &kmap_ops[i] : NULL);
-		if (ret)
-			goto out;
-	}
-
- out:
-	if (lazy)
-		arch_leave_lazy_mmu_mode();
-
-	return ret;
+	return clear_foreign_p2m_mapping(unmap_ops, kmap_ops, pages, count);
 }
 EXPORT_SYMBOL_GPL(gnttab_unmap_refs);
 

drivers/xen/manage.c

@@ -46,6 +46,20 @@ struct suspend_info {
 	void (*post)(int cancelled);
 };
 
+static RAW_NOTIFIER_HEAD(xen_resume_notifier);
+
+void xen_resume_notifier_register(struct notifier_block *nb)
+{
+	raw_notifier_chain_register(&xen_resume_notifier, nb);
+}
+EXPORT_SYMBOL_GPL(xen_resume_notifier_register);
+
+void xen_resume_notifier_unregister(struct notifier_block *nb)
+{
+	raw_notifier_chain_unregister(&xen_resume_notifier, nb);
+}
+EXPORT_SYMBOL_GPL(xen_resume_notifier_unregister);
+
 #ifdef CONFIG_HIBERNATE_CALLBACKS
 static void xen_hvm_post_suspend(int cancelled)
 {
@@ -152,6 +166,8 @@ static void do_suspend(void)
 
 	err = stop_machine(xen_suspend, &si, cpumask_of(0));
 
+	raw_notifier_call_chain(&xen_resume_notifier, 0, NULL);
+
 	dpm_resume_start(si.cancelled ? PMSG_THAW : PMSG_RESTORE);
 
 	if (err) {

drivers/xen/pcpu.c

@@ -40,6 +40,7 @@
 #include <linux/capability.h>
 
 #include <xen/xen.h>
+#include <xen/acpi.h>
 #include <xen/xenbus.h>
 #include <xen/events.h>
 #include <xen/interface/platform.h>

drivers/xen/platform-pci.c

@@ -45,7 +45,7 @@ static unsigned long platform_mmio_alloc;
 static unsigned long platform_mmiolen;
 static uint64_t callback_via;
 
-unsigned long alloc_xen_mmio(unsigned long len)
+static unsigned long alloc_xen_mmio(unsigned long len)
 {
 	unsigned long addr;
 

drivers/xen/xen-acpi-processor.c

@@ -27,10 +27,10 @@
 #include <linux/init.h>
 #include <linux/module.h>
 #include <linux/types.h>
-#include <linux/syscore_ops.h>
 #include <linux/acpi.h>
 #include <acpi/processor.h>
 #include <xen/xen.h>
+#include <xen/xen-ops.h>
 #include <xen/interface/platform.h>
 #include <asm/xen/hypercall.h>
 
@@ -495,14 +495,15 @@ static int xen_upload_processor_pm_data(void)
 	return rc;
 }
 
-static void xen_acpi_processor_resume(void)
+static int xen_acpi_processor_resume(struct notifier_block *nb,
+				     unsigned long action, void *data)
 {
 	bitmap_zero(acpi_ids_done, nr_acpi_bits);
-	xen_upload_processor_pm_data();
+	return xen_upload_processor_pm_data();
 }
 
-static struct syscore_ops xap_syscore_ops = {
-	.resume	= xen_acpi_processor_resume,
+struct notifier_block xen_acpi_processor_resume_nb = {
+	.notifier_call = xen_acpi_processor_resume,
 };
 
 static int __init xen_acpi_processor_init(void)
@@ -555,7 +556,7 @@ static int __init xen_acpi_processor_init(void)
 	if (rc)
 		goto err_unregister;
 
-	register_syscore_ops(&xap_syscore_ops);
+	xen_resume_notifier_register(&xen_acpi_processor_resume_nb);
 
 	return 0;
 err_unregister:
@@ -574,7 +575,7 @@ static void __exit xen_acpi_processor_exit(void)
 {
 	int i;
 
-	unregister_syscore_ops(&xap_syscore_ops);
+	xen_resume_notifier_unregister(&xen_acpi_processor_resume_nb);
 	kfree(acpi_ids_done);
 	kfree(acpi_id_present);
 	kfree(acpi_id_cst_present);

drivers/xen/xen-pciback/pciback_ops.c

@@ -213,8 +213,7 @@ int xen_pcibk_enable_msix(struct xen_pcibk_device *pdev,
 		entries[i].vector = op->msix_entries[i].vector;
 	}
 
-	result = pci_enable_msix(dev, entries, op->value);
-
+	result = pci_enable_msix_exact(dev, entries, op->value);
 	if (result == 0) {
 		for (i = 0; i < op->value; i++) {
 			op->msix_entries[i].entry = entries[i].entry;

drivers/xen/xen-selfballoon.c

@@ -170,6 +170,7 @@ static void frontswap_selfshrink(void)
 		tgt_frontswap_pages = cur_frontswap_pages -
 			(cur_frontswap_pages / frontswap_hysteresis);
 	frontswap_shrink(tgt_frontswap_pages);
+	frontswap_inertia_counter = frontswap_inertia;
 }
 
 #endif /* CONFIG_FRONTSWAP */

drivers/xen/xenbus/xenbus_client.c

@@ -400,33 +400,6 @@ int xenbus_alloc_evtchn(struct xenbus_device *dev, int *port)
 EXPORT_SYMBOL_GPL(xenbus_alloc_evtchn);
 
 
-/**
- * Bind to an existing interdomain event channel in another domain. Returns 0
- * on success and stores the local port in *port. On error, returns -errno,
- * switches the device to XenbusStateClosing, and saves the error in XenStore.
- */
-int xenbus_bind_evtchn(struct xenbus_device *dev, int remote_port, int *port)
-{
-	struct evtchn_bind_interdomain bind_interdomain;
-	int err;
-
-	bind_interdomain.remote_dom = dev->otherend_id;
-	bind_interdomain.remote_port = remote_port;
-
-	err = HYPERVISOR_event_channel_op(EVTCHNOP_bind_interdomain,
-					  &bind_interdomain);
-	if (err)
-		xenbus_dev_fatal(dev, err,
-				 "binding to event channel %d from domain %d",
-				 remote_port, dev->otherend_id);
-	else
-		*port = bind_interdomain.local_port;
-
-	return err;
-}
-EXPORT_SYMBOL_GPL(xenbus_bind_evtchn);
-
-
 /**
  * Free an existing event channel. Returns 0 on success or -errno on error.
  */