Merge tag 'for-linus-6.18-rc1-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip

Pull xen updates from Juergen Gross:

 - fix the migration of a Xen virq to another cpu plus some related
   cleanup work

 - clean up Xen-PV mode specific code, resulting in removing some of
   that code in the resulting binary in case CONFIG_XEN_PV is not set

 - fixes and cleanup for suspend handling under Xen

* tag 'for-linus-6.18-rc1-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip:
  xen: take system_transition_mutex on suspend
  xen/manage: Fix suspend error path
  xen/events: Update virq_to_irq on migration
  xen/events: Return -EEXIST for bound VIRQs
  xen/events: Cleanup find_virq() return codes
  x86/xen: select HIBERNATE_CALLBACKS more directly
  drivers/xen/gntdev: use xen_pv_domain() instead of cached value
  xen: replace XENFEAT_auto_translated_physmap with xen_pv_domain()
  xen: rework xen_pv_domain()

Author: torvalds

arch/x86/configs/xen.config

@@ -12,7 +12,6 @@ CONFIG_CPU_FREQ=y
 
 # x86 xen specific config options
 CONFIG_XEN_PVH=y
-CONFIG_XEN_SAVE_RESTORE=y
 # CONFIG_XEN_DEBUG_FS is not set
 CONFIG_XEN_MCE_LOG=y
 CONFIG_XEN_ACPI_PROCESSOR=m

arch/x86/include/asm/xen/page.h

@@ -12,9 +12,9 @@
 #include <asm/extable.h>
 #include <asm/page.h>
 
+#include <xen/xen.h>
 #include <xen/interface/xen.h>
 #include <xen/interface/grant_table.h>
-#include <xen/features.h>
 
 /* Xen machine address */
 typedef struct xmaddr {
@@ -162,7 +162,7 @@ static inline unsigned long pfn_to_mfn(unsigned long pfn)
 	 * pfn_to_mfn. This will have to be removed when we figured
 	 * out which call.
 	 */
-	if (xen_feature(XENFEAT_auto_translated_physmap))
+	if (!xen_pv_domain())
 		return pfn;
 
 	mfn = __pfn_to_mfn(pfn);
@@ -175,7 +175,7 @@ static inline unsigned long pfn_to_mfn(unsigned long pfn)
 
 static inline int phys_to_machine_mapping_valid(unsigned long pfn)
 {
-	if (xen_feature(XENFEAT_auto_translated_physmap))
+	if (!xen_pv_domain())
 		return 1;
 
 	return __pfn_to_mfn(pfn) != INVALID_P2M_ENTRY;
@@ -210,7 +210,7 @@ static inline unsigned long mfn_to_pfn(unsigned long mfn)
 	 * gfn_to_pfn. This will have to be removed when we figure
 	 * out which call.
 	 */
-	if (xen_feature(XENFEAT_auto_translated_physmap))
+	if (!xen_pv_domain())
 		return mfn;
 
 	pfn = mfn_to_pfn_no_overrides(mfn);
@@ -242,15 +242,15 @@ static inline xpaddr_t machine_to_phys(xmaddr_t machine)
 /* Pseudo-physical <-> Guest conversion */
 static inline unsigned long pfn_to_gfn(unsigned long pfn)
 {
-	if (xen_feature(XENFEAT_auto_translated_physmap))
+	if (!xen_pv_domain())
 		return pfn;
 	else
 		return pfn_to_mfn(pfn);
 }
 
 static inline unsigned long gfn_to_pfn(unsigned long gfn)
 {
-	if (xen_feature(XENFEAT_auto_translated_physmap))
+	if (!xen_pv_domain())
 		return gfn;
 	else
 		return mfn_to_pfn(gfn);
@@ -284,7 +284,7 @@ static inline unsigned long bfn_to_local_pfn(unsigned long mfn)
 {
 	unsigned long pfn;
 
-	if (xen_feature(XENFEAT_auto_translated_physmap))
+	if (!xen_pv_domain())
 		return mfn;
 
 	pfn = mfn_to_pfn(mfn);

arch/x86/xen/Kconfig

@@ -8,6 +8,7 @@ config XEN
 	depends on PARAVIRT
 	select PARAVIRT_CLOCK
 	select X86_HV_CALLBACK_VECTOR
+	select HIBERNATE_CALLBACKS
 	depends on X86_64 || (X86_32 && X86_PAE)
 	depends on X86_64 || (X86_GENERIC || MPENTIUM4 || MATOM)
 	depends on X86_LOCAL_APIC && X86_TSC
@@ -64,12 +65,6 @@ config XEN_PVHVM_GUEST
 	help
 	  Support running as a Xen PVHVM guest.
 
-config XEN_SAVE_RESTORE
-	bool
-	depends on XEN
-	select HIBERNATE_CALLBACKS
-	default y
-
 config XEN_DEBUG_FS
 	bool "Enable Xen debug and tuning parameters in debugfs"
 	depends on XEN && DEBUG_FS

arch/x86/xen/enlighten_pv.c

@@ -382,7 +382,6 @@ static bool __init xen_check_xsave(void)
 
 static void __init xen_init_capabilities(void)
 {
-	setup_force_cpu_cap(X86_FEATURE_XENPV);
 	setup_clear_cpu_cap(X86_FEATURE_DCA);
 	setup_clear_cpu_cap(X86_FEATURE_APERFMPERF);
 	setup_clear_cpu_cap(X86_FEATURE_MTRR);
@@ -1402,6 +1401,7 @@ asmlinkage __visible void __init xen_start_kernel(struct start_info *si)
 			JMP32_INSN_SIZE);
 
 	xen_domain_type = XEN_PV_DOMAIN;
+	setup_force_cpu_cap(X86_FEATURE_XENPV);
 	xen_start_flags = xen_start_info->flags;
 	/* Interrupts are guaranteed to be off initially. */
 	early_boot_irqs_disabled = true;

arch/x86/xen/mmu.c

@@ -41,7 +41,7 @@ EXPORT_SYMBOL_GPL(arbitrary_virt_to_machine);
 int xen_unmap_domain_gfn_range(struct vm_area_struct *vma,
 			       int nr, struct page **pages)
 {
-	if (xen_feature(XENFEAT_auto_translated_physmap))
+	if (!xen_pv_domain())
 		return xen_xlate_unmap_gfn_range(vma, nr, pages);
 
 	if (!pages)

arch/x86/xen/p2m.c

@@ -686,7 +686,7 @@ int set_foreign_p2m_mapping(struct gnttab_map_grant_ref *map_ops,
 	int i, ret = 0;
 	pte_t *pte;
 
-	if (xen_feature(XENFEAT_auto_translated_physmap))
+	if (!xen_pv_domain())
 		return 0;
 
 	if (kmap_ops) {
@@ -769,7 +769,7 @@ int clear_foreign_p2m_mapping(struct gnttab_unmap_grant_ref *unmap_ops,
 {
 	int i, ret = 0;
 
-	if (xen_feature(XENFEAT_auto_translated_physmap))
+	if (!xen_pv_domain())
 		return 0;
 
 	for (i = 0; i < count; i++) {

drivers/xen/balloon.c

@@ -302,7 +302,7 @@ static enum bp_state reserve_additional_memory(void)
          * are not restored since this region is now known not to
          * conflict with any devices.
          */ 
-	if (!xen_feature(XENFEAT_auto_translated_physmap)) {
+	if (xen_pv_domain()) {
 		unsigned long pfn, i;
 
 		pfn = PFN_DOWN(resource->start);
@@ -626,7 +626,7 @@ int xen_alloc_ballooned_pages(unsigned int nr_pages, struct page **pages)
 			 */
 			BUILD_BUG_ON(XEN_PAGE_SIZE != PAGE_SIZE);
 
-			if (!xen_feature(XENFEAT_auto_translated_physmap)) {
+			if (xen_pv_domain()) {
 				ret = xen_alloc_p2m_entry(page_to_pfn(page));
 				if (ret < 0)
 					goto out_undo;

drivers/xen/events/events_base.c

@@ -1314,27 +1314,34 @@ int bind_interdomain_evtchn_to_irq_lateeoi(struct xenbus_device *dev,
 }
 EXPORT_SYMBOL_GPL(bind_interdomain_evtchn_to_irq_lateeoi);
 
-static int find_virq(unsigned int virq, unsigned int cpu, evtchn_port_t *evtchn)
+static int find_virq(unsigned int virq, unsigned int cpu, evtchn_port_t *evtchn,
+		     bool percpu)
 {
 	struct evtchn_status status;
 	evtchn_port_t port;
-	int rc = -ENOENT;
+	bool exists = false;
 
 	memset(&status, 0, sizeof(status));
 	for (port = 0; port < xen_evtchn_max_channels(); port++) {
+		int rc;
+
 		status.dom = DOMID_SELF;
 		status.port = port;
 		rc = HYPERVISOR_event_channel_op(EVTCHNOP_status, &status);
 		if (rc < 0)
 			continue;
 		if (status.status != EVTCHNSTAT_virq)
 			continue;
-		if (status.u.virq == virq && status.vcpu == xen_vcpu_nr(cpu)) {
+		if (status.u.virq != virq)
+			continue;
+		if (status.vcpu == xen_vcpu_nr(cpu)) {
 			*evtchn = port;
-			break;
+			return 0;
+		} else if (!percpu) {
+			exists = true;
 		}
 	}
-	return rc;
+	return exists ? -EEXIST : -ENOENT;
 }
 
 /**
@@ -1381,8 +1388,11 @@ int bind_virq_to_irq(unsigned int virq, unsigned int cpu, bool percpu)
 			evtchn = bind_virq.port;
 		else {
 			if (ret == -EEXIST)
-				ret = find_virq(virq, cpu, &evtchn);
-			BUG_ON(ret < 0);
+				ret = find_virq(virq, cpu, &evtchn, percpu);
+			if (ret) {
+				__unbind_from_irq(info, info->irq);
+				goto out;
+			}
 		}
 
 		ret = xen_irq_info_virq_setup(info, cpu, evtchn, virq);
@@ -1787,9 +1797,20 @@ static int xen_rebind_evtchn_to_cpu(struct irq_info *info, unsigned int tcpu)
 	 * virq or IPI channel, which don't actually need to be rebound. Ignore
 	 * it, but don't do the xenlinux-level rebind in that case.
 	 */
-	if (HYPERVISOR_event_channel_op(EVTCHNOP_bind_vcpu, &bind_vcpu) >= 0)
+	if (HYPERVISOR_event_channel_op(EVTCHNOP_bind_vcpu, &bind_vcpu) >= 0) {
+		int old_cpu = info->cpu;
+
 		bind_evtchn_to_cpu(info, tcpu, false);
 
+		if (info->type == IRQT_VIRQ) {
+			int virq = info->u.virq;
+			int irq = per_cpu(virq_to_irq, old_cpu)[virq];
+
+			per_cpu(virq_to_irq, old_cpu)[virq] = -1;
+			per_cpu(virq_to_irq, tcpu)[virq] = irq;
+		}
+	}
+
 	do_unmask(info, EVT_MASK_REASON_TEMPORARY);
 
 	return 0;

drivers/xen/gntdev-dmabuf.c

@@ -720,16 +720,15 @@ static void dmabuf_imp_release_all(struct gntdev_dmabuf_priv *priv)
 
 /* DMA buffer IOCTL support. */
 
-long gntdev_ioctl_dmabuf_exp_from_refs(struct gntdev_priv *priv, int use_ptemod,
+long gntdev_ioctl_dmabuf_exp_from_refs(struct gntdev_priv *priv,
 				       struct ioctl_gntdev_dmabuf_exp_from_refs __user *u)
 {
 	struct ioctl_gntdev_dmabuf_exp_from_refs op;
 	u32 *refs;
 	long ret;
 
-	if (use_ptemod) {
-		pr_debug("Cannot provide dma-buf: use_ptemode %d\n",
-			 use_ptemod);
+	if (xen_pv_domain()) {
+		pr_debug("Cannot provide dma-buf in a PV domain\n");
 		return -EINVAL;
 	}
 

drivers/xen/gntdev-dmabuf.h

@@ -18,7 +18,7 @@ struct gntdev_dmabuf_priv *gntdev_dmabuf_init(struct file *filp);
 
 void gntdev_dmabuf_fini(struct gntdev_dmabuf_priv *priv);
 
-long gntdev_ioctl_dmabuf_exp_from_refs(struct gntdev_priv *priv, int use_ptemod,
+long gntdev_ioctl_dmabuf_exp_from_refs(struct gntdev_priv *priv,
 				       struct ioctl_gntdev_dmabuf_exp_from_refs __user *u);
 
 long gntdev_ioctl_dmabuf_exp_wait_released(struct gntdev_priv *priv,