Merge tag 's390-6.19-2' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux

Pull more s390 updates from Heiko Carstens:

 - Use the MSI parent domain API instead of the legacy API for setup and
   teardown of PCI MSI IRQs

 - Select POSIX_CPU_TIMERS_TASK_WORK now that VIRT_XFER_TO_GUEST_WORK
   has been implemented for s390

 - Fix a KVM bug which can lead to guest memory corruption

 - Fix KASAN shadow memory mapping for hotplugged memory

 - Minor bug fixes and improvements

* tag 's390-6.19-2' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux:
  s390/bug: Add missing alignment
  s390/bug: Add missing CONFIG_BUG ifdef again
  KVM: s390: Fix gmap_helper_zap_one_page() again
  s390/pci: Migrate s390 IRQ logic to IRQ domain API
  genirq: Change hwirq parameter to irq_hw_number_t
  s390: Select POSIX_CPU_TIMERS_TASK_WORK
  s390: Unmap early KASAN shadow on memory offlining
  s390/vmem: Support 2G page splitting for KASAN shadow freeing
  s390/boot: Use entire page for PTEs
  s390/vmur: Use scnprintf() instead of sprintf()

Author: torvalds

arch/s390/Kconfig

@@ -238,6 +238,7 @@ config S390
 	select HAVE_PERF_EVENTS
 	select HAVE_PERF_REGS
 	select HAVE_PERF_USER_STACK_DUMP
+	select HAVE_POSIX_CPU_TIMERS_TASK_WORK
 	select HAVE_PREEMPT_DYNAMIC_KEY
 	select HAVE_REGS_AND_STACK_ACCESS_API
 	select HAVE_RELIABLE_STACKTRACE
@@ -254,6 +255,7 @@ config S390
 	select HOTPLUG_SMT
 	select IOMMU_HELPER		if PCI
 	select IOMMU_SUPPORT		if PCI
+	select IRQ_MSI_LIB		if PCI
 	select KASAN_VMALLOC if KASAN
 	select LOCK_MM_AND_FIND_VMA
 	select MMU_GATHER_MERGE_VMAS

arch/s390/boot/vmem.c

@@ -244,22 +244,10 @@ static void *boot_crst_alloc(unsigned long val)
 
 static pte_t *boot_pte_alloc(void)
 {
-	static void *pte_leftover;
 	pte_t *pte;
 
-	/*
-	 * handling pte_leftovers this way helps to avoid memory fragmentation
-	 * during POPULATE_KASAN_MAP_SHADOW when EDAT is off
-	 */
-	if (!pte_leftover) {
-		pte_leftover = (void *)physmem_alloc_or_die(RR_VMEM, PAGE_SIZE, PAGE_SIZE);
-		pte = pte_leftover + _PAGE_TABLE_SIZE;
-		__arch_set_page_dat(pte, 1);
-	} else {
-		pte = pte_leftover;
-		pte_leftover = NULL;
-	}
-
+	pte = (void *)physmem_alloc_or_die(RR_VMEM, PAGE_SIZE, PAGE_SIZE);
+	__arch_set_page_dat(pte, 1);
 	memset64((u64 *)pte, _PAGE_INVALID, PTRS_PER_PTE);
 	return pte;
 }

arch/s390/include/asm/bug.h

@@ -4,11 +4,14 @@
 
 #include <linux/stringify.h>
 
+#ifdef CONFIG_BUG
+
 #ifndef CONFIG_DEBUG_BUGVERBOSE
 #define _BUGVERBOSE_LOCATION(file, line)
 #else
 #define __BUGVERBOSE_LOCATION(file, line)			\
 		.pushsection .rodata.str, "aMS", @progbits, 1;	\
+		.align 2;					\
 	10002:	.ascii file "\0";				\
 		.popsection;					\
 								\
@@ -52,6 +55,8 @@ do {								\
 
 #define HAVE_ARCH_BUG
 
+#endif /* CONFIG_BUG */
+
 #include <asm-generic/bug.h>
 
 #endif /* _ASM_S390_BUG_H */

arch/s390/include/asm/page.h

@@ -166,6 +166,8 @@ static inline int page_reset_referenced(unsigned long addr)
 	return CC_TRANSFORM(cc);
 }
 
+int split_pud_page(pud_t *pudp, unsigned long addr);
+
 /* Bits int the storage key */
 #define _PAGE_CHANGED		0x02	/* HW changed bit		*/
 #define _PAGE_REFERENCED	0x04	/* HW referenced bit		*/

arch/s390/include/asm/pci.h

@@ -5,6 +5,7 @@
 #include <linux/pci.h>
 #include <linux/mutex.h>
 #include <linux/iommu.h>
+#include <linux/irqdomain.h>
 #include <linux/pci_hotplug.h>
 #include <asm/pci_clp.h>
 #include <asm/pci_debug.h>
@@ -109,6 +110,7 @@ struct zpci_bus {
 	struct list_head	resources;
 	struct list_head	bus_next;
 	struct resource		bus_resource;
+	struct irq_domain	*msi_parent_domain;
 	int			topo;		/* TID if topo_is_tid, PCHID otherwise */
 	int			domain_nr;
 	u8			multifunction	: 1;
@@ -310,6 +312,9 @@ int zpci_dma_exit_device(struct zpci_dev *zdev);
 /* IRQ */
 int __init zpci_irq_init(void);
 void __init zpci_irq_exit(void);
+int zpci_set_irq(struct zpci_dev *zdev);
+int zpci_create_parent_msi_domain(struct zpci_bus *zbus);
+void zpci_remove_parent_msi_domain(struct zpci_bus *zbus);
 
 /* FMB */
 int zpci_fmb_enable_device(struct zpci_dev *);

arch/s390/mm/gmap_helpers.c

@@ -47,6 +47,7 @@ static void ptep_zap_softleaf_entry(struct mm_struct *mm, softleaf_t entry)
 void gmap_helper_zap_one_page(struct mm_struct *mm, unsigned long vmaddr)
 {
 	struct vm_area_struct *vma;
+	unsigned long pgstev;
 	spinlock_t *ptl;
 	pgste_t pgste;
 	pte_t *ptep;
@@ -65,9 +66,13 @@ void gmap_helper_zap_one_page(struct mm_struct *mm, unsigned long vmaddr)
 	if (pte_swap(*ptep)) {
 		preempt_disable();
 		pgste = pgste_get_lock(ptep);
+		pgstev = pgste_val(pgste);
 
-		ptep_zap_softleaf_entry(mm, softleaf_from_pte(*ptep));
-		pte_clear(mm, vmaddr, ptep);
+		if ((pgstev & _PGSTE_GPS_USAGE_MASK) == _PGSTE_GPS_USAGE_UNUSED ||
+		    (pgstev & _PGSTE_GPS_ZERO)) {
+			ptep_zap_softleaf_entry(mm, softleaf_from_pte(*ptep));
+			pte_clear(mm, vmaddr, ptep);
+		}
 
 		pgste_set_unlock(ptep, pgste);
 		preempt_enable();

arch/s390/mm/pageattr.c

@@ -204,7 +204,7 @@ static int walk_pmd_level(pud_t *pudp, unsigned long addr, unsigned long end,
 	return rc;
 }
 
-static int split_pud_page(pud_t *pudp, unsigned long addr)
+int split_pud_page(pud_t *pudp, unsigned long addr)
 {
 	unsigned long pmd_addr, prot;
 	pmd_t *pm_dir, *pmdp;

arch/s390/mm/vmem.c

@@ -330,10 +330,14 @@ static int modify_pud_table(p4d_t *p4d, unsigned long addr, unsigned long end,
 			if (pud_leaf(*pud)) {
 				if (IS_ALIGNED(addr, PUD_SIZE) &&
 				    IS_ALIGNED(next, PUD_SIZE)) {
+					if (!direct)
+						vmem_free_pages(pud_deref(*pud), get_order(PUD_SIZE), altmap);
 					pud_clear(pud);
 					pages++;
+					continue;
+				} else {
+					split_pud_page(pud, addr & PUD_MASK);
 				}
-				continue;
 			}
 		} else if (pud_none(*pud)) {
 			if (IS_ALIGNED(addr, PUD_SIZE) &&
@@ -433,9 +437,15 @@ static int modify_pagetable(unsigned long start, unsigned long end, bool add,
 
 	if (WARN_ON_ONCE(!PAGE_ALIGNED(start | end)))
 		return -EINVAL;
-	/* Don't mess with any tables not fully in 1:1 mapping & vmemmap area */
+	/* Don't mess with any tables not fully in 1:1 mapping, vmemmap & kasan area */
+#ifdef CONFIG_KASAN
+	if (WARN_ON_ONCE(!(start >= KASAN_SHADOW_START && end <= KASAN_SHADOW_END) &&
+			 end > __abs_lowcore))
+		return -EINVAL;
+#else
 	if (WARN_ON_ONCE(end > __abs_lowcore))
 		return -EINVAL;
+#endif
 	for (addr = start; addr < end; addr = next) {
 		next = pgd_addr_end(addr, end);
 		pgd = pgd_offset_k(addr);

arch/s390/pci/pci.c

@@ -708,6 +708,12 @@ int zpci_reenable_device(struct zpci_dev *zdev)
 	if (rc)
 		return rc;
 
+	if (zdev->msi_nr_irqs > 0) {
+		rc = zpci_set_irq(zdev);
+		if (rc)
+			return rc;
+	}
+
 	rc = zpci_iommu_register_ioat(zdev, &status);
 	if (rc)
 		zpci_disable_device(zdev);

arch/s390/pci/pci_bus.c

@@ -14,6 +14,7 @@
 #include <linux/err.h>
 #include <linux/delay.h>
 #include <linux/seq_file.h>
+#include <linux/irqdomain.h>
 #include <linux/jump_label.h>
 #include <linux/pci.h>
 #include <linux/printk.h>
@@ -198,19 +199,27 @@ static int zpci_bus_create_pci_bus(struct zpci_bus *zbus, struct zpci_dev *fr, s
 	zbus->multifunction = zpci_bus_is_multifunction_root(fr);
 	zbus->max_bus_speed = fr->max_bus_speed;
 
+	if (zpci_create_parent_msi_domain(zbus))
+		goto out_free_domain;
+
 	/*
 	 * Note that the zbus->resources are taken over and zbus->resources
 	 * is empty after a successful call
 	 */
 	bus = pci_create_root_bus(NULL, ZPCI_BUS_NR, ops, zbus, &zbus->resources);
-	if (!bus) {
-		zpci_free_domain(zbus->domain_nr);
-		return -EFAULT;
-	}
+	if (!bus)
+		goto out_remove_msi_domain;
 
 	zbus->bus = bus;
+	dev_set_msi_domain(&zbus->bus->dev, zbus->msi_parent_domain);
 
 	return 0;
+
+out_remove_msi_domain:
+	zpci_remove_parent_msi_domain(zbus);
+out_free_domain:
+	zpci_free_domain(zbus->domain_nr);
+	return -ENOMEM;
 }
 
 static void zpci_bus_release(struct kref *kref)
@@ -231,6 +240,7 @@ static void zpci_bus_release(struct kref *kref)
 	mutex_lock(&zbus_list_lock);
 	list_del(&zbus->bus_next);
 	mutex_unlock(&zbus_list_lock);
+	zpci_remove_parent_msi_domain(zbus);
 	kfree(zbus);
 }