Merge tag 's390-6.13-2' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux

Pull more s390 updates from Heiko Carstens:

 - Add swap entry for hugetlbfs support

 - Add PTE_MARKER support for hugetlbs mappings; this fixes a regression
   (possible page fault loop) which was introduced when support for
   UFFDIO_POISON for hugetlbfs was added

 - Add ARCH_HAS_PREEMPT_LAZY and PREEMPT_DYNAMIC support

 - Mark IRQ entries in entry code, so that stack tracers can filter out
   the non-IRQ parts of stack traces. This fixes stack depot capacity
   limit warnings, since without filtering the number of unique stack
   traces is huge

 - In PCI code fix leak of struct zpci_dev object, and fix potential
   double remove of hotplug slot

 - Fix pagefault_disable() / pagefault_enable() unbalance in
   arch_stack_user_walk_common()

 - A couple of inline assembly optimizations, more cmpxchg() to
   try_cmpxchg() conversions, and removal of usages of xchg() and
   cmpxchg() on one and two byte memory areas

 - Various other small improvements and cleanups

* tag 's390-6.13-2' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux: (27 commits)
  Revert "s390/mm: Allow large pages for KASAN shadow mapping"
  s390/spinlock: Use flag output constraint for arch_cmpxchg_niai8()
  s390/spinlock: Use R constraint for arch_load_niai4()
  s390/spinlock: Generate shorter code for arch_spin_unlock()
  s390/spinlock: Remove condition code clobber from arch_spin_unlock()
  s390/spinlock: Use symbolic names in inline assemblies
  s390: Support PREEMPT_DYNAMIC
  s390/pci: Fix potential double remove of hotplug slot
  s390/pci: Fix leak of struct zpci_dev when zpci_add_device() fails
  s390/mm/hugetlbfs: Add missing includes
  s390/mm: Add PTE_MARKER support for hugetlbfs mappings
  s390/mm: Introduce region-third and segment table swap entries
  s390/mm: Introduce region-third and segment table entry present bits
  s390/mm: Rearrange region-third and segment table entry SW bits
  KVM: s390: Increase size of union sca_utility to four bytes
  KVM: s390: Remove one byte cmpxchg() usage
  KVM: s390: Use try_cmpxchg() instead of cmpxchg() loops
  s390/ap: Replace xchg() with WRITE_ONCE()
  s390/mm: Allow large pages for KASAN shadow mapping
  s390: Add ARCH_HAS_PREEMPT_LAZY support
  ...

Author: torvalds

arch/s390/Kconfig

@@ -87,6 +87,7 @@ config S390
 	select ARCH_HAS_MEMBARRIER_SYNC_CORE
 	select ARCH_HAS_MEM_ENCRYPT
 	select ARCH_HAS_NMI_SAFE_THIS_CPU_OPS
+	select ARCH_HAS_PREEMPT_LAZY
 	select ARCH_HAS_PTE_SPECIAL
 	select ARCH_HAS_SCALED_CPUTIME
 	select ARCH_HAS_SET_DIRECT_MAP
@@ -218,6 +219,7 @@ config S390
 	select HAVE_PERF_EVENTS
 	select HAVE_PERF_REGS
 	select HAVE_PERF_USER_STACK_DUMP
+	select HAVE_PREEMPT_DYNAMIC_KEY
 	select HAVE_REGS_AND_STACK_ACCESS_API
 	select HAVE_RELIABLE_STACKTRACE
 	select HAVE_RETHOOK

arch/s390/include/asm/debug.h

@@ -66,22 +66,24 @@ typedef int (debug_header_proc_t) (debug_info_t *id,
 				   struct debug_view *view,
 				   int area,
 				   debug_entry_t *entry,
-				   char *out_buf);
+				   char *out_buf, size_t out_buf_size);
 
 typedef int (debug_format_proc_t) (debug_info_t *id,
 				   struct debug_view *view, char *out_buf,
+				   size_t out_buf_size,
 				   const char *in_buf);
 typedef int (debug_prolog_proc_t) (debug_info_t *id,
 				   struct debug_view *view,
-				   char *out_buf);
+				   char *out_buf, size_t out_buf_size);
 typedef int (debug_input_proc_t) (debug_info_t *id,
 				  struct debug_view *view,
 				  struct file *file,
 				  const char __user *user_buf,
 				  size_t in_buf_size, loff_t *offset);
 
 int debug_dflt_header_fn(debug_info_t *id, struct debug_view *view,
-			 int area, debug_entry_t *entry, char *out_buf);
+			 int area, debug_entry_t *entry,
+			 char *out_buf, size_t out_buf_size);
 
 struct debug_view {
 	char name[DEBUG_MAX_NAME_LEN];

arch/s390/include/asm/gmap.h

@@ -17,8 +17,8 @@
 #define GMAP_NOTIFY_MPROT	0x1
 
 /* Status bits only for huge segment entries */
-#define _SEGMENT_ENTRY_GMAP_IN		0x8000	/* invalidation notify bit */
-#define _SEGMENT_ENTRY_GMAP_UC		0x4000	/* dirty (migration) */
+#define _SEGMENT_ENTRY_GMAP_IN		0x0800	/* invalidation notify bit */
+#define _SEGMENT_ENTRY_GMAP_UC		0x0002	/* dirty (migration) */
 
 /**
  * struct gmap_struct - guest address space

arch/s390/include/asm/hugetlb.h

@@ -10,6 +10,8 @@
 #define _ASM_S390_HUGETLB_H
 
 #include <linux/pgtable.h>
+#include <linux/swap.h>
+#include <linux/swapops.h>
 #include <asm/page.h>
 
 #define hugepages_supported()			(MACHINE_HAS_EDAT1)
@@ -78,7 +80,7 @@ static inline int huge_pte_none(pte_t pte)
 #define __HAVE_ARCH_HUGE_PTE_NONE_MOSTLY
 static inline int huge_pte_none_mostly(pte_t pte)
 {
-	return huge_pte_none(pte);
+	return huge_pte_none(pte) || is_pte_marker(pte);
 }
 
 #define __HAVE_ARCH_HUGE_PTE_MKUFFD_WP

arch/s390/include/asm/kvm_host.h

@@ -94,11 +94,16 @@ union ipte_control {
 	};
 };
 
+/*
+ * Utility is defined as two bytes but having it four bytes wide
+ * generates more efficient code. Since the following bytes are
+ * reserved this makes no functional difference.
+ */
 union sca_utility {
-	__u16 val;
+	__u32 val;
 	struct {
-		__u16 mtcr : 1;
-		__u16 reserved : 15;
+		__u32 mtcr : 1;
+		__u32	   : 31;
 	};
 };
 
@@ -107,15 +112,15 @@ struct bsca_block {
 	__u64	reserved[5];
 	__u64	mcn;
 	union sca_utility utility;
-	__u8	reserved2[6];
+	__u8	reserved2[4];
 	struct bsca_entry cpu[KVM_S390_BSCA_CPU_SLOTS];
 };
 
 struct esca_block {
 	union ipte_control ipte_control;
 	__u64   reserved1[6];
 	union sca_utility utility;
-	__u8	reserved2[6];
+	__u8	reserved2[4];
 	__u64   mcn[4];
 	__u64   reserved3[20];
 	struct esca_entry cpu[KVM_S390_ESCA_CPU_SLOTS];

arch/s390/include/asm/pgtable.h

@@ -277,26 +277,36 @@ static inline int is_module_addr(void *addr)
 #define _REGION1_ENTRY_EMPTY	(_REGION_ENTRY_TYPE_R1 | _REGION_ENTRY_INVALID)
 #define _REGION2_ENTRY		(_REGION_ENTRY_TYPE_R2 | _REGION_ENTRY_LENGTH)
 #define _REGION2_ENTRY_EMPTY	(_REGION_ENTRY_TYPE_R2 | _REGION_ENTRY_INVALID)
-#define _REGION3_ENTRY		(_REGION_ENTRY_TYPE_R3 | _REGION_ENTRY_LENGTH)
+#define _REGION3_ENTRY		(_REGION_ENTRY_TYPE_R3 | _REGION_ENTRY_LENGTH | \
+				 _REGION3_ENTRY_PRESENT)
 #define _REGION3_ENTRY_EMPTY	(_REGION_ENTRY_TYPE_R3 | _REGION_ENTRY_INVALID)
 
 #define _REGION3_ENTRY_HARDWARE_BITS		0xfffffffffffff6ffUL
 #define _REGION3_ENTRY_HARDWARE_BITS_LARGE	0xffffffff8001073cUL
 #define _REGION3_ENTRY_ORIGIN_LARGE ~0x7fffffffUL /* large page address	     */
 #define _REGION3_ENTRY_DIRTY	0x2000	/* SW region dirty bit */
 #define _REGION3_ENTRY_YOUNG	0x1000	/* SW region young bit */
+#define _REGION3_ENTRY_COMM	0x0010	/* Common-Region, marks swap entry */
 #define _REGION3_ENTRY_LARGE	0x0400	/* RTTE-format control, large page  */
-#define _REGION3_ENTRY_WRITE	0x0002	/* SW region write bit */
-#define _REGION3_ENTRY_READ	0x0001	/* SW region read bit */
+#define _REGION3_ENTRY_WRITE	0x8000	/* SW region write bit */
+#define _REGION3_ENTRY_READ	0x4000	/* SW region read bit */
 
 #ifdef CONFIG_MEM_SOFT_DIRTY
-#define _REGION3_ENTRY_SOFT_DIRTY 0x4000 /* SW region soft dirty bit */
+#define _REGION3_ENTRY_SOFT_DIRTY 0x0002 /* SW region soft dirty bit */
 #else
 #define _REGION3_ENTRY_SOFT_DIRTY 0x0000 /* SW region soft dirty bit */
 #endif
 
 #define _REGION_ENTRY_BITS	 0xfffffffffffff22fUL
 
+/*
+ * SW region present bit. For non-leaf region-third-table entries, bits 62-63
+ * indicate the TABLE LENGTH and both must be set to 1. But such entries
+ * would always be considered as present, so it is safe to use bit 63 as
+ * PRESENT bit for PUD.
+ */
+#define _REGION3_ENTRY_PRESENT	0x0001
+
 /* Bits in the segment table entry */
 #define _SEGMENT_ENTRY_BITS			0xfffffffffffffe3fUL
 #define _SEGMENT_ENTRY_HARDWARE_BITS		0xfffffffffffffe3cUL
@@ -308,21 +318,29 @@ static inline int is_module_addr(void *addr)
 #define _SEGMENT_ENTRY_INVALID	0x20	/* invalid segment table entry	    */
 #define _SEGMENT_ENTRY_TYPE_MASK 0x0c	/* segment table type mask	    */
 
-#define _SEGMENT_ENTRY		(0)
+#define _SEGMENT_ENTRY		(_SEGMENT_ENTRY_PRESENT)
 #define _SEGMENT_ENTRY_EMPTY	(_SEGMENT_ENTRY_INVALID)
 
 #define _SEGMENT_ENTRY_DIRTY	0x2000	/* SW segment dirty bit */
 #define _SEGMENT_ENTRY_YOUNG	0x1000	/* SW segment young bit */
+
+#define _SEGMENT_ENTRY_COMM	0x0010	/* Common-Segment, marks swap entry */
 #define _SEGMENT_ENTRY_LARGE	0x0400	/* STE-format control, large page */
-#define _SEGMENT_ENTRY_WRITE	0x0002	/* SW segment write bit */
-#define _SEGMENT_ENTRY_READ	0x0001	/* SW segment read bit */
+#define _SEGMENT_ENTRY_WRITE	0x8000	/* SW segment write bit */
+#define _SEGMENT_ENTRY_READ	0x4000	/* SW segment read bit */
 
 #ifdef CONFIG_MEM_SOFT_DIRTY
-#define _SEGMENT_ENTRY_SOFT_DIRTY 0x4000 /* SW segment soft dirty bit */
+#define _SEGMENT_ENTRY_SOFT_DIRTY 0x0002 /* SW segment soft dirty bit */
 #else
 #define _SEGMENT_ENTRY_SOFT_DIRTY 0x0000 /* SW segment soft dirty bit */
 #endif
 
+#define _SEGMENT_ENTRY_PRESENT	0x0001	/* SW segment present bit */
+
+/* Common bits in region and segment table entries, for swap entries */
+#define _RST_ENTRY_COMM		0x0010	/* Common-Region/Segment, marks swap entry */
+#define _RST_ENTRY_INVALID	0x0020	/* invalid region/segment table entry */
+
 #define _CRST_ENTRIES	2048	/* number of region/segment table entries */
 #define _PAGE_ENTRIES	256	/* number of page table entries	*/
 
@@ -454,17 +472,22 @@ static inline int is_module_addr(void *addr)
 /*
  * Segment entry (large page) protection definitions.
  */
-#define SEGMENT_NONE	__pgprot(_SEGMENT_ENTRY_INVALID | \
+#define SEGMENT_NONE	__pgprot(_SEGMENT_ENTRY_PRESENT | \
+				 _SEGMENT_ENTRY_INVALID | \
 				 _SEGMENT_ENTRY_PROTECT)
-#define SEGMENT_RO	__pgprot(_SEGMENT_ENTRY_PROTECT | \
+#define SEGMENT_RO	__pgprot(_SEGMENT_ENTRY_PRESENT | \
+				 _SEGMENT_ENTRY_PROTECT | \
 				 _SEGMENT_ENTRY_READ | \
 				 _SEGMENT_ENTRY_NOEXEC)
-#define SEGMENT_RX	__pgprot(_SEGMENT_ENTRY_PROTECT | \
+#define SEGMENT_RX	__pgprot(_SEGMENT_ENTRY_PRESENT | \
+				 _SEGMENT_ENTRY_PROTECT | \
 				 _SEGMENT_ENTRY_READ)
-#define SEGMENT_RW	__pgprot(_SEGMENT_ENTRY_READ | \
+#define SEGMENT_RW	__pgprot(_SEGMENT_ENTRY_PRESENT | \
+				 _SEGMENT_ENTRY_READ | \
 				 _SEGMENT_ENTRY_WRITE | \
 				 _SEGMENT_ENTRY_NOEXEC)
-#define SEGMENT_RWX	__pgprot(_SEGMENT_ENTRY_READ | \
+#define SEGMENT_RWX	__pgprot(_SEGMENT_ENTRY_PRESENT | \
+				 _SEGMENT_ENTRY_READ | \
 				 _SEGMENT_ENTRY_WRITE)
 #define SEGMENT_KERNEL	__pgprot(_SEGMENT_ENTRY |	\
 				 _SEGMENT_ENTRY_LARGE |	\
@@ -491,19 +514,22 @@ static inline int is_module_addr(void *addr)
  */
 
 #define REGION3_KERNEL	__pgprot(_REGION_ENTRY_TYPE_R3 | \
+				 _REGION3_ENTRY_PRESENT | \
 				 _REGION3_ENTRY_LARGE |	 \
 				 _REGION3_ENTRY_READ |	 \
 				 _REGION3_ENTRY_WRITE |	 \
 				 _REGION3_ENTRY_YOUNG |	 \
 				 _REGION3_ENTRY_DIRTY | \
 				 _REGION_ENTRY_NOEXEC)
 #define REGION3_KERNEL_RO __pgprot(_REGION_ENTRY_TYPE_R3 | \
+				   _REGION3_ENTRY_PRESENT | \
 				   _REGION3_ENTRY_LARGE |  \
 				   _REGION3_ENTRY_READ |   \
 				   _REGION3_ENTRY_YOUNG |  \
 				   _REGION_ENTRY_PROTECT | \
 				   _REGION_ENTRY_NOEXEC)
 #define REGION3_KERNEL_EXEC __pgprot(_REGION_ENTRY_TYPE_R3 | \
+				 _REGION3_ENTRY_PRESENT | \
 				 _REGION3_ENTRY_LARGE |	 \
 				 _REGION3_ENTRY_READ |	 \
 				 _REGION3_ENTRY_WRITE |	 \
@@ -746,7 +772,7 @@ static inline int pud_present(pud_t pud)
 {
 	if (pud_folded(pud))
 		return 1;
-	return (pud_val(pud) & _REGION_ENTRY_ORIGIN) != 0UL;
+	return (pud_val(pud) & _REGION3_ENTRY_PRESENT) != 0;
 }
 
 static inline int pud_none(pud_t pud)
@@ -761,13 +787,18 @@ static inline bool pud_leaf(pud_t pud)
 {
 	if ((pud_val(pud) & _REGION_ENTRY_TYPE_MASK) != _REGION_ENTRY_TYPE_R3)
 		return 0;
-	return !!(pud_val(pud) & _REGION3_ENTRY_LARGE);
+	return (pud_present(pud) && (pud_val(pud) & _REGION3_ENTRY_LARGE) != 0);
+}
+
+static inline int pmd_present(pmd_t pmd)
+{
+	return (pmd_val(pmd) & _SEGMENT_ENTRY_PRESENT) != 0;
 }
 
 #define pmd_leaf pmd_leaf
 static inline bool pmd_leaf(pmd_t pmd)
 {
-	return (pmd_val(pmd) & _SEGMENT_ENTRY_LARGE) != 0;
+	return (pmd_present(pmd) && (pmd_val(pmd) & _SEGMENT_ENTRY_LARGE) != 0);
 }
 
 static inline int pmd_bad(pmd_t pmd)
@@ -799,11 +830,6 @@ static inline int p4d_bad(p4d_t p4d)
 	return (p4d_val(p4d) & ~_REGION_ENTRY_BITS) != 0;
 }
 
-static inline int pmd_present(pmd_t pmd)
-{
-	return pmd_val(pmd) != _SEGMENT_ENTRY_EMPTY;
-}
-
 static inline int pmd_none(pmd_t pmd)
 {
 	return pmd_val(pmd) == _SEGMENT_ENTRY_EMPTY;
@@ -1851,7 +1877,7 @@ static inline pmd_t pmdp_collapse_flush(struct vm_area_struct *vma,
 
 static inline int pmd_trans_huge(pmd_t pmd)
 {
-	return pmd_val(pmd) & _SEGMENT_ENTRY_LARGE;
+	return pmd_leaf(pmd);
 }
 
 #define has_transparent_hugepage has_transparent_hugepage
@@ -1911,6 +1937,53 @@ static inline swp_entry_t __swp_entry(unsigned long type, unsigned long offset)
 #define __pte_to_swp_entry(pte)	((swp_entry_t) { pte_val(pte) })
 #define __swp_entry_to_pte(x)	((pte_t) { (x).val })
 
+/*
+ * 64 bit swap entry format for REGION3 and SEGMENT table entries (RSTE)
+ * Bits 59 and 63 are used to indicate the swap entry. Bit 58 marks the rste
+ * as invalid.
+ * A swap entry is indicated by bit pattern (rste & 0x011) == 0x010
+ * |			  offset			|Xtype |11TT|S0|
+ * |0000000000111111111122222222223333333333444444444455|555555|5566|66|
+ * |0123456789012345678901234567890123456789012345678901|234567|8901|23|
+ *
+ * Bits 0-51 store the offset.
+ * Bits 53-57 store the type.
+ * Bit 62 (S) is used for softdirty tracking.
+ * Bits 60-61 (TT) indicate the table type: 0x01 for REGION3 and 0x00 for SEGMENT.
+ * Bit 52 (X) is unused.
+ */
+
+#define __SWP_OFFSET_MASK_RSTE	((1UL << 52) - 1)
+#define __SWP_OFFSET_SHIFT_RSTE	12
+#define __SWP_TYPE_MASK_RSTE		((1UL << 5) - 1)
+#define __SWP_TYPE_SHIFT_RSTE	6
+
+/*
+ * TT bits set to 0x00 == SEGMENT. For REGION3 entries, caller must add R3
+ * bits 0x01. See also __set_huge_pte_at().
+ */
+static inline unsigned long mk_swap_rste(unsigned long type, unsigned long offset)
+{
+	unsigned long rste;
+
+	rste = _RST_ENTRY_INVALID | _RST_ENTRY_COMM;
+	rste |= (offset & __SWP_OFFSET_MASK_RSTE) << __SWP_OFFSET_SHIFT_RSTE;
+	rste |= (type & __SWP_TYPE_MASK_RSTE) << __SWP_TYPE_SHIFT_RSTE;
+	return rste;
+}
+
+static inline unsigned long __swp_type_rste(swp_entry_t entry)
+{
+	return (entry.val >> __SWP_TYPE_SHIFT_RSTE) & __SWP_TYPE_MASK_RSTE;
+}
+
+static inline unsigned long __swp_offset_rste(swp_entry_t entry)
+{
+	return (entry.val >> __SWP_OFFSET_SHIFT_RSTE) & __SWP_OFFSET_MASK_RSTE;
+}
+
+#define __rste_to_swp_entry(rste)	((swp_entry_t) { rste })
+
 extern int vmem_add_mapping(unsigned long start, unsigned long size);
 extern void vmem_remove_mapping(unsigned long start, unsigned long size);
 extern int __vmem_map_4k_page(unsigned long addr, unsigned long phys, pgprot_t prot, bool alloc);

arch/s390/include/asm/preempt.h

@@ -130,10 +130,24 @@ static __always_inline bool should_resched(int preempt_offset)
 #define init_idle_preempt_count(p, cpu)	do { } while (0)
 
 #ifdef CONFIG_PREEMPTION
-extern void preempt_schedule(void);
-#define __preempt_schedule() preempt_schedule()
-extern void preempt_schedule_notrace(void);
-#define __preempt_schedule_notrace() preempt_schedule_notrace()
+
+void preempt_schedule(void);
+void preempt_schedule_notrace(void);
+
+#ifdef CONFIG_PREEMPT_DYNAMIC
+
+void dynamic_preempt_schedule(void);
+void dynamic_preempt_schedule_notrace(void);
+#define __preempt_schedule()		dynamic_preempt_schedule()
+#define __preempt_schedule_notrace()	dynamic_preempt_schedule_notrace()
+
+#else /* CONFIG_PREEMPT_DYNAMIC */
+
+#define __preempt_schedule()		preempt_schedule()
+#define __preempt_schedule_notrace()	preempt_schedule_notrace()
+
+#endif /* CONFIG_PREEMPT_DYNAMIC */
+
 #endif /* CONFIG_PREEMPTION */
 
 #endif /* __ASM_PREEMPT_H */

arch/s390/include/asm/spinlock.h

@@ -82,9 +82,10 @@ static inline void arch_spin_unlock(arch_spinlock_t *lp)
 	kcsan_release();
 	asm_inline volatile(
 		ALTERNATIVE("nop", ".insn rre,0xb2fa0000,7,0", ALT_FACILITY(49)) /* NIAI 7 */
-		"	sth	%1,%0\n"
-		: "=R" (((unsigned short *) &lp->lock)[1])
-		: "d" (0) : "cc", "memory");
+		"	mvhhi	%[lock],0\n"
+		: [lock] "=Q" (((unsigned short *)&lp->lock)[1])
+		:
+		: "memory");
 }
 
 /*