Merge tag 's390-6.11-2' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux

Pull more s390 updates from Vasily Gorbik:

 - Fix KMSAN build breakage caused by the conflict between s390 and
   mm-stable trees

 - Add KMSAN page markers for ptdump

 - Add runtime constant support

 - Fix __pa/__va for modules under non-GPL licenses by exporting
   necessary vm_layout struct with EXPORT_SYMBOL to prevent linkage
   problems

 - Fix an endless loop in the CF_DIAG event stop in the CPU Measurement
   Counter Facility code when the counter set size is zero

 - Remove the PROTECTED_VIRTUALIZATION_GUEST config option and enable
   its functionality by default

 - Support allocation of multiple MSI interrupts per device and improve
   logging of architecture-specific limitations

 - Add support for lowcore relocation as a debugging feature to catch
   all null ptr dereferences in the kernel address space, improving
   detection beyond the current implementation's limited write access
   protection

 - Clean up and rework CPU alternatives to allow for callbacks and early
   patching for the lowcore relocation

* tag 's390-6.11-2' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux: (39 commits)
  s390: Remove protvirt and kvm config guards for uv code
  s390/boot: Add cmdline option to relocate lowcore
  s390/kdump: Make kdump ready for lowcore relocation
  s390/entry: Make system_call() ready for lowcore relocation
  s390/entry: Make ret_from_fork() ready for lowcore relocation
  s390/entry: Make __switch_to() ready for lowcore relocation
  s390/entry: Make restart_int_handler() ready for lowcore relocation
  s390/entry: Make mchk_int_handler() ready for lowcore relocation
  s390/entry: Make int handlers ready for lowcore relocation
  s390/entry: Make pgm_check_handler() ready for lowcore relocation
  s390/entry: Add base register to CHECK_VMAP_STACK/CHECK_STACK macro
  s390/entry: Add base register to SIEEXIT macro
  s390/entry: Add base register to MBEAR macro
  s390/entry: Make __sie64a() ready for lowcore relocation
  s390/head64: Make startup code ready for lowcore relocation
  s390: Add infrastructure to patch lowcore accesses
  s390/atomic_ops: Disable flag outputs constraint for GCC versions below 14.2.0
  s390/entry: Move SIE indicator flag to thread info
  s390/nmi: Simplify ptregs setup
  s390/alternatives: Remove alternative facility list
  ...

Documentation/admin-guide/kernel-parameters.txt

@@ -3830,9 +3830,6 @@
 
 	noalign		[KNL,ARM]
 
-	noaltinstr	[S390,EARLY] Disables alternative instructions
-			patching (CPU alternatives feature).
-
 	noapic		[SMP,APIC,EARLY] Tells the kernel to not make use of any
 			IOAPICs that may be present in the system.
 

arch/s390/Kconfig

@@ -799,17 +799,6 @@ config HAVE_PNETID
 
 menu "Virtualization"
 
-config PROTECTED_VIRTUALIZATION_GUEST
-	def_bool n
-	prompt "Protected virtualization guest support"
-	help
-	  Select this option, if you want to be able to run this
-	  kernel as a protected virtualization KVM guest.
-	  Protected virtualization capable machines have a mini hypervisor
-	  located at machine level (an ultravisor). With help of the
-	  Ultravisor, KVM will be able to run "protected" VMs, special
-	  VMs whose memory and management data are unavailable to KVM.
-
 config PFAULT
 	def_bool y
 	prompt "Pseudo page fault support"

arch/s390/boot/Makefile

@@ -39,8 +39,7 @@ CFLAGS_sclp_early_core.o += -I$(srctree)/drivers/s390/char
 
 obj-y	:= head.o als.o startup.o physmem_info.o ipl_parm.o ipl_report.o vmem.o
 obj-y	+= string.o ebcdic.o sclp_early_core.o mem.o ipl_vmparm.o cmdline.o
-obj-y	+= version.o pgm_check_info.o ctype.o ipl_data.o relocs.o
-obj-$(findstring y, $(CONFIG_PROTECTED_VIRTUALIZATION_GUEST) $(CONFIG_PGSTE))	+= uv.o
+obj-y	+= version.o pgm_check_info.o ctype.o ipl_data.o relocs.o alternative.o uv.o
 obj-$(CONFIG_RANDOMIZE_BASE)	+= kaslr.o
 obj-y	+= $(if $(CONFIG_KERNEL_UNCOMPRESSED),,decompressor.o) info.o
 obj-$(CONFIG_KERNEL_ZSTD) += clz_ctz.o

arch/s390/boot/alternative.c

+// SPDX-License-Identifier: GPL-2.0
+
+#include "../kernel/alternative.c"

arch/s390/boot/boot.h

@@ -30,6 +30,8 @@ struct vmlinux_info {
 	unsigned long init_mm_off;
 	unsigned long swapper_pg_dir_off;
 	unsigned long invalid_pg_dir_off;
+	unsigned long alt_instructions;
+	unsigned long alt_instructions_end;
 #ifdef CONFIG_KASAN
 	unsigned long kasan_early_shadow_page_off;
 	unsigned long kasan_early_shadow_pte_off;
@@ -89,8 +91,10 @@ extern char _end[], _decompressor_end[];
 extern unsigned char _compressed_start[];
 extern unsigned char _compressed_end[];
 extern struct vmlinux_info _vmlinux_info;
+
 #define vmlinux _vmlinux_info
 
+#define __lowcore_pa(x)		((unsigned long)(x) % sizeof(struct lowcore))
 #define __abs_lowcore_pa(x)	(((unsigned long)(x) - __abs_lowcore) % sizeof(struct lowcore))
 #define __kernel_va(x)		((void *)((unsigned long)(x) - __kaslr_offset_phys + __kaslr_offset))
 #define __kernel_pa(x)		((unsigned long)(x) - __kaslr_offset + __kaslr_offset_phys)

arch/s390/boot/ipl_parm.c

@@ -3,6 +3,7 @@
 #include <linux/init.h>
 #include <linux/ctype.h>
 #include <linux/pgtable.h>
+#include <asm/abs_lowcore.h>
 #include <asm/page-states.h>
 #include <asm/ebcdic.h>
 #include <asm/sclp.h>
@@ -310,5 +311,7 @@ void parse_boot_command_line(void)
 				prot_virt_host = 1;
 		}
 #endif
+		if (!strcmp(param, "relocate_lowcore") && test_facility(193))
+			relocate_lowcore = 1;
 	}
 }

arch/s390/boot/startup.c

@@ -30,6 +30,7 @@ unsigned long __bootdata_preserved(vmemmap_size);
 unsigned long __bootdata_preserved(MODULES_VADDR);
 unsigned long __bootdata_preserved(MODULES_END);
 unsigned long __bootdata_preserved(max_mappable);
+int __bootdata_preserved(relocate_lowcore);
 
 u64 __bootdata_preserved(stfle_fac_list[16]);
 struct oldmem_data __bootdata_preserved(oldmem_data);
@@ -376,6 +377,8 @@ static void kaslr_adjust_vmlinux_info(long offset)
 	vmlinux.init_mm_off += offset;
 	vmlinux.swapper_pg_dir_off += offset;
 	vmlinux.invalid_pg_dir_off += offset;
+	vmlinux.alt_instructions += offset;
+	vmlinux.alt_instructions_end += offset;
 #ifdef CONFIG_KASAN
 	vmlinux.kasan_early_shadow_page_off += offset;
 	vmlinux.kasan_early_shadow_pte_off += offset;
@@ -478,8 +481,12 @@ void startup_kernel(void)
 	 * before the kernel started. Therefore, in case the two sections
 	 * overlap there is no risk of corrupting any data.
 	 */
-	if (kaslr_enabled())
-		amode31_lma = randomize_within_range(vmlinux.amode31_size, PAGE_SIZE, 0, SZ_2G);
+	if (kaslr_enabled()) {
+		unsigned long amode31_min;
+
+		amode31_min = (unsigned long)_decompressor_end;
+		amode31_lma = randomize_within_range(vmlinux.amode31_size, PAGE_SIZE, amode31_min, SZ_2G);
+	}
 	if (!amode31_lma)
 		amode31_lma = __kaslr_offset_phys - vmlinux.amode31_size;
 	physmem_reserve(RR_AMODE31, amode31_lma, vmlinux.amode31_size);
@@ -503,6 +510,9 @@ void startup_kernel(void)
 	kaslr_adjust_got(__kaslr_offset);
 	setup_vmem(__kaslr_offset, __kaslr_offset + kernel_size, asce_limit);
 	copy_bootdata();
+	__apply_alternatives((struct alt_instr *)_vmlinux_info.alt_instructions,
+			     (struct alt_instr *)_vmlinux_info.alt_instructions_end,
+			     ALT_CTX_EARLY);
 
 	/*
 	 * Save KASLR offset for early dumps, before vmcore_info is set.

arch/s390/boot/uv.c

@@ -8,12 +8,8 @@
 #include "uv.h"
 
 /* will be used in arch/s390/kernel/uv.c */
-#ifdef CONFIG_PROTECTED_VIRTUALIZATION_GUEST
 int __bootdata_preserved(prot_virt_guest);
-#endif
-#if IS_ENABLED(CONFIG_KVM)
 int __bootdata_preserved(prot_virt_host);
-#endif
 struct uv_info __bootdata_preserved(uv_info);
 
 void uv_query_info(void)
@@ -53,14 +49,11 @@ void uv_query_info(void)
 		uv_info.max_secrets = uvcb.max_secrets;
 	}
 
-#ifdef CONFIG_PROTECTED_VIRTUALIZATION_GUEST
 	if (test_bit_inv(BIT_UVC_CMD_SET_SHARED_ACCESS, (unsigned long *)uvcb.inst_calls_list) &&
 	    test_bit_inv(BIT_UVC_CMD_REMOVE_SHARED_ACCESS, (unsigned long *)uvcb.inst_calls_list))
 		prot_virt_guest = 1;
-#endif
 }
 
-#if IS_ENABLED(CONFIG_KVM)
 unsigned long adjust_to_uv_max(unsigned long limit)
 {
 	if (is_prot_virt_host() && uv_info.max_sec_stor_addr)
@@ -92,4 +85,3 @@ void sanitize_prot_virt_host(void)
 {
 	prot_virt_host = is_prot_virt_host_capable();
 }
-#endif

arch/s390/boot/uv.h

@@ -2,21 +2,8 @@
 #ifndef BOOT_UV_H
 #define BOOT_UV_H
 
-#if IS_ENABLED(CONFIG_KVM)
 unsigned long adjust_to_uv_max(unsigned long limit);
 void sanitize_prot_virt_host(void);
-#else
-static inline unsigned long adjust_to_uv_max(unsigned long limit)
-{
-	return limit;
-}
-static inline void sanitize_prot_virt_host(void) {}
-#endif
-
-#if defined(CONFIG_PROTECTED_VIRTUALIZATION_GUEST) || IS_ENABLED(CONFIG_KVM)
 void uv_query_info(void);
-#else
-static inline void uv_query_info(void) {}
-#endif
 
 #endif /* BOOT_UV_H */

arch/s390/boot/vmem.c

@@ -26,6 +26,7 @@ atomic_long_t __bootdata_preserved(direct_pages_count[PG_DIRECT_MAP_MAX]);
 enum populate_mode {
 	POPULATE_NONE,
 	POPULATE_DIRECT,
+	POPULATE_LOWCORE,
 	POPULATE_ABS_LOWCORE,
 	POPULATE_IDENTITY,
 	POPULATE_KERNEL,
@@ -242,6 +243,8 @@ static unsigned long _pa(unsigned long addr, unsigned long size, enum populate_m
 		return -1;
 	case POPULATE_DIRECT:
 		return addr;
+	case POPULATE_LOWCORE:
+		return __lowcore_pa(addr);
 	case POPULATE_ABS_LOWCORE:
 		return __abs_lowcore_pa(addr);
 	case POPULATE_KERNEL:
@@ -418,6 +421,7 @@ static void pgtable_populate(unsigned long addr, unsigned long end, enum populat
 
 void setup_vmem(unsigned long kernel_start, unsigned long kernel_end, unsigned long asce_limit)
 {
+	unsigned long lowcore_address = 0;
 	unsigned long start, end;
 	unsigned long asce_type;
 	unsigned long asce_bits;
@@ -455,12 +459,17 @@ void setup_vmem(unsigned long kernel_start, unsigned long kernel_end, unsigned l
 	__arch_set_page_dat((void *)swapper_pg_dir, 1UL << CRST_ALLOC_ORDER);
 	__arch_set_page_dat((void *)invalid_pg_dir, 1UL << CRST_ALLOC_ORDER);
 
+	if (relocate_lowcore)
+		lowcore_address = LOWCORE_ALT_ADDRESS;
+
 	/*
 	 * To allow prefixing the lowcore must be mapped with 4KB pages.
 	 * To prevent creation of a large page at address 0 first map
 	 * the lowcore and create the identity mapping only afterwards.
 	 */
-	pgtable_populate(0, sizeof(struct lowcore), POPULATE_DIRECT);
+	pgtable_populate(lowcore_address,
+			 lowcore_address + sizeof(struct lowcore),
+			 POPULATE_LOWCORE);
 	for_each_physmem_usable_range(i, &start, &end) {
 		pgtable_populate((unsigned long)__identity_va(start),
 				 (unsigned long)__identity_va(end),

arch/s390/configs/debug_defconfig

@@ -55,7 +55,6 @@ CONFIG_EXPOLINE_AUTO=y
 CONFIG_CHSC_SCH=y
 CONFIG_VFIO_CCW=m
 CONFIG_VFIO_AP=m
-CONFIG_PROTECTED_VIRTUALIZATION_GUEST=y
 CONFIG_CMM=m
 CONFIG_APPLDATA_BASE=y
 CONFIG_S390_HYPFS_FS=y

arch/s390/configs/defconfig

@@ -53,7 +53,6 @@ CONFIG_EXPOLINE_AUTO=y
 CONFIG_CHSC_SCH=y
 CONFIG_VFIO_CCW=m
 CONFIG_VFIO_AP=m
-CONFIG_PROTECTED_VIRTUALIZATION_GUEST=y
 CONFIG_CMM=m
 CONFIG_APPLDATA_BASE=y
 CONFIG_S390_HYPFS_FS=y

arch/s390/include/asm/abs_lowcore.h

@@ -2,6 +2,7 @@
 #ifndef _ASM_S390_ABS_LOWCORE_H
 #define _ASM_S390_ABS_LOWCORE_H
 
+#include <asm/sections.h>
 #include <asm/lowcore.h>
 
 #define ABS_LOWCORE_MAP_SIZE	(NR_CPUS * sizeof(struct lowcore))
@@ -24,4 +25,11 @@ static inline void put_abs_lowcore(struct lowcore *lc)
 	put_cpu();
 }
 
+extern int __bootdata_preserved(relocate_lowcore);
+
+static inline int have_relocated_lowcore(void)
+{
+	return relocate_lowcore;
+}
+
 #endif /* _ASM_S390_ABS_LOWCORE_H */

arch/s390/include/asm/alternative-asm.h

-/* SPDX-License-Identifier: GPL-2.0 */
-#ifndef _ASM_S390_ALTERNATIVE_ASM_H
-#define _ASM_S390_ALTERNATIVE_ASM_H
-
-#ifdef __ASSEMBLY__
-
-/*
- * Issue one struct alt_instr descriptor entry (need to put it into
- * the section .altinstructions, see below). This entry contains
- * enough information for the alternatives patching code to patch an
- * instruction. See apply_alternatives().
- */
-.macro alt_entry orig_start, orig_end, alt_start, alt_end, feature
-	.long	\orig_start - .
-	.long	\alt_start - .
-	.word	\feature
-	.byte	\orig_end - \orig_start
-	.org	. - ( \orig_end - \orig_start ) & 1
-	.org	. - ( \orig_end - \orig_start ) + ( \alt_end - \alt_start )
-	.org	. - ( \alt_end - \alt_start ) + ( \orig_end - \orig_start )
-.endm
-
-/*
- * Define an alternative between two instructions. If @feature is
- * present, early code in apply_alternatives() replaces @oldinstr with
- * @newinstr.
- */
-.macro ALTERNATIVE oldinstr, newinstr, feature
-	.pushsection .altinstr_replacement,"ax"
-770:	\newinstr
-771:	.popsection
-772:	\oldinstr
-773:	.pushsection .altinstructions,"a"
-	alt_entry 772b, 773b, 770b, 771b, \feature
-	.popsection
-.endm
-
-/*
- * Define an alternative between two instructions. If @feature is
- * present, early code in apply_alternatives() replaces @oldinstr with
- * @newinstr.
- */
-.macro ALTERNATIVE_2 oldinstr, newinstr1, feature1, newinstr2, feature2
-	.pushsection .altinstr_replacement,"ax"
-770:	\newinstr1
-771:	\newinstr2
-772:	.popsection
-773:	\oldinstr
-774:	.pushsection .altinstructions,"a"
-	alt_entry 773b, 774b, 770b, 771b,\feature1
-	alt_entry 773b, 774b, 771b, 772b,\feature2
-	.popsection
-.endm
-
-#endif	/*  __ASSEMBLY__  */
-
-#endif /* _ASM_S390_ALTERNATIVE_ASM_H */

arch/s390/include/asm/alternative.h

@@ -2,6 +2,58 @@
 #ifndef _ASM_S390_ALTERNATIVE_H
 #define _ASM_S390_ALTERNATIVE_H
 
+/*
+ * Each alternative comes with a 32 bit feature field:
+ *	union {
+ *		u32 feature;
+ *		struct {
+ *			u32 ctx	 : 4;
+ *			u32 type : 8;
+ *			u32 data : 20;
+ *		};
+ *	}
+ *
+ * @ctx is a bitfield, where only one bit must be set. Each bit defines
+ * in which context an alternative is supposed to be applied to the
+ * kernel image:
+ *
+ * - from the decompressor before the kernel itself is executed
+ * - from early kernel code from within the kernel
+ *
+ * @type is a number which defines the type and with that the type
+ * specific alternative patching.
+ *
+ * @data is additional type specific information which defines if an
+ * alternative should be applied.
+ */
+
+#define ALT_CTX_EARLY		1
+#define ALT_CTX_LATE		2
+#define ALT_CTX_ALL		(ALT_CTX_EARLY | ALT_CTX_LATE)
+
+#define ALT_TYPE_FACILITY	0
+#define ALT_TYPE_SPEC		1
+#define ALT_TYPE_LOWCORE	2
+
+#define ALT_DATA_SHIFT		0
+#define ALT_TYPE_SHIFT		20
+#define ALT_CTX_SHIFT		28
+
+#define ALT_FACILITY_EARLY(facility)	(ALT_CTX_EARLY << ALT_CTX_SHIFT		| \
+					 ALT_TYPE_FACILITY << ALT_TYPE_SHIFT	| \
+					 (facility) << ALT_DATA_SHIFT)
+
+#define ALT_FACILITY(facility)		(ALT_CTX_LATE << ALT_CTX_SHIFT		| \
+					 ALT_TYPE_FACILITY << ALT_TYPE_SHIFT	| \
+					 (facility) << ALT_DATA_SHIFT)
+
+#define ALT_SPEC(facility)		(ALT_CTX_LATE << ALT_CTX_SHIFT		| \
+					 ALT_TYPE_SPEC << ALT_TYPE_SHIFT	| \
+					 (facility) << ALT_DATA_SHIFT)
+
+#define ALT_LOWCORE			(ALT_CTX_EARLY << ALT_CTX_SHIFT		| \
+					 ALT_TYPE_LOWCORE << ALT_TYPE_SHIFT)
+
 #ifndef __ASSEMBLY__
 
 #include <linux/types.h>
@@ -11,12 +63,30 @@
 struct alt_instr {
 	s32 instr_offset;	/* original instruction */
 	s32 repl_offset;	/* offset to replacement instruction */
-	u16 facility;		/* facility bit set for replacement */
+	union {
+		u32 feature;	/* feature required for replacement */
+		struct {
+			u32 ctx	 : 4;  /* context */
+			u32 type : 8;  /* type of alternative */
+			u32 data : 20; /* patching information */
+		};
+	};
 	u8  instrlen;		/* length of original instruction */
 } __packed;
 
-void apply_alternative_instructions(void);
-void apply_alternatives(struct alt_instr *start, struct alt_instr *end);
+extern struct alt_instr __alt_instructions[], __alt_instructions_end[];
+
+void __apply_alternatives(struct alt_instr *start, struct alt_instr *end, unsigned int ctx);
+
+static inline void apply_alternative_instructions(void)
+{
+	__apply_alternatives(__alt_instructions, __alt_instructions_end, ALT_CTX_LATE);
+}
+
+static inline void apply_alternatives(struct alt_instr *start, struct alt_instr *end)
+{
+	__apply_alternatives(start, end, ALT_CTX_ALL);
+}
 
 /*
  * +---------------------------------+
@@ -48,10 +118,10 @@ void apply_alternatives(struct alt_instr *start, struct alt_instr *end);
 #define OLDINSTR(oldinstr) \
 	"661:\n\t" oldinstr "\n662:\n"
 
-#define ALTINSTR_ENTRY(facility, num)					\
+#define ALTINSTR_ENTRY(feature, num)					\
 	"\t.long 661b - .\n"			/* old instruction */	\
 	"\t.long " b_altinstr(num)"b - .\n"	/* alt instruction */	\
-	"\t.word " __stringify(facility) "\n"	/* facility bit    */	\
+	"\t.long " __stringify(feature) "\n"	/* feature	   */	\
 	"\t.byte " oldinstr_len "\n"		/* instruction len */	\
 	"\t.org . - (" oldinstr_len ") & 1\n"				\
 	"\t.org . - (" oldinstr_len ") + (" altinstr_len(num) ")\n"	\
@@ -61,24 +131,24 @@ void apply_alternatives(struct alt_instr *start, struct alt_instr *end);
 	b_altinstr(num)":\n\t" altinstr "\n" e_altinstr(num) ":\n"
 
 /* alternative assembly primitive: */
-#define ALTERNATIVE(oldinstr, altinstr, facility) \
+#define ALTERNATIVE(oldinstr, altinstr, feature) \
 	".pushsection .altinstr_replacement, \"ax\"\n"			\
 	ALTINSTR_REPLACEMENT(altinstr, 1)				\
 	".popsection\n"							\
 	OLDINSTR(oldinstr)						\
 	".pushsection .altinstructions,\"a\"\n"				\
-	ALTINSTR_ENTRY(facility, 1)					\
+	ALTINSTR_ENTRY(feature, 1)					\
 	".popsection\n"
 
-#define ALTERNATIVE_2(oldinstr, altinstr1, facility1, altinstr2, facility2)\
+#define ALTERNATIVE_2(oldinstr, altinstr1, feature1, altinstr2, feature2)\
 	".pushsection .altinstr_replacement, \"ax\"\n"			\
 	ALTINSTR_REPLACEMENT(altinstr1, 1)				\
 	ALTINSTR_REPLACEMENT(altinstr2, 2)				\
 	".popsection\n"							\
 	OLDINSTR(oldinstr)						\
 	".pushsection .altinstructions,\"a\"\n"				\
-	ALTINSTR_ENTRY(facility1, 1)					\
-	ALTINSTR_ENTRY(facility2, 2)					\
+	ALTINSTR_ENTRY(feature1, 1)					\
+	ALTINSTR_ENTRY(feature2, 2)					\
 	".popsection\n"
 
 /*
@@ -93,12 +163,12 @@ void apply_alternatives(struct alt_instr *start, struct alt_instr *end);
  * For non barrier like inlines please define new variants
  * without volatile and memory clobber.
  */
-#define alternative(oldinstr, altinstr, facility)			\
-	asm_inline volatile(ALTERNATIVE(oldinstr, altinstr, facility) : : : "memory")
+#define alternative(oldinstr, altinstr, feature)			\
+	asm_inline volatile(ALTERNATIVE(oldinstr, altinstr, feature) : : : "memory")
 
-#define alternative_2(oldinstr, altinstr1, facility1, altinstr2, facility2) \
-	asm_inline volatile(ALTERNATIVE_2(oldinstr, altinstr1, facility1,   \
-				   altinstr2, facility2) ::: "memory")
+#define alternative_2(oldinstr, altinstr1, feature1, altinstr2, feature2) \
+	asm_inline volatile(ALTERNATIVE_2(oldinstr, altinstr1, feature1,   \
+				   altinstr2, feature2) ::: "memory")
 
 /* Alternative inline assembly with input. */
 #define alternative_input(oldinstr, newinstr, feature, input...)	\
@@ -106,8 +176,8 @@ void apply_alternatives(struct alt_instr *start, struct alt_instr *end);
 		: : input)
 
 /* Like alternative_input, but with a single output argument */
-#define alternative_io(oldinstr, altinstr, facility, output, input...)	\
-	asm_inline volatile(ALTERNATIVE(oldinstr, altinstr, facility)	\
+#define alternative_io(oldinstr, altinstr, feature, output, input...)	\
+	asm_inline volatile(ALTERNATIVE(oldinstr, altinstr, feature)	\
 		: output : input)
 
 /* Use this macro if more than one output parameter is needed. */
@@ -116,6 +186,56 @@ void apply_alternatives(struct alt_instr *start, struct alt_instr *end);
 /* Use this macro if clobbers are needed without inputs. */
 #define ASM_NO_INPUT_CLOBBER(clobber...) : clobber
 
+#else  /* __ASSEMBLY__ */
+
+/*
+ * Issue one struct alt_instr descriptor entry (need to put it into
+ * the section .altinstructions, see below). This entry contains
+ * enough information for the alternatives patching code to patch an
+ * instruction. See apply_alternatives().
+ */
+.macro alt_entry orig_start, orig_end, alt_start, alt_end, feature
+	.long	\orig_start - .
+	.long	\alt_start - .
+	.long	\feature
+	.byte	\orig_end - \orig_start
+	.org	. - ( \orig_end - \orig_start ) & 1
+	.org	. - ( \orig_end - \orig_start ) + ( \alt_end - \alt_start )
+	.org	. - ( \alt_end - \alt_start ) + ( \orig_end - \orig_start )
+.endm
+
+/*
+ * Define an alternative between two instructions. If @feature is
+ * present, early code in apply_alternatives() replaces @oldinstr with
+ * @newinstr.
+ */
+.macro ALTERNATIVE oldinstr, newinstr, feature
+	.pushsection .altinstr_replacement,"ax"
+770:	\newinstr
+771:	.popsection
+772:	\oldinstr
+773:	.pushsection .altinstructions,"a"
+	alt_entry 772b, 773b, 770b, 771b, \feature
+	.popsection
+.endm
+
+/*
+ * Define an alternative between two instructions. If @feature is
+ * present, early code in apply_alternatives() replaces @oldinstr with
+ * @newinstr.
+ */
+.macro ALTERNATIVE_2 oldinstr, newinstr1, feature1, newinstr2, feature2
+	.pushsection .altinstr_replacement,"ax"
+770:	\newinstr1
+771:	\newinstr2
+772:	.popsection
+773:	\oldinstr
+774:	.pushsection .altinstructions,"a"
+	alt_entry 773b, 774b, 770b, 771b,\feature1
+	alt_entry 773b, 774b, 771b, 772b,\feature2
+	.popsection
+.endm
+
 #endif /* __ASSEMBLY__ */
 
 #endif /* _ASM_S390_ALTERNATIVE_H */

arch/s390/include/asm/atomic_ops.h

@@ -188,7 +188,8 @@ static __always_inline long __atomic64_cmpxchg(long *ptr, long old, long new)
 	return old;
 }
 
-#ifdef __GCC_ASM_FLAG_OUTPUTS__
+/* GCC versions before 14.2.0 may die with an ICE in some configurations. */
+#if defined(__GCC_ASM_FLAG_OUTPUTS__) && !(IS_ENABLED(CONFIG_CC_IS_GCC) && (GCC_VERSION < 140200))
 
 static __always_inline bool __atomic_cmpxchg_bool(int *ptr, int old, int new)
 {

arch/s390/include/asm/facility.h

@@ -20,7 +20,6 @@
 #define MAX_FACILITY_BIT (sizeof(stfle_fac_list) * 8)
 
 extern u64 stfle_fac_list[16];
-extern u64 alt_stfle_fac_list[16];
 
 static inline void __set_facility(unsigned long nr, void *facilities)
 {

arch/s390/include/asm/kmsan.h

@@ -12,8 +12,8 @@
 
 static inline bool is_lowcore_addr(void *addr)
 {
-	return addr >= (void *)&S390_lowcore &&
-	       addr < (void *)(&S390_lowcore + 1);
+	return addr >= (void *)get_lowcore() &&
+	       addr < (void *)(get_lowcore() + 1);
 }
 
 static inline void *arch_kmsan_get_meta_or_null(void *addr, bool is_origin)
@@ -25,7 +25,7 @@ static inline void *arch_kmsan_get_meta_or_null(void *addr, bool is_origin)
 		 * order to get a distinct struct page.
 		 */
 		addr += (void *)lowcore_ptr[raw_smp_processor_id()] -
-			(void *)&S390_lowcore;
+			(void *)get_lowcore();
 		if (KMSAN_WARN_ON(is_lowcore_addr(addr)))
 			return NULL;
 		return kmsan_get_metadata(addr, is_origin);

arch/s390/include/asm/lowcore.h

@@ -14,10 +14,15 @@
 #include <asm/ctlreg.h>
 #include <asm/cpu.h>
 #include <asm/types.h>
+#include <asm/alternative.h>
 
 #define LC_ORDER 1
 #define LC_PAGES 2
 
+#define LOWCORE_ALT_ADDRESS	_AC(0x70000, UL)
+
+#ifndef __ASSEMBLY__
+
 struct pgm_tdb {
 	u64 data[32];
 };
@@ -97,8 +102,7 @@ struct lowcore {
 	__u64	save_area_async[8];		/* 0x0240 */
 	__u64	save_area_restart[1];		/* 0x0280 */
 
-	/* CPU flags. */
-	__u64	cpu_flags;			/* 0x0288 */
+	__u64	pcpu;				/* 0x0288 */
 
 	/* Return psws. */
 	psw_t	return_psw;			/* 0x0290 */
@@ -215,7 +219,14 @@ struct lowcore {
 
 static __always_inline struct lowcore *get_lowcore(void)
 {
+	struct lowcore *lc;
+
+	if (__is_defined(__DECOMPRESSOR))
 		return NULL;
+	asm(ALTERNATIVE("llilh %[lc],0", "llilh %[lc],%[alt]", ALT_LOWCORE)
+	    : [lc] "=d" (lc)
+	    : [alt] "i" (LOWCORE_ALT_ADDRESS >> 16));
+	return lc;
 }
 
 extern struct lowcore *lowcore_ptr[];
@@ -225,4 +236,19 @@ static inline void set_prefix(__u32 address)
 	asm volatile("spx %0" : : "Q" (address) : "memory");
 }
 
+#else /* __ASSEMBLY__ */
+
+.macro GET_LC reg
+	ALTERNATIVE "llilh	\reg,0",					\
+		__stringify(llilh	\reg, LOWCORE_ALT_ADDRESS >> 16),	\
+		ALT_LOWCORE
+.endm
+
+.macro STMG_LC start, end, savearea
+	ALTERNATIVE "stmg	\start, \end, \savearea",				\
+		__stringify(stmg	\start, \end, LOWCORE_ALT_ADDRESS + \savearea),	\
+		ALT_LOWCORE
+.endm
+
+#endif /* __ASSEMBLY__ */
 #endif /* _ASM_S390_LOWCORE_H */

arch/s390/include/asm/nospec-branch.h

@@ -5,8 +5,17 @@
 #ifndef __ASSEMBLY__
 
 #include <linux/types.h>
+#include <asm/facility.h>
 
 extern int nospec_disable;
+extern int nobp;
+
+static inline bool nobp_enabled(void)
+{
+	if (__is_defined(__DECOMPRESSOR))
+		return false;
+	return nobp && test_facility(82);
+}
 
 void nospec_init_branches(void);
 void nospec_auto_detect(void);

arch/s390/include/asm/page.h

@@ -174,12 +174,10 @@ static inline int devmem_is_allowed(unsigned long pfn)
 #define HAVE_ARCH_FREE_PAGE
 #define HAVE_ARCH_ALLOC_PAGE
 
-#if IS_ENABLED(CONFIG_PGSTE)
 int arch_make_folio_accessible(struct folio *folio);
 #define HAVE_ARCH_MAKE_FOLIO_ACCESSIBLE
 int arch_make_page_accessible(struct page *page);
 #define HAVE_ARCH_MAKE_PAGE_ACCESSIBLE
-#endif
 
 struct vm_layout {
 	unsigned long kaslr_offset;

arch/s390/include/asm/processor.h

@@ -14,13 +14,11 @@
 
 #include <linux/bits.h>
 
-#define CIF_SIE			0	/* CPU needs SIE exit cleanup */
 #define CIF_NOHZ_DELAY		2	/* delay HZ disable for a tick */
 #define CIF_ENABLED_WAIT	5	/* in enabled wait state */
 #define CIF_MCCK_GUEST		6	/* machine check happening in guest */
 #define CIF_DEDICATED_CPU	7	/* this CPU is dedicated */
 
-#define _CIF_SIE		BIT(CIF_SIE)
 #define _CIF_NOHZ_DELAY		BIT(CIF_NOHZ_DELAY)
 #define _CIF_ENABLED_WAIT	BIT(CIF_ENABLED_WAIT)
 #define _CIF_MCCK_GUEST		BIT(CIF_MCCK_GUEST)
@@ -42,21 +40,37 @@
 #include <asm/irqflags.h>
 #include <asm/alternative.h>
 
+struct pcpu {
+	unsigned long ec_mask;		/* bit mask for ec_xxx functions */
+	unsigned long ec_clk;		/* sigp timestamp for ec_xxx */
+	unsigned long flags;		/* per CPU flags */
+	signed char state;		/* physical cpu state */
+	signed char polarization;	/* physical polarization */
+	u16 address;			/* physical cpu address */
+};
+
+DECLARE_PER_CPU(struct pcpu, pcpu_devices);
+
 typedef long (*sys_call_ptr_t)(struct pt_regs *regs);
 
+static __always_inline struct pcpu *this_pcpu(void)
+{
+	return (struct pcpu *)(get_lowcore()->pcpu);
+}
+
 static __always_inline void set_cpu_flag(int flag)
 {
-	get_lowcore()->cpu_flags |= (1UL << flag);
+	this_pcpu()->flags |= (1UL << flag);
 }
 
 static __always_inline void clear_cpu_flag(int flag)
 {
-	get_lowcore()->cpu_flags &= ~(1UL << flag);
+	this_pcpu()->flags &= ~(1UL << flag);
 }
 
 static __always_inline bool test_cpu_flag(int flag)
 {
-	return get_lowcore()->cpu_flags & (1UL << flag);
+	return this_pcpu()->flags & (1UL << flag);
 }
 
 static __always_inline bool test_and_set_cpu_flag(int flag)
@@ -81,9 +95,7 @@ static __always_inline bool test_and_clear_cpu_flag(int flag)
  */
 static __always_inline bool test_cpu_flag_of(int flag, int cpu)
 {
-	struct lowcore *lc = lowcore_ptr[cpu];
-
-	return lc->cpu_flags & (1UL << flag);
+	return per_cpu(pcpu_devices, cpu).flags & (1UL << flag);
 }
 
 #define arch_needs_cpu() test_cpu_flag(CIF_NOHZ_DELAY)
@@ -405,7 +417,7 @@ static __always_inline bool regs_irqs_disabled(struct pt_regs *regs)
 
 static __always_inline void bpon(void)
 {
-	asm volatile(ALTERNATIVE("nop", ".insn	rrf,0xb2e80000,0,0,13,0", 82));
+	asm volatile(ALTERNATIVE("nop", ".insn	rrf,0xb2e80000,0,0,13,0", ALT_SPEC(82)));
 }
 
 #endif /* __ASSEMBLY__ */

arch/s390/include/asm/runtime-const.h

+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _ASM_S390_RUNTIME_CONST_H
+#define _ASM_S390_RUNTIME_CONST_H
+
+#include <linux/uaccess.h>
+
+#define runtime_const_ptr(sym)					\
+({								\
+	typeof(sym) __ret;					\
+								\
+	asm_inline(						\
+		"0:	iihf	%[__ret],%[c1]\n"		\
+		"	iilf	%[__ret],%[c2]\n"		\
+		".pushsection runtime_ptr_" #sym ",\"a\"\n"	\
+		".long 0b - .\n"				\
+		".popsection"					\
+		: [__ret] "=d" (__ret)				\
+		: [c1] "i" (0x01234567UL),			\
+		  [c2] "i" (0x89abcdefUL));			\
+	__ret;							\
+})
+
+#define runtime_const_shift_right_32(val, sym)			\
+({								\
+	unsigned int __ret = (val);				\
+								\
+	asm_inline(						\
+		"0:	srl	%[__ret],12\n"			\
+		".pushsection runtime_shift_" #sym ",\"a\"\n"	\
+		".long 0b - .\n"				\
+		".popsection"					\
+		: [__ret] "+d" (__ret));			\
+	__ret;							\
+})
+
+#define runtime_const_init(type, sym) do {			\
+	extern s32 __start_runtime_##type##_##sym[];		\
+	extern s32 __stop_runtime_##type##_##sym[];		\
+								\
+	runtime_const_fixup(__runtime_fixup_##type,		\
+			    (unsigned long)(sym),		\
+			    __start_runtime_##type##_##sym,	\
+			    __stop_runtime_##type##_##sym);	\
+} while (0)
+
+/* 32-bit immediate for iihf and iilf in bits in I2 field */
+static inline void __runtime_fixup_32(u32 *p, unsigned int val)
+{
+	s390_kernel_write(p, &val, sizeof(val));
+}
+
+static inline void __runtime_fixup_ptr(void *where, unsigned long val)
+{
+	__runtime_fixup_32(where + 2, val >> 32);
+	__runtime_fixup_32(where + 8, val);
+}
+
+/* Immediate value is lower 12 bits of D2 field of srl */
+static inline void __runtime_fixup_shift(void *where, unsigned long val)
+{
+	u32 insn = *(u32 *)where;
+
+	insn &= 0xfffff000;
+	insn |= (val & 63);
+	s390_kernel_write(where, &insn, sizeof(insn));
+}
+
+static inline void runtime_const_fixup(void (*fn)(void *, unsigned long),
+				       unsigned long val, s32 *start, s32 *end)
+{
+	while (start < end) {
+		fn(*start + (void *)start, val);
+		start++;
+	}
+}
+
+#endif /* _ASM_S390_RUNTIME_CONST_H */

arch/s390/include/asm/smp.h

@@ -24,7 +24,6 @@ extern int __cpu_up(unsigned int cpu, struct task_struct *tidle);
 extern void arch_send_call_function_single_ipi(int cpu);
 extern void arch_send_call_function_ipi_mask(const struct cpumask *mask);
 
-extern void smp_call_online_cpu(void (*func)(void *), void *);
 extern void smp_call_ipl_cpu(void (*func)(void *), void *);
 extern void smp_emergency_stop(void);
 

arch/s390/include/asm/spinlock.h

@@ -79,7 +79,7 @@ static inline void arch_spin_unlock(arch_spinlock_t *lp)
 	typecheck(int, lp->lock);
 	kcsan_release();
 	asm_inline volatile(
-		ALTERNATIVE("nop", ".insn rre,0xb2fa0000,7,0", 49) /* NIAI 7 */
+		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");

arch/s390/include/asm/thread_info.h

@@ -40,6 +40,7 @@ struct thread_info {
 	unsigned long		flags;		/* low level flags */
 	unsigned long		syscall_work;	/* SYSCALL_WORK_ flags */
 	unsigned int		cpu;		/* current CPU */
+	unsigned char		sie;		/* running in SIE context */
 };
 
 /*

arch/s390/include/asm/uaccess.h

@@ -332,7 +332,14 @@ static inline unsigned long __must_check clear_user(void __user *to, unsigned lo
 	return __clear_user(to, n);
 }
 
-void *s390_kernel_write(void *dst, const void *src, size_t size);
+void *__s390_kernel_write(void *dst, const void *src, size_t size);
+
+static inline void *s390_kernel_write(void *dst, const void *src, size_t size)
+{
+	if (__is_defined(__DECOMPRESSOR))
+		return memcpy(dst, src, size);
+	return __s390_kernel_write(dst, src, size);
+}
 
 int __noreturn __put_kernel_bad(void);
 

arch/s390/include/asm/uv.h

@@ -414,7 +414,6 @@ static inline bool uv_has_feature(u8 feature_bit)
 	return test_bit_inv(feature_bit, &uv_info.uv_feature_indications);
 }
 
-#ifdef CONFIG_PROTECTED_VIRTUALIZATION_GUEST
 extern int prot_virt_guest;
 
 static inline int is_prot_virt_guest(void)
@@ -466,13 +465,6 @@ static inline int uv_remove_shared(unsigned long addr)
 	return share(addr, UVC_CMD_REMOVE_SHARED_ACCESS);
 }
 
-#else
-#define is_prot_virt_guest() 0
-static inline int uv_set_shared(unsigned long addr) { return 0; }
-static inline int uv_remove_shared(unsigned long addr) { return 0; }
-#endif
-
-#if IS_ENABLED(CONFIG_KVM)
 extern int prot_virt_host;
 
 static inline int is_prot_virt_host(void)
@@ -489,29 +481,5 @@ int uv_convert_from_secure_pte(pte_t pte);
 int gmap_convert_to_secure(struct gmap *gmap, unsigned long gaddr);
 
 void setup_uv(void);
-#else
-#define is_prot_virt_host() 0
-static inline void setup_uv(void) {}
-
-static inline int uv_pin_shared(unsigned long paddr)
-{
-	return 0;
-}
-
-static inline int uv_destroy_folio(struct folio *folio)
-{
-	return 0;
-}
-
-static inline int uv_destroy_pte(pte_t pte)
-{
-	return 0;
-}
-
-static inline int uv_convert_from_secure_pte(pte_t pte)
-{
-	return 0;
-}
-#endif
 
 #endif /* _ASM_S390_UV_H */

arch/s390/kernel/Makefile

@@ -43,7 +43,7 @@ obj-y	+= sysinfo.o lgr.o os_info.o ctlreg.o
 obj-y	+= runtime_instr.o cache.o fpu.o dumpstack.o guarded_storage.o sthyi.o
 obj-y	+= entry.o reipl.o kdebugfs.o alternative.o
 obj-y	+= nospec-branch.o ipl_vmparm.o machine_kexec_reloc.o unwind_bc.o
-obj-y	+= smp.o text_amode31.o stacktrace.o abs_lowcore.o facility.o
+obj-y	+= smp.o text_amode31.o stacktrace.o abs_lowcore.o facility.o uv.o
 
 extra-y				+= vmlinux.lds
 
@@ -80,7 +80,6 @@ obj-$(CONFIG_PERF_EVENTS)	+= perf_cpum_cf_events.o perf_regs.o
 obj-$(CONFIG_PERF_EVENTS)	+= perf_pai_crypto.o perf_pai_ext.o
 
 obj-$(CONFIG_TRACEPOINTS)	+= trace.o
-obj-$(findstring y, $(CONFIG_PROTECTED_VIRTUALIZATION_GUEST) $(CONFIG_PGSTE))	+= uv.o
 
 # vdso
 obj-y				+= vdso64/

arch/s390/kernel/abs_lowcore.c

@@ -4,6 +4,7 @@
 #include <asm/abs_lowcore.h>
 
 unsigned long __bootdata_preserved(__abs_lowcore);
+int __bootdata_preserved(relocate_lowcore);
 
 int abs_lowcore_map(int cpu, struct lowcore *lc, bool alloc)
 {

arch/s390/kernel/alternative.c

 // SPDX-License-Identifier: GPL-2.0
-#include <linux/module.h>
-#include <linux/cpu.h>
-#include <linux/smp.h>
-#include <asm/text-patching.h>
+
+#include <linux/uaccess.h>
+#include <asm/nospec-branch.h>
+#include <asm/abs_lowcore.h>
 #include <asm/alternative.h>
 #include <asm/facility.h>
-#include <asm/nospec-branch.h>
-
-static int __initdata_or_module alt_instr_disabled;
-
-static int __init disable_alternative_instructions(char *str)
-{
-	alt_instr_disabled = 1;
-	return 0;
-}
-
-early_param("noaltinstr", disable_alternative_instructions);
 
-static void __init_or_module __apply_alternatives(struct alt_instr *start,
-						  struct alt_instr *end)
+void __apply_alternatives(struct alt_instr *start, struct alt_instr *end, unsigned int ctx)
 {
-	struct alt_instr *a;
 	u8 *instr, *replacement;
+	struct alt_instr *a;
+	bool replace;
 
 	/*
 	 * The scan order should be from start to end. A later scanned
 	 * alternative code can overwrite previously scanned alternative code.
 	 */
 	for (a = start; a < end; a++) {
+		if (!(a->ctx & ctx))
+			continue;
+		switch (a->type) {
+		case ALT_TYPE_FACILITY:
+			replace = test_facility(a->data);
+			break;
+		case ALT_TYPE_SPEC:
+			replace = nobp_enabled();
+			break;
+		case ALT_TYPE_LOWCORE:
+			replace = have_relocated_lowcore();
+			break;
+		default:
+			replace = false;
+		}
+		if (!replace)
+			continue;
 		instr = (u8 *)&a->instr_offset + a->instr_offset;
 		replacement = (u8 *)&a->repl_offset + a->repl_offset;
-
-		if (!__test_facility(a->facility, alt_stfle_fac_list))
-			continue;
 		s390_kernel_write(instr, replacement, a->instrlen);
 	}
 }
-
-void __init_or_module apply_alternatives(struct alt_instr *start,
-					 struct alt_instr *end)
-{
-	if (!alt_instr_disabled)
-		__apply_alternatives(start, end);
-}
-
-extern struct alt_instr __alt_instructions[], __alt_instructions_end[];
-void __init apply_alternative_instructions(void)
-{
-	apply_alternatives(__alt_instructions, __alt_instructions_end);
-}
-
-static void do_sync_core(void *info)
-{
-	sync_core();
-}
-
-void text_poke_sync(void)
-{
-	on_each_cpu(do_sync_core, NULL, 1);
-}
-
-void text_poke_sync_lock(void)
-{
-	cpus_read_lock();
-	text_poke_sync();
-	cpus_read_unlock();
-}

arch/s390/kernel/asm-offsets.c

@@ -28,6 +28,7 @@ int main(void)
 	BLANK();
 	/* thread info offsets */
 	OFFSET(__TI_flags, task_struct, thread_info.flags);
+	OFFSET(__TI_sie, task_struct, thread_info.sie);
 	BLANK();
 	/* pt_regs offsets */
 	OFFSET(__PT_PSW, pt_regs, psw);
@@ -114,7 +115,7 @@ int main(void)
 	OFFSET(__LC_SAVE_AREA_SYNC, lowcore, save_area_sync);
 	OFFSET(__LC_SAVE_AREA_ASYNC, lowcore, save_area_async);
 	OFFSET(__LC_SAVE_AREA_RESTART, lowcore, save_area_restart);
-	OFFSET(__LC_CPU_FLAGS, lowcore, cpu_flags);
+	OFFSET(__LC_PCPU, lowcore, pcpu);
 	OFFSET(__LC_RETURN_PSW, lowcore, return_psw);
 	OFFSET(__LC_RETURN_MCCK_PSW, lowcore, return_mcck_psw);
 	OFFSET(__LC_SYS_ENTER_TIMER, lowcore, sys_enter_timer);
@@ -186,5 +187,7 @@ int main(void)
 #endif
 	OFFSET(__FTRACE_REGS_PT_REGS, ftrace_regs, regs);
 	DEFINE(__FTRACE_REGS_SIZE, sizeof(struct ftrace_regs));
+
+	OFFSET(__PCPU_FLAGS, pcpu, flags);
 	return 0;
 }

arch/s390/kernel/early.c

@@ -48,6 +48,7 @@ decompressor_handled_param(dfltcc);
 decompressor_handled_param(facilities);
 decompressor_handled_param(nokaslr);
 decompressor_handled_param(cmma);
+decompressor_handled_param(relocate_lowcore);
 #if IS_ENABLED(CONFIG_KVM)
 decompressor_handled_param(prot_virt);
 #endif
@@ -190,13 +191,6 @@ static noinline __init void setup_lowcore_early(void)
 	get_lowcore()->preempt_count = INIT_PREEMPT_COUNT;
 }
 
-static noinline __init void setup_facility_list(void)
-{
-	memcpy(alt_stfle_fac_list, stfle_fac_list, sizeof(alt_stfle_fac_list));
-	if (!IS_ENABLED(CONFIG_KERNEL_NOBP))
-		__clear_facility(82, alt_stfle_fac_list);
-}
-
 static __init void detect_diag9c(void)
 {
 	unsigned int cpu_address;
@@ -291,7 +285,6 @@ void __init startup_init(void)
 	lockdep_off();
 	sort_amode31_extable();
 	setup_lowcore_early();
-	setup_facility_list();
 	detect_machine_type();
 	setup_arch_string();
 	setup_boot_command_line();

arch/s390/kernel/head64.S

@@ -10,6 +10,7 @@
 
 #include <linux/init.h>
 #include <linux/linkage.h>
+#include <asm/lowcore.h>
 #include <asm/asm-offsets.h>
 #include <asm/thread_info.h>
 #include <asm/page.h>
@@ -18,14 +19,15 @@
 __HEAD
 SYM_CODE_START(startup_continue)
 	larl	%r1,tod_clock_base
-	mvc	0(16,%r1),__LC_BOOT_CLOCK
+	GET_LC	%r2
+	mvc	0(16,%r1),__LC_BOOT_CLOCK(%r2)
 #
 # Setup stack
 #
 	larl	%r14,init_task
-	stg	%r14,__LC_CURRENT
+	stg	%r14,__LC_CURRENT(%r2)
 	larl	%r15,init_thread_union+STACK_INIT_OFFSET
-	stg	%r15,__LC_KERNEL_STACK
+	stg	%r15,__LC_KERNEL_STACK(%r2)
 	brasl	%r14,sclp_early_adjust_va	# allow sclp_early_printk
 	brasl	%r14,startup_init		# s390 specific early init
 	brasl	%r14,start_kernel		# common init code

arch/s390/kernel/ipl.c

@@ -2112,7 +2112,7 @@ void do_restart(void *arg)
 	tracing_off();
 	debug_locks_off();
 	lgr_info_log();
-	smp_call_online_cpu(__do_restart, arg);
+	smp_call_ipl_cpu(__do_restart, arg);
 }
 
 /* on halt */

arch/s390/kernel/machine_kexec.c

@@ -62,7 +62,7 @@ static void __do_machine_kdump(void *data)
 	 * This need to be done *after* s390_reset_system set the
 	 * prefix register of this CPU to zero
 	 */
-	memcpy(absolute_pointer(__LC_FPREGS_SAVE_AREA),
+	memcpy(absolute_pointer(get_lowcore()->floating_pt_save_area),
 	       phys_to_virt(prefix + __LC_FPREGS_SAVE_AREA), 512);
 
 	call_nodat(1, int, purgatory, int, 1);

arch/s390/kernel/nospec-branch.c

@@ -4,6 +4,8 @@
 #include <linux/cpu.h>
 #include <asm/nospec-branch.h>
 
+int nobp = IS_ENABLED(CONFIG_KERNEL_NOBP);
+
 static int __init nobp_setup_early(char *str)
 {
 	bool enabled;
@@ -17,11 +19,11 @@ static int __init nobp_setup_early(char *str)
 		 * The user explicitly requested nobp=1, enable it and
 		 * disable the expoline support.
 		 */
-		__set_facility(82, alt_stfle_fac_list);
+		nobp = 1;
 		if (IS_ENABLED(CONFIG_EXPOLINE))
 			nospec_disable = 1;
 	} else {
-		__clear_facility(82, alt_stfle_fac_list);
+		nobp = 0;
 	}
 	return 0;
 }
@@ -29,7 +31,7 @@ early_param("nobp", nobp_setup_early);
 
 static int __init nospec_setup_early(char *str)
 {
-	__clear_facility(82, alt_stfle_fac_list);
+	nobp = 0;
 	return 0;
 }
 early_param("nospec", nospec_setup_early);
@@ -40,7 +42,7 @@ static int __init nospec_report(void)
 		pr_info("Spectre V2 mitigation: etokens\n");
 	if (nospec_uses_trampoline())
 		pr_info("Spectre V2 mitigation: execute trampolines\n");
-	if (__test_facility(82, alt_stfle_fac_list))
+	if (nobp_enabled())
 		pr_info("Spectre V2 mitigation: limited branch prediction\n");
 	return 0;
 }
@@ -66,14 +68,14 @@ void __init nospec_auto_detect(void)
 		 */
 		if (__is_defined(CC_USING_EXPOLINE))
 			nospec_disable = 1;
-		__clear_facility(82, alt_stfle_fac_list);
+		nobp = 0;
 	} else if (__is_defined(CC_USING_EXPOLINE)) {
 		/*
 		 * The kernel has been compiled with expolines.
 		 * Keep expolines enabled and disable nobp.
 		 */
 		nospec_disable = 0;
-		__clear_facility(82, alt_stfle_fac_list);
+		nobp = 0;
 	}
 	/*
 	 * If the kernel has not been compiled with expolines the
@@ -86,7 +88,7 @@ static int __init spectre_v2_setup_early(char *str)
 {
 	if (str && !strncmp(str, "on", 2)) {
 		nospec_disable = 0;
-		__clear_facility(82, alt_stfle_fac_list);
+		nobp = 0;
 	}
 	if (str && !strncmp(str, "off", 3))
 		nospec_disable = 1;

arch/s390/kernel/nospec-sysfs.c

@@ -17,7 +17,7 @@ ssize_t cpu_show_spectre_v2(struct device *dev,
 		return sprintf(buf, "Mitigation: etokens\n");
 	if (nospec_uses_trampoline())
 		return sprintf(buf, "Mitigation: execute trampolines\n");
-	if (__test_facility(82, alt_stfle_fac_list))
+	if (nobp_enabled())
 		return sprintf(buf, "Mitigation: limited branch prediction\n");
 	return sprintf(buf, "Vulnerable\n");
 }

arch/s390/kernel/perf_cpum_cf.c

@@ -556,25 +556,31 @@ static int cfdiag_diffctr(struct cpu_cf_events *cpuhw, unsigned long auth)
 	struct cf_trailer_entry *trailer_start, *trailer_stop;
 	struct cf_ctrset_entry *ctrstart, *ctrstop;
 	size_t offset = 0;
+	int i;
 
-	auth &= (1 << CPUMF_LCCTL_ENABLE_SHIFT) - 1;
-	do {
+	for (i = CPUMF_CTR_SET_BASIC; i < CPUMF_CTR_SET_MAX; ++i) {
 		ctrstart = (struct cf_ctrset_entry *)(cpuhw->start + offset);
 		ctrstop = (struct cf_ctrset_entry *)(cpuhw->stop + offset);
 
+		/* Counter set not authorized */
+		if (!(auth & cpumf_ctr_ctl[i]))
+			continue;
+		/* Counter set size zero was not saved */
+		if (!cpum_cf_read_setsize(i))
+			continue;
+
 		if (memcmp(ctrstop, ctrstart, sizeof(*ctrstop))) {
 			pr_err_once("cpum_cf_diag counter set compare error "
 				    "in set %i\n", ctrstart->set);
 			return 0;
 		}
-		auth &= ~cpumf_ctr_ctl[ctrstart->set];
 		if (ctrstart->def == CF_DIAG_CTRSET_DEF) {
 			cfdiag_diffctrset((u64 *)(ctrstart + 1),
 					  (u64 *)(ctrstop + 1), ctrstart->ctr);
 			offset += ctrstart->ctr * sizeof(u64) +
 							sizeof(*ctrstart);
 		}
-	} while (ctrstart->def && auth);
+	}
 
 	/* Save time_stamp from start of event in stop's trailer */
 	trailer_start = (struct cf_trailer_entry *)(cpuhw->start + offset);

arch/s390/kernel/processor.c

@@ -17,7 +17,8 @@
 #include <linux/mm_types.h>
 #include <linux/delay.h>
 #include <linux/cpu.h>
-
+#include <linux/smp.h>
+#include <asm/text-patching.h>
 #include <asm/diag.h>
 #include <asm/facility.h>
 #include <asm/elf.h>
@@ -79,6 +80,23 @@ void notrace stop_machine_yield(const struct cpumask *cpumask)
 	}
 }
 
+static void do_sync_core(void *info)
+{
+	sync_core();
+}
+
+void text_poke_sync(void)
+{
+	on_each_cpu(do_sync_core, NULL, 1);
+}
+
+void text_poke_sync_lock(void)
+{
+	cpus_read_lock();
+	text_poke_sync();
+	cpus_read_unlock();
+}
+
 /*
  * cpu_init - initializes state that is per-CPU.
  */

arch/s390/kernel/reipl.S

@@ -9,6 +9,7 @@
 #include <asm/asm-offsets.h>
 #include <asm/nospec-insn.h>
 #include <asm/sigp.h>
+#include <asm/lowcore.h>
 
 	GEN_BR_THUNK %r9
 
@@ -20,20 +21,15 @@
 # r3 = Parameter for function
 #
 SYM_CODE_START(store_status)
-	/* Save register one and load save area base */
-	stg	%r1,__LC_SAVE_AREA_RESTART
+	STMG_LC	%r0,%r15,__LC_GPREGS_SAVE_AREA
 	/* General purpose registers */
-	lghi	%r1,__LC_GPREGS_SAVE_AREA
-	stmg	%r0,%r15,0(%r1)
-	mvc	8(8,%r1),__LC_SAVE_AREA_RESTART
+	GET_LC	%r13
 	/* Control registers */
-	lghi	%r1,__LC_CREGS_SAVE_AREA
-	stctg	%c0,%c15,0(%r1)
+	stctg	%c0,%c15,__LC_CREGS_SAVE_AREA(%r13)
 	/* Access registers */
-	lghi	%r1,__LC_AREGS_SAVE_AREA
-	stam	%a0,%a15,0(%r1)
+	stamy	%a0,%a15,__LC_AREGS_SAVE_AREA(%r13)
 	/* Floating point registers */
-	lghi	%r1,__LC_FPREGS_SAVE_AREA
+	lay	%r1,__LC_FPREGS_SAVE_AREA(%r13)
 	std	%f0, 0x00(%r1)
 	std	%f1, 0x08(%r1)
 	std	%f2, 0x10(%r1)
@@ -51,21 +47,21 @@ SYM_CODE_START(store_status)
 	std	%f14,0x70(%r1)
 	std	%f15,0x78(%r1)
 	/* Floating point control register */
-	lghi	%r1,__LC_FP_CREG_SAVE_AREA
+	lay	%r1,__LC_FP_CREG_SAVE_AREA(%r13)
 	stfpc	0(%r1)
 	/* CPU timer */
-	lghi	%r1,__LC_CPU_TIMER_SAVE_AREA
+	lay	%r1,__LC_CPU_TIMER_SAVE_AREA(%r13)
 	stpt	0(%r1)
 	/* Store prefix register */
-	lghi	%r1,__LC_PREFIX_SAVE_AREA
+	lay	%r1,__LC_PREFIX_SAVE_AREA(%r13)
 	stpx	0(%r1)
 	/* Clock comparator - seven bytes */
-	lghi	%r1,__LC_CLOCK_COMP_SAVE_AREA
 	larl	%r4,clkcmp
 	stckc	0(%r4)
+	lay	%r1,__LC_CLOCK_COMP_SAVE_AREA(%r13)
 	mvc	1(7,%r1),1(%r4)
 	/* Program status word */
-	lghi	%r1,__LC_PSW_SAVE_AREA
+	lay	%r1,__LC_PSW_SAVE_AREA(%r13)
 	epsw	%r4,%r5
 	st	%r4,0(%r1)
 	st	%r5,4(%r1)

arch/s390/kernel/setup.c

@@ -149,13 +149,12 @@ unsigned long __bootdata_preserved(max_mappable);
 struct physmem_info __bootdata(physmem_info);
 
 struct vm_layout __bootdata_preserved(vm_layout);
-EXPORT_SYMBOL_GPL(vm_layout);
+EXPORT_SYMBOL(vm_layout);
 int __bootdata_preserved(__kaslr_enabled);
 unsigned int __bootdata_preserved(zlib_dfltcc_support);
 EXPORT_SYMBOL(zlib_dfltcc_support);
 u64 __bootdata_preserved(stfle_fac_list[16]);
 EXPORT_SYMBOL(stfle_fac_list);
-u64 alt_stfle_fac_list[16];
 struct oldmem_data __bootdata_preserved(oldmem_data);
 
 unsigned long VMALLOC_START;
@@ -406,6 +405,7 @@ static void __init setup_lowcore(void)
 		panic("%s: Failed to allocate %zu bytes align=%zx\n",
 		      __func__, sizeof(*lc), sizeof(*lc));
 
+	lc->pcpu = (unsigned long)per_cpu_ptr(&pcpu_devices, 0);
 	lc->restart_psw.mask = PSW_KERNEL_BITS & ~PSW_MASK_DAT;
 	lc->restart_psw.addr = __pa(restart_int_handler);
 	lc->external_new_psw.mask = PSW_KERNEL_BITS;
@@ -889,6 +889,9 @@ void __init setup_arch(char **cmdline_p)
 	else
 		pr_info("Linux is running as a guest in 64-bit mode\n");
 
+	if (have_relocated_lowcore())
+		pr_info("Lowcore relocated to 0x%px\n", get_lowcore());
+
 	log_component_list();
 
 	/* Have one command line that is parsed and saved in /proc/cmdline */

arch/s390/kernel/uv.c

@@ -18,11 +18,22 @@
 #include <asm/sections.h>
 #include <asm/uv.h>
 
+#if !IS_ENABLED(CONFIG_KVM)
+unsigned long __gmap_translate(struct gmap *gmap, unsigned long gaddr)
+{
+	return 0;
+}
+
+int gmap_fault(struct gmap *gmap, unsigned long gaddr,
+	       unsigned int fault_flags)
+{
+	return 0;
+}
+#endif
+
 /* the bootdata_preserved fields come from ones in arch/s390/boot/uv.c */
-#ifdef CONFIG_PROTECTED_VIRTUALIZATION_GUEST
 int __bootdata_preserved(prot_virt_guest);
 EXPORT_SYMBOL(prot_virt_guest);
-#endif
 
 /*
  * uv_info contains both host and guest information but it's currently only
@@ -35,7 +46,6 @@ EXPORT_SYMBOL(prot_virt_guest);
 struct uv_info __bootdata_preserved(uv_info);
 EXPORT_SYMBOL(uv_info);
 
-#if IS_ENABLED(CONFIG_KVM)
 int __bootdata_preserved(prot_virt_host);
 EXPORT_SYMBOL(prot_virt_host);
 
@@ -543,9 +553,6 @@ int arch_make_page_accessible(struct page *page)
 	return arch_make_folio_accessible(page_folio(page));
 }
 EXPORT_SYMBOL_GPL(arch_make_page_accessible);
-#endif
-
-#if defined(CONFIG_PROTECTED_VIRTUALIZATION_GUEST) || IS_ENABLED(CONFIG_KVM)
 static ssize_t uv_query_facilities(struct kobject *kobj,
 				   struct kobj_attribute *attr, char *buf)
 {
@@ -721,24 +728,13 @@ static struct attribute_group uv_query_attr_group = {
 static ssize_t uv_is_prot_virt_guest(struct kobject *kobj,
 				     struct kobj_attribute *attr, char *buf)
 {
-	int val = 0;
-
-#ifdef CONFIG_PROTECTED_VIRTUALIZATION_GUEST
-	val = prot_virt_guest;
-#endif
-	return sysfs_emit(buf, "%d\n", val);
+	return sysfs_emit(buf, "%d\n", prot_virt_guest);
 }
 
 static ssize_t uv_is_prot_virt_host(struct kobject *kobj,
 				    struct kobj_attribute *attr, char *buf)
 {
-	int val = 0;
-
-#if IS_ENABLED(CONFIG_KVM)
-	val = prot_virt_host;
-#endif
-
-	return sysfs_emit(buf, "%d\n", val);
+	return sysfs_emit(buf, "%d\n", prot_virt_host);
 }
 
 static struct kobj_attribute uv_prot_virt_guest =
@@ -790,4 +786,3 @@ static int __init uv_info_init(void)
 	return rc;
 }
 device_initcall(uv_info_init);
-#endif

arch/s390/kernel/vmlinux.lds.S

@@ -190,6 +190,9 @@ SECTIONS
 	. = ALIGN(PAGE_SIZE);
 	INIT_DATA_SECTION(0x100)
 
+	RUNTIME_CONST(shift, d_hash_shift)
+	RUNTIME_CONST(ptr, dentry_hashtable)
+
 	PERCPU_SECTION(0x100)
 
 	. = ALIGN(PAGE_SIZE);
@@ -219,6 +222,8 @@ SECTIONS
 		QUAD(init_mm)
 		QUAD(swapper_pg_dir)
 		QUAD(invalid_pg_dir)
+		QUAD(__alt_instructions)
+		QUAD(__alt_instructions_end)
 #ifdef CONFIG_KASAN
 		QUAD(kasan_early_shadow_page)
 		QUAD(kasan_early_shadow_pte)

arch/s390/lib/spinlock.c

@@ -75,7 +75,7 @@ static inline int arch_load_niai4(int *lock)
 	int owner;
 
 	asm_inline volatile(
-		ALTERNATIVE("nop", ".insn rre,0xb2fa0000,4,0", 49) /* NIAI 4 */
+		ALTERNATIVE("nop", ".insn rre,0xb2fa0000,4,0", ALT_FACILITY(49)) /* NIAI 4 */
 		"	l	%0,%1\n"
 		: "=d" (owner) : "Q" (*lock) : "memory");
 	return owner;
@@ -86,7 +86,7 @@ static inline int arch_cmpxchg_niai8(int *lock, int old, int new)
 	int expected = old;
 
 	asm_inline volatile(
-		ALTERNATIVE("nop", ".insn rre,0xb2fa0000,8,0", 49) /* NIAI 8 */
+		ALTERNATIVE("nop", ".insn rre,0xb2fa0000,8,0", ALT_FACILITY(49)) /* NIAI 8 */
 		"	cs	%0,%3,%1\n"
 		: "=d" (old), "=Q" (*lock)
 		: "0" (old), "d" (new), "Q" (*lock)

arch/s390/mm/dump_pagetables.c

@@ -36,6 +36,16 @@ enum address_markers_idx {
 	VMEMMAP_END_NR,
 	VMALLOC_NR,
 	VMALLOC_END_NR,
+#ifdef CONFIG_KMSAN
+	KMSAN_VMALLOC_SHADOW_START_NR,
+	KMSAN_VMALLOC_SHADOW_END_NR,
+	KMSAN_VMALLOC_ORIGIN_START_NR,
+	KMSAN_VMALLOC_ORIGIN_END_NR,
+	KMSAN_MODULES_SHADOW_START_NR,
+	KMSAN_MODULES_SHADOW_END_NR,
+	KMSAN_MODULES_ORIGIN_START_NR,
+	KMSAN_MODULES_ORIGIN_END_NR,
+#endif
 	MODULES_NR,
 	MODULES_END_NR,
 	ABS_LOWCORE_NR,
@@ -65,6 +75,16 @@ static struct addr_marker address_markers[] = {
 	[VMEMMAP_END_NR]	= {0, "vmemmap Area End"},
 	[VMALLOC_NR]		= {0, "vmalloc Area Start"},
 	[VMALLOC_END_NR]	= {0, "vmalloc Area End"},
+#ifdef CONFIG_KMSAN
+	[KMSAN_VMALLOC_SHADOW_START_NR]	= {0, "Kmsan vmalloc Shadow Start"},
+	[KMSAN_VMALLOC_SHADOW_END_NR]	= {0, "Kmsan vmalloc Shadow End"},
+	[KMSAN_VMALLOC_ORIGIN_START_NR]	= {0, "Kmsan vmalloc Origins Start"},
+	[KMSAN_VMALLOC_ORIGIN_END_NR]	= {0, "Kmsan vmalloc Origins End"},
+	[KMSAN_MODULES_SHADOW_START_NR]	= {0, "Kmsan Modules Shadow Start"},
+	[KMSAN_MODULES_SHADOW_END_NR]	= {0, "Kmsan Modules Shadow End"},
+	[KMSAN_MODULES_ORIGIN_START_NR]	= {0, "Kmsan Modules Origins Start"},
+	[KMSAN_MODULES_ORIGIN_END_NR]	= {0, "Kmsan Modules Origins End"},
+#endif
 	[MODULES_NR]		= {0, "Modules Area Start"},
 	[MODULES_END_NR]	= {0, "Modules Area End"},
 	[ABS_LOWCORE_NR]	= {0, "Lowcore Area Start"},
@@ -306,6 +326,16 @@ static int pt_dump_init(void)
 #ifdef CONFIG_KFENCE
 	address_markers[KFENCE_START_NR].start_address = kfence_start;
 	address_markers[KFENCE_END_NR].start_address = kfence_start + KFENCE_POOL_SIZE;
+#endif
+#ifdef CONFIG_KMSAN
+	address_markers[KMSAN_VMALLOC_SHADOW_START_NR].start_address = KMSAN_VMALLOC_SHADOW_START;
+	address_markers[KMSAN_VMALLOC_SHADOW_END_NR].start_address = KMSAN_VMALLOC_SHADOW_END;
+	address_markers[KMSAN_VMALLOC_ORIGIN_START_NR].start_address = KMSAN_VMALLOC_ORIGIN_START;
+	address_markers[KMSAN_VMALLOC_ORIGIN_END_NR].start_address = KMSAN_VMALLOC_ORIGIN_END;
+	address_markers[KMSAN_MODULES_SHADOW_START_NR].start_address = KMSAN_MODULES_SHADOW_START;
+	address_markers[KMSAN_MODULES_SHADOW_END_NR].start_address = KMSAN_MODULES_SHADOW_END;
+	address_markers[KMSAN_MODULES_ORIGIN_START_NR].start_address = KMSAN_MODULES_ORIGIN_START;
+	address_markers[KMSAN_MODULES_ORIGIN_END_NR].start_address = KMSAN_MODULES_ORIGIN_END;
 #endif
 	sort_address_markers();
 #ifdef CONFIG_PTDUMP_DEBUGFS

arch/s390/mm/maccess.c

@@ -48,7 +48,7 @@ static notrace long s390_kernel_write_odd(void *dst, const void *src, size_t siz
 }
 
 /*
- * s390_kernel_write - write to kernel memory bypassing DAT
+ * __s390_kernel_write - write to kernel memory bypassing DAT
  * @dst: destination address
  * @src: source address
  * @size: number of bytes to copy
@@ -61,7 +61,7 @@ static notrace long s390_kernel_write_odd(void *dst, const void *src, size_t siz
  */
 static DEFINE_SPINLOCK(s390_kernel_write_lock);
 
-notrace void *s390_kernel_write(void *dst, const void *src, size_t size)
+notrace void *__s390_kernel_write(void *dst, const void *src, size_t size)
 {
 	void *tmp = dst;
 	unsigned long flags;

arch/s390/pci/pci_irq.c

@@ -268,33 +268,20 @@ static void zpci_floating_irq_handler(struct airq_struct *airq,
 	}
 }
 
-int arch_setup_msi_irqs(struct pci_dev *pdev, int nvec, int type)
+static int __alloc_airq(struct zpci_dev *zdev, int msi_vecs,
+			unsigned long *bit)
 {
-	struct zpci_dev *zdev = to_zpci(pdev);
-	unsigned int hwirq, msi_vecs, cpu;
-	unsigned long bit;
-	struct msi_desc *msi;
-	struct msi_msg msg;
-	int cpu_addr;
-	int rc, irq;
-
-	zdev->aisb = -1UL;
-	zdev->msi_first_bit = -1U;
-	if (type == PCI_CAP_ID_MSI && nvec > 1)
-		return 1;
-	msi_vecs = min_t(unsigned int, nvec, zdev->max_msi);
-
 	if (irq_delivery == DIRECTED) {
 		/* Allocate cpu vector bits */
-		bit = airq_iv_alloc(zpci_ibv[0], msi_vecs);
-		if (bit == -1UL)
+		*bit = airq_iv_alloc(zpci_ibv[0], msi_vecs);
+		if (*bit == -1UL)
 			return -EIO;
 	} else {
 		/* Allocate adapter summary indicator bit */
-		bit = airq_iv_alloc_bit(zpci_sbv);
-		if (bit == -1UL)
+		*bit = airq_iv_alloc_bit(zpci_sbv);
+		if (*bit == -1UL)
 			return -EIO;
-		zdev->aisb = bit;
+		zdev->aisb = *bit;
 
 		/* Create adapter interrupt vector */
 		zdev->aibv = airq_iv_create(msi_vecs, AIRQ_IV_DATA | AIRQ_IV_BITLOCK, NULL);
@@ -302,27 +289,66 @@ int arch_setup_msi_irqs(struct pci_dev *pdev, int nvec, int type)
 			return -ENOMEM;
 
 		/* Wire up shortcut pointer */
-		zpci_ibv[bit] = zdev->aibv;
+		zpci_ibv[*bit] = zdev->aibv;
 		/* Each function has its own interrupt vector */
-		bit = 0;
+		*bit = 0;
+	}
+	return 0;
+}
+
+int arch_setup_msi_irqs(struct pci_dev *pdev, int nvec, int type)
+{
+	unsigned int hwirq, msi_vecs, irqs_per_msi, i, cpu;
+	struct zpci_dev *zdev = to_zpci(pdev);
+	struct msi_desc *msi;
+	struct msi_msg msg;
+	unsigned long bit;
+	int cpu_addr;
+	int rc, irq;
+
+	zdev->aisb = -1UL;
+	zdev->msi_first_bit = -1U;
+
+	msi_vecs = min_t(unsigned int, nvec, zdev->max_msi);
+	if (msi_vecs < nvec) {
+		pr_info("%s requested %d irqs, allocate system limit of %d",
+			pci_name(pdev), nvec, zdev->max_msi);
 	}
 
-	/* Request MSI interrupts */
+	rc = __alloc_airq(zdev, msi_vecs, &bit);
+	if (rc < 0)
+		return rc;
+
+	/*
+	 * Request MSI interrupts:
+	 * When using MSI, nvec_used interrupt sources and their irq
+	 * descriptors are controlled through one msi descriptor.
+	 * Thus the outer loop over msi descriptors shall run only once,
+	 * while two inner loops iterate over the interrupt vectors.
+	 * When using MSI-X, each interrupt vector/irq descriptor
+	 * is bound to exactly one msi descriptor (nvec_used is one).
+	 * So the inner loops are executed once, while the outer iterates
+	 * over the MSI-X descriptors.
+	 */
 	hwirq = bit;
 	msi_for_each_desc(msi, &pdev->dev, MSI_DESC_NOTASSOCIATED) {
-		rc = -EIO;
 		if (hwirq - bit >= msi_vecs)
 			break;
-		irq = __irq_alloc_descs(-1, 0, 1, 0, THIS_MODULE,
+		irqs_per_msi = min_t(unsigned int, msi_vecs, msi->nvec_used);
+		irq = __irq_alloc_descs(-1, 0, irqs_per_msi, 0, THIS_MODULE,
 					(irq_delivery == DIRECTED) ?
 					msi->affinity : NULL);
 		if (irq < 0)
 			return -ENOMEM;
-		rc = irq_set_msi_desc(irq, msi);
+
+		for (i = 0; i < irqs_per_msi; i++) {
+			rc = irq_set_msi_desc_off(irq, i, msi);
 			if (rc)
 				return rc;
-		irq_set_chip_and_handler(irq, &zpci_irq_chip,
+			irq_set_chip_and_handler(irq + i, &zpci_irq_chip,
 						 handle_percpu_irq);
+		}
+
 		msg.data = hwirq - bit;
 		if (irq_delivery == DIRECTED) {
 			if (msi->affinity)
@@ -335,31 +361,35 @@ int arch_setup_msi_irqs(struct pci_dev *pdev, int nvec, int type)
 			msg.address_lo |= (cpu_addr << 8);
 
 			for_each_possible_cpu(cpu) {
-				airq_iv_set_data(zpci_ibv[cpu], hwirq, irq);
+				for (i = 0; i < irqs_per_msi; i++)
+					airq_iv_set_data(zpci_ibv[cpu],
+							 hwirq + i, irq + i);
 			}
 		} else {
 			msg.address_lo = zdev->msi_addr & 0xffffffff;
-			airq_iv_set_data(zdev->aibv, hwirq, irq);
+			for (i = 0; i < irqs_per_msi; i++)
+				airq_iv_set_data(zdev->aibv, hwirq + i, irq + i);
 		}
 		msg.address_hi = zdev->msi_addr >> 32;
 		pci_write_msi_msg(irq, &msg);
-		hwirq++;
+		hwirq += irqs_per_msi;
 	}
 
 	zdev->msi_first_bit = bit;
-	zdev->msi_nr_irqs = msi_vecs;
+	zdev->msi_nr_irqs = hwirq - bit;
 
 	rc = zpci_set_irq(zdev);
 	if (rc)
 		return rc;
 
-	return (msi_vecs == nvec) ? 0 : msi_vecs;
+	return (zdev->msi_nr_irqs == nvec) ? 0 : zdev->msi_nr_irqs;
 }
 
 void arch_teardown_msi_irqs(struct pci_dev *pdev)
 {
 	struct zpci_dev *zdev = to_zpci(pdev);
 	struct msi_desc *msi;
+	unsigned int i;
 	int rc;
 
 	/* Disable interrupts */
@@ -369,8 +399,10 @@ void arch_teardown_msi_irqs(struct pci_dev *pdev)
 
 	/* Release MSI interrupts */
 	msi_for_each_desc(msi, &pdev->dev, MSI_DESC_ASSOCIATED) {
-		irq_set_msi_desc(msi->irq, NULL);
-		irq_free_desc(msi->irq);
+		for (i = 0; i < msi->nvec_used; i++) {
+			irq_set_msi_desc(msi->irq + i, NULL);
+			irq_free_desc(msi->irq + i);
+		}
 		msi->msg.address_lo = 0;
 		msi->msg.address_hi = 0;
 		msi->msg.data = 0;

drivers/s390/char/Kconfig

@@ -96,7 +96,7 @@ config SCLP_OFB
 config S390_UV_UAPI
 	def_tristate m
 	prompt "Ultravisor userspace API"
-	depends on S390 && (KVM || PROTECTED_VIRTUALIZATION_GUEST)
+	depends on S390
 	help
 	  Selecting exposes parts of the UV interface to userspace
 	  by providing a misc character device at /dev/uv.