16 files changed, 129 insertions, 46 deletions

diff --git a/arch/arm64/kvm/mmu.c b/arch/arm64/kvm/mmu.c
index 6981b1bc0946..a509b63bd4dd 100644
--- a/arch/arm64/kvm/mmu.c
+++ b/arch/arm64/kvm/mmu.c
@@ -1540,8 +1540,15 @@ static int user_mem_abort(struct kvm_vcpu *vcpu, phys_addr_t fault_ipa,
 		vma_pagesize = min(vma_pagesize, (long)max_map_size);
 	}
 
-	if (vma_pagesize == PMD_SIZE || vma_pagesize == PUD_SIZE)
+	/*
+	 * Both the canonical IPA and fault IPA must be hugepage-aligned to
+	 * ensure we find the right PFN and lay down the mapping in the right
+	 * place.
+	 */
+	if (vma_pagesize == PMD_SIZE || vma_pagesize == PUD_SIZE) {
 		fault_ipa &= ~(vma_pagesize - 1);
+		ipa &= ~(vma_pagesize - 1);
+	}
 
 	gfn = ipa >> PAGE_SHIFT;
 	mte_allowed = kvm_vma_mte_allowed(vma);
diff --git a/arch/arm64/kvm/sys_regs.c b/arch/arm64/kvm/sys_regs.c
index c90324060436..31e49da867ff 100644
--- a/arch/arm64/kvm/sys_regs.c
+++ b/arch/arm64/kvm/sys_regs.c
@@ -33,6 +33,7 @@
 #include <trace/events/kvm.h>
 
 #include "sys_regs.h"
+#include "vgic/vgic.h"
 
 #include "trace.h"
 
@@ -435,6 +436,11 @@ static bool access_gic_sgi(struct kvm_vcpu *vcpu,
 {
 	bool g1;
 
+	if (!kvm_has_gicv3(vcpu->kvm)) {
+		kvm_inject_undefined(vcpu);
+		return false;
+	}
+
 	if (!p->is_write)
 		return read_from_write_only(vcpu, p, r);
 
diff --git a/arch/arm64/kvm/vgic/vgic-debug.c b/arch/arm64/kvm/vgic/vgic-debug.c
index bc74d06398ef..e1397ab2072a 100644
--- a/arch/arm64/kvm/vgic/vgic-debug.c
+++ b/arch/arm64/kvm/vgic/vgic-debug.c
@@ -85,7 +85,7 @@ static void iter_unmark_lpis(struct kvm *kvm)
 	struct vgic_irq *irq;
 	unsigned long intid;
 
-	xa_for_each(&dist->lpi_xa, intid, irq) {
+	xa_for_each_marked(&dist->lpi_xa, intid, irq, LPI_XA_MARK_DEBUG_ITER) {
 		xa_clear_mark(&dist->lpi_xa, intid, LPI_XA_MARK_DEBUG_ITER);
 		vgic_put_irq(kvm, irq);
 	}
diff --git a/arch/arm64/kvm/vgic/vgic-init.c b/arch/arm64/kvm/vgic/vgic-init.c
index 41feb858ff9a..e7c53e8af3d1 100644
--- a/arch/arm64/kvm/vgic/vgic-init.c
+++ b/arch/arm64/kvm/vgic/vgic-init.c
@@ -417,10 +417,8 @@ static void __kvm_vgic_vcpu_destroy(struct kvm_vcpu *vcpu)
 	kfree(vgic_cpu->private_irqs);
 	vgic_cpu->private_irqs = NULL;
 
-	if (vcpu->kvm->arch.vgic.vgic_model == KVM_DEV_TYPE_ARM_VGIC_V3) {
-		vgic_unregister_redist_iodev(vcpu);
+	if (vcpu->kvm->arch.vgic.vgic_model == KVM_DEV_TYPE_ARM_VGIC_V3)
 		vgic_cpu->rd_iodev.base_addr = VGIC_ADDR_UNDEF;
-	}
 }
 
 void kvm_vgic_vcpu_destroy(struct kvm_vcpu *vcpu)
@@ -448,6 +446,11 @@ void kvm_vgic_destroy(struct kvm *kvm)
 	kvm_vgic_dist_destroy(kvm);
 
 	mutex_unlock(&kvm->arch.config_lock);
+
+	if (kvm->arch.vgic.vgic_model == KVM_DEV_TYPE_ARM_VGIC_V3)
+		kvm_for_each_vcpu(i, vcpu, kvm)
+			vgic_unregister_redist_iodev(vcpu);
+
 	mutex_unlock(&kvm->slots_lock);
 }
 
diff --git a/arch/arm64/kvm/vgic/vgic.c b/arch/arm64/kvm/vgic/vgic.c
index 974849ea7101..abe29c7d85d0 100644
--- a/arch/arm64/kvm/vgic/vgic.c
+++ b/arch/arm64/kvm/vgic/vgic.c
@@ -36,6 +36,11 @@ struct vgic_global kvm_vgic_global_state __ro_after_init = {
  * we have to disable IRQs before taking this lock and everything lower
  * than it.
  *
+ * The config_lock has additional ordering requirements:
+ * kvm->slots_lock
+ *   kvm->srcu
+ *     kvm->arch.config_lock
+ *
  * If you need to take multiple locks, always take the upper lock first,
  * then the lower ones, e.g. first take the its_lock, then the irq_lock.
  * If you are already holding a lock and need to take a higher one, you
diff --git a/arch/arm64/kvm/vgic/vgic.h b/arch/arm64/kvm/vgic/vgic.h
index ba8f790431bd..8532bfe3fed4 100644
--- a/arch/arm64/kvm/vgic/vgic.h
+++ b/arch/arm64/kvm/vgic/vgic.h
@@ -346,4 +346,11 @@ void vgic_v4_configure_vsgis(struct kvm *kvm);
 void vgic_v4_get_vlpi_state(struct vgic_irq *irq, bool *val);
 int vgic_v4_request_vpe_irq(struct kvm_vcpu *vcpu, int irq);
 
+static inline bool kvm_has_gicv3(struct kvm *kvm)
+{
+	return (static_branch_unlikely(&kvm_vgic_global_state.gicv3_cpuif) &&
+		irqchip_in_kernel(kvm) &&
+		kvm->arch.vgic.vgic_model == KVM_DEV_TYPE_ARM_VGIC_V3);
+}
+
 #endif
diff --git a/arch/mips/kernel/cevt-r4k.c b/arch/mips/kernel/cevt-r4k.c
index 368e8475870f..5f6e9e2ebbdb 100644
--- a/arch/mips/kernel/cevt-r4k.c
+++ b/arch/mips/kernel/cevt-r4k.c
@@ -303,13 +303,6 @@ int r4k_clockevent_init(void)
 	if (!c0_compare_int_usable())
 		return -ENXIO;
 
-	/*
-	 * With vectored interrupts things are getting platform specific.
-	 * get_c0_compare_int is a hook to allow a platform to return the
-	 * interrupt number of its liking.
-	 */
-	irq = get_c0_compare_int();
-
 	cd = &per_cpu(mips_clockevent_device, cpu);
 
 	cd->name		= "MIPS";
@@ -320,7 +313,6 @@ int r4k_clockevent_init(void)
 	min_delta		= calculate_min_delta();
 
 	cd->rating		= 300;
-	cd->irq			= irq;
 	cd->cpumask		= cpumask_of(cpu);
 	cd->set_next_event	= mips_next_event;
 	cd->event_handler	= mips_event_handler;
@@ -332,6 +324,13 @@ int r4k_clockevent_init(void)
 
 	cp0_timer_irq_installed = 1;
 
+	/*
+	 * With vectored interrupts things are getting platform specific.
+	 * get_c0_compare_int is a hook to allow a platform to return the
+	 * interrupt number of its liking.
+	 */
+	irq = get_c0_compare_int();
+
 	if (request_irq(irq, c0_compare_interrupt, flags, "timer",
 			c0_compare_interrupt))
 		pr_err("Failed to request irq %d (timer)\n", irq);
diff --git a/arch/mips/kernel/cpu-probe.c b/arch/mips/kernel/cpu-probe.c
index bda7f193baab..af7412549e6e 100644
--- a/arch/mips/kernel/cpu-probe.c
+++ b/arch/mips/kernel/cpu-probe.c
@@ -1724,12 +1724,16 @@ static inline void cpu_probe_loongson(struct cpuinfo_mips *c, unsigned int cpu)
 		c->ases |= (MIPS_ASE_LOONGSON_MMI | MIPS_ASE_LOONGSON_CAM |
 			MIPS_ASE_LOONGSON_EXT | MIPS_ASE_LOONGSON_EXT2);
 		c->ases &= ~MIPS_ASE_VZ; /* VZ of Loongson-3A2000/3000 is incomplete */
+		change_c0_config6(LOONGSON_CONF6_EXTIMER | LOONGSON_CONF6_INTIMER,
+				  LOONGSON_CONF6_INTIMER);
 		break;
 	case PRID_IMP_LOONGSON_64G:
 		__cpu_name[cpu] = "ICT Loongson-3";
 		set_elf_platform(cpu, "loongson3a");
 		set_isa(c, MIPS_CPU_ISA_M64R2);
 		decode_cpucfg(c);
+		change_c0_config6(LOONGSON_CONF6_EXTIMER | LOONGSON_CONF6_INTIMER,
+				  LOONGSON_CONF6_INTIMER);
 		break;
 	default:
 		panic("Unknown Loongson Processor ID!");
diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
index a822f952f64a..c60e699e99f5 100644
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -604,6 +604,19 @@ config RANDOMIZE_BASE
 	  as a security feature that deters exploit attempts relying on
 	  knowledge of the location of kernel internals.
 
+config RANDOMIZE_IDENTITY_BASE
+	bool "Randomize the address of the identity mapping base"
+	depends on RANDOMIZE_BASE
+	default DEBUG_VM
+	help
+	  The identity mapping base address is pinned to zero by default.
+	  Allow randomization of that base to expose otherwise missed
+	  notion of physical and virtual addresses of data structures.
+	  That does not have any impact on the base address at which the
+	  kernel image is loaded.
+
+	  If unsure, say N
+
 config KERNEL_IMAGE_BASE
 	hex "Kernel image base address"
 	range 0x100000 0x1FFFFFE0000000 if !KASAN
diff --git a/arch/s390/boot/startup.c b/arch/s390/boot/startup.c
index ce232552bc1c..c73b5118ad42 100644
--- a/arch/s390/boot/startup.c
+++ b/arch/s390/boot/startup.c
@@ -162,7 +162,7 @@ static void kaslr_adjust_relocs(unsigned long min_addr, unsigned long max_addr,
 		loc = (long)*reloc + phys_offset;
 		if (loc < min_addr || loc > max_addr)
 			error("64-bit relocation outside of kernel!\n");
-		*(u64 *)loc += offset - __START_KERNEL;
+		*(u64 *)loc += offset;
 	}
 }
 
@@ -177,7 +177,7 @@ static void kaslr_adjust_got(unsigned long offset)
 	 */
 	for (entry = (u64 *)vmlinux.got_start; entry < (u64 *)vmlinux.got_end; entry++) {
 		if (*entry)
-			*entry += offset - __START_KERNEL;
+			*entry += offset;
 	}
 }
 
@@ -252,7 +252,7 @@ static unsigned long setup_kernel_memory_layout(unsigned long kernel_size)
 	vmemmap_size = SECTION_ALIGN_UP(pages) * sizeof(struct page);
 
 	/* choose kernel address space layout: 4 or 3 levels. */
-	BUILD_BUG_ON(!IS_ALIGNED(__START_KERNEL, THREAD_SIZE));
+	BUILD_BUG_ON(!IS_ALIGNED(TEXT_OFFSET, THREAD_SIZE));
 	BUILD_BUG_ON(!IS_ALIGNED(__NO_KASLR_START_KERNEL, THREAD_SIZE));
 	BUILD_BUG_ON(__NO_KASLR_END_KERNEL > _REGION1_SIZE);
 	vsize = get_vmem_size(ident_map_size, vmemmap_size, vmalloc_size, _REGION3_SIZE);
@@ -341,7 +341,8 @@ static unsigned long setup_kernel_memory_layout(unsigned long kernel_size)
 	BUILD_BUG_ON(MAX_DCSS_ADDR > (1UL << MAX_PHYSMEM_BITS));
 	max_mappable = max(ident_map_size, MAX_DCSS_ADDR);
 	max_mappable = min(max_mappable, vmemmap_start);
-	__identity_base = round_down(vmemmap_start - max_mappable, rte_size);
+	if (IS_ENABLED(CONFIG_RANDOMIZE_IDENTITY_BASE))
+		__identity_base = round_down(vmemmap_start - max_mappable, rte_size);
 
 	return asce_limit;
 }
@@ -388,31 +389,25 @@ static void kaslr_adjust_vmlinux_info(long offset)
 #endif
 }
 
-static void fixup_vmlinux_info(void)
-{
-	vmlinux.entry -= __START_KERNEL;
-	kaslr_adjust_vmlinux_info(-__START_KERNEL);
-}
-
 void startup_kernel(void)
 {
-	unsigned long kernel_size = vmlinux.image_size + vmlinux.bss_size;
-	unsigned long nokaslr_offset_phys, kaslr_large_page_offset;
-	unsigned long amode31_lma = 0;
+	unsigned long vmlinux_size = vmlinux.image_size + vmlinux.bss_size;
+	unsigned long nokaslr_text_lma, text_lma = 0, amode31_lma = 0;
+	unsigned long kernel_size = TEXT_OFFSET + vmlinux_size;
+	unsigned long kaslr_large_page_offset;
 	unsigned long max_physmem_end;
 	unsigned long asce_limit;
 	unsigned long safe_addr;
 	psw_t psw;
 
-	fixup_vmlinux_info();
 	setup_lpp();
 
 	/*
 	 * Non-randomized kernel physical start address must be _SEGMENT_SIZE
 	 * aligned (see blow).
 	 */
-	nokaslr_offset_phys = ALIGN(mem_safe_offset(), _SEGMENT_SIZE);
-	safe_addr = PAGE_ALIGN(nokaslr_offset_phys + kernel_size);
+	nokaslr_text_lma = ALIGN(mem_safe_offset(), _SEGMENT_SIZE);
+	safe_addr = PAGE_ALIGN(nokaslr_text_lma + vmlinux_size);
 
 	/*
 	 * Reserve decompressor memory together with decompression heap,
@@ -456,16 +451,27 @@ void startup_kernel(void)
 	 */
 	kaslr_large_page_offset = __kaslr_offset & ~_SEGMENT_MASK;
 	if (kaslr_enabled()) {
-		unsigned long end = ident_map_size - kaslr_large_page_offset;
+		unsigned long size = vmlinux_size + kaslr_large_page_offset;
 
-		__kaslr_offset_phys = randomize_within_range(kernel_size, _SEGMENT_SIZE, 0, end);
+		text_lma = randomize_within_range(size, _SEGMENT_SIZE, TEXT_OFFSET, ident_map_size);
 	}
-	if (!__kaslr_offset_phys)
-		__kaslr_offset_phys = nokaslr_offset_phys;
-	__kaslr_offset_phys |= kaslr_large_page_offset;
+	if (!text_lma)
+		text_lma = nokaslr_text_lma;
+	text_lma |= kaslr_large_page_offset;
+
+	/*
+	 * [__kaslr_offset_phys..__kaslr_offset_phys + TEXT_OFFSET] region is
+	 * never accessed via the kernel image mapping as per the linker script:
+	 *
+	 *	. = TEXT_OFFSET;
+	 *
+	 * Therefore, this region could be used for something else and does
+	 * not need to be reserved. See how it is skipped in setup_vmem().
+	 */
+	__kaslr_offset_phys = text_lma - TEXT_OFFSET;
 	kaslr_adjust_vmlinux_info(__kaslr_offset_phys);
-	physmem_reserve(RR_VMLINUX, __kaslr_offset_phys, kernel_size);
-	deploy_kernel((void *)__kaslr_offset_phys);
+	physmem_reserve(RR_VMLINUX, text_lma, vmlinux_size);
+	deploy_kernel((void *)text_lma);
 
 	/* vmlinux decompression is done, shrink reserved low memory */
 	physmem_reserve(RR_DECOMPRESSOR, 0, (unsigned long)_decompressor_end);
@@ -488,7 +494,7 @@ void startup_kernel(void)
 		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;
+		amode31_lma = text_lma - vmlinux.amode31_size;
 	physmem_reserve(RR_AMODE31, amode31_lma, vmlinux.amode31_size);
 
 	/*
@@ -504,8 +510,8 @@ void startup_kernel(void)
 	 * - copy_bootdata() must follow setup_vmem() to propagate changes
 	 *   to bootdata made by setup_vmem()
 	 */
-	clear_bss_section(__kaslr_offset_phys);
-	kaslr_adjust_relocs(__kaslr_offset_phys, __kaslr_offset_phys + vmlinux.image_size,
+	clear_bss_section(text_lma);
+	kaslr_adjust_relocs(text_lma, text_lma + vmlinux.image_size,
 			    __kaslr_offset, __kaslr_offset_phys);
 	kaslr_adjust_got(__kaslr_offset);
 	setup_vmem(__kaslr_offset, __kaslr_offset + kernel_size, asce_limit);
diff --git a/arch/s390/boot/vmem.c b/arch/s390/boot/vmem.c
index 2847cc059ab7..145035f84a0e 100644
--- a/arch/s390/boot/vmem.c
+++ b/arch/s390/boot/vmem.c
@@ -90,7 +90,7 @@ static void kasan_populate_shadow(unsigned long kernel_start, unsigned long kern
 		}
 		memgap_start = end;
 	}
-	kasan_populate(kernel_start, kernel_end, POPULATE_KASAN_MAP_SHADOW);
+	kasan_populate(kernel_start + TEXT_OFFSET, kernel_end, POPULATE_KASAN_MAP_SHADOW);
 	kasan_populate(0, (unsigned long)__identity_va(0), POPULATE_KASAN_ZERO_SHADOW);
 	kasan_populate(AMODE31_START, AMODE31_END, POPULATE_KASAN_ZERO_SHADOW);
 	if (IS_ENABLED(CONFIG_KASAN_VMALLOC)) {
@@ -475,7 +475,17 @@ void setup_vmem(unsigned long kernel_start, unsigned long kernel_end, unsigned l
 				 (unsigned long)__identity_va(end),
 				 POPULATE_IDENTITY);
 	}
-	pgtable_populate(kernel_start, kernel_end, POPULATE_KERNEL);
+
+	/*
+	 * [kernel_start..kernel_start + TEXT_OFFSET] region is never
+	 * accessed as per the linker script:
+	 *
+	 *	. = TEXT_OFFSET;
+	 *
+	 * Therefore, skip mapping TEXT_OFFSET bytes to prevent access to
+	 * [__kaslr_offset_phys..__kaslr_offset_phys + TEXT_OFFSET] region.
+	 */
+	pgtable_populate(kernel_start + TEXT_OFFSET, kernel_end, POPULATE_KERNEL);
 	pgtable_populate(AMODE31_START, AMODE31_END, POPULATE_DIRECT);
 	pgtable_populate(__abs_lowcore, __abs_lowcore + sizeof(struct lowcore),
 			 POPULATE_ABS_LOWCORE);
diff --git a/arch/s390/boot/vmlinux.lds.S b/arch/s390/boot/vmlinux.lds.S
index a750711d44c8..66670212a361 100644
--- a/arch/s390/boot/vmlinux.lds.S
+++ b/arch/s390/boot/vmlinux.lds.S
@@ -109,7 +109,12 @@ SECTIONS
 #ifdef CONFIG_KERNEL_UNCOMPRESSED
 	. = ALIGN(PAGE_SIZE);
 	. += AMODE31_SIZE;		/* .amode31 section */
-	. = ALIGN(1 << 20);		/* _SEGMENT_SIZE */
+
+	/*
+	 * Make sure the location counter is not less than TEXT_OFFSET.
+	 * _SEGMENT_SIZE is not available, use ALIGN(1 << 20) instead.
+	 */
+	. = MAX(TEXT_OFFSET, ALIGN(1 << 20));
 #else
 	. = ALIGN(8);
 #endif
diff --git a/arch/s390/include/asm/page.h b/arch/s390/include/asm/page.h
index 06416b3f94f5..16e4caa931f1 100644
--- a/arch/s390/include/asm/page.h
+++ b/arch/s390/include/asm/page.h
@@ -279,8 +279,9 @@ static inline unsigned long virt_to_pfn(const void *kaddr)
 #define AMODE31_SIZE		(3 * PAGE_SIZE)
 
 #define KERNEL_IMAGE_SIZE	(512 * 1024 * 1024)
-#define __START_KERNEL		0x100000
 #define __NO_KASLR_START_KERNEL	CONFIG_KERNEL_IMAGE_BASE
 #define __NO_KASLR_END_KERNEL	(__NO_KASLR_START_KERNEL + KERNEL_IMAGE_SIZE)
 
+#define TEXT_OFFSET		0x100000
+
 #endif /* _S390_PAGE_H */
diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index 4ec99f73fa27..a3fea683b227 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -734,7 +734,23 @@ static void __init memblock_add_physmem_info(void)
 }
 
 /*
- * Reserve memory used for lowcore/command line/kernel image.
+ * Reserve memory used for lowcore.
+ */
+static void __init reserve_lowcore(void)
+{
+	void *lowcore_start = get_lowcore();
+	void *lowcore_end = lowcore_start + sizeof(struct lowcore);
+	void *start, *end;
+
+	if ((void *)__identity_base < lowcore_end) {
+		start = max(lowcore_start, (void *)__identity_base);
+		end = min(lowcore_end, (void *)(__identity_base + ident_map_size));
+		memblock_reserve(__pa(start), __pa(end));
+	}
+}
+
+/*
+ * Reserve memory used for absolute lowcore/command line/kernel image.
  */
 static void __init reserve_kernel(void)
 {
@@ -918,6 +934,7 @@ void __init setup_arch(char **cmdline_p)
 
 	/* Do some memory reservations *before* memory is added to memblock */
 	reserve_pgtables();
+	reserve_lowcore();
 	reserve_kernel();
 	reserve_initrd();
 	reserve_certificate_list();
diff --git a/arch/s390/kernel/vmlinux.lds.S b/arch/s390/kernel/vmlinux.lds.S
index e67cd409b858..ae5d0a9d6911 100644
--- a/arch/s390/kernel/vmlinux.lds.S
+++ b/arch/s390/kernel/vmlinux.lds.S
@@ -39,7 +39,7 @@ PHDRS {
 
 SECTIONS
 {
-	. = __START_KERNEL;
+	. = TEXT_OFFSET;
 	.text : {
 		_stext = .;		/* Start of text section */
 		_text = .;		/* Text and read-only data */
diff --git a/arch/s390/tools/relocs.c b/arch/s390/tools/relocs.c
index a74dbd5c9896..30a732c808f3 100644
--- a/arch/s390/tools/relocs.c
+++ b/arch/s390/tools/relocs.c
@@ -280,7 +280,7 @@ static int do_reloc(struct section *sec, Elf_Rel *rel)
 	case R_390_GOTOFF64:
 		break;
 	case R_390_64:
-		add_reloc(&relocs64, offset - ehdr.e_entry);
+		add_reloc(&relocs64, offset);
 		break;
 	default:
 		die("Unsupported relocation type: %d\n", r_type);