Merge tag 'loongarch-fixes-6.12-2' of git://git.kernel.org/pub/scm/linux/kernel/git/chenhuacai/linux-loongson

Pull LoongArch fixes from Huacai Chen:

 - fix possible CPUs setup logical-physical CPU mapping, in order to
   avoid CPU hotplug issue

 - fix some KASAN bugs

 - fix AP booting issue in VM mode

 - some trivial cleanups

* tag 'loongarch-fixes-6.12-2' of git://git.kernel.org/pub/scm/linux/kernel/git/chenhuacai/linux-loongson:
  LoongArch: Fix AP booting issue in VM mode
  LoongArch: Add WriteCombine shadow mapping in KASAN
  LoongArch: Disable KASAN if PGDIR_SIZE is too large for cpu_vabits
  LoongArch: Make KASAN work with 5-level page-tables
  LoongArch: Define a default value for VM_DATA_DEFAULT_FLAGS
  LoongArch: Fix early_numa_add_cpu() usage for FDT systems
  LoongArch: For all possible CPUs setup logical-physical CPU mapping

Author: torvalds

arch/loongarch/include/asm/kasan.h

@@ -25,6 +25,7 @@
 /* 64-bit segment value. */
 #define XKPRANGE_UC_SEG		(0x8000)
 #define XKPRANGE_CC_SEG		(0x9000)
+#define XKPRANGE_WC_SEG		(0xa000)
 #define XKVRANGE_VC_SEG		(0xffff)
 
 /* Cached */
@@ -41,20 +42,28 @@
 #define XKPRANGE_UC_SHADOW_SIZE		(XKPRANGE_UC_SIZE >> KASAN_SHADOW_SCALE_SHIFT)
 #define XKPRANGE_UC_SHADOW_END		(XKPRANGE_UC_KASAN_OFFSET + XKPRANGE_UC_SHADOW_SIZE)
 
+/* WriteCombine */
+#define XKPRANGE_WC_START		WRITECOMBINE_BASE
+#define XKPRANGE_WC_SIZE		XRANGE_SIZE
+#define XKPRANGE_WC_KASAN_OFFSET	XKPRANGE_UC_SHADOW_END
+#define XKPRANGE_WC_SHADOW_SIZE		(XKPRANGE_WC_SIZE >> KASAN_SHADOW_SCALE_SHIFT)
+#define XKPRANGE_WC_SHADOW_END		(XKPRANGE_WC_KASAN_OFFSET + XKPRANGE_WC_SHADOW_SIZE)
+
 /* VMALLOC (Cached or UnCached)  */
 #define XKVRANGE_VC_START		MODULES_VADDR
 #define XKVRANGE_VC_SIZE		round_up(KFENCE_AREA_END - MODULES_VADDR + 1, PGDIR_SIZE)
-#define XKVRANGE_VC_KASAN_OFFSET	XKPRANGE_UC_SHADOW_END
+#define XKVRANGE_VC_KASAN_OFFSET	XKPRANGE_WC_SHADOW_END
 #define XKVRANGE_VC_SHADOW_SIZE		(XKVRANGE_VC_SIZE >> KASAN_SHADOW_SCALE_SHIFT)
 #define XKVRANGE_VC_SHADOW_END		(XKVRANGE_VC_KASAN_OFFSET + XKVRANGE_VC_SHADOW_SIZE)
 
 /* KAsan shadow memory start right after vmalloc. */
 #define KASAN_SHADOW_START		round_up(KFENCE_AREA_END, PGDIR_SIZE)
 #define KASAN_SHADOW_SIZE		(XKVRANGE_VC_SHADOW_END - XKPRANGE_CC_KASAN_OFFSET)
-#define KASAN_SHADOW_END		round_up(KASAN_SHADOW_START + KASAN_SHADOW_SIZE, PGDIR_SIZE)
+#define KASAN_SHADOW_END		(round_up(KASAN_SHADOW_START + KASAN_SHADOW_SIZE, PGDIR_SIZE) - 1)
 
 #define XKPRANGE_CC_SHADOW_OFFSET	(KASAN_SHADOW_START + XKPRANGE_CC_KASAN_OFFSET)
 #define XKPRANGE_UC_SHADOW_OFFSET	(KASAN_SHADOW_START + XKPRANGE_UC_KASAN_OFFSET)
+#define XKPRANGE_WC_SHADOW_OFFSET	(KASAN_SHADOW_START + XKPRANGE_WC_KASAN_OFFSET)
 #define XKVRANGE_VC_SHADOW_OFFSET	(KASAN_SHADOW_START + XKVRANGE_VC_KASAN_OFFSET)
 
 extern bool kasan_early_stage;

arch/loongarch/include/asm/page.h

@@ -113,10 +113,7 @@ struct page *tlb_virt_to_page(unsigned long kaddr);
 extern int __virt_addr_valid(volatile void *kaddr);
 #define virt_addr_valid(kaddr)	__virt_addr_valid((volatile void *)(kaddr))
 
-#define VM_DATA_DEFAULT_FLAGS \
-	(VM_READ | VM_WRITE | \
-	 ((current->personality & READ_IMPLIES_EXEC) ? VM_EXEC : 0) | \
-	 VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC)
+#define VM_DATA_DEFAULT_FLAGS	VM_DATA_FLAGS_TSK_EXEC
 
 #include <asm-generic/memory_model.h>
 #include <asm-generic/getorder.h>

arch/loongarch/kernel/acpi.c

@@ -58,48 +58,48 @@ void __iomem *acpi_os_ioremap(acpi_physical_address phys, acpi_size size)
 		return ioremap_cache(phys, size);
 }
 
-static int cpu_enumerated = 0;
-
 #ifdef CONFIG_SMP
-static int set_processor_mask(u32 id, u32 flags)
+static int set_processor_mask(u32 id, u32 pass)
 {
-	int nr_cpus;
-	int cpu, cpuid = id;
-
-	if (!cpu_enumerated)
-		nr_cpus = NR_CPUS;
-	else
-		nr_cpus = nr_cpu_ids;
+	int cpu = -1, cpuid = id;
 
-	if (num_processors >= nr_cpus) {
+	if (num_processors >= NR_CPUS) {
 		pr_warn(PREFIX "nr_cpus limit of %i reached."
-			" processor 0x%x ignored.\n", nr_cpus, cpuid);
+			" processor 0x%x ignored.\n", NR_CPUS, cpuid);
 
 		return -ENODEV;
 
 	}
+
 	if (cpuid == loongson_sysconf.boot_cpu_id)
 		cpu = 0;
-	else
-		cpu = find_first_zero_bit(cpumask_bits(cpu_present_mask), NR_CPUS);
-
-	if (!cpu_enumerated)
-		set_cpu_possible(cpu, true);
 
-	if (flags & ACPI_MADT_ENABLED) {
+	switch (pass) {
+	case 1: /* Pass 1 handle enabled processors */
+		if (cpu < 0)
+			cpu = find_first_zero_bit(cpumask_bits(cpu_present_mask), NR_CPUS);
 		num_processors++;
 		set_cpu_present(cpu, true);
-		__cpu_number_map[cpuid] = cpu;
-		__cpu_logical_map[cpu] = cpuid;
-	} else
+		break;
+	case 2: /* Pass 2 handle disabled processors */
+		if (cpu < 0)
+			cpu = find_first_zero_bit(cpumask_bits(cpu_possible_mask), NR_CPUS);
 		disabled_cpus++;
+		break;
+	default:
+		return cpu;
+	}
+
+	set_cpu_possible(cpu, true);
+	__cpu_number_map[cpuid] = cpu;
+	__cpu_logical_map[cpu] = cpuid;
 
 	return cpu;
 }
 #endif
 
 static int __init
-acpi_parse_processor(union acpi_subtable_headers *header, const unsigned long end)
+acpi_parse_p1_processor(union acpi_subtable_headers *header, const unsigned long end)
 {
 	struct acpi_madt_core_pic *processor = NULL;
 
@@ -110,12 +110,29 @@ acpi_parse_processor(union acpi_subtable_headers *header, const unsigned long en
 	acpi_table_print_madt_entry(&header->common);
 #ifdef CONFIG_SMP
 	acpi_core_pic[processor->core_id] = *processor;
-	set_processor_mask(processor->core_id, processor->flags);
+	if (processor->flags & ACPI_MADT_ENABLED)
+		set_processor_mask(processor->core_id, 1);
 #endif
 
 	return 0;
 }
 
+static int __init
+acpi_parse_p2_processor(union acpi_subtable_headers *header, const unsigned long end)
+{
+	struct acpi_madt_core_pic *processor = NULL;
+
+	processor = (struct acpi_madt_core_pic *)header;
+	if (BAD_MADT_ENTRY(processor, end))
+		return -EINVAL;
+
+#ifdef CONFIG_SMP
+	if (!(processor->flags & ACPI_MADT_ENABLED))
+		set_processor_mask(processor->core_id, 2);
+#endif
+
+	return 0;
+}
 static int __init
 acpi_parse_eio_master(union acpi_subtable_headers *header, const unsigned long end)
 {
@@ -143,12 +160,14 @@ static void __init acpi_process_madt(void)
 	}
 #endif
 	acpi_table_parse_madt(ACPI_MADT_TYPE_CORE_PIC,
-			acpi_parse_processor, MAX_CORE_PIC);
+			acpi_parse_p1_processor, MAX_CORE_PIC);
+
+	acpi_table_parse_madt(ACPI_MADT_TYPE_CORE_PIC,
+			acpi_parse_p2_processor, MAX_CORE_PIC);
 
 	acpi_table_parse_madt(ACPI_MADT_TYPE_EIO_PIC,
 			acpi_parse_eio_master, MAX_IO_PICS);
 
-	cpu_enumerated = 1;
 	loongson_sysconf.nr_cpus = num_processors;
 }
 
@@ -310,6 +329,10 @@ static int __ref acpi_map_cpu2node(acpi_handle handle, int cpu, int physid)
 	int nid;
 
 	nid = acpi_get_node(handle);
+
+	if (nid != NUMA_NO_NODE)
+		nid = early_cpu_to_node(cpu);
+
 	if (nid != NUMA_NO_NODE) {
 		set_cpuid_to_node(physid, nid);
 		node_set(nid, numa_nodes_parsed);
@@ -324,12 +347,14 @@ int acpi_map_cpu(acpi_handle handle, phys_cpuid_t physid, u32 acpi_id, int *pcpu
 {
 	int cpu;
 
-	cpu = set_processor_mask(physid, ACPI_MADT_ENABLED);
-	if (cpu < 0) {
+	cpu = cpu_number_map(physid);
+	if (cpu < 0 || cpu >= nr_cpu_ids) {
 		pr_info(PREFIX "Unable to map lapic to logical cpu number\n");
-		return cpu;
+		return -ERANGE;
 	}
 
+	num_processors++;
+	set_cpu_present(cpu, true);
 	acpi_map_cpu2node(handle, cpu, physid);
 
 	*pcpu = cpu;

arch/loongarch/kernel/paravirt.c

@@ -51,11 +51,18 @@ static u64 paravt_steal_clock(int cpu)
 }
 
 #ifdef CONFIG_SMP
+static struct smp_ops native_ops;
+
 static void pv_send_ipi_single(int cpu, unsigned int action)
 {
 	int min, old;
 	irq_cpustat_t *info = &per_cpu(irq_stat, cpu);
 
+	if (unlikely(action == ACTION_BOOT_CPU)) {
+		native_ops.send_ipi_single(cpu, action);
+		return;
+	}
+
 	old = atomic_fetch_or(BIT(action), &info->message);
 	if (old)
 		return;
@@ -75,6 +82,11 @@ static void pv_send_ipi_mask(const struct cpumask *mask, unsigned int action)
 	if (cpumask_empty(mask))
 		return;
 
+	if (unlikely(action == ACTION_BOOT_CPU)) {
+		native_ops.send_ipi_mask(mask, action);
+		return;
+	}
+
 	action = BIT(action);
 	for_each_cpu(i, mask) {
 		info = &per_cpu(irq_stat, i);
@@ -147,6 +159,8 @@ static void pv_init_ipi(void)
 {
 	int r, swi;
 
+	/* Init native ipi irq for ACTION_BOOT_CPU */
+	native_ops.init_ipi();
 	swi = get_percpu_irq(INT_SWI0);
 	if (swi < 0)
 		panic("SWI0 IRQ mapping failed\n");
@@ -193,6 +207,7 @@ int __init pv_ipi_init(void)
 		return 0;
 
 #ifdef CONFIG_SMP
+	native_ops		= mp_ops;
 	mp_ops.init_ipi		= pv_init_ipi;
 	mp_ops.send_ipi_single	= pv_send_ipi_single;
 	mp_ops.send_ipi_mask	= pv_send_ipi_mask;

arch/loongarch/kernel/smp.c

@@ -302,7 +302,7 @@ static void __init fdt_smp_setup(void)
 		__cpu_number_map[cpuid] = cpu;
 		__cpu_logical_map[cpu] = cpuid;
 
-		early_numa_add_cpu(cpu, 0);
+		early_numa_add_cpu(cpuid, 0);
 		set_cpuid_to_node(cpuid, 0);
 	}
 
@@ -331,11 +331,11 @@ void __init loongson_prepare_cpus(unsigned int max_cpus)
 	int i = 0;
 
 	parse_acpi_topology();
+	cpu_data[0].global_id = cpu_logical_map(0);
 
 	for (i = 0; i < loongson_sysconf.nr_cpus; i++) {
 		set_cpu_present(i, true);
 		csr_mail_send(0, __cpu_logical_map[i], 0);
-		cpu_data[i].global_id = __cpu_logical_map[i];
 	}
 
 	per_cpu(cpu_state, smp_processor_id()) = CPU_ONLINE;
@@ -380,6 +380,7 @@ void loongson_init_secondary(void)
 		     cpu_logical_map(cpu) / loongson_sysconf.cores_per_package;
 	cpu_data[cpu].core = pptt_enabled ? cpu_data[cpu].core :
 		     cpu_logical_map(cpu) % loongson_sysconf.cores_per_package;
+	cpu_data[cpu].global_id = cpu_logical_map(cpu);
 }
 
 void loongson_smp_finish(void)

arch/loongarch/mm/kasan_init.c

@@ -13,6 +13,13 @@
 
 static pgd_t kasan_pg_dir[PTRS_PER_PGD] __initdata __aligned(PAGE_SIZE);
 
+#ifdef __PAGETABLE_P4D_FOLDED
+#define __pgd_none(early, pgd) (0)
+#else
+#define __pgd_none(early, pgd) (early ? (pgd_val(pgd) == 0) : \
+(__pa(pgd_val(pgd)) == (unsigned long)__pa(kasan_early_shadow_p4d)))
+#endif
+
 #ifdef __PAGETABLE_PUD_FOLDED
 #define __p4d_none(early, p4d) (0)
 #else
@@ -55,6 +62,9 @@ void *kasan_mem_to_shadow(const void *addr)
 		case XKPRANGE_UC_SEG:
 			offset = XKPRANGE_UC_SHADOW_OFFSET;
 			break;
+		case XKPRANGE_WC_SEG:
+			offset = XKPRANGE_WC_SHADOW_OFFSET;
+			break;
 		case XKVRANGE_VC_SEG:
 			offset = XKVRANGE_VC_SHADOW_OFFSET;
 			break;
@@ -79,6 +89,8 @@ const void *kasan_shadow_to_mem(const void *shadow_addr)
 
 	if (addr >= XKVRANGE_VC_SHADOW_OFFSET)
 		return (void *)(((addr - XKVRANGE_VC_SHADOW_OFFSET) << KASAN_SHADOW_SCALE_SHIFT) + XKVRANGE_VC_START);
+	else if (addr >= XKPRANGE_WC_SHADOW_OFFSET)
+		return (void *)(((addr - XKPRANGE_WC_SHADOW_OFFSET) << KASAN_SHADOW_SCALE_SHIFT) + XKPRANGE_WC_START);
 	else if (addr >= XKPRANGE_UC_SHADOW_OFFSET)
 		return (void *)(((addr - XKPRANGE_UC_SHADOW_OFFSET) << KASAN_SHADOW_SCALE_SHIFT) + XKPRANGE_UC_START);
 	else if (addr >= XKPRANGE_CC_SHADOW_OFFSET)
@@ -142,6 +154,19 @@ static pud_t *__init kasan_pud_offset(p4d_t *p4dp, unsigned long addr, int node,
 	return pud_offset(p4dp, addr);
 }
 
+static p4d_t *__init kasan_p4d_offset(pgd_t *pgdp, unsigned long addr, int node, bool early)
+{
+	if (__pgd_none(early, pgdp_get(pgdp))) {
+		phys_addr_t p4d_phys = early ?
+			__pa_symbol(kasan_early_shadow_p4d) : kasan_alloc_zeroed_page(node);
+		if (!early)
+			memcpy(__va(p4d_phys), kasan_early_shadow_p4d, sizeof(kasan_early_shadow_p4d));
+		pgd_populate(&init_mm, pgdp, (p4d_t *)__va(p4d_phys));
+	}
+
+	return p4d_offset(pgdp, addr);
+}
+
 static void __init kasan_pte_populate(pmd_t *pmdp, unsigned long addr,
 				      unsigned long end, int node, bool early)
 {
@@ -178,19 +203,19 @@ static void __init kasan_pud_populate(p4d_t *p4dp, unsigned long addr,
 	do {
 		next = pud_addr_end(addr, end);
 		kasan_pmd_populate(pudp, addr, next, node, early);
-	} while (pudp++, addr = next, addr != end);
+	} while (pudp++, addr = next, addr != end && __pud_none(early, READ_ONCE(*pudp)));
 }
 
 static void __init kasan_p4d_populate(pgd_t *pgdp, unsigned long addr,
 					    unsigned long end, int node, bool early)
 {
 	unsigned long next;
-	p4d_t *p4dp = p4d_offset(pgdp, addr);
+	p4d_t *p4dp = kasan_p4d_offset(pgdp, addr, node, early);
 
 	do {
 		next = p4d_addr_end(addr, end);
 		kasan_pud_populate(p4dp, addr, next, node, early);
-	} while (p4dp++, addr = next, addr != end);
+	} while (p4dp++, addr = next, addr != end && __p4d_none(early, READ_ONCE(*p4dp)));
 }
 
 static void __init kasan_pgd_populate(unsigned long addr, unsigned long end,
@@ -218,7 +243,7 @@ static void __init kasan_map_populate(unsigned long start, unsigned long end,
 asmlinkage void __init kasan_early_init(void)
 {
 	BUILD_BUG_ON(!IS_ALIGNED(KASAN_SHADOW_START, PGDIR_SIZE));
-	BUILD_BUG_ON(!IS_ALIGNED(KASAN_SHADOW_END, PGDIR_SIZE));
+	BUILD_BUG_ON(!IS_ALIGNED(KASAN_SHADOW_END + 1, PGDIR_SIZE));
 }
 
 static inline void kasan_set_pgd(pgd_t *pgdp, pgd_t pgdval)
@@ -233,7 +258,7 @@ static void __init clear_pgds(unsigned long start, unsigned long end)
 	 * swapper_pg_dir. pgd_clear() can't be used
 	 * here because it's nop on 2,3-level pagetable setups
 	 */
-	for (; start < end; start += PGDIR_SIZE)
+	for (; start < end; start = pgd_addr_end(start, end))
 		kasan_set_pgd((pgd_t *)pgd_offset_k(start), __pgd(0));
 }
 
@@ -242,6 +267,17 @@ void __init kasan_init(void)
 	u64 i;
 	phys_addr_t pa_start, pa_end;
 
+	/*
+	 * If PGDIR_SIZE is too large for cpu_vabits, KASAN_SHADOW_END will
+	 * overflow UINTPTR_MAX and then looks like a user space address.
+	 * For example, PGDIR_SIZE of CONFIG_4KB_4LEVEL is 2^39, which is too
+	 * large for Loongson-2K series whose cpu_vabits = 39.
+	 */
+	if (KASAN_SHADOW_END < vm_map_base) {
+		pr_warn("PGDIR_SIZE too large for cpu_vabits, KernelAddressSanitizer disabled.\n");
+		return;
+	}
+
 	/*
 	 * PGD was populated as invalid_pmd_table or invalid_pud_table
 	 * in pagetable_init() which depends on how many levels of page