Merge tag 'loongarch-fixes-6.12-1' of git://git.kernel.org/pub/scm/linux/kernel/git/chenhuacai/linux-loongson

Pull LoongArch fixes from Huacai Chen:
 "Get correct cores_per_package for SMT systems, enable IRQ if do_ale()
  triggered in irq-enabled context, and fix some bugs about vDSO, memory
  managenent, hrtimer in KVM, etc"

* tag 'loongarch-fixes-6.12-1' of git://git.kernel.org/pub/scm/linux/kernel/git/chenhuacai/linux-loongson:
  LoongArch: KVM: Mark hrtimer to expire in hard interrupt context
  LoongArch: Make KASAN usable for variable cpu_vabits
  LoongArch: Set initial pte entry with PAGE_GLOBAL for kernel space
  LoongArch: Don't crash in stack_top() for tasks without vDSO
  LoongArch: Set correct size for vDSO code mapping
  LoongArch: Enable IRQ if do_ale() triggered in irq-enabled context
  LoongArch: Get correct cores_per_package for SMT systems
  LoongArch: Use "Exception return address" to comment ERA

Author: torvalds

arch/loongarch/include/asm/bootinfo.h

@@ -26,6 +26,10 @@ struct loongson_board_info {
 
 #define NR_WORDS DIV_ROUND_UP(NR_CPUS, BITS_PER_LONG)
 
+/*
+ * The "core" of cores_per_node and cores_per_package stands for a
+ * logical core, which means in a SMT system it stands for a thread.
+ */
 struct loongson_system_configuration {
 	int nr_cpus;
 	int nr_nodes;

arch/loongarch/include/asm/kasan.h

@@ -16,7 +16,7 @@
 #define XRANGE_SHIFT (48)
 
 /* Valid address length */
-#define XRANGE_SHADOW_SHIFT	(PGDIR_SHIFT + PAGE_SHIFT - 3)
+#define XRANGE_SHADOW_SHIFT	min(cpu_vabits, VA_BITS)
 /* Used for taking out the valid address */
 #define XRANGE_SHADOW_MASK	GENMASK_ULL(XRANGE_SHADOW_SHIFT - 1, 0)
 /* One segment whole address space size */

arch/loongarch/include/asm/loongarch.h

@@ -250,7 +250,7 @@
 #define  CSR_ESTAT_IS_WIDTH		15
 #define  CSR_ESTAT_IS			(_ULCAST_(0x7fff) << CSR_ESTAT_IS_SHIFT)
 
-#define LOONGARCH_CSR_ERA		0x6	/* ERA */
+#define LOONGARCH_CSR_ERA		0x6	/* Exception return address */
 
 #define LOONGARCH_CSR_BADV		0x7	/* Bad virtual address */
 

arch/loongarch/include/asm/pgalloc.h

@@ -10,6 +10,7 @@
 
 #define __HAVE_ARCH_PMD_ALLOC_ONE
 #define __HAVE_ARCH_PUD_ALLOC_ONE
+#define __HAVE_ARCH_PTE_ALLOC_ONE_KERNEL
 #include <asm-generic/pgalloc.h>
 
 static inline void pmd_populate_kernel(struct mm_struct *mm,
@@ -44,6 +45,16 @@ extern void pagetable_init(void);
 
 extern pgd_t *pgd_alloc(struct mm_struct *mm);
 
+static inline pte_t *pte_alloc_one_kernel(struct mm_struct *mm)
+{
+	pte_t *pte = __pte_alloc_one_kernel(mm);
+
+	if (pte)
+		kernel_pte_init(pte);
+
+	return pte;
+}
+
 #define __pte_free_tlb(tlb, pte, address)			\
 do {								\
 	pagetable_pte_dtor(page_ptdesc(pte));			\

arch/loongarch/include/asm/pgtable.h

@@ -269,6 +269,7 @@ extern void set_pmd_at(struct mm_struct *mm, unsigned long addr, pmd_t *pmdp, pm
 extern void pgd_init(void *addr);
 extern void pud_init(void *addr);
 extern void pmd_init(void *addr);
+extern void kernel_pte_init(void *addr);
 
 /*
  * Encode/decode swap entries and swap PTEs. Swap PTEs are all PTEs that
@@ -325,39 +326,17 @@ static inline void set_pte(pte_t *ptep, pte_t pteval)
 {
 	WRITE_ONCE(*ptep, pteval);
 
-	if (pte_val(pteval) & _PAGE_GLOBAL) {
-		pte_t *buddy = ptep_buddy(ptep);
-		/*
-		 * Make sure the buddy is global too (if it's !none,
-		 * it better already be global)
-		 */
-		if (pte_none(ptep_get(buddy))) {
 #ifdef CONFIG_SMP
-			/*
-			 * For SMP, multiple CPUs can race, so we need
-			 * to do this atomically.
-			 */
-			__asm__ __volatile__(
-			__AMOR "$zero, %[global], %[buddy] \n"
-			: [buddy] "+ZB" (buddy->pte)
-			: [global] "r" (_PAGE_GLOBAL)
-			: "memory");
-
-			DBAR(0b11000); /* o_wrw = 0b11000 */
-#else /* !CONFIG_SMP */
-			WRITE_ONCE(*buddy, __pte(pte_val(ptep_get(buddy)) | _PAGE_GLOBAL));
-#endif /* CONFIG_SMP */
-		}
-	}
+	if (pte_val(pteval) & _PAGE_GLOBAL)
+		DBAR(0b11000); /* o_wrw = 0b11000 */
+#endif
 }
 
 static inline void pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
 {
-	/* Preserve global status for the pair */
-	if (pte_val(ptep_get(ptep_buddy(ptep))) & _PAGE_GLOBAL)
-		set_pte(ptep, __pte(_PAGE_GLOBAL));
-	else
-		set_pte(ptep, __pte(0));
+	pte_t pte = ptep_get(ptep);
+	pte_val(pte) &= _PAGE_GLOBAL;
+	set_pte(ptep, pte);
 }
 
 #define PGD_T_LOG2	(__builtin_ffs(sizeof(pgd_t)) - 1)

arch/loongarch/kernel/process.c

@@ -293,13 +293,15 @@ unsigned long stack_top(void)
 {
 	unsigned long top = TASK_SIZE & PAGE_MASK;
 
-	/* Space for the VDSO & data page */
-	top -= PAGE_ALIGN(current->thread.vdso->size);
-	top -= VVAR_SIZE;
-
-	/* Space to randomize the VDSO base */
-	if (current->flags & PF_RANDOMIZE)
-		top -= VDSO_RANDOMIZE_SIZE;
+	if (current->thread.vdso) {
+		/* Space for the VDSO & data page */
+		top -= PAGE_ALIGN(current->thread.vdso->size);
+		top -= VVAR_SIZE;
+
+		/* Space to randomize the VDSO base */
+		if (current->flags & PF_RANDOMIZE)
+			top -= VDSO_RANDOMIZE_SIZE;
+	}
 
 	return top;
 }

arch/loongarch/kernel/setup.c

@@ -55,6 +55,7 @@
 #define SMBIOS_FREQHIGH_OFFSET		0x17
 #define SMBIOS_FREQLOW_MASK		0xFF
 #define SMBIOS_CORE_PACKAGE_OFFSET	0x23
+#define SMBIOS_THREAD_PACKAGE_OFFSET	0x25
 #define LOONGSON_EFI_ENABLE		(1 << 3)
 
 unsigned long fw_arg0, fw_arg1, fw_arg2;
@@ -125,7 +126,7 @@ static void __init parse_cpu_table(const struct dmi_header *dm)
 	cpu_clock_freq = freq_temp * 1000000;
 
 	loongson_sysconf.cpuname = (void *)dmi_string_parse(dm, dmi_data[16]);
-	loongson_sysconf.cores_per_package = *(dmi_data + SMBIOS_CORE_PACKAGE_OFFSET);
+	loongson_sysconf.cores_per_package = *(dmi_data + SMBIOS_THREAD_PACKAGE_OFFSET);
 
 	pr_info("CpuClock = %llu\n", cpu_clock_freq);
 }

arch/loongarch/kernel/traps.c

@@ -555,6 +555,9 @@ asmlinkage void noinstr do_ale(struct pt_regs *regs)
 #else
 	unsigned int *pc;
 
+	if (regs->csr_prmd & CSR_PRMD_PIE)
+		local_irq_enable();
+
 	perf_sw_event(PERF_COUNT_SW_ALIGNMENT_FAULTS, 1, regs, regs->csr_badvaddr);
 
 	/*
@@ -579,6 +582,8 @@ asmlinkage void noinstr do_ale(struct pt_regs *regs)
 	die_if_kernel("Kernel ale access", regs);
 	force_sig_fault(SIGBUS, BUS_ADRALN, (void __user *)regs->csr_badvaddr);
 out:
+	if (regs->csr_prmd & CSR_PRMD_PIE)
+		local_irq_disable();
 #endif
 	irqentry_exit(regs, state);
 }

arch/loongarch/kernel/vdso.c

@@ -34,7 +34,6 @@ static union {
 	struct loongarch_vdso_data vdata;
 } loongarch_vdso_data __page_aligned_data;
 
-static struct page *vdso_pages[] = { NULL };
 struct vdso_data *vdso_data = generic_vdso_data.data;
 struct vdso_pcpu_data *vdso_pdata = loongarch_vdso_data.vdata.pdata;
 struct vdso_rng_data *vdso_rng_data = &loongarch_vdso_data.vdata.rng_data;
@@ -85,10 +84,8 @@ static vm_fault_t vvar_fault(const struct vm_special_mapping *sm,
 
 struct loongarch_vdso_info vdso_info = {
 	.vdso = vdso_start,
-	.size = PAGE_SIZE,
 	.code_mapping = {
 		.name = "[vdso]",
-		.pages = vdso_pages,
 		.mremap = vdso_mremap,
 	},
 	.data_mapping = {
@@ -103,11 +100,14 @@ static int __init init_vdso(void)
 	unsigned long i, cpu, pfn;
 
 	BUG_ON(!PAGE_ALIGNED(vdso_info.vdso));
-	BUG_ON(!PAGE_ALIGNED(vdso_info.size));
 
 	for_each_possible_cpu(cpu)
 		vdso_pdata[cpu].node = cpu_to_node(cpu);
 
+	vdso_info.size = PAGE_ALIGN(vdso_end - vdso_start);
+	vdso_info.code_mapping.pages =
+		kcalloc(vdso_info.size / PAGE_SIZE, sizeof(struct page *), GFP_KERNEL);
+
 	pfn = __phys_to_pfn(__pa_symbol(vdso_info.vdso));
 	for (i = 0; i < vdso_info.size / PAGE_SIZE; i++)
 		vdso_info.code_mapping.pages[i] = pfn_to_page(pfn + i);

arch/loongarch/kvm/timer.c

@@ -161,10 +161,11 @@ static void _kvm_save_timer(struct kvm_vcpu *vcpu)
 	if (kvm_vcpu_is_blocking(vcpu)) {
 
 		/*
-		 * HRTIMER_MODE_PINNED is suggested since vcpu may run in
-		 * the same physical cpu in next time
+		 * HRTIMER_MODE_PINNED_HARD is suggested since vcpu may run in
+		 * the same physical cpu in next time, and the timer should run
+		 * in hardirq context even in the PREEMPT_RT case.
 		 */
-		hrtimer_start(&vcpu->arch.swtimer, expire, HRTIMER_MODE_ABS_PINNED);
+		hrtimer_start(&vcpu->arch.swtimer, expire, HRTIMER_MODE_ABS_PINNED_HARD);
 	}
 }
 

arch/loongarch/kvm/vcpu.c

@@ -1457,7 +1457,7 @@ int kvm_arch_vcpu_create(struct kvm_vcpu *vcpu)
 	vcpu->arch.vpid = 0;
 	vcpu->arch.flush_gpa = INVALID_GPA;
 
-	hrtimer_init(&vcpu->arch.swtimer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS_PINNED);
+	hrtimer_init(&vcpu->arch.swtimer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS_PINNED_HARD);
 	vcpu->arch.swtimer.function = kvm_swtimer_wakeup;
 
 	vcpu->arch.handle_exit = kvm_handle_exit;

arch/loongarch/mm/init.c

@@ -201,7 +201,9 @@ pte_t * __init populate_kernel_pte(unsigned long addr)
 		pte = memblock_alloc(PAGE_SIZE, PAGE_SIZE);
 		if (!pte)
 			panic("%s: Failed to allocate memory\n", __func__);
+
 		pmd_populate_kernel(&init_mm, pmd, pte);
+		kernel_pte_init(pte);
 	}
 
 	return pte_offset_kernel(pmd, addr);

arch/loongarch/mm/pgtable.c

@@ -116,6 +116,26 @@ void pud_init(void *addr)
 EXPORT_SYMBOL_GPL(pud_init);
 #endif
 
+void kernel_pte_init(void *addr)
+{
+	unsigned long *p, *end;
+
+	p = (unsigned long *)addr;
+	end = p + PTRS_PER_PTE;
+
+	do {
+		p[0] = _PAGE_GLOBAL;
+		p[1] = _PAGE_GLOBAL;
+		p[2] = _PAGE_GLOBAL;
+		p[3] = _PAGE_GLOBAL;
+		p[4] = _PAGE_GLOBAL;
+		p += 8;
+		p[-3] = _PAGE_GLOBAL;
+		p[-2] = _PAGE_GLOBAL;
+		p[-1] = _PAGE_GLOBAL;
+	} while (p != end);
+}
+
 pmd_t mk_pmd(struct page *page, pgprot_t prot)
 {
 	pmd_t pmd;

include/linux/mm.h

@@ -3818,8 +3818,9 @@ void *sparse_buffer_alloc(unsigned long size);
 struct page * __populate_section_memmap(unsigned long pfn,
 		unsigned long nr_pages, int nid, struct vmem_altmap *altmap,
 		struct dev_pagemap *pgmap);
-void pmd_init(void *addr);
 void pud_init(void *addr);
+void pmd_init(void *addr);
+void kernel_pte_init(void *addr);
 pgd_t *vmemmap_pgd_populate(unsigned long addr, int node);
 p4d_t *vmemmap_p4d_populate(pgd_t *pgd, unsigned long addr, int node);
 pud_t *vmemmap_pud_populate(p4d_t *p4d, unsigned long addr, int node);

mm/kasan/init.c

@@ -106,6 +106,10 @@ static void __ref zero_pte_populate(pmd_t *pmd, unsigned long addr,
 	}
 }
 
+void __weak __meminit kernel_pte_init(void *addr)
+{
+}
+
 static int __ref zero_pmd_populate(pud_t *pud, unsigned long addr,
 				unsigned long end)
 {
@@ -126,8 +130,10 @@ static int __ref zero_pmd_populate(pud_t *pud, unsigned long addr,
 
 			if (slab_is_available())
 				p = pte_alloc_one_kernel(&init_mm);
-			else
+			else {
 				p = early_alloc(PAGE_SIZE, NUMA_NO_NODE);
+				kernel_pte_init(p);
+			}
 			if (!p)
 				return -ENOMEM;
 

mm/sparse-vmemmap.c

@@ -184,13 +184,18 @@ static void * __meminit vmemmap_alloc_block_zero(unsigned long size, int node)
 	return p;
 }
 
+void __weak __meminit kernel_pte_init(void *addr)
+{
+}
+
 pmd_t * __meminit vmemmap_pmd_populate(pud_t *pud, unsigned long addr, int node)
 {
 	pmd_t *pmd = pmd_offset(pud, addr);
 	if (pmd_none(*pmd)) {
 		void *p = vmemmap_alloc_block_zero(PAGE_SIZE, node);
 		if (!p)
 			return NULL;
+		kernel_pte_init(p);
 		pmd_populate_kernel(&init_mm, pmd, p);
 	}
 	return pmd;