1 files changed, 86 insertions, 184 deletions

diff --git a/arch/s390/mm/vmem.c b/arch/s390/mm/vmem.c
index b26649233d12..d96587b84e81 100644
--- a/arch/s390/mm/vmem.c
+++ b/arch/s390/mm/vmem.c
@@ -4,8 +4,9 @@
  */
 
 #include <linux/memory_hotplug.h>
+#include <linux/bootmem_info.h>
+#include <linux/cpufeature.h>
 #include <linux/memblock.h>
-#include <linux/kasan.h>
 #include <linux/pfn.h>
 #include <linux/mm.h>
 #include <linux/init.h>
@@ -13,13 +14,18 @@
 #include <linux/hugetlb.h>
 #include <linux/slab.h>
 #include <linux/sort.h>
+#include <asm/page-states.h>
+#include <asm/abs_lowcore.h>
 #include <asm/cacheflush.h>
+#include <asm/maccess.h>
 #include <asm/nospec-branch.h>
+#include <asm/ctlreg.h>
 #include <asm/pgalloc.h>
 #include <asm/setup.h>
 #include <asm/tlbflush.h>
 #include <asm/sections.h>
 #include <asm/set_memory.h>
+#include <asm/physmem_info.h>
 
 static DEFINE_MUTEX(vmem_mutex);
 
@@ -32,13 +38,23 @@ static void __ref *vmem_alloc_pages(unsigned int order)
 	return memblock_alloc(size, size);
 }
 
-static void vmem_free_pages(unsigned long addr, int order)
+static void vmem_free_pages(unsigned long addr, int order, struct vmem_altmap *altmap)
 {
-	/* We don't expect boot memory to be removed ever. */
-	if (!slab_is_available() ||
-	    WARN_ON_ONCE(PageReserved(virt_to_page(addr))))
+	unsigned int nr_pages = 1 << order;
+	struct page *page;
+
+	if (altmap) {
+		vmem_altmap_free(altmap, 1 << order);
 		return;
-	free_pages(addr, order);
+	}
+	page = virt_to_page((void *)addr);
+	if (PageReserved(page)) {
+		/* allocated from memblock */
+		while (nr_pages--)
+			free_bootmem_page(page++);
+	} else {
+		free_pages(addr, order);
+	}
 }
 
 void *vmem_crst_alloc(unsigned long val)
@@ -46,32 +62,30 @@ void *vmem_crst_alloc(unsigned long val)
 	unsigned long *table;
 
 	table = vmem_alloc_pages(CRST_ALLOC_ORDER);
-	if (table)
-		crst_table_init(table, val);
+	if (!table)
+		return NULL;
+	crst_table_init(table, val);
+	__arch_set_page_dat(table, 1UL << CRST_ALLOC_ORDER);
 	return table;
 }
 
 pte_t __ref *vmem_pte_alloc(void)
 {
-	unsigned long size = PTRS_PER_PTE * sizeof(pte_t);
 	pte_t *pte;
 
 	if (slab_is_available())
-		pte = (pte_t *) page_table_alloc(&init_mm);
+		pte = (pte_t *)page_table_alloc(&init_mm);
 	else
-		pte = (pte_t *) memblock_alloc(size, size);
+		pte = (pte_t *)memblock_alloc(PAGE_SIZE, PAGE_SIZE);
 	if (!pte)
 		return NULL;
 	memset64((u64 *)pte, _PAGE_INVALID, PTRS_PER_PTE);
+	__arch_set_page_dat(pte, 1);
 	return pte;
 }
 
 static void vmem_pte_free(unsigned long *table)
 {
-	/* We don't expect boot memory to be removed ever. */
-	if (!slab_is_available() ||
-	    WARN_ON_ONCE(PageReserved(virt_to_page(table))))
-		return;
 	page_table_free(&init_mm, table);
 }
 
@@ -152,27 +166,25 @@ static bool vmemmap_unuse_sub_pmd(unsigned long start, unsigned long end)
 
 /* __ref: we'll only call vmemmap_alloc_block() via vmemmap_populate() */
 static int __ref modify_pte_table(pmd_t *pmd, unsigned long addr,
-				  unsigned long end, bool add, bool direct)
+				  unsigned long end, bool add, bool direct,
+				  struct vmem_altmap *altmap)
 {
 	unsigned long prot, pages = 0;
 	int ret = -ENOMEM;
 	pte_t *pte;
 
 	prot = pgprot_val(PAGE_KERNEL);
-	if (!MACHINE_HAS_NX)
-		prot &= ~_PAGE_NOEXEC;
-
 	pte = pte_offset_kernel(pmd, addr);
 	for (; addr < end; addr += PAGE_SIZE, pte++) {
 		if (!add) {
 			if (pte_none(*pte))
 				continue;
 			if (!direct)
-				vmem_free_pages((unsigned long) pfn_to_virt(pte_pfn(*pte)), 0);
+				vmem_free_pages((unsigned long)pfn_to_virt(pte_pfn(*pte)), get_order(PAGE_SIZE), altmap);
 			pte_clear(&init_mm, addr, pte);
 		} else if (pte_none(*pte)) {
 			if (!direct) {
-				void *new_page = vmemmap_alloc_block(PAGE_SIZE, NUMA_NO_NODE);
+				void *new_page = vmemmap_alloc_block_buf(PAGE_SIZE, NUMA_NO_NODE, altmap);
 
 				if (!new_page)
 					goto out;
@@ -209,7 +221,8 @@ static void try_free_pte_table(pmd_t *pmd, unsigned long start)
 
 /* __ref: we'll only call vmemmap_alloc_block() via vmemmap_populate() */
 static int __ref modify_pmd_table(pud_t *pud, unsigned long addr,
-				  unsigned long end, bool add, bool direct)
+				  unsigned long end, bool add, bool direct,
+				  struct vmem_altmap *altmap)
 {
 	unsigned long next, prot, pages = 0;
 	int ret = -ENOMEM;
@@ -217,24 +230,21 @@ static int __ref modify_pmd_table(pud_t *pud, unsigned long addr,
 	pte_t *pte;
 
 	prot = pgprot_val(SEGMENT_KERNEL);
-	if (!MACHINE_HAS_NX)
-		prot &= ~_SEGMENT_ENTRY_NOEXEC;
-
 	pmd = pmd_offset(pud, addr);
 	for (; addr < end; addr = next, pmd++) {
 		next = pmd_addr_end(addr, end);
 		if (!add) {
 			if (pmd_none(*pmd))
 				continue;
-			if (pmd_large(*pmd)) {
+			if (pmd_leaf(*pmd)) {
 				if (IS_ALIGNED(addr, PMD_SIZE) &&
 				    IS_ALIGNED(next, PMD_SIZE)) {
 					if (!direct)
-						vmem_free_pages(pmd_deref(*pmd), get_order(PMD_SIZE));
+						vmem_free_pages(pmd_deref(*pmd), get_order(PMD_SIZE), altmap);
 					pmd_clear(pmd);
 					pages++;
 				} else if (!direct && vmemmap_unuse_sub_pmd(addr, next)) {
-					vmem_free_pages(pmd_deref(*pmd), get_order(PMD_SIZE));
+					vmem_free_pages(pmd_deref(*pmd), get_order(PMD_SIZE), altmap);
 					pmd_clear(pmd);
 				}
 				continue;
@@ -242,12 +252,12 @@ static int __ref modify_pmd_table(pud_t *pud, unsigned long addr,
 		} else if (pmd_none(*pmd)) {
 			if (IS_ALIGNED(addr, PMD_SIZE) &&
 			    IS_ALIGNED(next, PMD_SIZE) &&
-			    MACHINE_HAS_EDAT1 && direct &&
+			    cpu_has_edat1() && direct &&
 			    !debug_pagealloc_enabled()) {
 				set_pmd(pmd, __pmd(__pa(addr) | prot));
 				pages++;
 				continue;
-			} else if (!direct && MACHINE_HAS_EDAT1) {
+			} else if (!direct && cpu_has_edat1()) {
 				void *new_page;
 
 				/*
@@ -257,7 +267,7 @@ static int __ref modify_pmd_table(pud_t *pud, unsigned long addr,
 				 * page tables since vmemmap_populate gets
 				 * called for each section separately.
 				 */
-				new_page = vmemmap_alloc_block(PMD_SIZE, NUMA_NO_NODE);
+				new_page = vmemmap_alloc_block_buf(PMD_SIZE, NUMA_NO_NODE, altmap);
 				if (new_page) {
 					set_pmd(pmd, __pmd(__pa(new_page) | prot));
 					if (!IS_ALIGNED(addr, PMD_SIZE) ||
@@ -271,12 +281,12 @@ static int __ref modify_pmd_table(pud_t *pud, unsigned long addr,
 			if (!pte)
 				goto out;
 			pmd_populate(&init_mm, pmd, pte);
-		} else if (pmd_large(*pmd)) {
+		} else if (pmd_leaf(*pmd)) {
 			if (!direct)
 				vmemmap_use_sub_pmd(addr, next);
 			continue;
 		}
-		ret = modify_pte_table(pmd, addr, next, add, direct);
+		ret = modify_pte_table(pmd, addr, next, add, direct, altmap);
 		if (ret)
 			goto out;
 		if (!add)
@@ -291,24 +301,19 @@ out:
 
 static void try_free_pmd_table(pud_t *pud, unsigned long start)
 {
-	const unsigned long end = start + PUD_SIZE;
 	pmd_t *pmd;
 	int i;
 
-	/* Don't mess with any tables not fully in 1:1 mapping & vmemmap area */
-	if (end > VMALLOC_START)
-		return;
-
 	pmd = pmd_offset(pud, start);
 	for (i = 0; i < PTRS_PER_PMD; i++, pmd++)
 		if (!pmd_none(*pmd))
 			return;
-	vmem_free_pages(pud_deref(*pud), CRST_ALLOC_ORDER);
+	vmem_free_pages(pud_deref(*pud), CRST_ALLOC_ORDER, NULL);
 	pud_clear(pud);
 }
 
 static int modify_pud_table(p4d_t *p4d, unsigned long addr, unsigned long end,
-			    bool add, bool direct)
+			    bool add, bool direct, struct vmem_altmap *altmap)
 {
 	unsigned long next, prot, pages = 0;
 	int ret = -ENOMEM;
@@ -316,15 +321,13 @@ static int modify_pud_table(p4d_t *p4d, unsigned long addr, unsigned long end,
 	pmd_t *pmd;
 
 	prot = pgprot_val(REGION3_KERNEL);
-	if (!MACHINE_HAS_NX)
-		prot &= ~_REGION_ENTRY_NOEXEC;
 	pud = pud_offset(p4d, addr);
 	for (; addr < end; addr = next, pud++) {
 		next = pud_addr_end(addr, end);
 		if (!add) {
 			if (pud_none(*pud))
 				continue;
-			if (pud_large(*pud)) {
+			if (pud_leaf(*pud)) {
 				if (IS_ALIGNED(addr, PUD_SIZE) &&
 				    IS_ALIGNED(next, PUD_SIZE)) {
 					pud_clear(pud);
@@ -335,7 +338,7 @@ static int modify_pud_table(p4d_t *p4d, unsigned long addr, unsigned long end,
 		} else if (pud_none(*pud)) {
 			if (IS_ALIGNED(addr, PUD_SIZE) &&
 			    IS_ALIGNED(next, PUD_SIZE) &&
-			    MACHINE_HAS_EDAT2 && direct &&
+			    cpu_has_edat2() && direct &&
 			    !debug_pagealloc_enabled()) {
 				set_pud(pud, __pud(__pa(addr) | prot));
 				pages++;
@@ -345,10 +348,10 @@ static int modify_pud_table(p4d_t *p4d, unsigned long addr, unsigned long end,
 			if (!pmd)
 				goto out;
 			pud_populate(&init_mm, pud, pmd);
-		} else if (pud_large(*pud)) {
+		} else if (pud_leaf(*pud)) {
 			continue;
 		}
-		ret = modify_pmd_table(pud, addr, next, add, direct);
+		ret = modify_pmd_table(pud, addr, next, add, direct, altmap);
 		if (ret)
 			goto out;
 		if (!add)
@@ -363,25 +366,20 @@ out:
 
 static void try_free_pud_table(p4d_t *p4d, unsigned long start)
 {
-	const unsigned long end = start + P4D_SIZE;
 	pud_t *pud;
 	int i;
 
-	/* Don't mess with any tables not fully in 1:1 mapping & vmemmap area */
-	if (end > VMALLOC_START)
-		return;
-
 	pud = pud_offset(p4d, start);
 	for (i = 0; i < PTRS_PER_PUD; i++, pud++) {
 		if (!pud_none(*pud))
 			return;
 	}
-	vmem_free_pages(p4d_deref(*p4d), CRST_ALLOC_ORDER);
+	vmem_free_pages(p4d_deref(*p4d), CRST_ALLOC_ORDER, NULL);
 	p4d_clear(p4d);
 }
 
 static int modify_p4d_table(pgd_t *pgd, unsigned long addr, unsigned long end,
-			    bool add, bool direct)
+			    bool add, bool direct, struct vmem_altmap *altmap)
 {
 	unsigned long next;
 	int ret = -ENOMEM;
@@ -400,7 +398,7 @@ static int modify_p4d_table(pgd_t *pgd, unsigned long addr, unsigned long end,
 				goto out;
 			p4d_populate(&init_mm, p4d, pud);
 		}
-		ret = modify_pud_table(p4d, addr, next, add, direct);
+		ret = modify_pud_table(p4d, addr, next, add, direct, altmap);
 		if (ret)
 			goto out;
 		if (!add)
@@ -413,25 +411,20 @@ out:
 
 static void try_free_p4d_table(pgd_t *pgd, unsigned long start)
 {
-	const unsigned long end = start + PGDIR_SIZE;
 	p4d_t *p4d;
 	int i;
 
-	/* Don't mess with any tables not fully in 1:1 mapping & vmemmap area */
-	if (end > VMALLOC_START)
-		return;
-
 	p4d = p4d_offset(pgd, start);
 	for (i = 0; i < PTRS_PER_P4D; i++, p4d++) {
 		if (!p4d_none(*p4d))
 			return;
 	}
-	vmem_free_pages(pgd_deref(*pgd), CRST_ALLOC_ORDER);
+	vmem_free_pages(pgd_deref(*pgd), CRST_ALLOC_ORDER, NULL);
 	pgd_clear(pgd);
 }
 
 static int modify_pagetable(unsigned long start, unsigned long end, bool add,
-			    bool direct)
+			    bool direct, struct vmem_altmap *altmap)
 {
 	unsigned long addr, next;
 	int ret = -ENOMEM;
@@ -440,6 +433,9 @@ static int modify_pagetable(unsigned long start, unsigned long end, bool add,
 
 	if (WARN_ON_ONCE(!PAGE_ALIGNED(start | end)))
 		return -EINVAL;
+	/* Don't mess with any tables not fully in 1:1 mapping & vmemmap area */
+	if (WARN_ON_ONCE(end > __abs_lowcore))
+		return -EINVAL;
 	for (addr = start; addr < end; addr = next) {
 		next = pgd_addr_end(addr, end);
 		pgd = pgd_offset_k(addr);
@@ -453,7 +449,7 @@ static int modify_pagetable(unsigned long start, unsigned long end, bool add,
 				goto out;
 			pgd_populate(&init_mm, pgd, p4d);
 		}
-		ret = modify_p4d_table(pgd, addr, next, add, direct);
+		ret = modify_p4d_table(pgd, addr, next, add, direct, altmap);
 		if (ret)
 			goto out;
 		if (!add)
@@ -466,14 +462,16 @@ out:
 	return ret;
 }
 
-static int add_pagetable(unsigned long start, unsigned long end, bool direct)
+static int add_pagetable(unsigned long start, unsigned long end, bool direct,
+			 struct vmem_altmap *altmap)
 {
-	return modify_pagetable(start, end, true, direct);
+	return modify_pagetable(start, end, true, direct, altmap);
 }
 
-static int remove_pagetable(unsigned long start, unsigned long end, bool direct)
+static int remove_pagetable(unsigned long start, unsigned long end, bool direct,
+			    struct vmem_altmap *altmap)
 {
-	return modify_pagetable(start, end, false, direct);
+	return modify_pagetable(start, end, false, direct, altmap);
 }
 
 /*
@@ -482,7 +480,7 @@ static int remove_pagetable(unsigned long start, unsigned long end, bool direct)
 static int vmem_add_range(unsigned long start, unsigned long size)
 {
 	start = (unsigned long)__va(start);
-	return add_pagetable(start, start + size, true);
+	return add_pagetable(start, start + size, true, NULL);
 }
 
 /*
@@ -491,7 +489,7 @@ static int vmem_add_range(unsigned long start, unsigned long size)
 static void vmem_remove_range(unsigned long start, unsigned long size)
 {
 	start = (unsigned long)__va(start);
-	remove_pagetable(start, start + size, true);
+	remove_pagetable(start, start + size, true, NULL);
 }
 
 /*
@@ -504,21 +502,25 @@ int __meminit vmemmap_populate(unsigned long start, unsigned long end, int node,
 
 	mutex_lock(&vmem_mutex);
 	/* We don't care about the node, just use NUMA_NO_NODE on allocations */
-	ret = add_pagetable(start, end, false);
+	ret = add_pagetable(start, end, false, altmap);
 	if (ret)
-		remove_pagetable(start, end, false);
+		remove_pagetable(start, end, false, altmap);
 	mutex_unlock(&vmem_mutex);
 	return ret;
 }
 
+#ifdef CONFIG_MEMORY_HOTPLUG
+
 void vmemmap_free(unsigned long start, unsigned long end,
 		  struct vmem_altmap *altmap)
 {
 	mutex_lock(&vmem_mutex);
-	remove_pagetable(start, end, false);
+	remove_pagetable(start, end, false, altmap);
 	mutex_unlock(&vmem_mutex);
 }
 
+#endif
+
 void vmem_remove_mapping(unsigned long start, unsigned long size)
 {
 	mutex_lock(&vmem_mutex);
@@ -531,7 +533,7 @@ struct range arch_get_mappable_range(void)
 	struct range mhp_range;
 
 	mhp_range.start = 0;
-	mhp_range.end =  VMEM_MAX_PHYS - 1;
+	mhp_range.end = max_mappable - 1;
 	return mhp_range;
 }
 
@@ -595,7 +597,7 @@ pte_t *vmem_get_alloc_pte(unsigned long addr, bool alloc)
 		if (!pmd)
 			goto out;
 		pud_populate(&init_mm, pud, pmd);
-	} else if (WARN_ON_ONCE(pud_large(*pud))) {
+	} else if (WARN_ON_ONCE(pud_leaf(*pud))) {
 		goto out;
 	}
 	pmd = pmd_offset(pud, addr);
@@ -606,7 +608,7 @@ pte_t *vmem_get_alloc_pte(unsigned long addr, bool alloc)
 		if (!pte)
 			goto out;
 		pmd_populate(&init_mm, pmd, pte);
-	} else if (WARN_ON_ONCE(pmd_large(*pmd))) {
+	} else if (WARN_ON_ONCE(pmd_leaf(*pmd))) {
 		goto out;
 	}
 	ptep = pte_offset_kernel(pmd, addr);
@@ -650,120 +652,20 @@ void vmem_unmap_4k_page(unsigned long addr)
 	mutex_unlock(&vmem_mutex);
 }
 
-static int __init memblock_region_cmp(const void *a, const void *b)
-{
-	const struct memblock_region *r1 = a;
-	const struct memblock_region *r2 = b;
-
-	if (r1->base < r2->base)
-		return -1;
-	if (r1->base > r2->base)
-		return 1;
-	return 0;
-}
-
-static void __init memblock_region_swap(void *a, void *b, int size)
-{
-	swap(*(struct memblock_region *)a, *(struct memblock_region *)b);
-}
-
-#ifdef CONFIG_KASAN
-#define __sha(x)	((unsigned long)kasan_mem_to_shadow((void *)x))
-
-static inline int set_memory_kasan(unsigned long start, unsigned long end)
-{
-	start = PAGE_ALIGN_DOWN(__sha(start));
-	end = PAGE_ALIGN(__sha(end));
-	return set_memory_rwnx(start, (end - start) >> PAGE_SHIFT);
-}
-#endif
-
-/*
- * map whole physical memory to virtual memory (identity mapping)
- * we reserve enough space in the vmalloc area for vmemmap to hotplug
- * additional memory segments.
- */
 void __init vmem_map_init(void)
 {
-	struct memblock_region memory_rwx_regions[] = {
-		{
-			.base	= 0,
-			.size	= sizeof(struct lowcore),
-			.flags	= MEMBLOCK_NONE,
-#ifdef CONFIG_NUMA
-			.nid	= NUMA_NO_NODE,
-#endif
-		},
-		{
-			.base	= __pa(_stext),
-			.size	= _etext - _stext,
-			.flags	= MEMBLOCK_NONE,
-#ifdef CONFIG_NUMA
-			.nid	= NUMA_NO_NODE,
-#endif
-		},
-		{
-			.base	= __pa(_sinittext),
-			.size	= _einittext - _sinittext,
-			.flags	= MEMBLOCK_NONE,
-#ifdef CONFIG_NUMA
-			.nid	= NUMA_NO_NODE,
-#endif
-		},
-		{
-			.base	= __stext_amode31,
-			.size	= __etext_amode31 - __stext_amode31,
-			.flags	= MEMBLOCK_NONE,
-#ifdef CONFIG_NUMA
-			.nid	= NUMA_NO_NODE,
-#endif
-		},
-	};
-	struct memblock_type memory_rwx = {
-		.regions	= memory_rwx_regions,
-		.cnt		= ARRAY_SIZE(memory_rwx_regions),
-		.max		= ARRAY_SIZE(memory_rwx_regions),
-	};
-	phys_addr_t base, end;
-	u64 i;
-
+	__set_memory_rox(_stext, _etext);
+	__set_memory_ro(_etext, __end_rodata);
+	__set_memory_rox(__stext_amode31, __etext_amode31);
 	/*
-	 * Set RW+NX attribute on all memory, except regions enumerated with
-	 * memory_rwx exclude type. These regions need different attributes,
-	 * which are enforced afterwards.
-	 *
-	 * __for_each_mem_range() iterate and exclude types should be sorted.
-	 * The relative location of _stext and _sinittext is hardcoded in the
-	 * linker script. However a location of __stext_amode31 and the kernel
-	 * image itself are chosen dynamically. Thus, sort the exclude type.
+	 * If the BEAR-enhancement facility is not installed the first
+	 * prefix page is used to return to the previous context with
+	 * an LPSWE instruction and therefore must be executable.
 	 */
-	sort(&memory_rwx_regions,
-	     ARRAY_SIZE(memory_rwx_regions), sizeof(memory_rwx_regions[0]),
-	     memblock_region_cmp, memblock_region_swap);
-	__for_each_mem_range(i, &memblock.memory, &memory_rwx,
-			     NUMA_NO_NODE, MEMBLOCK_NONE, &base, &end, NULL) {
-		set_memory_rwnx((unsigned long)__va(base),
-				(end - base) >> PAGE_SHIFT);
-	}
-
-#ifdef CONFIG_KASAN
-	for_each_mem_range(i, &base, &end)
-		set_memory_kasan(base, end);
-#endif
-	set_memory_rox((unsigned long)_stext,
-		       (unsigned long)(_etext - _stext) >> PAGE_SHIFT);
-	set_memory_ro((unsigned long)_etext,
-		      (unsigned long)(__end_rodata - _etext) >> PAGE_SHIFT);
-	set_memory_rox((unsigned long)_sinittext,
-		       (unsigned long)(_einittext - _sinittext) >> PAGE_SHIFT);
-	set_memory_rox(__stext_amode31,
-		       (__etext_amode31 - __stext_amode31) >> PAGE_SHIFT);
-
-	/* lowcore must be executable for LPSWE */
-	if (static_key_enabled(&cpu_has_bear))
-		set_memory_nx(0, 1);
-	set_memory_nx(PAGE_SIZE, 1);
-
+	if (!cpu_has_bear())
+		set_memory_x(0, 1);
+	if (debug_pagealloc_enabled())
+		__set_memory_4k(__va(0), absolute_pointer(__va(0)) + ident_map_size);
 	pr_info("Write protected kernel read-only data: %luk\n",
 		(unsigned long)(__end_rodata - _stext) >> 10);
 }