x86/cpu_entry_area: Move it out of the fixmap

Put the cpu_entry_area into a separate P4D entry. The fixmap gets too big
and 0-day already hit a case where the fixmap PTEs were cleared by
cleanup_highmap().

Aside of that the fixmap API is a pain as it's all backwards.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Juergen Gross <jgross@suse.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>

5 files changed, 78 insertions, 30 deletions

diff --git a/arch/x86/mm/cpu_entry_area.c b/arch/x86/mm/cpu_entry_area.c
index 235ff9cfaaf4..21e8b595cbb1 100644
--- a/arch/x86/mm/cpu_entry_area.c
+++ b/arch/x86/mm/cpu_entry_area.c
@@ -15,11 +15,27 @@ static DEFINE_PER_CPU_PAGE_ALIGNED(char, exception_stacks
 	[(N_EXCEPTION_STACKS - 1) * EXCEPTION_STKSZ + DEBUG_STKSZ]);
 #endif
 
+struct cpu_entry_area *get_cpu_entry_area(int cpu)
+{
+	unsigned long va = CPU_ENTRY_AREA_PER_CPU + cpu * CPU_ENTRY_AREA_SIZE;
+	BUILD_BUG_ON(sizeof(struct cpu_entry_area) % PAGE_SIZE != 0);
+
+	return (struct cpu_entry_area *) va;
+}
+EXPORT_SYMBOL(get_cpu_entry_area);
+
+void cea_set_pte(void *cea_vaddr, phys_addr_t pa, pgprot_t flags)
+{
+	unsigned long va = (unsigned long) cea_vaddr;
+
+	set_pte_vaddr(va, pfn_pte(pa >> PAGE_SHIFT, flags));
+}
+
 static void __init
-set_percpu_fixmap_pages(int idx, void *ptr, int pages, pgprot_t prot)
+cea_map_percpu_pages(void *cea_vaddr, void *ptr, int pages, pgprot_t prot)
 {
-	for ( ; pages; pages--, idx--, ptr += PAGE_SIZE)
-		__set_fixmap(idx, per_cpu_ptr_to_phys(ptr), prot);
+	for ( ; pages; pages--, cea_vaddr+= PAGE_SIZE, ptr += PAGE_SIZE)
+		cea_set_pte(cea_vaddr, per_cpu_ptr_to_phys(ptr), prot);
 }
 
 /* Setup the fixmap mappings only once per-processor */
@@ -47,10 +63,12 @@ static void __init setup_cpu_entry_area(int cpu)
 	pgprot_t tss_prot = PAGE_KERNEL;
 #endif
 
-	__set_fixmap(get_cpu_entry_area_index(cpu, gdt), get_cpu_gdt_paddr(cpu), gdt_prot);
-	set_percpu_fixmap_pages(get_cpu_entry_area_index(cpu, entry_stack_page),
-				per_cpu_ptr(&entry_stack_storage, cpu), 1,
-				PAGE_KERNEL);
+	cea_set_pte(&get_cpu_entry_area(cpu)->gdt, get_cpu_gdt_paddr(cpu),
+		    gdt_prot);
+
+	cea_map_percpu_pages(&get_cpu_entry_area(cpu)->entry_stack_page,
+			     per_cpu_ptr(&entry_stack_storage, cpu), 1,
+			     PAGE_KERNEL);
 
 	/*
 	 * The Intel SDM says (Volume 3, 7.2.1):
@@ -72,10 +90,9 @@ static void __init setup_cpu_entry_area(int cpu)
 	BUILD_BUG_ON((offsetof(struct tss_struct, x86_tss) ^
 		      offsetofend(struct tss_struct, x86_tss)) & PAGE_MASK);
 	BUILD_BUG_ON(sizeof(struct tss_struct) % PAGE_SIZE != 0);
-	set_percpu_fixmap_pages(get_cpu_entry_area_index(cpu, tss),
-				&per_cpu(cpu_tss_rw, cpu),
-				sizeof(struct tss_struct) / PAGE_SIZE,
-				tss_prot);
+	cea_map_percpu_pages(&get_cpu_entry_area(cpu)->tss,
+			     &per_cpu(cpu_tss_rw, cpu),
+			     sizeof(struct tss_struct) / PAGE_SIZE, tss_prot);
 
 #ifdef CONFIG_X86_32
 	per_cpu(cpu_entry_area, cpu) = get_cpu_entry_area(cpu);
@@ -85,20 +102,37 @@ static void __init setup_cpu_entry_area(int cpu)
 	BUILD_BUG_ON(sizeof(exception_stacks) % PAGE_SIZE != 0);
 	BUILD_BUG_ON(sizeof(exception_stacks) !=
 		     sizeof(((struct cpu_entry_area *)0)->exception_stacks));
-	set_percpu_fixmap_pages(get_cpu_entry_area_index(cpu, exception_stacks),
-				&per_cpu(exception_stacks, cpu),
-				sizeof(exception_stacks) / PAGE_SIZE,
-				PAGE_KERNEL);
+	cea_map_percpu_pages(&get_cpu_entry_area(cpu)->exception_stacks,
+			     &per_cpu(exception_stacks, cpu),
+			     sizeof(exception_stacks) / PAGE_SIZE, PAGE_KERNEL);
 
-	__set_fixmap(get_cpu_entry_area_index(cpu, entry_trampoline),
+	cea_set_pte(&get_cpu_entry_area(cpu)->entry_trampoline,
 		     __pa_symbol(_entry_trampoline), PAGE_KERNEL_RX);
 #endif
 }
 
+static __init void setup_cpu_entry_area_ptes(void)
+{
+#ifdef CONFIG_X86_32
+	unsigned long start, end;
+
+	BUILD_BUG_ON(CPU_ENTRY_AREA_PAGES * PAGE_SIZE < CPU_ENTRY_AREA_MAP_SIZE);
+	BUG_ON(CPU_ENTRY_AREA_BASE & ~PMD_MASK);
+
+	start = CPU_ENTRY_AREA_BASE;
+	end = start + CPU_ENTRY_AREA_MAP_SIZE;
+
+	for (; start < end; start += PMD_SIZE)
+		populate_extra_pte(start);
+#endif
+}
+
 void __init setup_cpu_entry_areas(void)
 {
 	unsigned int cpu;
 
+	setup_cpu_entry_area_ptes();
+
 	for_each_possible_cpu(cpu)
 		setup_cpu_entry_area(cpu);
 }
diff --git a/arch/x86/mm/dump_pagetables.c b/arch/x86/mm/dump_pagetables.c
index fdf09d8f98da..43dedbfb7257 100644
--- a/arch/x86/mm/dump_pagetables.c
+++ b/arch/x86/mm/dump_pagetables.c
@@ -58,6 +58,7 @@ enum address_markers_idx {
 	KASAN_SHADOW_START_NR,
 	KASAN_SHADOW_END_NR,
 #endif
+	CPU_ENTRY_AREA_NR,
 #ifdef CONFIG_X86_ESPFIX64
 	ESPFIX_START_NR,
 #endif
@@ -81,6 +82,7 @@ static struct addr_marker address_markers[] = {
 	[KASAN_SHADOW_START_NR]	= { KASAN_SHADOW_START,	"KASAN shadow" },
 	[KASAN_SHADOW_END_NR]	= { KASAN_SHADOW_END,	"KASAN shadow end" },
 #endif
+	[CPU_ENTRY_AREA_NR]	= { CPU_ENTRY_AREA_BASE,"CPU entry Area" },
 #ifdef CONFIG_X86_ESPFIX64
 	[ESPFIX_START_NR]	= { ESPFIX_BASE_ADDR,	"ESPfix Area", 16 },
 #endif
@@ -104,6 +106,7 @@ enum address_markers_idx {
 #ifdef CONFIG_HIGHMEM
 	PKMAP_BASE_NR,
 #endif
+	CPU_ENTRY_AREA_NR,
 	FIXADDR_START_NR,
 	END_OF_SPACE_NR,
 };
@@ -116,6 +119,7 @@ static struct addr_marker address_markers[] = {
 #ifdef CONFIG_HIGHMEM
 	[PKMAP_BASE_NR]		= { 0UL,		"Persistent kmap() Area" },
 #endif
+	[CPU_ENTRY_AREA_NR]	= { 0UL,		"CPU entry area" },
 	[FIXADDR_START_NR]	= { 0UL,		"Fixmap area" },
 	[END_OF_SPACE_NR]	= { -1,			NULL }
 };
@@ -541,8 +545,8 @@ static int __init pt_dump_init(void)
 	address_markers[PKMAP_BASE_NR].start_address = PKMAP_BASE;
 # endif
 	address_markers[FIXADDR_START_NR].start_address = FIXADDR_START;
+	address_markers[CPU_ENTRY_AREA_NR].start_address = CPU_ENTRY_AREA_BASE;
 #endif
-
 	return 0;
 }
 __initcall(pt_dump_init);
diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c
index 8a64a6f2848d..135c9a7898c7 100644
--- a/arch/x86/mm/init_32.c
+++ b/arch/x86/mm/init_32.c
@@ -50,6 +50,7 @@
 #include <asm/setup.h>
 #include <asm/set_memory.h>
 #include <asm/page_types.h>
+#include <asm/cpu_entry_area.h>
 #include <asm/init.h>
 
 #include "mm_internal.h"
@@ -766,6 +767,7 @@ void __init mem_init(void)
 	mem_init_print_info(NULL);
 	printk(KERN_INFO "virtual kernel memory layout:\n"
 		"    fixmap  : 0x%08lx - 0x%08lx   (%4ld kB)\n"
+		"  cpu_entry : 0x%08lx - 0x%08lx   (%4ld kB)\n"
 #ifdef CONFIG_HIGHMEM
 		"    pkmap   : 0x%08lx - 0x%08lx   (%4ld kB)\n"
 #endif
@@ -777,6 +779,10 @@ void __init mem_init(void)
 		FIXADDR_START, FIXADDR_TOP,
 		(FIXADDR_TOP - FIXADDR_START) >> 10,
 
+		CPU_ENTRY_AREA_BASE,
+		CPU_ENTRY_AREA_BASE + CPU_ENTRY_AREA_MAP_SIZE,
+		CPU_ENTRY_AREA_MAP_SIZE >> 10,
+
 #ifdef CONFIG_HIGHMEM
 		PKMAP_BASE, PKMAP_BASE+LAST_PKMAP*PAGE_SIZE,
 		(LAST_PKMAP*PAGE_SIZE) >> 10,
diff --git a/arch/x86/mm/kasan_init_64.c b/arch/x86/mm/kasan_init_64.c
index 9ec70d780f1f..47388f0c0e59 100644
--- a/arch/x86/mm/kasan_init_64.c
+++ b/arch/x86/mm/kasan_init_64.c
@@ -15,6 +15,7 @@
 #include <asm/tlbflush.h>
 #include <asm/sections.h>
 #include <asm/pgtable.h>
+#include <asm/cpu_entry_area.h>
 
 extern struct range pfn_mapped[E820_MAX_ENTRIES];
 
@@ -322,31 +323,33 @@ void __init kasan_init(void)
 		map_range(&pfn_mapped[i]);
 	}
 
-	kasan_populate_zero_shadow(
-		kasan_mem_to_shadow((void *)PAGE_OFFSET + MAXMEM),
-		kasan_mem_to_shadow((void *)__START_KERNEL_map));
-
-	kasan_populate_shadow((unsigned long)kasan_mem_to_shadow(_stext),
-			      (unsigned long)kasan_mem_to_shadow(_end),
-			      early_pfn_to_nid(__pa(_stext)));
-
-	shadow_cpu_entry_begin = (void *)__fix_to_virt(FIX_CPU_ENTRY_AREA_BOTTOM);
+	shadow_cpu_entry_begin = (void *)CPU_ENTRY_AREA_BASE;
 	shadow_cpu_entry_begin = kasan_mem_to_shadow(shadow_cpu_entry_begin);
 	shadow_cpu_entry_begin = (void *)round_down((unsigned long)shadow_cpu_entry_begin,
 						PAGE_SIZE);
 
-	shadow_cpu_entry_end = (void *)(__fix_to_virt(FIX_CPU_ENTRY_AREA_TOP) + PAGE_SIZE);
+	shadow_cpu_entry_end = (void *)(CPU_ENTRY_AREA_BASE +
+					CPU_ENTRY_AREA_MAP_SIZE);
 	shadow_cpu_entry_end = kasan_mem_to_shadow(shadow_cpu_entry_end);
 	shadow_cpu_entry_end = (void *)round_up((unsigned long)shadow_cpu_entry_end,
 					PAGE_SIZE);
 
-	kasan_populate_zero_shadow(kasan_mem_to_shadow((void *)MODULES_END),
-				   shadow_cpu_entry_begin);
+	kasan_populate_zero_shadow(
+		kasan_mem_to_shadow((void *)PAGE_OFFSET + MAXMEM),
+		shadow_cpu_entry_begin);
 
 	kasan_populate_shadow((unsigned long)shadow_cpu_entry_begin,
 			      (unsigned long)shadow_cpu_entry_end, 0);
 
-	kasan_populate_zero_shadow(shadow_cpu_entry_end, (void *)KASAN_SHADOW_END);
+	kasan_populate_zero_shadow(shadow_cpu_entry_end,
+				kasan_mem_to_shadow((void *)__START_KERNEL_map));
+
+	kasan_populate_shadow((unsigned long)kasan_mem_to_shadow(_stext),
+			      (unsigned long)kasan_mem_to_shadow(_end),
+			      early_pfn_to_nid(__pa(_stext)));
+
+	kasan_populate_zero_shadow(kasan_mem_to_shadow((void *)MODULES_END),
+				(void *)KASAN_SHADOW_END);
 
 	load_cr3(init_top_pgt);
 	__flush_tlb_all();
diff --git a/arch/x86/mm/pgtable_32.c b/arch/x86/mm/pgtable_32.c
index 6b9bf023a700..c3c5274410a9 100644
--- a/arch/x86/mm/pgtable_32.c
+++ b/arch/x86/mm/pgtable_32.c
@@ -10,6 +10,7 @@
 #include <linux/pagemap.h>
 #include <linux/spinlock.h>
 
+#include <asm/cpu_entry_area.h>
 #include <asm/pgtable.h>
 #include <asm/pgalloc.h>
 #include <asm/fixmap.h>