x86: replace hard coded reservations in 64-bit early boot code with dynamic table

On x86-64 there are several memory allocations before bootmem. To avoid
them stomping on each other they used to be all hard coded in bad_area().
Replace this with an array that is filled as needed.

This cleans up the code considerably and allows to expand its use.

Cc: peterz@infradead.org
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

arch/x86/kernel/e820_64.c

@@ -47,56 +47,65 @@ unsigned long end_pfn_map;
  */
 static unsigned long __initdata end_user_pfn = MAXMEM>>PAGE_SHIFT;
 
-/* Check for some hardcoded bad areas that early boot is not allowed to touch */
-static inline int bad_addr(unsigned long *addrp, unsigned long size)
-{
-	unsigned long addr = *addrp, last = addr + size;
+/*
+ * Early reserved memory areas.
+ */
+#define MAX_EARLY_RES 20
+
+struct early_res {
+	unsigned long start, end;
+};
+static struct early_res early_res[MAX_EARLY_RES] __initdata = {
+	{ 0, PAGE_SIZE },			/* BIOS data page */
+#ifdef CONFIG_SMP
+	{ SMP_TRAMPOLINE_BASE, SMP_TRAMPOLINE_BASE + 2*PAGE_SIZE },
+#endif
+	{}
+};
 
-	/* various gunk below that needed for SMP startup */
-	if (addr < 0x8000) {
-		*addrp = PAGE_ALIGN(0x8000);
-		return 1;
+void __init reserve_early(unsigned long start, unsigned long end)
+{
+	int i;
+	struct early_res *r;
+	for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) {
+		r = &early_res[i];
+		if (end > r->start && start < r->end)
+			panic("Duplicated early reservation %lx-%lx\n",
+			      start, end);
 	}
+	if (i >= MAX_EARLY_RES)
+		panic("Too many early reservations");
+	r = &early_res[i];
+	r->start = start;
+	r->end = end;
+}
 
-	/* direct mapping tables of the kernel */
-	if (last >= table_start<<PAGE_SHIFT && addr < table_end<<PAGE_SHIFT) {
-		*addrp = PAGE_ALIGN(table_end << PAGE_SHIFT);
-		return 1;
+void __init early_res_to_bootmem(void)
+{
+	int i;
+	for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) {
+		struct early_res *r = &early_res[i];
+		reserve_bootmem_generic(r->start, r->end - r->start);
 	}
+}
 
-	/* initrd */
-#ifdef CONFIG_BLK_DEV_INITRD
-	if (boot_params.hdr.type_of_loader && boot_params.hdr.ramdisk_image) {
-		unsigned long ramdisk_image = boot_params.hdr.ramdisk_image;
-		unsigned long ramdisk_size  = boot_params.hdr.ramdisk_size;
-		unsigned long ramdisk_end   = ramdisk_image+ramdisk_size;
-
-		if (last >= ramdisk_image && addr < ramdisk_end) {
-			*addrp = PAGE_ALIGN(ramdisk_end);
-			return 1;
+/* Check for already reserved areas */
+static inline int bad_addr(unsigned long *addrp, unsigned long size)
+{
+	int i;
+	unsigned long addr = *addrp, last;
+	int changed = 0;
+again:
+	last = addr + size;
+	for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) {
+		struct early_res *r = &early_res[i];
+		if (last >= r->start && addr < r->end) {
+			*addrp = addr = r->end;
+			changed = 1;
+			goto again;
 		}
 	}
-#endif
-	/* kernel code */
-	if (last >= __pa_symbol(&_text) && addr < __pa_symbol(&_end)) {
-		*addrp = PAGE_ALIGN(__pa_symbol(&_end));
-		return 1;
-	}
-
-	if (last >= ebda_addr && addr < ebda_addr + ebda_size) {
-		*addrp = PAGE_ALIGN(ebda_addr + ebda_size);
-		return 1;
-	}
-
-#ifdef CONFIG_NUMA
-	/* NUMA memory to node map */
-	if (last >= nodemap_addr && addr < nodemap_addr + nodemap_size) {
-		*addrp = nodemap_addr + nodemap_size;
-		return 1;
-	}
-#endif
-	/* XXX ramdisk image here? */
-	return 0;
+	return changed;
 }
 
 /*

arch/x86/kernel/head64.c

@@ -21,6 +21,7 @@
 #include <asm/tlbflush.h>
 #include <asm/sections.h>
 #include <asm/kdebug.h>
+#include <asm/e820.h>
 
 static void __init zap_identity_mappings(void)
 {
@@ -48,6 +49,35 @@ static void __init copy_bootdata(char *real_mode_data)
 	}
 }
 
+#define EBDA_ADDR_POINTER 0x40E
+
+static __init void reserve_ebda(void)
+{
+	unsigned ebda_addr, ebda_size;
+
+	/*
+	 * there is a real-mode segmented pointer pointing to the
+	 * 4K EBDA area at 0x40E
+	 */
+	ebda_addr = *(unsigned short *)__va(EBDA_ADDR_POINTER);
+	ebda_addr <<= 4;
+
+	if (!ebda_addr)
+		return;
+
+	ebda_size = *(unsigned short *)__va(ebda_addr);
+
+	/* Round EBDA up to pages */
+	if (ebda_size == 0)
+		ebda_size = 1;
+	ebda_size <<= 10;
+	ebda_size = round_up(ebda_size + (ebda_addr & ~PAGE_MASK), PAGE_SIZE);
+	if (ebda_size > 64*1024)
+		ebda_size = 64*1024;
+
+	reserve_early(ebda_addr, ebda_addr + ebda_size);
+}
+
 void __init x86_64_start_kernel(char * real_mode_data)
 {
 	int i;
@@ -75,5 +105,23 @@ void __init x86_64_start_kernel(char * real_mode_data)
 	pda_init(0);
 	copy_bootdata(__va(real_mode_data));
 
+	reserve_early(__pa_symbol(&_text), __pa_symbol(&_end));
+
+	/* Reserve INITRD */
+	if (boot_params.hdr.type_of_loader && boot_params.hdr.ramdisk_image) {
+		unsigned long ramdisk_image = boot_params.hdr.ramdisk_image;
+		unsigned long ramdisk_size  = boot_params.hdr.ramdisk_size;
+		unsigned long ramdisk_end   = ramdisk_image + ramdisk_size;
+		reserve_early(ramdisk_image, ramdisk_end);
+	}
+
+	reserve_ebda();
+
+	/*
+	 * At this point everything still needed from the boot loader
+	 * or BIOS or kernel text should be early reserved or marked not
+	 * RAM in e820. All other memory is free game.
+	 */
+
 	start_kernel();
 }

arch/x86/kernel/setup_64.c

@@ -245,41 +245,6 @@ static inline void __init reserve_crashkernel(void)
 {}
 #endif
 
-#define EBDA_ADDR_POINTER 0x40E
-
-unsigned __initdata ebda_addr;
-unsigned __initdata ebda_size;
-
-static void __init discover_ebda(void)
-{
-	/*
-	 * there is a real-mode segmented pointer pointing to the
-	 * 4K EBDA area at 0x40E
-	 */
-	ebda_addr = *(unsigned short *)__va(EBDA_ADDR_POINTER);
-	/*
-	 * There can be some situations, like paravirtualized guests,
-	 * in which there is no available ebda information. In such
-	 * case, just skip it
-	 */
-	if (!ebda_addr) {
-		ebda_size = 0;
-		return;
-	}
-
-	ebda_addr <<= 4;
-
-	ebda_size = *(unsigned short *)__va(ebda_addr);
-
-	/* Round EBDA up to pages */
-	if (ebda_size == 0)
-		ebda_size = 1;
-	ebda_size <<= 10;
-	ebda_size = round_up(ebda_size + (ebda_addr & ~PAGE_MASK), PAGE_SIZE);
-	if (ebda_size > 64*1024)
-		ebda_size = 64*1024;
-}
-
 /* Overridden in paravirt.c if CONFIG_PARAVIRT */
 void __attribute__((weak)) __init memory_setup(void)
 {
@@ -349,8 +314,6 @@ void __init setup_arch(char **cmdline_p)
 
 	check_efer();
 
-	discover_ebda();
-
 	init_memory_mapping(0, (end_pfn_map << PAGE_SHIFT));
 	if (efi_enabled)
 		efi_init();
@@ -397,33 +360,7 @@ void __init setup_arch(char **cmdline_p)
 	contig_initmem_init(0, end_pfn);
 #endif
 
-	/* Reserve direct mapping */
-	reserve_bootmem_generic(table_start << PAGE_SHIFT,
-				(table_end - table_start) << PAGE_SHIFT);
-
-	/* reserve kernel */
-	reserve_bootmem_generic(__pa_symbol(&_text),
-				__pa_symbol(&_end) - __pa_symbol(&_text));
-
-	/*
-	 * reserve physical page 0 - it's a special BIOS page on many boxes,
-	 * enabling clean reboots, SMP operation, laptop functions.
-	 */
-	reserve_bootmem_generic(0, PAGE_SIZE);
-
-	/* reserve ebda region */
-	if (ebda_addr)
-		reserve_bootmem_generic(ebda_addr, ebda_size);
-#ifdef CONFIG_NUMA
-	/* reserve nodemap region */
-	if (nodemap_addr)
-		reserve_bootmem_generic(nodemap_addr, nodemap_size);
-#endif
-
-#ifdef CONFIG_SMP
-	/* Reserve SMP trampoline */
-	reserve_bootmem_generic(SMP_TRAMPOLINE_BASE, 2*PAGE_SIZE);
-#endif
+	early_res_to_bootmem();
 
 #ifdef CONFIG_ACPI_SLEEP
 	/*
@@ -453,6 +390,8 @@ void __init setup_arch(char **cmdline_p)
 			initrd_start = ramdisk_image + PAGE_OFFSET;
 			initrd_end = initrd_start+ramdisk_size;
 		} else {
+			/* Assumes everything on node 0 */
+			free_bootmem(ramdisk_image, ramdisk_size);
 			printk(KERN_ERR "initrd extends beyond end of memory "
 			       "(0x%08lx > 0x%08lx)\ndisabling initrd\n",
 			       ramdisk_end, end_of_mem);

arch/x86/mm/init_64.c

@@ -176,7 +176,8 @@ __set_fixmap (enum fixed_addresses idx, unsigned long phys, pgprot_t prot)
 	set_pte_phys(address, phys, prot);
 }
 
-unsigned long __meminitdata table_start, table_end;
+static unsigned long __initdata table_start;
+static unsigned long __meminitdata table_end;
 
 static __meminit void *alloc_low_page(unsigned long *phys)
 { 
@@ -387,6 +388,8 @@ void __init_refok init_memory_mapping(unsigned long start, unsigned long end)
 	if (!after_bootmem)
 		mmu_cr4_features = read_cr4();
 	__flush_tlb_all();
+
+	reserve_early(table_start << PAGE_SHIFT, table_end << PAGE_SHIFT);
 }
 
 #ifndef CONFIG_NUMA

arch/x86/mm/numa_64.c

@@ -102,6 +102,7 @@ static int __init allocate_cachealigned_memnodemap(void)
 	}
 	pad_addr = (nodemap_addr + pad) & ~pad;
 	memnodemap = phys_to_virt(pad_addr);
+	reserve_early(nodemap_addr, nodemap_addr + nodemap_size);
 
 	printk(KERN_DEBUG "NUMA: Allocated memnodemap from %lx - %lx\n",
 	       nodemap_addr, nodemap_addr + nodemap_size);

include/asm-x86/e820_64.h

@@ -41,8 +41,8 @@ extern void finish_e820_parsing(void);
 extern struct e820map e820;
 extern void update_e820(void);
 
-extern unsigned ebda_addr, ebda_size;
-extern unsigned long nodemap_addr, nodemap_size;
+extern void reserve_early(unsigned long start, unsigned long end);
+extern void early_res_to_bootmem(void);
 
 #endif/*!__ASSEMBLY__*/
 

include/asm-x86/proto.h

@@ -22,8 +22,6 @@ extern void syscall32_cpu_init(void);
 
 extern void check_efer(void);
 
-extern unsigned long table_start, table_end;
-
 extern int reboot_force;
 
 long do_arch_prctl(struct task_struct *task, int code, unsigned long addr);