x86, 64bit: Use a #PF handler to materialize early mappings on demand

Linear mode (CR0.PG = 0) is mutually exclusive with 64-bit mode; all
64-bit code has to use page tables.  This makes it awkward before we
have first set up properly all-covering page tables to access objects
that are outside the static kernel range.

So far we have dealt with that simply by mapping a fixed amount of
low memory, but that fails in at least two upcoming use cases:

1. We will support load and run kernel, struct boot_params, ramdisk,
   command line, etc. above the 4 GiB mark.
2. need to access ramdisk early to get microcode to update that as
   early possible.

We could use early_iomap to access them too, but it will make code to
messy and hard to be unified with 32 bit.

Hence, set up a #PF table and use a fixed number of buffers to set up
page tables on demand.  If the buffers fill up then we simply flush
them and start over.  These buffers are all in __initdata, so it does
not increase RAM usage at runtime.

Thus, with the help of the #PF handler, we can set the final kernel
mapping from blank, and switch to init_level4_pgt later.

During the switchover in head_64.S, before #PF handler is available,
we use three pages to handle kernel crossing 1G, 512G boundaries with
sharing page by playing games with page aliasing: the same page is
mapped twice in the higher-level tables with appropriate wraparound.
The kernel region itself will be properly mapped; other mappings may
be spurious.

early_make_pgtable is using kernel high mapping address to access pages
to set page table.

-v4: Add phys_base offset to make kexec happy, and add
	init_mapping_kernel()   - Yinghai
-v5: fix compiling with xen, and add back ident level3 and level2 for xen
     also move back init_level4_pgt from BSS to DATA again.
     because we have to clear it anyway.  - Yinghai
-v6: switch to init_level4_pgt in init_mem_mapping. - Yinghai
-v7: remove not needed clear_page for init_level4_page
     it is with fill 512,8,0 already in head_64.S  - Yinghai
-v8: we need to keep that handler alive until init_mem_mapping and don't
     let early_trap_init to trash that early #PF handler.
     So split early_trap_pf_init out and move it down. - Yinghai
-v9: switchover only cover kernel space instead of 1G so could avoid
     touch possible mem holes. - Yinghai
-v11: change far jmp back to far return to initial_code, that is needed
     to fix failure that is reported by Konrad on AMD systems.  - Yinghai
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Link: http://lkml.kernel.org/r/1359058816-7615-12-git-send-email-yinghai@kernel.orgSigned-off-by: H. Peter Anvin <hpa@linux.intel.com>

arch/x86/include/asm/pgtable_64_types.h

 #ifndef _ASM_X86_PGTABLE_64_DEFS_H
 #define _ASM_X86_PGTABLE_64_DEFS_H
 
+#include <asm/sparsemem.h>
+
 #ifndef __ASSEMBLY__
 #include <linux/types.h>
 
@@ -60,4 +62,6 @@ typedef struct { pteval_t pte; } pte_t;
 #define MODULES_END      _AC(0xffffffffff000000, UL)
 #define MODULES_LEN   (MODULES_END - MODULES_VADDR)
 
+#define EARLY_DYNAMIC_PAGE_TABLES	64
+
 #endif /* _ASM_X86_PGTABLE_64_DEFS_H */

arch/x86/include/asm/processor.h

@@ -731,6 +731,7 @@ extern void enable_sep_cpu(void);
 extern int sysenter_setup(void);
 
 extern void early_trap_init(void);
+void early_trap_pf_init(void);
 
 /* Defined in head.S */
 extern struct desc_ptr		early_gdt_descr;

arch/x86/kernel/head64.c

@@ -27,11 +27,73 @@
 #include <asm/bios_ebda.h>
 #include <asm/bootparam_utils.h>
 
-static void __init zap_identity_mappings(void)
+/*
+ * Manage page tables very early on.
+ */
+extern pgd_t early_level4_pgt[PTRS_PER_PGD];
+extern pmd_t early_dynamic_pgts[EARLY_DYNAMIC_PAGE_TABLES][PTRS_PER_PMD];
+static unsigned int __initdata next_early_pgt = 2;
+
+/* Wipe all early page tables except for the kernel symbol map */
+static void __init reset_early_page_tables(void)
 {
-	pgd_t *pgd = pgd_offset_k(0UL);
-	pgd_clear(pgd);
-	__flush_tlb_all();
+	unsigned long i;
+
+	for (i = 0; i < PTRS_PER_PGD-1; i++)
+		early_level4_pgt[i].pgd = 0;
+
+	next_early_pgt = 0;
+
+	write_cr3(__pa(early_level4_pgt));
+}
+
+/* Create a new PMD entry */
+int __init early_make_pgtable(unsigned long address)
+{
+	unsigned long physaddr = address - __PAGE_OFFSET;
+	unsigned long i;
+	pgdval_t pgd, *pgd_p;
+	pudval_t *pud_p;
+	pmdval_t pmd, *pmd_p;
+
+	/* Invalid address or early pgt is done ?  */
+	if (physaddr >= MAXMEM || read_cr3() != __pa(early_level4_pgt))
+		return -1;
+
+	i = (address >> PGDIR_SHIFT) & (PTRS_PER_PGD - 1);
+	pgd_p = &early_level4_pgt[i].pgd;
+	pgd = *pgd_p;
+
+	/*
+	 * The use of __START_KERNEL_map rather than __PAGE_OFFSET here is
+	 * critical -- __PAGE_OFFSET would point us back into the dynamic
+	 * range and we might end up looping forever...
+	 */
+	if (pgd && next_early_pgt < EARLY_DYNAMIC_PAGE_TABLES) {
+		pud_p = (pudval_t *)((pgd & PTE_PFN_MASK) + __START_KERNEL_map - phys_base);
+	} else {
+		if (next_early_pgt >= EARLY_DYNAMIC_PAGE_TABLES-1)
+			reset_early_page_tables();
+
+		pud_p = (pudval_t *)early_dynamic_pgts[next_early_pgt++];
+		for (i = 0; i < PTRS_PER_PUD; i++)
+			pud_p[i] = 0;
+
+		*pgd_p = (pgdval_t)pud_p - __START_KERNEL_map + phys_base + _KERNPG_TABLE;
+	}
+	i = (address >> PUD_SHIFT) & (PTRS_PER_PUD - 1);
+	pud_p += i;
+
+	pmd_p = (pmdval_t *)early_dynamic_pgts[next_early_pgt++];
+	pmd = (physaddr & PUD_MASK) + (__PAGE_KERNEL_LARGE & ~_PAGE_GLOBAL);
+	for (i = 0; i < PTRS_PER_PMD; i++) {
+		pmd_p[i] = pmd;
+		pmd += PMD_SIZE;
+	}
+
+	*pud_p = (pudval_t)pmd_p - __START_KERNEL_map + phys_base + _KERNPG_TABLE;
+
+	return 0;
 }
 
 /* Don't add a printk in there. printk relies on the PDA which is not initialized 
@@ -72,12 +134,13 @@ void __init x86_64_start_kernel(char * real_mode_data)
 				(__START_KERNEL & PGDIR_MASK)));
 	BUILD_BUG_ON(__fix_to_virt(__end_of_fixed_addresses) <= MODULES_END);
 
+	/* Kill off the identity-map trampoline */
+	reset_early_page_tables();
+
 	/* clear bss before set_intr_gate with early_idt_handler */
 	clear_bss();
 
-	/* Make NULL pointers segfault */
-	zap_identity_mappings();
-
+	/* XXX - this is wrong... we need to build page tables from scratch */
 	max_pfn_mapped = KERNEL_IMAGE_SIZE >> PAGE_SHIFT;
 
 	for (i = 0; i < NUM_EXCEPTION_VECTORS; i++) {
@@ -94,6 +157,10 @@ void __init x86_64_start_kernel(char * real_mode_data)
 	if (console_loglevel == 10)
 		early_printk("Kernel alive\n");
 
+	clear_page(init_level4_pgt);
+	/* set init_level4_pgt kernel high mapping*/
+	init_level4_pgt[511] = early_level4_pgt[511];
+
 	x86_64_start_reservations(real_mode_data);
 }
 

arch/x86/kernel/setup.c

@@ -1005,6 +1005,8 @@ void __init setup_arch(char **cmdline_p)
 
 	init_mem_mapping();
 
+	early_trap_pf_init();
+
 	setup_real_mode();
 
 	memblock.current_limit = get_max_mapped();

arch/x86/kernel/traps.c

@@ -688,10 +688,19 @@ void __init early_trap_init(void)
 	set_intr_gate_ist(X86_TRAP_DB, &debug, DEBUG_STACK);
 	/* int3 can be called from all */
 	set_system_intr_gate_ist(X86_TRAP_BP, &int3, DEBUG_STACK);
+#ifdef CONFIG_X86_32
 	set_intr_gate(X86_TRAP_PF, &page_fault);
+#endif
 	load_idt(&idt_descr);
 }
 
+void __init early_trap_pf_init(void)
+{
+#ifdef CONFIG_X86_64
+	set_intr_gate(X86_TRAP_PF, &page_fault);
+#endif
+}
+
 void __init trap_init(void)
 {
 	int i;

arch/x86/mm/init.c

@@ -446,9 +446,10 @@ void __init init_mem_mapping(void)
 	}
 #else
 	early_ioremap_page_table_range_init();
+#endif
+
 	load_cr3(swapper_pg_dir);
 	__flush_tlb_all();
-#endif
 
 	early_memtest(0, max_pfn_mapped << PAGE_SHIFT);
 }