[PATCH] x86: tighten kernel image page access rights

On x86-64, kernel memory freed after init can be entirely unmapped instead
of just getting 'poisoned' by overwriting with a debug pattern.

On i386 and x86-64 (under CONFIG_DEBUG_RODATA), kernel text and bug table
can also be write-protected.

Compared to the first version, this one prevents re-creating deleted
mappings in the kernel image range on x86-64, if those got removed
previously. This, together with the original changes, prevents temporarily
having inconsistent mappings when cacheability attributes are being
changed on such pages (e.g. from AGP code). While on i386 such duplicate
mappings don't exist, the same change is done there, too, both for
consistency and because checking pte_present() before using various other
pte_XXX functions is a requirement anyway. At once, i386 code gets
adjusted to use pte_huge() instead of open coding this.

AK: split out cpa() changes
Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Andi Kleen <ak@suse.de>

arch/i386/kernel/vmlinux.lds.S

@@ -61,8 +61,6 @@ SECTIONS
   	__stop___ex_table = .;
   }
 
-  RODATA
-
   BUG_TABLE
 
   . = ALIGN(4);
@@ -72,6 +70,8 @@ SECTIONS
   	__tracedata_end = .;
   }
 
+  RODATA
+
   /* writeable */
   . = ALIGN(4096);
   .data : AT(ADDR(.data) - LOAD_OFFSET) {	/* Data */

arch/i386/mm/init.c

@@ -22,6 +22,7 @@
 #include <linux/init.h>
 #include <linux/highmem.h>
 #include <linux/pagemap.h>
+#include <linux/pfn.h>
 #include <linux/poison.h>
 #include <linux/bootmem.h>
 #include <linux/slab.h>
@@ -751,13 +752,25 @@ static int noinline do_test_wp_bit(void)
 
 void mark_rodata_ro(void)
 {
-	unsigned long addr = (unsigned long)__start_rodata;
+	unsigned long start = PFN_ALIGN(_text);
+	unsigned long size = PFN_ALIGN(_etext) - start;
 
-	for (; addr < (unsigned long)__end_rodata; addr += PAGE_SIZE)
-		change_page_attr(virt_to_page(addr), 1, PAGE_KERNEL_RO);
+#ifdef CONFIG_HOTPLUG_CPU
+	/* It must still be possible to apply SMP alternatives. */
+	if (num_possible_cpus() <= 1)
+#endif
+	{
+		change_page_attr(virt_to_page(start),
+		                 size >> PAGE_SHIFT, PAGE_KERNEL_RX);
+		printk("Write protecting the kernel text: %luk\n", size >> 10);
+	}
 
-	printk("Write protecting the kernel read-only data: %uk\n",
-			(__end_rodata - __start_rodata) >> 10);
+	start += size;
+	size = (unsigned long)__end_rodata - start;
+	change_page_attr(virt_to_page(start),
+	                 size >> PAGE_SHIFT, PAGE_KERNEL_RO);
+	printk("Write protecting the kernel read-only data: %luk\n",
+	       size >> 10);
 
 	/*
 	 * change_page_attr() requires a global_flush_tlb() call after it.
@@ -781,7 +794,7 @@ void free_init_pages(char *what, unsigned long begin, unsigned long end)
 		__free_page(page);
 		totalram_pages++;
 	}
-	printk(KERN_INFO "Freeing %s: %ldk freed\n", what, (end - begin) >> 10);
+	printk(KERN_INFO "Freeing %s: %luk freed\n", what, (end - begin) >> 10);
 }
 
 void free_initmem(void)

arch/x86_64/kernel/head.S

@@ -280,7 +280,6 @@ early_idt_ripmsg:
 
 .balign PAGE_SIZE
 ENTRY(stext)
-ENTRY(_stext)
 
 #define NEXT_PAGE(name) \
 	.balign	PAGE_SIZE; \

arch/x86_64/kernel/vmlinux.lds.S

@@ -29,6 +29,7 @@ SECTIONS
   .text :  AT(ADDR(.text) - LOAD_OFFSET) {
 	/* First the code that has to be first for bootstrapping */
 	*(.bootstrap.text)
+	_stext = .;
 	/* Then all the functions that are "hot" in profiles, to group them
            onto the same hugetlb entry */
 	#include "functionlist"
@@ -50,10 +51,10 @@ SECTIONS
   __ex_table : AT(ADDR(__ex_table) - LOAD_OFFSET) { *(__ex_table) }
   __stop___ex_table = .;
 
-  RODATA
-
   BUG_TABLE
 
+  RODATA
+
   . = ALIGN(PAGE_SIZE);        /* Align data segment to page size boundary */
 				/* Data */
   .data : AT(ADDR(.data) - LOAD_OFFSET) {

arch/x86_64/mm/init.c

@@ -22,6 +22,7 @@
 #include <linux/bootmem.h>
 #include <linux/proc_fs.h>
 #include <linux/pci.h>
+#include <linux/pfn.h>
 #include <linux/poison.h>
 #include <linux/dma-mapping.h>
 #include <linux/module.h>
@@ -563,21 +564,23 @@ void free_init_pages(char *what, unsigned long begin, unsigned long end)
 	if (begin >= end)
 		return;
 
-	printk(KERN_INFO "Freeing %s: %ldk freed\n", what, (end - begin) >> 10);
+	printk(KERN_INFO "Freeing %s: %luk freed\n", what, (end - begin) >> 10);
 	for (addr = begin; addr < end; addr += PAGE_SIZE) {
 		struct page *page = pfn_to_page(addr >> PAGE_SHIFT);
 		ClearPageReserved(page);
 		init_page_count(page);
 		memset(page_address(page), POISON_FREE_INITMEM, PAGE_SIZE);
+		if (addr >= __START_KERNEL_map)
+			change_page_attr_addr(addr, 1, __pgprot(0));
 		__free_page(page);
 		totalram_pages++;
 	}
+	if (addr > __START_KERNEL_map)
+		global_flush_tlb();
 }
 
 void free_initmem(void)
 {
-	memset(__initdata_begin, POISON_FREE_INITDATA,
-		__initdata_end - __initdata_begin);
 	free_init_pages("unused kernel memory",
 			__pa_symbol(&__init_begin),
 			__pa_symbol(&__init_end));
@@ -587,14 +590,18 @@ void free_initmem(void)
 
 void mark_rodata_ro(void)
 {
-	unsigned long addr = (unsigned long)__va(__pa_symbol(&__start_rodata));
-	unsigned long end  = (unsigned long)__va(__pa_symbol(&__end_rodata));
+	unsigned long start = PFN_ALIGN(__va(__pa_symbol(&_stext))), size;
 
-	for (; addr < end; addr += PAGE_SIZE)
-		change_page_attr_addr(addr, 1, PAGE_KERNEL_RO);
+#ifdef CONFIG_HOTPLUG_CPU
+	/* It must still be possible to apply SMP alternatives. */
+	if (num_possible_cpus() > 1)
+		start = PFN_ALIGN(__va(__pa_symbol(&_etext)));
+#endif
+	size = (unsigned long)__va(__pa_symbol(&__end_rodata)) - start;
+	change_page_attr_addr(start, size >> PAGE_SHIFT, PAGE_KERNEL_RO);
 
-	printk ("Write protecting the kernel read-only data: %luk\n",
-			(__end_rodata - __start_rodata) >> 10);
+	printk(KERN_INFO "Write protecting the kernel read-only data: %luk\n",
+	       size >> 10);
 
 	/*
 	 * change_page_attr_addr() requires a global_flush_tlb() call after it.

include/linux/poison.h

@@ -26,9 +26,6 @@
 /********** arch/$ARCH/mm/init.c **********/
 #define POISON_FREE_INITMEM	0xcc
 
-/********** arch/x86_64/mm/init.c **********/
-#define	POISON_FREE_INITDATA	0xba
-
 /********** arch/ia64/hp/common/sba_iommu.c **********/
 /*
  * arch/ia64/hp/common/sba_iommu.c uses a 16-byte poison string with a