Revert "[PATCH] x86: __pa and __pa_symbol address space separation"

This was broken.  It adds complexity, for no good reason.  Rather than
separate __pa() and __pa_symbol(), we should deprecate __pa_symbol(),
and preferably __pa() too - and just use "virt_to_phys()" instead, which
is more readable and has nicer semantics.

However, right now, just undo the separation, and make __pa_symbol() be
the exact same as __pa().  That fixes the bugs this patch introduced,
and we can do the fairly obvious cleanups later.

Do the new __phys_addr() function (which is now the actual workhorse for
the unified __pa()/__pa_symbol()) as a real external function, that way
all the potential issues with compile/link-time optimizations of
constant symbol addresses go away, and we can also, if we choose to, add
more sanity-checking of the argument.

Cc: Eric W. Biederman <ebiederm@xmission.com>
Cc: Vivek Goyal <vgoyal@in.ibm.com>
Cc: Andi Kleen <ak@suse.de>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

10 files changed, 66 insertions, 58 deletions

diff --git a/arch/i386/kernel/alternative.c b/arch/i386/kernel/alternative.c
index e5cec6685cc..d8cda14fff8 100644
--- a/arch/i386/kernel/alternative.c
+++ b/arch/i386/kernel/alternative.c
@@ -390,8 +390,8 @@ void __init alternative_instructions(void)
 						_text, _etext);
 		}
 		free_init_pages("SMP alternatives",
-				__pa_symbol(&__smp_locks),
-				__pa_symbol(&__smp_locks_end));
+				(unsigned long)__smp_locks,
+				(unsigned long)__smp_locks_end);
 	} else {
 		alternatives_smp_module_add(NULL, "core kernel",
 					    __smp_locks, __smp_locks_end,
diff --git a/arch/i386/mm/init.c b/arch/i386/mm/init.c
index dbe16f63a56..1a7197e89eb 100644
--- a/arch/i386/mm/init.c
+++ b/arch/i386/mm/init.c
@@ -843,11 +843,10 @@ void free_init_pages(char *what, unsigned long begin, unsigned long end)
 	unsigned long addr;
 
 	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);
-		__free_page(page);
+		ClearPageReserved(virt_to_page(addr));
+		init_page_count(virt_to_page(addr));
+		memset((void *)addr, POISON_FREE_INITMEM, PAGE_SIZE);
+		free_page(addr);
 		totalram_pages++;
 	}
 	printk(KERN_INFO "Freeing %s: %luk freed\n", what, (end - begin) >> 10);
@@ -856,14 +855,14 @@ void free_init_pages(char *what, unsigned long begin, unsigned long end)
 void free_initmem(void)
 {
 	free_init_pages("unused kernel memory",
-			__pa_symbol(&__init_begin),
-			__pa_symbol(&__init_end));
+			(unsigned long)(&__init_begin),
+			(unsigned long)(&__init_end));
 }
 
 #ifdef CONFIG_BLK_DEV_INITRD
 void free_initrd_mem(unsigned long start, unsigned long end)
 {
-	free_init_pages("initrd memory", __pa(start), __pa(end));
+	free_init_pages("initrd memory", start, end);
 }
 #endif
 
diff --git a/arch/x86_64/kernel/machine_kexec.c b/arch/x86_64/kernel/machine_kexec.c
index a8bb33c1a8f..c3a55470367 100644
--- a/arch/x86_64/kernel/machine_kexec.c
+++ b/arch/x86_64/kernel/machine_kexec.c
@@ -189,21 +189,21 @@ NORET_TYPE void machine_kexec(struct kimage *image)
 	control_page = page_address(image->control_code_page) + PAGE_SIZE;
 	memcpy(control_page, relocate_kernel, PAGE_SIZE);
 
-	page_list[PA_CONTROL_PAGE] = __pa(control_page);
+	page_list[PA_CONTROL_PAGE] = virt_to_phys(control_page);
 	page_list[VA_CONTROL_PAGE] = (unsigned long)relocate_kernel;
-	page_list[PA_PGD] = __pa_symbol(&kexec_pgd);
+	page_list[PA_PGD] = virt_to_phys(&kexec_pgd);
 	page_list[VA_PGD] = (unsigned long)kexec_pgd;
-	page_list[PA_PUD_0] = __pa_symbol(&kexec_pud0);
+	page_list[PA_PUD_0] = virt_to_phys(&kexec_pud0);
 	page_list[VA_PUD_0] = (unsigned long)kexec_pud0;
-	page_list[PA_PMD_0] = __pa_symbol(&kexec_pmd0);
+	page_list[PA_PMD_0] = virt_to_phys(&kexec_pmd0);
 	page_list[VA_PMD_0] = (unsigned long)kexec_pmd0;
-	page_list[PA_PTE_0] = __pa_symbol(&kexec_pte0);
+	page_list[PA_PTE_0] = virt_to_phys(&kexec_pte0);
 	page_list[VA_PTE_0] = (unsigned long)kexec_pte0;
-	page_list[PA_PUD_1] = __pa_symbol(&kexec_pud1);
+	page_list[PA_PUD_1] = virt_to_phys(&kexec_pud1);
 	page_list[VA_PUD_1] = (unsigned long)kexec_pud1;
-	page_list[PA_PMD_1] = __pa_symbol(&kexec_pmd1);
+	page_list[PA_PMD_1] = virt_to_phys(&kexec_pmd1);
 	page_list[VA_PMD_1] = (unsigned long)kexec_pmd1;
-	page_list[PA_PTE_1] = __pa_symbol(&kexec_pte1);
+	page_list[PA_PTE_1] = virt_to_phys(&kexec_pte1);
 	page_list[VA_PTE_1] = (unsigned long)kexec_pte1;
 
 	page_list[PA_TABLE_PAGE] =
diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c
index db30b5bcef6..db51577bda3 100644
--- a/arch/x86_64/kernel/setup.c
+++ b/arch/x86_64/kernel/setup.c
@@ -245,12 +245,11 @@ void __init setup_arch(char **cmdline_p)
 	init_mm.end_code = (unsigned long) &_etext;
 	init_mm.end_data = (unsigned long) &_edata;
 	init_mm.brk = (unsigned long) &_end;
-	init_mm.pgd = __va(__pa_symbol(&init_level4_pgt));
 
-	code_resource.start = __pa_symbol(&_text);
-	code_resource.end = __pa_symbol(&_etext)-1;
-	data_resource.start = __pa_symbol(&_etext);
-	data_resource.end = __pa_symbol(&_edata)-1;
+	code_resource.start = virt_to_phys(&_text);
+	code_resource.end = virt_to_phys(&_etext)-1;
+	data_resource.start = virt_to_phys(&_etext);
+	data_resource.end = virt_to_phys(&_edata)-1;
 
 	early_identify_cpu(&boot_cpu_data);
 
diff --git a/arch/x86_64/kernel/smp.c b/arch/x86_64/kernel/smp.c
index 22abae4e9f3..bd1d123947c 100644
--- a/arch/x86_64/kernel/smp.c
+++ b/arch/x86_64/kernel/smp.c
@@ -76,7 +76,7 @@ static inline void leave_mm(int cpu)
 	if (read_pda(mmu_state) == TLBSTATE_OK)
 		BUG();
 	cpu_clear(cpu, read_pda(active_mm)->cpu_vm_mask);
-	load_cr3(init_mm.pgd);
+	load_cr3(swapper_pg_dir);
 }
 
 /*
diff --git a/arch/x86_64/mm/init.c b/arch/x86_64/mm/init.c
index 282b0a8f00a..c0822683b91 100644
--- a/arch/x86_64/mm/init.c
+++ b/arch/x86_64/mm/init.c
@@ -572,13 +572,13 @@ void free_init_pages(char *what, unsigned long begin, unsigned long end)
 
 	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);
+		ClearPageReserved(virt_to_page(addr));
+		init_page_count(virt_to_page(addr));
+		memset((void *)(addr & ~(PAGE_SIZE-1)),
+			POISON_FREE_INITMEM, PAGE_SIZE);
 		if (addr >= __START_KERNEL_map)
 			change_page_attr_addr(addr, 1, __pgprot(0));
-		__free_page(page);
+		free_page(addr);
 		totalram_pages++;
 	}
 	if (addr > __START_KERNEL_map)
@@ -588,26 +588,31 @@ void free_init_pages(char *what, unsigned long begin, unsigned long end)
 void free_initmem(void)
 {
 	free_init_pages("unused kernel memory",
-			__pa_symbol(&__init_begin),
-			__pa_symbol(&__init_end));
+			(unsigned long)(&__init_begin),
+			(unsigned long)(&__init_end));
 }
 
 #ifdef CONFIG_DEBUG_RODATA
 
 void mark_rodata_ro(void)
 {
-	unsigned long start = PFN_ALIGN(__va(__pa_symbol(&_stext))), size;
+	unsigned long start = (unsigned long)_stext, end;
 
 #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)));
+		start = (unsigned long)_etext;
 #endif
-	size = (unsigned long)__va(__pa_symbol(&__end_rodata)) - start;
-	change_page_attr_addr(start, size >> PAGE_SHIFT, PAGE_KERNEL_RO);
+	end = (unsigned long)__end_rodata;
+	start = (start + PAGE_SIZE - 1) & PAGE_MASK;
+	end &= PAGE_MASK;
+	if (end <= start)
+		return;
+
+	change_page_attr_addr(start, (end - start) >> PAGE_SHIFT, PAGE_KERNEL_RO);
 
 	printk(KERN_INFO "Write protecting the kernel read-only data: %luk\n",
-	       size >> 10);
+	       (end - start) >> 10);
 
 	/*
 	 * change_page_attr_addr() requires a global_flush_tlb() call after it.
@@ -622,7 +627,7 @@ void mark_rodata_ro(void)
 #ifdef CONFIG_BLK_DEV_INITRD
 void free_initrd_mem(unsigned long start, unsigned long end)
 {
-	free_init_pages("initrd memory", __pa(start), __pa(end));
+	free_init_pages("initrd memory", start, end);
 }
 #endif
 
diff --git a/arch/x86_64/mm/ioremap.c b/arch/x86_64/mm/ioremap.c
index c6e5e8d401a..6cac90aa503 100644
--- a/arch/x86_64/mm/ioremap.c
+++ b/arch/x86_64/mm/ioremap.c
@@ -13,12 +13,21 @@
 #include <linux/slab.h>
 #include <linux/module.h>
 #include <linux/io.h>
+
 #include <asm/pgalloc.h>
 #include <asm/fixmap.h>
 #include <asm/tlbflush.h>
 #include <asm/cacheflush.h>
 #include <asm/proto.h>
 
+unsigned long __phys_addr(unsigned long x)
+{
+	if (x >= __START_KERNEL_map)
+		return x - __START_KERNEL_map + phys_base;
+	return x - PAGE_OFFSET;
+}
+EXPORT_SYMBOL(__phys_addr);
+
 #define ISA_START_ADDRESS      0xa0000
 #define ISA_END_ADDRESS                0x100000
 
diff --git a/arch/x86_64/mm/pageattr.c b/arch/x86_64/mm/pageattr.c
index bf4aa8dd425..d653d0bf3df 100644
--- a/arch/x86_64/mm/pageattr.c
+++ b/arch/x86_64/mm/pageattr.c
@@ -51,6 +51,7 @@ static struct page *split_large_page(unsigned long address, pgprot_t prot,
 	SetPagePrivate(base);
 	page_private(base) = 0;
 
+	address = __pa(address);
 	addr = address & LARGE_PAGE_MASK; 
 	pbase = (pte_t *)page_address(base);
 	for (i = 0; i < PTRS_PER_PTE; i++, addr += PAGE_SIZE) {
@@ -100,12 +101,13 @@ static inline void save_page(struct page *fpage)
  * No more special protections in this 2/4MB area - revert to a
  * large page again. 
  */
-static void revert_page(unsigned long address, unsigned long pfn, pgprot_t ref_prot)
+static void revert_page(unsigned long address, pgprot_t ref_prot)
 {
 	pgd_t *pgd;
 	pud_t *pud;
 	pmd_t *pmd;
 	pte_t large_pte;
+	unsigned long pfn;
 
 	pgd = pgd_offset_k(address);
 	BUG_ON(pgd_none(*pgd));
@@ -113,6 +115,7 @@ static void revert_page(unsigned long address, unsigned long pfn, pgprot_t ref_p
 	BUG_ON(pud_none(*pud));
 	pmd = pmd_offset(pud, address);
 	BUG_ON(pmd_val(*pmd) & _PAGE_PSE);
+	pfn = (__pa(address) & LARGE_PAGE_MASK) >> PAGE_SHIFT;
 	large_pte = pfn_pte(pfn, ref_prot);
 	large_pte = pte_mkhuge(large_pte);
 	set_pte((pte_t *)pmd, large_pte);
@@ -138,8 +141,7 @@ __change_page_attr(unsigned long address, unsigned long pfn, pgprot_t prot,
  			 */
 			struct page *split;
 			ref_prot2 = pte_pgprot(pte_clrhuge(*kpte));
-			split = split_large_page(pfn << PAGE_SHIFT, prot,
-							ref_prot2);
+			split = split_large_page(address, prot, ref_prot2);
 			if (!split)
 				return -ENOMEM;
 			set_pte(kpte, mk_pte(split, ref_prot2));
@@ -158,7 +160,7 @@ __change_page_attr(unsigned long address, unsigned long pfn, pgprot_t prot,
 
 	if (page_private(kpte_page) == 0) {
 		save_page(kpte_page);
-		revert_page(address, pfn, ref_prot);
+		revert_page(address, ref_prot);
  	}
 	return 0;
 } 
@@ -178,7 +180,6 @@ __change_page_attr(unsigned long address, unsigned long pfn, pgprot_t prot,
  */
 int change_page_attr_addr(unsigned long address, int numpages, pgprot_t prot)
 {
-	unsigned long phys_base_pfn = __pa_symbol(__START_KERNEL_map) >> PAGE_SHIFT;
 	int err = 0, kernel_map = 0;
 	int i; 
 
@@ -199,11 +200,10 @@ int change_page_attr_addr(unsigned long address, int numpages, pgprot_t prot)
 		}
 		/* Handle kernel mapping too which aliases part of the
 		 * lowmem */
-		if ((pfn >= phys_base_pfn) &&
-			((pfn - phys_base_pfn) < (KERNEL_TEXT_SIZE >> PAGE_SHIFT))) {
+		if (__pa(address) < KERNEL_TEXT_SIZE) {
 			unsigned long addr2;
 			pgprot_t prot2;
-			addr2 = __START_KERNEL_map + ((pfn - phys_base_pfn) << PAGE_SHIFT);
+			addr2 = __START_KERNEL_map + __pa(address);
 			/* Make sure the kernel mappings stay executable */
 			prot2 = pte_pgprot(pte_mkexec(pfn_pte(0, prot)));
 			err = __change_page_attr(addr2, pfn, prot2,
diff --git a/include/asm-x86_64/page.h b/include/asm-x86_64/page.h
index b17fc16ec2e..4d04e247956 100644
--- a/include/asm-x86_64/page.h
+++ b/include/asm-x86_64/page.h
@@ -94,26 +94,22 @@ extern unsigned long phys_base;
 
 #define KERNEL_TEXT_SIZE  (40*1024*1024)
 #define KERNEL_TEXT_START 0xffffffff80000000
+#define PAGE_OFFSET		__PAGE_OFFSET
 
 #ifndef __ASSEMBLY__
 
 #include <asm/bug.h>
 
-#endif /* __ASSEMBLY__ */
+extern unsigned long __phys_addr(unsigned long);
 
-#define PAGE_OFFSET		__PAGE_OFFSET
+#endif /* __ASSEMBLY__ */
 
-/* Note: __pa(&symbol_visible_to_c) should be always replaced with __pa_symbol.
-   Otherwise you risk miscompilation. */
-#define __pa(x)			((unsigned long)(x) - PAGE_OFFSET)
-/* __pa_symbol should be used for C visible symbols.
-   This seems to be the official gcc blessed way to do such arithmetic. */ 
-#define __pa_symbol(x)		\
-	({unsigned long v;  \
-	  asm("" : "=r" (v) : "0" (x)); \
-	  ((v - __START_KERNEL_map) + phys_base); })
+#define __pa(x)		__phys_addr((unsigned long)(x))
+#define __pa_symbol(x)	__phys_addr((unsigned long)(x))
 
 #define __va(x)			((void *)((unsigned long)(x)+PAGE_OFFSET))
+#define __boot_va(x)		__va(x)
+#define __boot_pa(x)		__pa(x)
 #ifdef CONFIG_FLATMEM
 #define pfn_valid(pfn)		((pfn) < end_pfn)
 #endif
diff --git a/include/asm-x86_64/pgtable.h b/include/asm-x86_64/pgtable.h
index 599993f6ba8..da3390faaea 100644
--- a/include/asm-x86_64/pgtable.h
+++ b/include/asm-x86_64/pgtable.h
@@ -19,7 +19,7 @@ extern pmd_t level2_kernel_pgt[512];
 extern pgd_t init_level4_pgt[];
 extern unsigned long __supported_pte_mask;
 
-#define swapper_pg_dir ((pgd_t *)NULL)
+#define swapper_pg_dir init_level4_pgt
 
 extern void paging_init(void);
 extern void clear_kernel_mapping(unsigned long addr, unsigned long size);
@@ -29,7 +29,7 @@ extern void clear_kernel_mapping(unsigned long addr, unsigned long size);
  * for zero-mapped memory areas etc..
  */
 extern unsigned long empty_zero_page[PAGE_SIZE/sizeof(unsigned long)];
-#define ZERO_PAGE(vaddr) (pfn_to_page(__pa_symbol(&empty_zero_page) >> PAGE_SHIFT))
+#define ZERO_PAGE(vaddr) (virt_to_page(empty_zero_page))
 
 #endif /* !__ASSEMBLY__ */