sh: Enable shared page fault handler for _32/_64.

This moves the now generic _32 page fault handling code to a shared place
and adapts the _64 implementation to make use of it.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>

5 files changed, 93 insertions, 314 deletions

diff --git a/arch/sh/mm/Makefile b/arch/sh/mm/Makefile
index 2228c8cee4d..ba819108631 100644
--- a/arch/sh/mm/Makefile
+++ b/arch/sh/mm/Makefile
@@ -15,8 +15,8 @@ cacheops-$(CONFIG_CPU_SHX3)		+= cache-shx3.o
 obj-y			+= $(cacheops-y)
 
 mmu-y			:= nommu.o extable_32.o
-mmu-$(CONFIG_MMU)	:= extable_$(BITS).o fault_$(BITS).o gup.o \
-			   ioremap.o kmap.o pgtable.o tlbflush_$(BITS).o
+mmu-$(CONFIG_MMU)	:= extable_$(BITS).o fault.o gup.o ioremap.o kmap.o \
+			   pgtable.o tlbex_$(BITS).o tlbflush_$(BITS).o
 
 obj-y			+= $(mmu-y)
 
diff --git a/arch/sh/mm/fault_32.c b/arch/sh/mm/fault.c
index 39e291c6b35..16799f920f9 100644
--- a/arch/sh/mm/fault_32.c
+++ b/arch/sh/mm/fault.c
@@ -66,7 +66,7 @@ static void show_pte(struct mm_struct *mm, unsigned long addr)
 	printk(KERN_ALERT "pgd = %p\n", pgd);
 	pgd += pgd_index(addr);
 	printk(KERN_ALERT "[%08lx] *pgd=%0*Lx", addr,
-	       sizeof(*pgd) * 2, (u64)pgd_val(*pgd));
+	       (u32)(sizeof(*pgd) * 2), (u64)pgd_val(*pgd));
 
 	do {
 		pud_t *pud;
@@ -83,7 +83,7 @@ static void show_pte(struct mm_struct *mm, unsigned long addr)
 
 		pud = pud_offset(pgd, addr);
 		if (PTRS_PER_PUD != 1)
-			printk(", *pud=%0*Lx", sizeof(*pud) * 2,
+			printk(", *pud=%0*Lx", (u32)(sizeof(*pud) * 2),
 			       (u64)pud_val(*pud));
 
 		if (pud_none(*pud))
@@ -96,7 +96,7 @@ static void show_pte(struct mm_struct *mm, unsigned long addr)
 
 		pmd = pmd_offset(pud, addr);
 		if (PTRS_PER_PMD != 1)
-			printk(", *pmd=%0*Lx", sizeof(*pmd) * 2,
+			printk(", *pmd=%0*Lx", (u32)(sizeof(*pmd) * 2),
 			       (u64)pmd_val(*pmd));
 
 		if (pmd_none(*pmd))
@@ -112,7 +112,8 @@ static void show_pte(struct mm_struct *mm, unsigned long addr)
 			break;
 
 		pte = pte_offset_kernel(pmd, addr);
-		printk(", *pte=%0*Lx", sizeof(*pte) * 2, (u64)pte_val(*pte));
+		printk(", *pte=%0*Lx", (u32)(sizeof(*pte) * 2),
+		       (u64)pte_val(*pte));
 	} while (0);
 
 	printk("\n");
@@ -354,15 +355,20 @@ mm_fault_error(struct pt_regs *regs, unsigned long error_code,
 	return 1;
 }
 
-static inline int access_error(int write, struct vm_area_struct *vma)
+static inline int access_error(int error_code, struct vm_area_struct *vma)
 {
-	if (write) {
+	if (error_code & FAULT_CODE_WRITE) {
 		/* write, present and write, not present: */
 		if (unlikely(!(vma->vm_flags & VM_WRITE)))
 			return 1;
 		return 0;
 	}
 
+	/* ITLB miss on NX page */
+	if (unlikely((error_code & FAULT_CODE_ITLB) &&
+		     !(vma->vm_flags & VM_EXEC)))
+		return 1;
+
 	/* read, not present: */
 	if (unlikely(!(vma->vm_flags & (VM_READ | VM_EXEC | VM_WRITE))))
 		return 1;
@@ -500,65 +506,3 @@ good_area:
 
 	up_read(&mm->mmap_sem);
 }
-
-/*
- * Called with interrupts disabled.
- */
-asmlinkage int __kprobes
-handle_tlbmiss(struct pt_regs *regs, unsigned long error_code,
-	       unsigned long address)
-{
-	pgd_t *pgd;
-	pud_t *pud;
-	pmd_t *pmd;
-	pte_t *pte;
-	pte_t entry;
-
-	/*
-	 * We don't take page faults for P1, P2, and parts of P4, these
-	 * are always mapped, whether it be due to legacy behaviour in
-	 * 29-bit mode, or due to PMB configuration in 32-bit mode.
-	 */
-	if (address >= P3SEG && address < P3_ADDR_MAX) {
-		pgd = pgd_offset_k(address);
-	} else {
-		if (unlikely(address >= TASK_SIZE || !current->mm))
-			return 1;
-
-		pgd = pgd_offset(current->mm, address);
-	}
-
-	pud = pud_offset(pgd, address);
-	if (pud_none_or_clear_bad(pud))
-		return 1;
-	pmd = pmd_offset(pud, address);
-	if (pmd_none_or_clear_bad(pmd))
-		return 1;
-	pte = pte_offset_kernel(pmd, address);
-	entry = *pte;
-	if (unlikely(pte_none(entry) || pte_not_present(entry)))
-		return 1;
-	if (unlikely(error_code && !pte_write(entry)))
-		return 1;
-
-	if (error_code)
-		entry = pte_mkdirty(entry);
-	entry = pte_mkyoung(entry);
-
-	set_pte(pte, entry);
-
-#if defined(CONFIG_CPU_SH4) && !defined(CONFIG_SMP)
-	/*
-	 * SH-4 does not set MMUCR.RC to the corresponding TLB entry in
-	 * the case of an initial page write exception, so we need to
-	 * flush it in order to avoid potential TLB entry duplication.
-	 */
-	if (error_code == FAULT_CODE_INITIAL)
-		local_flush_tlb_one(get_asid(), address & PAGE_MASK);
-#endif
-
-	set_thread_fault_code(error_code);
-	update_mmu_cache(NULL, address, pte);
-
-	return 0;
-}
diff --git a/arch/sh/mm/tlbex_32.c b/arch/sh/mm/tlbex_32.c
new file mode 100644
index 00000000000..382262dc0c4
--- /dev/null
+++ b/arch/sh/mm/tlbex_32.c
@@ -0,0 +1,78 @@
+/*
+ * TLB miss handler for SH with an MMU.
+ *
+ *  Copyright (C) 1999  Niibe Yutaka
+ *  Copyright (C) 2003 - 2012  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/kernel.h>
+#include <linux/mm.h>
+#include <linux/kprobes.h>
+#include <linux/kdebug.h>
+#include <asm/mmu_context.h>
+#include <asm/thread_info.h>
+
+/*
+ * Called with interrupts disabled.
+ */
+asmlinkage int __kprobes
+handle_tlbmiss(struct pt_regs *regs, unsigned long error_code,
+	       unsigned long address)
+{
+	pgd_t *pgd;
+	pud_t *pud;
+	pmd_t *pmd;
+	pte_t *pte;
+	pte_t entry;
+
+	/*
+	 * We don't take page faults for P1, P2, and parts of P4, these
+	 * are always mapped, whether it be due to legacy behaviour in
+	 * 29-bit mode, or due to PMB configuration in 32-bit mode.
+	 */
+	if (address >= P3SEG && address < P3_ADDR_MAX) {
+		pgd = pgd_offset_k(address);
+	} else {
+		if (unlikely(address >= TASK_SIZE || !current->mm))
+			return 1;
+
+		pgd = pgd_offset(current->mm, address);
+	}
+
+	pud = pud_offset(pgd, address);
+	if (pud_none_or_clear_bad(pud))
+		return 1;
+	pmd = pmd_offset(pud, address);
+	if (pmd_none_or_clear_bad(pmd))
+		return 1;
+	pte = pte_offset_kernel(pmd, address);
+	entry = *pte;
+	if (unlikely(pte_none(entry) || pte_not_present(entry)))
+		return 1;
+	if (unlikely(error_code && !pte_write(entry)))
+		return 1;
+
+	if (error_code)
+		entry = pte_mkdirty(entry);
+	entry = pte_mkyoung(entry);
+
+	set_pte(pte, entry);
+
+#if defined(CONFIG_CPU_SH4) && !defined(CONFIG_SMP)
+	/*
+	 * SH-4 does not set MMUCR.RC to the corresponding TLB entry in
+	 * the case of an initial page write exception, so we need to
+	 * flush it in order to avoid potential TLB entry duplication.
+	 */
+	if (error_code == FAULT_CODE_INITIAL)
+		local_flush_tlb_one(get_asid(), address & PAGE_MASK);
+#endif
+
+	set_thread_fault_code(error_code);
+	update_mmu_cache(NULL, address, pte);
+
+	return 0;
+}
diff --git a/arch/sh/mm/fault_64.c b/arch/sh/mm/tlbex_64.c
index 33a921962d0..d15b9946650 100644
--- a/arch/sh/mm/fault_64.c
+++ b/arch/sh/mm/tlbex_64.c
@@ -246,7 +246,7 @@ asmlinkage int do_fast_page_fault(unsigned long long ssr_md,
 	tsk = current;
 	mm = tsk->mm;
 
-	if ((address >= VMALLOC_START && address < VMALLOC_END)) {
+	if (is_vmalloc_addr((void *)address)) {
 		if (ssr_md)
 			/*
 			 * Process-contexts can never have this address
diff --git a/arch/sh/mm/tlbflush_64.c b/arch/sh/mm/tlbflush_64.c
index 99c5833036b..908167bdfc0 100644
--- a/arch/sh/mm/tlbflush_64.c
+++ b/arch/sh/mm/tlbflush_64.c
@@ -28,249 +28,6 @@
 #include <asm/pgalloc.h>
 #include <asm/mmu_context.h>
 
-static pte_t *lookup_pte(struct mm_struct *mm, unsigned long address)
-{
-	pgd_t *dir;
-	pud_t *pud;
-	pmd_t *pmd;
-	pte_t *pte;
-	pte_t entry;
-
-	dir = pgd_offset(mm, address);
-	if (pgd_none(*dir))
-		return NULL;
-
-	pud = pud_offset(dir, address);
-	if (pud_none(*pud))
-		return NULL;
-
-	pmd = pmd_offset(pud, address);
-	if (pmd_none(*pmd))
-		return NULL;
-
-	pte = pte_offset_kernel(pmd, address);
-	entry = *pte;
-	if (pte_none(entry) || !pte_present(entry))
-		return NULL;
-
-	return pte;
-}
-
-/*
- * This routine handles page faults.  It determines the address,
- * and the problem, and then passes it off to one of the appropriate
- * routines.
- */
-asmlinkage void do_page_fault(struct pt_regs *regs, unsigned long error_code,
-			      unsigned long address)
-{
-	struct task_struct *tsk;
-	struct mm_struct *mm;
-	struct vm_area_struct * vma;
-	const struct exception_table_entry *fixup;
-	int write = error_code & FAULT_CODE_WRITE;
-	int textaccess = error_code & FAULT_CODE_ITLB;
-	unsigned int flags = (FAULT_FLAG_ALLOW_RETRY | FAULT_FLAG_KILLABLE |
-			      (write ? FAULT_FLAG_WRITE : 0));
-	pte_t *pte;
-	int fault;
-
-	/* SIM
-	 * Note this is now called with interrupts still disabled
-	 * This is to cope with being called for a missing IO port
-	 * address with interrupts disabled. This should be fixed as
-	 * soon as we have a better 'fast path' miss handler.
-	 *
-	 * Plus take care how you try and debug this stuff.
-	 * For example, writing debug data to a port which you
-	 * have just faulted on is not going to work.
-	 */
-
-	tsk = current;
-	mm = tsk->mm;
-
-	/* Not an IO address, so reenable interrupts */
-	local_irq_enable();
-
-	perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS, 1, regs, address);
-
-	/*
-	 * If we're in an interrupt or have no user
-	 * context, we must not take the fault..
-	 */
-	if (in_atomic() || !mm)
-		goto no_context;
-
-retry:
-	/* TLB misses upon some cache flushes get done under cli() */
-	down_read(&mm->mmap_sem);
-
-	vma = find_vma(mm, address);
-	if (!vma)
-		goto bad_area;
-	if (vma->vm_start <= address)
-		goto good_area;
-	if (!(vma->vm_flags & VM_GROWSDOWN))
-		goto bad_area;
-	if (expand_stack(vma, address))
-		goto bad_area;
-
-/*
- * Ok, we have a good vm_area for this memory access, so
- * we can handle it..
- */
-good_area:
-	if (textaccess) {
-		if (!(vma->vm_flags & VM_EXEC))
-			goto bad_area;
-	} else {
-		if (write) {
-			if (!(vma->vm_flags & VM_WRITE))
-				goto bad_area;
-		} else {
-			if (!(vma->vm_flags & VM_READ))
-				goto bad_area;
-		}
-	}
-
-	/*
-	 * If for any reason at all we couldn't handle the fault,
-	 * make sure we exit gracefully rather than endlessly redo
-	 * the fault.
-	 */
-	fault = handle_mm_fault(mm, vma, address, flags);
-
-	if ((fault & VM_FAULT_RETRY) && fatal_signal_pending(current))
-		return;
-
-	if (unlikely(fault & VM_FAULT_ERROR)) {
-		if (fault & VM_FAULT_OOM)
-			goto out_of_memory;
-		else if (fault & VM_FAULT_SIGBUS)
-			goto do_sigbus;
-		BUG();
-	}
-
-	if (flags & FAULT_FLAG_ALLOW_RETRY) {
-		if (fault & VM_FAULT_MAJOR) {
-			tsk->maj_flt++;
-			perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS_MAJ, 1,
-				      regs, address);
-		} else {
-			tsk->min_flt++;
-			perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS_MIN, 1,
-				      regs, address);
-		}
-
-		if (fault & VM_FAULT_RETRY) {
-			flags &= ~FAULT_FLAG_ALLOW_RETRY;
-
-			/*
-			 * No need to up_read(&mm->mmap_sem) as we would
-			 * have already released it in __lock_page_or_retry
-			 * in mm/filemap.c.
-			 */
-			goto retry;
-		}
-	}
-
-	/* If we get here, the page fault has been handled.  Do the TLB refill
-	   now from the newly-setup PTE, to avoid having to fault again right
-	   away on the same instruction. */
-	pte = lookup_pte (mm, address);
-	if (!pte) {
-		/* From empirical evidence, we can get here, due to
-		   !pte_present(pte).  (e.g. if a swap-in occurs, and the page
-		   is swapped back out again before the process that wanted it
-		   gets rescheduled?) */
-		goto no_pte;
-	}
-
-	__do_tlb_refill(address, textaccess, pte);
-
-no_pte:
-
-	up_read(&mm->mmap_sem);
-	return;
-
-/*
- * Something tried to access memory that isn't in our memory map..
- * Fix it, but check if it's kernel or user first..
- */
-bad_area:
-	up_read(&mm->mmap_sem);
-
-	if (user_mode(regs)) {
-		static int count=0;
-		siginfo_t info;
-		if (count < 4) {
-			/* This is really to help debug faults when starting
-			 * usermode, so only need a few */
-			count++;
-			printk("user mode bad_area address=%08lx pid=%d (%s) pc=%08lx\n",
-				address, task_pid_nr(current), current->comm,
-				(unsigned long) regs->pc);
-		}
-		if (is_global_init(tsk)) {
-			panic("INIT had user mode bad_area\n");
-		}
-		tsk->thread.address = address;
-		info.si_signo = SIGSEGV;
-		info.si_errno = 0;
-		info.si_addr = (void *) address;
-		force_sig_info(SIGSEGV, &info, tsk);
-		return;
-	}
-
-no_context:
-	/* Are we prepared to handle this kernel fault?  */
-	fixup = search_exception_tables(regs->pc);
-	if (fixup) {
-		regs->pc = fixup->fixup;
-		return;
-	}
-
-/*
- * Oops. The kernel tried to access some bad page. We'll have to
- * terminate things with extreme prejudice.
- *
- */
-	if (address < PAGE_SIZE)
-		printk(KERN_ALERT "Unable to handle kernel NULL pointer dereference");
-	else
-		printk(KERN_ALERT "Unable to handle kernel paging request");
-	printk(" at virtual address %08lx\n", address);
-	printk(KERN_ALERT "pc = %08Lx%08Lx\n", regs->pc >> 32, regs->pc & 0xffffffff);
-	die("Oops", regs, error_code);
-	do_exit(SIGKILL);
-
-/*
- * We ran out of memory, or some other thing happened to us that made
- * us unable to handle the page fault gracefully.
- */
-out_of_memory:
-	up_read(&mm->mmap_sem);
-	if (!user_mode(regs))
-		goto no_context;
-	pagefault_out_of_memory();
-	return;
-
-do_sigbus:
-	printk("fault:Do sigbus\n");
-	up_read(&mm->mmap_sem);
-
-	/*
-	 * Send a sigbus, regardless of whether we were in kernel
-	 * or user mode.
-	 */
-	tsk->thread.address = address;
-	force_sig(SIGBUS, tsk);
-
-	/* Kernel mode? Handle exceptions or die */
-	if (!user_mode(regs))
-		goto no_context;
-}
-
 void local_flush_tlb_one(unsigned long asid, unsigned long page)
 {
 	unsigned long long match, pteh=0, lpage;