s390: add no-execute support

Bit 0x100 of a page table, segment table of region table entry
can be used to disallow code execution for the virtual addresses
associated with the entry.

There is one tricky bit, the system call to return from a signal
is part of the signal frame written to the user stack. With a
non-executable stack this would stop working. To avoid breaking
things the protection fault handler checks the opcode that caused
the fault for 0x0a77 (sys_sigreturn) and 0x0aad (sys_rt_sigreturn)
and injects a system call. This is preferable to the alternative
solution with a stub function in the vdso because it works for
vdso=off and statically linked binaries as well.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

1 files changed, 54 insertions, 63 deletions

diff --git a/arch/s390/mm/pageattr.c b/arch/s390/mm/pageattr.c
index 44f150312a16..a1543b74ee00 100644
--- a/arch/s390/mm/pageattr.c
+++ b/arch/s390/mm/pageattr.c
@@ -81,24 +81,24 @@ static void pgt_set(unsigned long *old, unsigned long new, unsigned long addr,
 	}
 }
 
-struct cpa {
-	unsigned int set_ro	: 1;
-	unsigned int clear_ro	: 1;
-};
-
 static int walk_pte_level(pmd_t *pmdp, unsigned long addr, unsigned long end,
-			  struct cpa cpa)
+			  unsigned long flags)
 {
 	pte_t *ptep, new;
 
 	ptep = pte_offset(pmdp, addr);
 	do {
-		if (pte_none(*ptep))
+		new = *ptep;
+		if (pte_none(new))
 			return -EINVAL;
-		if (cpa.set_ro)
-			new = pte_wrprotect(*ptep);
-		else if (cpa.clear_ro)
-			new = pte_mkwrite(pte_mkdirty(*ptep));
+		if (flags & SET_MEMORY_RO)
+			new = pte_wrprotect(new);
+		else if (flags & SET_MEMORY_RW)
+			new = pte_mkwrite(pte_mkdirty(new));
+		if ((flags & SET_MEMORY_NX) && MACHINE_HAS_NX)
+			pte_val(new) |= _PAGE_NOEXEC;
+		else if (flags & SET_MEMORY_X)
+			pte_val(new) &= ~_PAGE_NOEXEC;
 		pgt_set((unsigned long *)ptep, pte_val(new), addr, CRDTE_DTT_PAGE);
 		ptep++;
 		addr += PAGE_SIZE;
@@ -112,14 +112,17 @@ static int split_pmd_page(pmd_t *pmdp, unsigned long addr)
 	unsigned long pte_addr, prot;
 	pte_t *pt_dir, *ptep;
 	pmd_t new;
-	int i, ro;
+	int i, ro, nx;
 
 	pt_dir = vmem_pte_alloc();
 	if (!pt_dir)
 		return -ENOMEM;
 	pte_addr = pmd_pfn(*pmdp) << PAGE_SHIFT;
 	ro = !!(pmd_val(*pmdp) & _SEGMENT_ENTRY_PROTECT);
+	nx = !!(pmd_val(*pmdp) & _SEGMENT_ENTRY_NOEXEC);
 	prot = pgprot_val(ro ? PAGE_KERNEL_RO : PAGE_KERNEL);
+	if (!nx)
+		prot &= ~_PAGE_NOEXEC;
 	ptep = pt_dir;
 	for (i = 0; i < PTRS_PER_PTE; i++) {
 		pte_val(*ptep) = pte_addr | prot;
@@ -133,19 +136,24 @@ static int split_pmd_page(pmd_t *pmdp, unsigned long addr)
 	return 0;
 }
 
-static void modify_pmd_page(pmd_t *pmdp, unsigned long addr, struct cpa cpa)
+static void modify_pmd_page(pmd_t *pmdp, unsigned long addr,
+			    unsigned long flags)
 {
-	pmd_t new;
-
-	if (cpa.set_ro)
-		new = pmd_wrprotect(*pmdp);
-	else if (cpa.clear_ro)
-		new = pmd_mkwrite(pmd_mkdirty(*pmdp));
+	pmd_t new = *pmdp;
+
+	if (flags & SET_MEMORY_RO)
+		new = pmd_wrprotect(new);
+	else if (flags & SET_MEMORY_RW)
+		new = pmd_mkwrite(pmd_mkdirty(new));
+	if ((flags & SET_MEMORY_NX) && MACHINE_HAS_NX)
+		pmd_val(new) |= _SEGMENT_ENTRY_NOEXEC;
+	else if (flags & SET_MEMORY_X)
+		pmd_val(new) &= ~_SEGMENT_ENTRY_NOEXEC;
 	pgt_set((unsigned long *)pmdp, pmd_val(new), addr, CRDTE_DTT_SEGMENT);
 }
 
 static int walk_pmd_level(pud_t *pudp, unsigned long addr, unsigned long end,
-			  struct cpa cpa)
+			  unsigned long flags)
 {
 	unsigned long next;
 	pmd_t *pmdp;
@@ -163,9 +171,9 @@ static int walk_pmd_level(pud_t *pudp, unsigned long addr, unsigned long end,
 					return rc;
 				continue;
 			}
-			modify_pmd_page(pmdp, addr, cpa);
+			modify_pmd_page(pmdp, addr, flags);
 		} else {
-			rc = walk_pte_level(pmdp, addr, next, cpa);
+			rc = walk_pte_level(pmdp, addr, next, flags);
 			if (rc)
 				return rc;
 		}
@@ -181,14 +189,17 @@ static int split_pud_page(pud_t *pudp, unsigned long addr)
 	unsigned long pmd_addr, prot;
 	pmd_t *pm_dir, *pmdp;
 	pud_t new;
-	int i, ro;
+	int i, ro, nx;
 
 	pm_dir = vmem_pmd_alloc();
 	if (!pm_dir)
 		return -ENOMEM;
 	pmd_addr = pud_pfn(*pudp) << PAGE_SHIFT;
 	ro = !!(pud_val(*pudp) & _REGION_ENTRY_PROTECT);
+	nx = !!(pud_val(*pudp) & _REGION_ENTRY_NOEXEC);
 	prot = pgprot_val(ro ? SEGMENT_KERNEL_RO : SEGMENT_KERNEL);
+	if (!nx)
+		prot &= ~_SEGMENT_ENTRY_NOEXEC;
 	pmdp = pm_dir;
 	for (i = 0; i < PTRS_PER_PMD; i++) {
 		pmd_val(*pmdp) = pmd_addr | prot;
@@ -202,19 +213,24 @@ static int split_pud_page(pud_t *pudp, unsigned long addr)
 	return 0;
 }
 
-static void modify_pud_page(pud_t *pudp, unsigned long addr, struct cpa cpa)
+static void modify_pud_page(pud_t *pudp, unsigned long addr,
+			    unsigned long flags)
 {
-	pud_t new;
-
-	if (cpa.set_ro)
-		new = pud_wrprotect(*pudp);
-	else if (cpa.clear_ro)
-		new = pud_mkwrite(pud_mkdirty(*pudp));
+	pud_t new = *pudp;
+
+	if (flags & SET_MEMORY_RO)
+		new = pud_wrprotect(new);
+	else if (flags & SET_MEMORY_RW)
+		new = pud_mkwrite(pud_mkdirty(new));
+	if ((flags & SET_MEMORY_NX) && MACHINE_HAS_NX)
+		pud_val(new) |= _REGION_ENTRY_NOEXEC;
+	else if (flags & SET_MEMORY_X)
+		pud_val(new) &= ~_REGION_ENTRY_NOEXEC;
 	pgt_set((unsigned long *)pudp, pud_val(new), addr, CRDTE_DTT_REGION3);
 }
 
 static int walk_pud_level(pgd_t *pgd, unsigned long addr, unsigned long end,
-			  struct cpa cpa)
+			  unsigned long flags)
 {
 	unsigned long next;
 	pud_t *pudp;
@@ -232,9 +248,9 @@ static int walk_pud_level(pgd_t *pgd, unsigned long addr, unsigned long end,
 					break;
 				continue;
 			}
-			modify_pud_page(pudp, addr, cpa);
+			modify_pud_page(pudp, addr, flags);
 		} else {
-			rc = walk_pmd_level(pudp, addr, next, cpa);
+			rc = walk_pmd_level(pudp, addr, next, flags);
 		}
 		pudp++;
 		addr = next;
@@ -246,7 +262,7 @@ static int walk_pud_level(pgd_t *pgd, unsigned long addr, unsigned long end,
 static DEFINE_MUTEX(cpa_mutex);
 
 static int change_page_attr(unsigned long addr, unsigned long end,
-			    struct cpa cpa)
+			    unsigned long flags)
 {
 	unsigned long next;
 	int rc = -EINVAL;
@@ -262,7 +278,7 @@ static int change_page_attr(unsigned long addr, unsigned long end,
 		if (pgd_none(*pgdp))
 			break;
 		next = pgd_addr_end(addr, end);
-		rc = walk_pud_level(pgdp, addr, next, cpa);
+		rc = walk_pud_level(pgdp, addr, next, flags);
 		if (rc)
 			break;
 		cond_resched();
@@ -271,35 +287,10 @@ static int change_page_attr(unsigned long addr, unsigned long end,
 	return rc;
 }
 
-int set_memory_ro(unsigned long addr, int numpages)
+int __set_memory(unsigned long addr, int numpages, unsigned long flags)
 {
-	struct cpa cpa = {
-		.set_ro = 1,
-	};
-
 	addr &= PAGE_MASK;
-	return change_page_attr(addr, addr + numpages * PAGE_SIZE, cpa);
-}
-
-int set_memory_rw(unsigned long addr, int numpages)
-{
-	struct cpa cpa = {
-		.clear_ro = 1,
-	};
-
-	addr &= PAGE_MASK;
-	return change_page_attr(addr, addr + numpages * PAGE_SIZE, cpa);
-}
-
-/* not possible */
-int set_memory_nx(unsigned long addr, int numpages)
-{
-	return 0;
-}
-
-int set_memory_x(unsigned long addr, int numpages)
-{
-	return 0;
+	return change_page_attr(addr, addr + numpages * PAGE_SIZE, flags);
 }
 
 #ifdef CONFIG_DEBUG_PAGEALLOC
@@ -339,7 +330,7 @@ void __kernel_map_pages(struct page *page, int numpages, int enable)
 		nr = min(numpages - i, nr);
 		if (enable) {
 			for (j = 0; j < nr; j++) {
-				pte_val(*pte) = address | pgprot_val(PAGE_KERNEL);
+				pte_val(*pte) &= ~_PAGE_INVALID;
 				address += PAGE_SIZE;
 				pte++;
 			}