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>

arch/s390/kernel/early.c

@@ -354,6 +354,10 @@ static __init void detect_machine_facilities(void)
 		S390_lowcore.machine_flags |= MACHINE_FLAG_VX;
 		__ctl_set_bit(0, 17);
 	}
+	if (test_facility(130)) {
+		S390_lowcore.machine_flags |= MACHINE_FLAG_NX;
+		__ctl_set_bit(0, 20);
+	}
 }
 
 static inline void save_vector_registers(void)
@@ -384,6 +388,21 @@ static int __init disable_vector_extension(char *str)
 }
 early_param("novx", disable_vector_extension);
 
+static int __init noexec_setup(char *str)
+{
+	bool enabled;
+	int rc;
+
+	rc = kstrtobool(str, &enabled);
+	if (!rc && !enabled) {
+		/* Disable no-execute support */
+		S390_lowcore.machine_flags &= ~MACHINE_FLAG_NX;
+		__ctl_clear_bit(0, 20);
+	}
+	return rc;
+}
+early_param("noexec", noexec_setup);
+
 static int __init cad_setup(char *str)
 {
 	int val;

arch/s390/kernel/entry.S

@@ -380,13 +380,11 @@ ENTRY(system_call)
 	brasl	%r14,do_signal
 	TSTMSK	__PT_FLAGS(%r11),_PIF_SYSCALL
 	jno	.Lsysc_return
+.Lsysc_do_syscall:
+	lghi	%r13,__TASK_thread
 	lmg	%r2,%r7,__PT_R2(%r11)	# load svc arguments
-	lghi	%r8,0			# svc 0 returns -ENOSYS
-	llgh	%r1,__PT_INT_CODE+2(%r11)	# load new svc number
-	cghi	%r1,NR_syscalls
-	jnl	.Lsysc_nr_ok		# invalid svc number -> do svc 0
-	slag	%r8,%r1,2
-	j	.Lsysc_nr_ok		# restart svc
+	lghi	%r1,0			# svc 0 returns -ENOSYS
+	j	.Lsysc_do_svc
 
 #
 # _TIF_NOTIFY_RESUME is set, call do_notify_resume
@@ -528,6 +526,8 @@ ENTRY(pgm_check_handler)
 	LOCKDEP_SYS_EXIT
 	tm	__PT_PSW+1(%r11),0x01	# returning to user ?
 	jno	.Lsysc_restore
+	TSTMSK	__PT_FLAGS(%r11),_PIF_SYSCALL
+	jo	.Lsysc_do_syscall
 	j	.Lsysc_tif
 
 #

arch/s390/kernel/kprobes.c

@@ -45,11 +45,17 @@ DEFINE_INSN_CACHE_OPS(dmainsn);
 
 static void *alloc_dmainsn_page(void)
 {
-	return (void *)__get_free_page(GFP_KERNEL | GFP_DMA);
+	void *page;
+
+	page = (void *) __get_free_page(GFP_KERNEL | GFP_DMA);
+	if (page)
+		set_memory_x((unsigned long) page, 1);
+	return page;
 }
 
 static void free_dmainsn_page(void *page)
 {
+	set_memory_nx((unsigned long) page, 1);
 	free_page((unsigned long)page);
 }
 

arch/s390/kernel/module.c

@@ -45,7 +45,8 @@ void *module_alloc(unsigned long size)
 	if (PAGE_ALIGN(size) > MODULES_LEN)
 		return NULL;
 	return __vmalloc_node_range(size, 1, MODULES_VADDR, MODULES_END,
-				    GFP_KERNEL, PAGE_KERNEL, 0, NUMA_NO_NODE,
+				    GFP_KERNEL, PAGE_KERNEL_EXEC,
+				    0, NUMA_NO_NODE,
 				    __builtin_return_address(0));
 }
 

arch/s390/kernel/vmlinux.lds.S

@@ -44,6 +44,7 @@ SECTIONS
 		*(.gnu.warning)
 	} :text = 0x0700
 
+	. = ALIGN(PAGE_SIZE);
 	_etext = .;		/* End of text section */
 
 	NOTES :text :note
@@ -79,7 +80,13 @@ SECTIONS
 	. = ALIGN(PAGE_SIZE);	/* Init code and data */
 	__init_begin = .;
 
-	INIT_TEXT_SECTION(PAGE_SIZE)
+	. = ALIGN(PAGE_SIZE);
+	.init.text : AT(ADDR(.init.text) - LOAD_OFFSET) {
+		VMLINUX_SYMBOL(_sinittext) = . ;
+		INIT_TEXT
+		. = ALIGN(PAGE_SIZE);
+		VMLINUX_SYMBOL(_einittext) = . ;
+	}
 
 	/*
 	 * .exit.text is discarded at runtime, not link time,