[PATCH] s390: general update

From: Martin Schwidefsky <schwidefsky@de.ibm.com>

 - Add console_unblank in machine_{restart,halt,power_off} to get
   all messages on the screen.
 - Set console_irq to -1 if condev= parameter is present.
 - Fix write_trylock for 64 bit.
 - Fix svc restarting.
 - System call number on 64 bit is an int. Fix compare in entry64.S.
 - Fix tlb flush problem.
 - Use the idte instruction to flush tlbs of a particular mm.
 - Fix ptrace.
 - Add fadvise64_64 system call wrapper.
 - Fix pfault handling.
 - Do not clobber _PAGE_INVALID_NONE pages in pte_wrprotect.
 - Fix siginfo_t size problem (needs to be 128 for s390x, not 136).
 - Avoid direct assignment to tsk->state, use __set_task_state.
 - Always make any pending restarted system call return -EINTR.
 - Add panic_on_oops.
 - Display symbol for psw address in show_trace.
 - Don't discard sections .exit.text, .exit.data and .eh_frame,
   otherwise stabs information for kerntypes will get lost. 
 - Add memory clobber to assembler inline in ip_fast_checksum for gcc 3.3.
 - Fix softirq_pending calls for the current cpu (cpu == smp_processor_id()).
 - Remove BUG_ON in irq_enter. Two irq_enters are possible.

arch/s390/defconfig

@@ -29,6 +29,7 @@ CONFIG_EPOLL=y
 CONFIG_IOSCHED_NOOP=y
 CONFIG_IOSCHED_AS=y
 CONFIG_IOSCHED_DEADLINE=y
+# CONFIG_CC_OPTIMIZE_FOR_SIZE is not set
 
 #
 # Loadable module support
@@ -109,8 +110,8 @@ CONFIG_SCSI_LOGGING=y
 #
 # SCSI low-level drivers
 #
-# CONFIG_SCSI_AIC7XXX is not set
 # CONFIG_SCSI_AIC7XXX_OLD is not set
+# CONFIG_SCSI_SATA is not set
 # CONFIG_SCSI_EATA_PIO is not set
 # CONFIG_SCSI_DEBUG is not set
 CONFIG_ZFCP=y
@@ -317,6 +318,21 @@ CONFIG_QETH=y
 # CONFIG_QETH_PERF_STATS is not set
 CONFIG_CCWGROUP=y
 
+#
+# Amateur Radio support
+#
+# CONFIG_HAMRADIO is not set
+
+#
+# IrDA (infrared) support
+#
+# CONFIG_IRDA is not set
+
+#
+# Bluetooth support
+#
+# CONFIG_BT is not set
+
 #
 # File systems
 #
@@ -358,8 +374,7 @@ CONFIG_PROC_KCORE=y
 # CONFIG_DEVFS_FS is not set
 CONFIG_DEVPTS_FS=y
 # CONFIG_DEVPTS_FS_XATTR is not set
-# CONFIG_TMPFS is not set
-# CONFIG_HUGETLBFS is not set
+CONFIG_TMPFS=y
 # CONFIG_HUGETLB_PAGE is not set
 CONFIG_RAMFS=y
 
@@ -385,6 +400,7 @@ CONFIG_RAMFS=y
 CONFIG_NFS_FS=y
 CONFIG_NFS_V3=y
 # CONFIG_NFS_V4 is not set
+# CONFIG_NFS_DIRECTIO is not set
 CONFIG_NFSD=y
 CONFIG_NFSD_V3=y
 # CONFIG_NFSD_V4 is not set
@@ -423,6 +439,11 @@ CONFIG_MSDOS_PARTITION=y
 # CONFIG_SUN_PARTITION is not set
 # CONFIG_EFI_PARTITION is not set
 
+#
+# Native Language Support
+#
+# CONFIG_NLS is not set
+
 #
 # Kernel hacking
 #

arch/s390/kernel/entry.S

@@ -249,6 +249,8 @@ sysc_sigpending:
         l       %r1,BASED(.Ldo_signal)
 	basr	%r14,%r1               # call do_signal
         stnsm   24(%r15),0xfc          # disable I/O and ext. interrupts
+	tm	__TI_flags+3(%r9),_TIF_RESTART_SVC
+	bo	BASED(sysc_restart)
 	b	BASED(sysc_leave)      # out of here, do NOT recheck
 
 #
@@ -258,7 +260,7 @@ sysc_restart:
 	ni	__TI_flags+3(%r9),255-_TIF_RESTART_SVC # clear TIF_RESTART_SVC
 	stosm	24(%r15),0x03          # reenable interrupts
 	l	%r7,SP_R2(%r15)        # load new svc number
-	sla	%r2,2
+	sla	%r7,2
 	mvc	SP_R2(4,%r15),SP_ORIG_R2(%r15) # restore first argument
 	lm	%r2,%r6,SP_R2(%r15)    # load svc arguments
 	b	BASED(sysc_do_restart) # restart svc
@@ -541,7 +543,7 @@ io_preempt:
 io_resume_loop:
 	tm	__TI_flags+3(%r9),_TIF_NEED_RESCHED
 	bno	BASED(io_leave)
-	mvc     __TI_precount(4,%r9),.Lc_pactive
+	mvc     __TI_precount(4,%r9),BASED(.Lc_pactive)
 	# hmpf, we are on the async. stack but to call schedule
 	# we have to move the interrupt frame to the process stack
 	l	%r1,SP_R15(%r15)

arch/s390/kernel/entry64.S

@@ -169,9 +169,10 @@ system_call:
         slag    %r7,%r7,2         # *4 and test for svc 0
 	jnz	sysc_do_restart
 	# svc 0: system call number in %r1
-	clg	%r1,.Lnr_syscalls-.Lconst(%r14)
+	cl	%r1,.Lnr_syscalls-.Lconst(%r14)
 	jnl	sysc_do_restart
-	slag    %r7,%r1,2         # svc 0: system call number in %r1
+	lgfr	%r7,%r1           # clear high word in r1
+	slag    %r7,%r7,2         # svc 0: system call number in %r1
 sysc_do_restart:
 	larl    %r10,sys_call_table
 #ifdef CONFIG_S390_SUPPORT
@@ -235,16 +236,18 @@ sysc_sigpending:
         sgr     %r3,%r3           # clear *oldset
 	brasl	%r14,do_signal    # call do_signal
 	stnsm   48(%r15),0xfc     # disable I/O and ext. interrupts
+	tm	__TI_flags+7(%r9),_TIF_RESTART_SVC
+	jo	sysc_restart
 	j	sysc_leave        # out of here, do NOT recheck
 
 #
 # _TIF_RESTART_SVC is set, set up registers and restart svc
 #
 sysc_restart:
-	ni	__TI_flags+3(%r9),255-_TIF_RESTART_SVC # clear TIF_RESTART_SVC
+	ni	__TI_flags+7(%r9),255-_TIF_RESTART_SVC # clear TIF_RESTART_SVC
 	stosm	48(%r15),0x03          # reenable interrupts
 	lg	%r7,SP_R2(%r15)        # load new svc number
-        slag    %r7,%r7,3              # *8
+        slag    %r7,%r7,2              # *4
 	mvc	SP_R2(8,%r15),SP_ORIG_R2(%r15) # restore first argument
 	lmg	%r2,%r6,SP_R2(%r15)    # load svc arguments
 	j	sysc_do_restart        # restart svc
@@ -503,7 +506,7 @@ pgm_svcstd:
 	larl    %r10,sys_call_table_emu # use 31 bit emulation system calls
 pgm_svcper_noemu:
 #endif
-	tm	__TI_flags+3(%r9),_TIF_SYSCALL_TRACE
+	tm	__TI_flags+7(%r9),_TIF_SYSCALL_TRACE
         lgf     %r8,0(%r7,%r10)   # load address of system call routine
         jo      pgm_tracesys
         basr    %r14,%r8          # call sys_xxxx
@@ -512,7 +515,7 @@ pgm_svcper_noemu:
                                   # changing anything here !!
 
 pgm_svcret:
-	tm	__TI_flags+3(%r9),_TIF_SIGPENDING
+	tm	__TI_flags+7(%r9),_TIF_SIGPENDING
 	jo	pgm_svcper_nosig
         la      %r2,SP_PTREGS(%r15) # load pt_regs
         sgr     %r3,%r3             # clear *oldset

arch/s390/kernel/head.S

@@ -569,6 +569,19 @@ startup:basr  %r13,0                     # get base
        oi     3(%r12),16                # set MVPG flag
 .Lchkmvpg:
 
+#
+# find out if we have the IDTE instruction
+#
+	mvc	__LC_PGM_NEW_PSW(8),.Lpcidte-.LPG1(%r13)
+	.long	0xb2b10000		# store facility list
+	tm	0xc8,0x08		# check bit for clearing-by-ASCE
+	bno	.Lchkidte-.LPG1(%r13)
+	lhi	%r1,2094
+	lhi	%r2,0
+	.long	0xb98e2001
+	oi	3(%r12),0x80		# set IDTE flag
+.Lchkidte:
+
         lpsw  .Lentry-.LPG1(13)         # jump to _stext in primary-space,
                                         # virtual and never return ...
         .align 8
@@ -593,6 +606,7 @@ startup:basr  %r13,0                     # get base
 .Lpcfpu:.long  0x00080000,0x80000000 + .Lchkfpu
 .Lpccsp:.long  0x00080000,0x80000000 + .Lchkcsp
 .Lpcmvpg:.long 0x00080000,0x80000000 + .Lchkmvpg
+.Lpcidte:.long 0x00080000,0x80000000 + .Lchkidte
 .Lmemsize:.long memory_size
 .Lmchunk:.long memory_chunk
 .Lmflags:.long machine_flags

arch/s390/kernel/head64.S

@@ -582,6 +582,20 @@ startup:basr  %r13,0                     # get base
 	mvc    __LC_DIAG44_OPCODE(8),.Ldiag44-.LPG1(%r13)
 0:	
 
+#
+# find out if we have the IDTE instruction
+#
+	la     %r1,0f-.LPG1(%r13)	# set program check address
+	stg    %r1,__LC_PGM_NEW_PSW+8
+	.long	0xb2b10000		# store facility list
+	tm	0xc8,0x08		# check bit for clearing-by-ASCE
+	bno	0f-.LPG1(%r13)
+	lhi	%r1,2094
+	lhi	%r2,0
+	.long	0xb98e2001
+	oi	7(%r12),0x80		# set IDTE flag
+0:
+
         lpswe .Lentry-.LPG1(13)         # jump to _stext in primary-space,
                                         # virtual and never return ...
         .align 16

arch/s390/kernel/ptrace.c

@@ -130,7 +130,11 @@ peek_user(struct task_struct *child, addr_t addr, addr_t data)
 	struct user *dummy = NULL;
 	addr_t offset, tmp;
 
-	if ((addr & __ADDR_MASK) || addr > sizeof(struct user) - __ADDR_MASK)
+	/*
+	 * Stupid gdb peeks/pokes the access registers in 64 bit with
+	 * an alignment of 4. Programmers from hell...
+	 */
+	if ((addr & 3) || addr > sizeof(struct user) - __ADDR_MASK)
 		return -EIO;
 
 	if (addr <= (addr_t) &dummy->regs.orig_gpr2) {
@@ -138,6 +142,9 @@ peek_user(struct task_struct *child, addr_t addr, addr_t data)
 		 * psw, gprs, acrs and orig_gpr2 are stored on the stack
 		 */
 		tmp = *(addr_t *)((addr_t) __KSTK_PTREGS(child) + addr);
+		if (addr == (addr_t) &dummy->regs.psw.mask)
+			/* Remove per bit from user psw. */
+			tmp &= ~PSW_MASK_PER;
 
 	} else if (addr >= (addr_t) &dummy->regs.fp_regs &&
 		   addr < (addr_t) (&dummy->regs.fp_regs + 1)) {
@@ -173,7 +180,11 @@ poke_user(struct task_struct *child, addr_t addr, addr_t data)
 	struct user *dummy = NULL;
 	addr_t offset;
 
-	if ((addr & __ADDR_MASK) || addr > sizeof(struct user) - __ADDR_MASK)
+	/*
+	 * Stupid gdb peeks/pokes the access registers in 64 bit with
+	 * an alignment of 4. Programmers from hell indeed...
+	 */
+	if ((addr & 3) || addr > sizeof(struct user) - __ADDR_MASK)
 		return -EIO;
 
 	if (addr <= (addr_t) &dummy->regs.orig_gpr2) {
@@ -258,7 +269,7 @@ do_ptrace_normal(struct task_struct *child, long request, long addr, long data)
 
 	case PTRACE_PEEKUSR_AREA:
 	case PTRACE_POKEUSR_AREA:
-		if (!copy_from_user(&parea, (void *) addr, sizeof(parea)))
+		if (copy_from_user(&parea, (void *) addr, sizeof(parea)))
 			return -EFAULT;
 		addr = parea.kernel_addr;
 		data = parea.process_addr;
@@ -266,8 +277,12 @@ do_ptrace_normal(struct task_struct *child, long request, long addr, long data)
 		while (copied < parea.len) {
 			if (request == PTRACE_PEEKUSR_AREA)
 				ret = peek_user(child, addr, data);
-			else
-				ret = poke_user(child, addr, data);
+			else {
+				addr_t tmp;
+				if (get_user (tmp, (addr_t *) data))
+					return -EFAULT;
+				ret = poke_user(child, addr, tmp);
+			}
 			if (ret)
 				return ret;
 			addr += sizeof(unsigned long);
@@ -390,7 +405,7 @@ poke_user_emu31(struct task_struct *child, addr_t addr, addr_t data)
 			if ((tmp & ~PSW32_MASK_CC) != PSW32_USER_BITS)
 				/* Invalid psw mask. */
 				return -EINVAL;
-			__KSTK_PTREGS(child)->psw.mask = PSW_USER_BITS |
+			__KSTK_PTREGS(child)->psw.mask = PSW_USER32_BITS |
 				((tmp & PSW32_MASK_CC) << 32);
 		} else if (addr == (addr_t) &dummy32->regs.psw.addr) {
 			/* Build a 64 bit psw address from 31 bit address. */
@@ -484,7 +499,7 @@ do_ptrace_emu31(struct task_struct *child, long request, long addr, long data)
 
 	case PTRACE_PEEKUSR_AREA:
 	case PTRACE_POKEUSR_AREA:
-		if (!copy_from_user(&parea, (void *) addr, sizeof(parea)))
+		if (copy_from_user(&parea, (void *) addr, sizeof(parea)))
 			return -EFAULT;
 		addr = parea.kernel_addr;
 		data = parea.process_addr;
@@ -492,8 +507,12 @@ do_ptrace_emu31(struct task_struct *child, long request, long addr, long data)
 		while (copied < parea.len) {
 			if (request == PTRACE_PEEKUSR_AREA)
 				ret = peek_user_emu31(child, addr, data);
-			else
-				ret = poke_user_emu31(child, addr, data);
+			else {
+				__u32 tmp;
+				if (get_user (tmp, (__u32 *) data))
+					return -EFAULT;
+				ret = poke_user_emu31(child, addr, tmp);
+			}
 			if (ret)
 				return ret;
 			addr += sizeof(unsigned int);

arch/s390/kernel/semaphore.c

@@ -64,14 +64,14 @@ void __down(struct semaphore * sem)
 	struct task_struct *tsk = current;
 	DECLARE_WAITQUEUE(wait, tsk);
 
-	tsk->state = TASK_UNINTERRUPTIBLE;
+	__set_task_state(tsk, TASK_UNINTERRUPTIBLE);
 	add_wait_queue_exclusive(&sem->wait, &wait);
 	while (__sem_update_count(sem, -1) <= 0) {
 		schedule();
-		tsk->state = TASK_UNINTERRUPTIBLE;
+		set_task_state(tsk, TASK_UNINTERRUPTIBLE);
 	}
 	remove_wait_queue(&sem->wait, &wait);
-	tsk->state = TASK_RUNNING;
+	__set_task_state(tsk, TASK_RUNNING);
 	wake_up(&sem->wait);
 }
 
@@ -87,7 +87,7 @@ int __down_interruptible(struct semaphore * sem)
 	struct task_struct *tsk = current;
 	DECLARE_WAITQUEUE(wait, tsk);
 
-	tsk->state = TASK_INTERRUPTIBLE;
+	__set_task_state(tsk, TASK_INTERRUPTIBLE);
 	add_wait_queue_exclusive(&sem->wait, &wait);
 	while (__sem_update_count(sem, -1) <= 0) {
 		if (signal_pending(current)) {
@@ -96,10 +96,10 @@ int __down_interruptible(struct semaphore * sem)
 			break;
 		}
 		schedule();
-		tsk->state = TASK_INTERRUPTIBLE;
+		set_task_state(tsk, TASK_INTERRUPTIBLE);
 	}
 	remove_wait_queue(&sem->wait, &wait);
-	tsk->state = TASK_RUNNING;
+	__set_task_state(tsk, TASK_RUNNING);
 	wake_up(&sem->wait);
 	return retval;
 }

arch/s390/kernel/setup.c

@@ -158,8 +158,10 @@ static int __init condev_setup(char *str)
 	int vdev;
 
 	vdev = simple_strtoul(str, &str, 0);
-	if (vdev >= 0 && vdev < 65536)
+	if (vdev >= 0 && vdev < 65536) {
 		console_device = vdev;
+		console_irq = -1;
+	}
 	return 1;
 }
 
@@ -287,6 +289,7 @@ void (*_machine_power_off)(void) = do_machine_power_off_nonsmp;
 
 void machine_restart(char *command)
 {
+	console_unblank();
 	_machine_restart(command);
 }
 
@@ -294,6 +297,7 @@ EXPORT_SYMBOL(machine_restart);
 
 void machine_halt(void)
 {
+	console_unblank();
 	_machine_halt();
 }
 
@@ -301,6 +305,7 @@ EXPORT_SYMBOL(machine_halt);
 
 void machine_power_off(void)
 {
+	console_unblank();
 	_machine_power_off();
 }
 

arch/s390/kernel/signal.c

@@ -167,6 +167,9 @@ static int restore_sigregs(struct pt_regs *regs, _sigregs *sregs)
 {
 	int err;
 
+	/* Alwys make any pending restarted system call return -EINTR */
+	current_thread_info()->restart_block.fn = do_no_restart_syscall;
+
 	err = __copy_from_user(regs, &sregs->regs, sizeof(_s390_regs_common));
 	regs->psw.mask = PSW_USER_BITS | (regs->psw.mask & PSW_MASK_CC);
 	regs->psw.addr |= PSW_ADDR_AMODE;

arch/s390/kernel/sys_s390.c

@@ -325,3 +325,39 @@ asmlinkage int s390x_personality(unsigned long personality)
 	return ret;
 }
 #endif /* CONFIG_ARCH_S390X */
+
+/*
+ * Wrapper function for sys_fadvise64/fadvise64_64
+ */
+#ifndef CONFIG_ARCH_S390X
+
+extern asmlinkage long sys_fadvise64(int, loff_t, size_t, int);
+
+asmlinkage long
+s390_fadvise64(int fd, u32 offset_high, u32 offset_low, size_t len, int advice)
+{
+	return sys_fadvise64(fd, (u64) offset_high << 32 | offset_low,
+			len, advice);
+}
+
+#endif
+
+extern asmlinkage long sys_fadvise64_64(int, loff_t, loff_t, int);
+
+struct fadvise64_64_args {
+	int fd;
+	long long offset;
+	long long len;
+	int advice;
+};
+
+asmlinkage long
+s390_fadvise64_64(struct fadvise64_64_args *args)
+{
+	struct fadvise64_64_args a;
+
+	if ( copy_from_user(&a, args, sizeof(a)) )
+		return -EFAULT;
+	return sys_fadvise64_64(a.fd, a.offset, a.len, a.advice);
+}
+

arch/s390/kernel/syscalls.S

@@ -15,7 +15,7 @@ SYSCALL(sys_read,sys_read,sys32_read_wrapper)
 SYSCALL(sys_write,sys_write,sys32_write_wrapper)
 SYSCALL(sys_open,sys_open,sys32_open_wrapper)			/* 5 */
 SYSCALL(sys_close,sys_close,sys32_close_wrapper)
-SYSCALL(sys_restart_syscall,sys_restart_syscall,sys_ni_syscall)
+SYSCALL(sys_restart_syscall,sys_restart_syscall,sys_restart_syscall)
 SYSCALL(sys_creat,sys_creat,sys32_creat_wrapper)
 SYSCALL(sys_link,sys_link,sys32_link_wrapper)
 SYSCALL(sys_unlink,sys_unlink,sys32_unlink_wrapper)		/* 10 */
@@ -261,7 +261,7 @@ SYSCALL(sys_epoll_create,sys_epoll_create,sys_epoll_create_wrapper)
 SYSCALL(sys_epoll_ctl,sys_epoll_ctl,sys_epoll_ctl_wrapper)	/* 250 */
 SYSCALL(sys_epoll_wait,sys_epoll_wait,sys_epoll_wait_wrapper)
 SYSCALL(sys_set_tid_address,sys_set_tid_address,sys32_set_tid_address_wrapper)
-SYSCALL(sys_fadvise64,sys_fadvise64,sys_ni_syscall)
+SYSCALL(s390_fadvise64,sys_fadvise64_64,sys_ni_syscall)
 SYSCALL(sys_timer_create,sys_timer_create,sys_ni_syscall)
 SYSCALL(sys_timer_settime,sys_timer_settime,sys_ni_syscall)	/* 255 */
 SYSCALL(sys_timer_gettime,sys_timer_gettime,sys_ni_syscall)
@@ -272,3 +272,4 @@ SYSCALL(sys_clock_gettime,sys_clock_gettime,sys_ni_syscall)	/* 260 */
 SYSCALL(sys_clock_getres,sys_clock_getres,sys_ni_syscall)
 SYSCALL(sys_clock_nanosleep,sys_clock_nanosleep,sys_ni_syscall)
 NI_SYSCALL							/* reserved for vserver */
+SYSCALL(s390_fadvise64_64,sys_ni_syscall,sys_ni_syscall)

arch/s390/kernel/traps.c

@@ -25,6 +25,7 @@
 #include <linux/smp.h>
 #include <linux/smp_lock.h>
 #include <linux/init.h>
+#include <linux/interrupt.h>
 #include <linux/delay.h>
 #include <linux/module.h>
 #include <linux/kallsyms.h>
@@ -156,9 +157,10 @@ void show_registers(struct pt_regs *regs)
 	int i;
 
 	mode = (regs->psw.mask & PSW_MASK_PSTATE) ? "User" : "Krnl";
-	printk("%s PSW : %p %p\n",
+	printk("%s PSW : %p %p",
 	       mode, (void *) regs->psw.mask,
 	       (void *) regs->psw.addr);
+	print_symbol(" (%s)\n", regs->psw.addr & PSW_ADDR_INSN);
 	printk("%s GPRS: " FOURLONG, mode,
 	       regs->gprs[0], regs->gprs[1], regs->gprs[2], regs->gprs[3]);
 	printk("           " FOURLONG,
@@ -250,6 +252,10 @@ void die(const char * str, struct pt_regs * regs, long err)
         show_regs(regs);
 	bust_spinlocks(0);
         spin_unlock_irq(&die_lock);
+	if (in_interrupt())
+		panic("Fatal exception in interrupt");
+	if (panic_on_oops)
+		panic("Fatal exception: panic_on_oops");
         do_exit(SIGSEGV);
 }
 

arch/s390/kernel/vmlinux.lds.S

@@ -117,10 +117,7 @@ SECTIONS
 
   /* Sections to be discarded */
   /DISCARD/ : {
-	*(.exit.text)
-	*(.exit.data)
 	*(.exitcall.exit)
-	*(.eh_frame)
 	}
 
   /* Stabs debugging sections.  */

arch/s390/mm/fault.c

@@ -538,8 +538,6 @@ asmlinkage void
 pfault_interrupt(struct pt_regs *regs, __u16 error_code)
 {
 	struct task_struct *tsk;
-	wait_queue_head_t queue;
-	wait_queue_head_t *qp;
 	__u16 subcode;
 
 	/*
@@ -553,46 +551,34 @@ pfault_interrupt(struct pt_regs *regs, __u16 error_code)
 		return;
 
 	/*
-	 * Get the token (= address of kernel stack of affected task).
+	 * Get the token (= address of the task structure of the affected task).
 	 */
 	tsk = *(struct task_struct **) __LC_PFAULT_INTPARM;
 
-	/*
-	 * We got all needed information from the lowcore and can
-	 * now safely switch on interrupts.
-	 */
-	if (regs->psw.mask & PSW_MASK_PSTATE)
-		local_irq_enable();
-
 	if (subcode & 0x0080) {
 		/* signal bit is set -> a page has been swapped in by VM */
-		qp = (wait_queue_head_t *)
-			xchg(&tsk->thread.pfault_wait, -1);
-		if (qp != NULL) {
+		if (xchg(&tsk->thread.pfault_wait, -1) != 0) {
 			/* Initial interrupt was faster than the completion
 			 * interrupt. pfault_wait is valid. Set pfault_wait
 			 * back to zero and wake up the process. This can
 			 * safely be done because the task is still sleeping
 			 * and can't procude new pfaults. */
-			tsk->thread.pfault_wait = 0ULL;
-			wake_up(qp);
+			tsk->thread.pfault_wait = 0;
+			wake_up_process(tsk);
 		}
 	} else {
 		/* signal bit not set -> a real page is missing. */
-                init_waitqueue_head (&queue);
-		qp = (wait_queue_head_t *)
-			xchg(&tsk->thread.pfault_wait, (addr_t) &queue);
-		if (qp != NULL) {
+		set_task_state(tsk, TASK_UNINTERRUPTIBLE);
+		if (xchg(&tsk->thread.pfault_wait, 1) != 0) {
 			/* Completion interrupt was faster than the initial
 			 * interrupt (swapped in a -1 for pfault_wait). Set
 			 * pfault_wait back to zero and exit. This can be
 			 * done safely because tsk is running in kernel 
 			 * mode and can't produce new pfaults. */
-			tsk->thread.pfault_wait = 0ULL;
-		}
-
-                /* go to sleep */
-                wait_event(queue, tsk->thread.pfault_wait == 0ULL);
+			tsk->thread.pfault_wait = 0;
+			set_task_state(tsk, TASK_RUNNING);
+		} else
+			set_tsk_need_resched(tsk);
 	}
 }
 #endif

include/asm-s390/bitops.h

@@ -526,10 +526,10 @@ __constant_test_bit(unsigned long nr, const volatile unsigned long *addr) {
  * Find-bit routines..
  */
 static inline int
-find_first_zero_bit(unsigned long * addr, unsigned size)
+find_first_zero_bit(unsigned long * addr, unsigned int size)
 {
 	unsigned long cmp, count;
-        int res;
+        unsigned int res;
 
         if (!size)
                 return 0;
@@ -565,10 +565,10 @@ find_first_zero_bit(unsigned long * addr, unsigned size)
 }
 
 static inline int
-find_first_bit(unsigned long * addr, unsigned size)
+find_first_bit(unsigned long * addr, unsigned int size)
 {
 	unsigned long cmp, count;
-        int res;
+        unsigned int res;
 
         if (!size)
                 return 0;
@@ -1022,7 +1022,7 @@ extern inline int ffs (int x)
 /*
  * fls: find last bit set.
  */
-extern __inline__ int fls(int x)
+static __inline__ int fls(int x)
 {
 	int r = 32;
 
@@ -1095,10 +1095,10 @@ extern __inline__ int fls(int x)
 #ifndef __s390x__
 
 static inline int 
-ext2_find_first_zero_bit(void *vaddr, unsigned size)
+ext2_find_first_zero_bit(void *vaddr, unsigned int size)
 {
 	unsigned long cmp, count;
-        int res;
+        unsigned int res;
 
         if (!size)
                 return 0;
@@ -1135,12 +1135,12 @@ ext2_find_first_zero_bit(void *vaddr, unsigned size)
 }
 
 static inline int 
-ext2_find_next_zero_bit(void *vaddr, unsigned size, unsigned offset)
+ext2_find_next_zero_bit(void *vaddr, unsigned int size, unsigned offset)
 {
         unsigned long *addr = vaddr;
         unsigned long *p = addr + (offset >> 5);
         unsigned long word, reg;
-        int bit = offset & 31UL, res;
+        unsigned int bit = offset & 31UL, res;
 
         if (offset >= size)
                 return size;

include/asm-s390/bug.h

 #ifndef _S390_BUG_H
 #define _S390_BUG_H
 
+#include <linux/kernel.h>
+
 #define BUG() do { \
         printk("kernel BUG at %s:%d!\n", __FILE__, __LINE__); \
         __asm__ __volatile__(".long 0"); \

include/asm-s390/byteorder.h

@@ -14,7 +14,7 @@
 #ifdef __GNUC__
 
 #ifdef __s390x__
-static __inline__ __const__ __u64 ___arch__swab64p(__u64 *x)
+static __inline__ __u64 ___arch__swab64p(__u64 *x)
 {
 	__u64 result;
 
@@ -24,7 +24,7 @@ static __inline__ __const__ __u64 ___arch__swab64p(__u64 *x)
 	return result;
 }
 
-static __inline__ __const__ __u64 ___arch__swab64(__u64 x)
+static __inline__ __u64 ___arch__swab64(__u64 x)
 {
 	__u64 result;
 
@@ -40,7 +40,7 @@ static __inline__ void ___arch__swab64s(__u64 *x)
 }
 #endif /* __s390x__ */
 
-static __inline__ __const__ __u32 ___arch__swab32p(__u32 *x)
+static __inline__ __u32 ___arch__swab32p(__u32 *x)
 {
 	__u32 result;
 	
@@ -58,7 +58,7 @@ static __inline__ __const__ __u32 ___arch__swab32p(__u32 *x)
 	return result;
 }
 
-static __inline__ __const__ __u32 ___arch__swab32(__u32 x)
+static __inline__ __u32 ___arch__swab32(__u32 x)
 {
 #ifndef __s390x__
 	return ___arch__swab32p(&x);
@@ -77,7 +77,7 @@ static __inline__ void ___arch__swab32s(__u32 *x)
 	*x = ___arch__swab32p(x);
 }
 
-static __inline__ __const__ __u16 ___arch__swab16p(__u16 *x)
+static __inline__ __u16 ___arch__swab16p(__u16 *x)
 {
 	__u16 result;
 	
@@ -93,7 +93,7 @@ static __inline__ __const__ __u16 ___arch__swab16p(__u16 *x)
 	return result;
 }
 
-static __inline__ __const__ __u16 ___arch__swab16(__u16 x)
+static __inline__ __u16 ___arch__swab16(__u16 x)
 {
 	return ___arch__swab16p(&x);
 }

include/asm-s390/checksum.h

@@ -70,7 +70,7 @@ csum_partial_inline(const unsigned char * buff, int len, unsigned int sum)
 	__asm__ __volatile__ (
 		"0:  cksm %0,%1\n"    /* do checksum on longs */
 		"    jo   0b\n"
-                : "+&d" (sum), "+&a" (rp) : : "cc" );
+                : "+&d" (sum), "+&a" (rp) : : "cc", "memory" );
 #else /* __s390x__ */
 	__asm__ __volatile__ (
 		"    lgr  2,%1\n"    /* address in gpr 2 */
@@ -79,7 +79,7 @@ csum_partial_inline(const unsigned char * buff, int len, unsigned int sum)
 		"    jo   0b\n"
                 : "+&d" (sum)
 		: "d" (buff), "d" (len)
-                : "cc", "2", "3" );
+                : "cc", "memory", "2", "3" );
 #endif /* __s390x__ */
 	return sum;
 }
@@ -165,7 +165,7 @@ ip_fast_csum(unsigned char *iph, unsigned int ihl)
 		"    sr   %0,%0\n"   /* set sum to zero */
                 "0:  cksm %0,%1\n"   /* do checksum on longs */
                 "    jo   0b\n"
-                : "=&d" (sum), "+&a" (rp) : : "cc" );
+                : "=&d" (sum), "+&a" (rp) : : "cc", "memory" );
 #else /* __s390x__ */
         __asm__ __volatile__ (
 		"    slgr %0,%0\n"   /* set sum to zero */
@@ -175,7 +175,7 @@ ip_fast_csum(unsigned char *iph, unsigned int ihl)
                 "    jo   0b\n"
                 : "=&d" (sum)
                 : "d" (iph), "d" (ihl*4)
-                : "cc", "2", "3" );
+                : "cc", "memory", "2", "3" );
 #endif /* __s390x__ */
         return csum_fold(sum);
 }

include/asm-s390/hardirq.h

@@ -25,9 +25,17 @@ typedef struct {
 	unsigned int __softirq_pending;
 } irq_cpustat_t;
 
-#define softirq_pending(cpu) (lowcore_ptr[(cpu)]->softirq_pending)
 #define local_softirq_pending() (S390_lowcore.softirq_pending)
 
+/* this is always called with cpu == smp_processor_id() at the moment */
+static inline __u32
+softirq_pending(unsigned int cpu)
+{
+	if (cpu == smp_processor_id())
+		return local_softirq_pending();
+	return lowcore_ptr[cpu]->softirq_pending;
+}
+
 #define __ARCH_IRQ_STAT
 
 /*
@@ -42,12 +50,12 @@ typedef struct {
  *
  * PREEMPT_MASK: 0x000000ff
  * SOFTIRQ_MASK: 0x0000ff00
- * HARDIRQ_MASK: 0x00010000
+ * HARDIRQ_MASK: 0x00ff0000
  */
 
 #define PREEMPT_BITS	8
 #define SOFTIRQ_BITS	8
-#define HARDIRQ_BITS	1
+#define HARDIRQ_BITS	8
 
 #define PREEMPT_SHIFT	0
 #define SOFTIRQ_SHIFT	(PREEMPT_SHIFT + PREEMPT_BITS)
@@ -81,7 +89,6 @@ typedef struct {
 
 #define irq_enter()							\
 do {									\
-	BUG_ON( hardirq_count() );					\
 	(preempt_count() += HARDIRQ_OFFSET);				\
 } while(0)
 	

include/asm-s390/ioctl.h

@@ -55,11 +55,21 @@
 	 ((nr)   << _IOC_NRSHIFT) | \
 	 ((size) << _IOC_SIZESHIFT))
 
+/* provoke compile error for invalid uses of size argument */
+extern unsigned long __invalid_size_argument_for_IOC;
+#define _IOC_TYPECHECK(t) \
+	((sizeof(t) == sizeof(t[1]) && \
+	  sizeof(t) < (1 << _IOC_SIZEBITS)) ? \
+	  sizeof(t) : __invalid_size_argument_for_IOC)
+
 /* used to create numbers */
 #define _IO(type,nr)		_IOC(_IOC_NONE,(type),(nr),0)
-#define _IOR(type,nr,size)	_IOC(_IOC_READ,(type),(nr),sizeof(size))
-#define _IOW(type,nr,size)	_IOC(_IOC_WRITE,(type),(nr),sizeof(size))
-#define _IOWR(type,nr,size)	_IOC(_IOC_READ|_IOC_WRITE,(type),(nr),sizeof(size))
+#define _IOR(type,nr,size)	_IOC(_IOC_READ,(type),(nr),(_IOC_TYPECHECK(size)))
+#define _IOW(type,nr,size)	_IOC(_IOC_WRITE,(type),(nr),(_IOC_TYPECHECK(size)))
+#define _IOWR(type,nr,size)	_IOC(_IOC_READ|_IOC_WRITE,(type),(nr),(_IOC_TYPECHECK(size)))
+#define _IOR_BAD(type,nr,size)	_IOC(_IOC_READ,(type),(nr),sizeof(size))
+#define _IOW_BAD(type,nr,size)	_IOC(_IOC_WRITE,(type),(nr),sizeof(size))
+#define _IOWR_BAD(type,nr,size)	_IOC(_IOC_READ|_IOC_WRITE,(type),(nr),sizeof(size))
 
 /* used to decode ioctl numbers.. */
 #define _IOC_DIR(nr)		(((nr) >> _IOC_DIRSHIFT) & _IOC_DIRMASK)

include/asm-s390/mmu_context.h

@@ -14,7 +14,7 @@
  */
 #define init_new_context(tsk,mm)        0
 
-#define destroy_context(mm)             flush_tlb_mm(mm)
+#define destroy_context(mm)             do { } while (0)
 
 static inline void enter_lazy_tlb(struct mm_struct *mm,
                                   struct task_struct *tsk)

include/asm-s390/pgalloc.h

@@ -84,7 +84,11 @@ static inline void pmd_free (pmd_t *pmd)
 	free_pages((unsigned long) pmd, 2);
 }
 
-#define __pmd_free_tlb(tlb,pmd) pmd_free(pmd)
+#define __pmd_free_tlb(tlb,pmd)			\
+	do {					\
+		tlb_flush_mmu(tlb, 0, 0);	\
+		pmd_free(pmd);			\
+	 } while (0)
 
 static inline void pgd_populate(struct mm_struct *mm, pgd_t *pgd, pmd_t *pmd)
 {
@@ -146,7 +150,7 @@ static inline void pte_free(struct page *pte)
         __free_page(pte);
 }
 
-#define __pte_free_tlb(tlb,pte) tlb_remove_page((tlb),(pte))
+#define __pte_free_tlb(tlb,pte) tlb_remove_page(tlb,pte)
 
 /*
  * This establishes kernel virtual mappings (e.g., as a result of a

include/asm-s390/pgtable.h

@@ -29,6 +29,7 @@
  * the S390 page table tree.
  */
 #ifndef __ASSEMBLY__
+#include <asm/bug.h>
 #include <asm/processor.h>
 #include <linux/threads.h>
 
@@ -465,6 +466,8 @@ extern inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 
 extern inline pte_t pte_wrprotect(pte_t pte)
 {
+	/* Do not clobber _PAGE_INVALID_NONE pages!  */
+	if (!(pte_val(pte) & _PAGE_INVALID))
 		pte_val(pte) |= _PAGE_RO;
 	return pte;
 }
@@ -682,9 +685,9 @@ extern inline pte_t mk_swap_pte(unsigned long type, unsigned long offset)
 	pte_t pte;
 	pte_val(pte) = (type << 1) | (offset << 12) | _PAGE_INVALID_SWAP;
 #ifndef __s390x__
-	pte_val(pte) &= 0x7ffff6fe;  /* better to be paranoid */
+	BUG_ON((pte_val(pte) & 0x80000901) != 0);
 #else /* __s390x__ */
-	pte_val(pte) &= 0xfffffffffffff6fe;  /* better to be paranoid */
+	BUG_ON((pte_val(pte) & 0x901) != 0);
 #endif /* __s390x__ */
 	return pte;
 }

include/asm-s390/scatterlist.h

@@ -7,6 +7,10 @@ struct scatterlist {
     unsigned int length;
 };
 
-#define ISA_DMA_THRESHOLD (0xffffffffffffffff)
+#ifdef __s390x__
+#define ISA_DMA_THRESHOLD (0xffffffffffffffffUL)
+#else
+#define ISA_DMA_THRESHOLD (0xffffffffUL)
+#endif
 
 #endif /* _ASMS390X_SCATTERLIST_H */

include/asm-s390/setup.h

@@ -36,6 +36,7 @@ extern unsigned long machine_flags;
 #define MACHINE_HAS_MVPG	(machine_flags & 16)
 #define MACHINE_HAS_DIAG44	(machine_flags & 32)
 #define MACHINE_NEW_STIDP	(machine_flags & 64)
+#define MACHINE_HAS_IDTE	(machine_flags & 128)
 
 #ifndef __s390x__
 #define MACHINE_HAS_IEEE	(machine_flags & 2)

include/asm-s390/siginfo.h

@@ -14,6 +14,12 @@
 #define __ARCH_SI_PREAMBLE_SIZE (4 * sizeof(int))
 #endif
 
+#ifdef CONFIG_ARCH_S390X
+#define SIGEV_PAD_SIZE ((SIGEV_MAX_SIZE/sizeof(int)) - 4)
+#else
+#define SIGEV_PAD_SIZE ((SIGEV_MAX_SIZE/sizeof(int)) - 3)
+#endif
+
 #include <asm-generic/siginfo.h>
 
 /*

include/asm-s390/spinlock.h

@@ -217,7 +217,7 @@ typedef struct {
 
 extern inline int _raw_write_trylock(rwlock_t *rw)
 {
-	unsigned int result, reg;
+	unsigned long result, reg;
 	
 	__asm__ __volatile__(
 #ifndef __s390x__

include/asm-s390/tlbflush.h

@@ -99,17 +99,21 @@ static inline void global_flush_tlb(void)
 static inline void __flush_tlb_mm(struct mm_struct * mm)
 {
 	cpumask_t local_cpumask;
+
+	if (unlikely(cpus_empty(mm->cpu_vm_mask)))
+		return;
+	if (MACHINE_HAS_IDTE) {
+		asm volatile (".insn rrf,0xb98e0000,0,%0,%1,0"
+			      : : "a" (2048),
+			      "a" (__pa(mm->pgd)&PAGE_MASK) : "cc" );
+		return;
+	}
 	preempt_disable();
 	local_cpumask = cpumask_of_cpu(smp_processor_id());
-	if (cpus_equal(mm->cpu_vm_mask, local_cpumask)) {
-		/* mm was active on more than one cpu. */
-		if (mm == current->active_mm &&
-		    atomic_read(&mm->mm_users) == 1)
-			/* this cpu is the only one using the mm. */
-			mm->cpu_vm_mask = local_cpumask;
-		global_flush_tlb();
-	} else
+	if (cpus_equal(mm->cpu_vm_mask, local_cpumask))
 		local_flush_tlb();
+	else
+		global_flush_tlb();
 	preempt_enable();
 }
 
@@ -136,8 +140,7 @@ static inline void flush_tlb_range(struct vm_area_struct *vma,
 	__flush_tlb_mm(vma->vm_mm); 
 }
 
-#define flush_tlb_kernel_range(start, end) \
-	__flush_tlb_mm(&init_mm)
+#define flush_tlb_kernel_range(start, end) global_flush_tlb()
 
 #endif
 

include/asm-s390/unistd.h

@@ -259,8 +259,9 @@
 /*
  * Number 263 is reserved for vserver
  */
+#define __NR_fadvise64_64	264
 
-#define NR_syscalls 264
+#define NR_syscalls 265
 
 /* 
  * There are some system calls that are not present on 64 bit, some
@@ -322,6 +323,7 @@
 #undef  __NR_getdents64
 #undef  __NR_fcntl64
 #undef  __NR_sendfile64
+#undef  __NR_fadvise64_64
 
 #define __NR_select		142
 #define __NR_getrlimit		191	/* SuS compliant getrlimit */