13 files changed, 43 insertions, 104 deletions

diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
index d385f396dfee..3cbb8757704e 100644
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -138,8 +138,10 @@ config S390
         select KTIME_SCALAR if 32BIT
         select HAVE_ARCH_SECCOMP_FILTER
         select GENERIC_KERNEL_THREAD
+        select GENERIC_KERNEL_EXECVE
         select HAVE_MOD_ARCH_SPECIFIC
         select MODULES_USE_ELF_RELA
+        select CLONE_BACKWARDS2
 
 config SCHED_OMIT_FRAME_POINTER
         def_bool y
diff --git a/arch/s390/include/asm/Kbuild b/arch/s390/include/asm/Kbuild
index 0633dc6d254d..f313f9cbcf44 100644
--- a/arch/s390/include/asm/Kbuild
+++ b/arch/s390/include/asm/Kbuild
@@ -1,3 +1,4 @@
 
 
 generic-y += clkdev.h
+generic-y += trace_clock.h
diff --git a/arch/s390/include/asm/cputime.h b/arch/s390/include/asm/cputime.h
index 023d5ae24482..d2ff41370c0c 100644
--- a/arch/s390/include/asm/cputime.h
+++ b/arch/s390/include/asm/cputime.h
@@ -14,6 +14,7 @@
 
 
 #define __ARCH_HAS_VTIME_ACCOUNT
+#define __ARCH_HAS_VTIME_TASK_SWITCH
 
 /* We want to use full resolution of the CPU timer: 2**-12 micro-seconds. */
 
diff --git a/arch/s390/include/asm/page.h b/arch/s390/include/asm/page.h
index 6d5367060a56..39faa4ac9660 100644
--- a/arch/s390/include/asm/page.h
+++ b/arch/s390/include/asm/page.h
@@ -158,6 +158,9 @@ static inline int page_reset_referenced(unsigned long addr)
  * race against modification of the referenced bit. This function
  * should therefore only be called if it is not mapped in any
  * address space.
+ *
+ * Note that the bit gets set whenever page content is changed. That means
+ * also when the page is modified by DMA or from inside the kernel.
  */
 #define __HAVE_ARCH_PAGE_TEST_AND_CLEAR_DIRTY
 static inline int page_test_and_clear_dirty(unsigned long pfn, int mapped)
diff --git a/arch/s390/include/asm/signal.h b/arch/s390/include/asm/signal.h
index bffdbdd5b3d7..db7ddfaf5b79 100644
--- a/arch/s390/include/asm/signal.h
+++ b/arch/s390/include/asm/signal.h
@@ -39,6 +39,4 @@ struct k_sigaction {
         struct sigaction sa;
 };
 
-#define ptrace_signal_deliver(regs, cookie) do { } while (0)
-
 #endif
diff --git a/arch/s390/include/asm/unistd.h b/arch/s390/include/asm/unistd.h
index bbbae41fa9a5..086bb8eaf6ab 100644
--- a/arch/s390/include/asm/unistd.h
+++ b/arch/s390/include/asm/unistd.h
@@ -54,7 +54,9 @@
 #   define __ARCH_WANT_COMPAT_SYS_RT_SIGSUSPEND
 # endif
 #define __ARCH_WANT_SYS_EXECVE
-#define __ARCH_WANT_KERNEL_EXECVE
+#define __ARCH_WANT_SYS_FORK
+#define __ARCH_WANT_SYS_VFORK
+#define __ARCH_WANT_SYS_CLONE
 
 /*
  * "Conditional" syscalls
diff --git a/arch/s390/kernel/compat_wrapper.S b/arch/s390/kernel/compat_wrapper.S
index ad79b846535c..827e094a2f49 100644
--- a/arch/s390/kernel/compat_wrapper.S
+++ b/arch/s390/kernel/compat_wrapper.S
@@ -28,7 +28,7 @@ ENTRY(sys32_open_wrapper)
         llgtr   %r2,%r2                 # const char *
         lgfr    %r3,%r3                 # int
         lgfr    %r4,%r4                 # int
-        jg      sys_open                # branch to system call
+        jg      compat_sys_open         # branch to system call
 
 ENTRY(sys32_close_wrapper)
         llgfr   %r2,%r2                 # unsigned int
diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S
index ef46f66bc0d6..aa8f2ba6289b 100644
--- a/arch/s390/kernel/entry.S
+++ b/arch/s390/kernel/entry.S
@@ -330,40 +330,18 @@ ENTRY(ret_from_fork)
         la      %r11,STACK_FRAME_OVERHEAD(%r15)
         l       %r12,__LC_THREAD_INFO
         l       %r13,__LC_SVC_NEW_PSW+4
-        tm      __PT_PSW+1(%r11),0x01   # forking a kernel thread ?
-        je      1f
         l       %r1,BASED(.Lschedule_tail)
         basr    %r14,%r1                # call schedule_tail
         TRACE_IRQS_ON
         ssm     __LC_SVC_NEW_PSW        # reenable interrupts
-        j       sysc_tracenogo
-
-1:      # it's a kernel thread
-        st      %r15,__PT_R15(%r11)     # store stack pointer for new kthread
-        l       %r1,BASED(.Lschedule_tail)
-        basr    %r14,%r1                # call schedule_tail
-        TRACE_IRQS_ON
-        ssm     __LC_SVC_NEW_PSW        # reenable interrupts
-        lm      %r9,%r11,__PT_R9(%r11)  # load gprs
+        tm      __PT_PSW+1(%r11),0x01   # forking a kernel thread ?
+        jne     sysc_tracenogo
+        # it's a kernel thread
+        lm      %r9,%r10,__PT_R9(%r11)  # load gprs
 ENTRY(kernel_thread_starter)
         la      %r2,0(%r10)
         basr    %r14,%r9
-        la      %r2,0
-        br      %r11                    # do_exit
-
-#
-# kernel_execve function needs to deal with pt_regs that is not
-# at the usual place
-#
-ENTRY(ret_from_kernel_execve)
-        ssm     __LC_PGM_NEW_PSW        # disable I/O and ext. interrupts
-        lr      %r15,%r2
-        lr      %r11,%r2
-        ahi     %r15,-STACK_FRAME_OVERHEAD
-        xc      __SF_BACKCHAIN(4,%r15),__SF_BACKCHAIN(%r15)
-        l       %r12,__LC_THREAD_INFO
-        ssm     __LC_SVC_NEW_PSW        # reenable interrupts
-        j       sysc_return
+        j       sysc_tracenogo
 
 /*
  * Program check handler routine
diff --git a/arch/s390/kernel/entry.h b/arch/s390/kernel/entry.h
index d0d3f69a7346..d8251b98f17a 100644
--- a/arch/s390/kernel/entry.h
+++ b/arch/s390/kernel/entry.h
@@ -54,10 +54,6 @@ long sys_s390_fadvise64(int fd, u32 offset_high, u32 offset_low,
 long sys_s390_fadvise64_64(struct fadvise64_64_args __user *args);
 long sys_s390_fallocate(int fd, int mode, loff_t offset, u32 len_high,
                         u32 len_low);
-long sys_fork(void);
-long sys_clone(unsigned long newsp, unsigned long clone_flags,
-               int __user *parent_tidptr, int __user *child_tidptr);
-long sys_vfork(void);
 long sys_sigsuspend(int history0, int history1, old_sigset_t mask);
 long sys_sigaction(int sig, const struct old_sigaction __user *act,
                    struct old_sigaction __user *oact);
diff --git a/arch/s390/kernel/entry64.S b/arch/s390/kernel/entry64.S
index 07d8de353984..499e95e90f38 100644
--- a/arch/s390/kernel/entry64.S
+++ b/arch/s390/kernel/entry64.S
@@ -352,33 +352,17 @@ sysc_tracenogo:
 ENTRY(ret_from_fork)
         la      %r11,STACK_FRAME_OVERHEAD(%r15)
         lg      %r12,__LC_THREAD_INFO
-        tm      __PT_PSW+1(%r11),0x01   # forking a kernel thread ?
-        je      1f
         brasl   %r14,schedule_tail
         TRACE_IRQS_ON
         ssm     __LC_SVC_NEW_PSW        # reenable interrupts
-        j       sysc_tracenogo
-1:      # it's a kernel thread
-        stg     %r15,__PT_R15(%r11)     # store stack pointer for new kthread
-        brasl   %r14,schedule_tail
-        TRACE_IRQS_ON
-        ssm     __LC_SVC_NEW_PSW        # reenable interrupts
-        lmg     %r9,%r11,__PT_R9(%r11)  # load gprs
+        tm      __PT_PSW+1(%r11),0x01   # forking a kernel thread ?
+        jne     sysc_tracenogo
+        # it's a kernel thread
+        lmg     %r9,%r10,__PT_R9(%r11)  # load gprs
 ENTRY(kernel_thread_starter)
         la      %r2,0(%r10)
         basr    %r14,%r9
-        la      %r2,0
-        br      %r11                    # do_exit
-
-ENTRY(ret_from_kernel_execve)
-        ssm     __LC_PGM_NEW_PSW        # disable I/O and ext. interrupts
-        lgr     %r15,%r2
-        lgr     %r11,%r2
-        aghi    %r15,-STACK_FRAME_OVERHEAD
-        xc      __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15)
-        lg      %r12,__LC_THREAD_INFO
-        ssm     __LC_SVC_NEW_PSW        # reenable interrupts
-        j       sysc_return
+        j       sysc_tracenogo
 
 /*
  * Program check handler routine
diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c
index cd31ad457a9b..536d64579d9a 100644
--- a/arch/s390/kernel/process.c
+++ b/arch/s390/kernel/process.c
@@ -117,8 +117,7 @@ void release_thread(struct task_struct *dead_task)
 }
 
 int copy_thread(unsigned long clone_flags, unsigned long new_stackp,
-                unsigned long arg,
-                struct task_struct *p, struct pt_regs *regs)
+                unsigned long arg, struct task_struct *p)
 {
         struct thread_info *ti;
         struct fake_frame
@@ -150,7 +149,7 @@ int copy_thread(unsigned long clone_flags, unsigned long new_stackp,
         frame->sf.gprs[9] = (unsigned long) frame;
 
         /* Store access registers to kernel stack of new process. */
-        if (unlikely(!regs)) {
+        if (unlikely(p->flags & PF_KTHREAD)) {
                 /* kernel thread */
                 memset(&frame->childregs, 0, sizeof(struct pt_regs));
                 frame->childregs.psw.mask = psw_kernel_bits | PSW_MASK_DAT |
@@ -164,9 +163,10 @@ int copy_thread(unsigned long clone_flags, unsigned long new_stackp,
 
                 return 0;
         }
-        frame->childregs = *regs;
+        frame->childregs = *current_pt_regs();
         frame->childregs.gprs[2] = 0;   /* child returns 0 on fork. */
-        frame->childregs.gprs[15] = new_stackp;
+        if (new_stackp)
+                frame->childregs.gprs[15] = new_stackp;
 
         /* Don't copy runtime instrumentation info */
         p->thread.ri_cb = NULL;
@@ -183,57 +183,24 @@ int copy_thread(unsigned long clone_flags, unsigned long new_stackp,
                sizeof(s390_fp_regs));
         /* Set a new TLS ?  */
         if (clone_flags & CLONE_SETTLS)
-                p->thread.acrs[0] = regs->gprs[6];
+                p->thread.acrs[0] = frame->childregs.gprs[6];
 #else /* CONFIG_64BIT */
         /* Save the fpu registers to new thread structure. */
         save_fp_regs(&p->thread.fp_regs);
         /* Set a new TLS ?  */
         if (clone_flags & CLONE_SETTLS) {
+                unsigned long tls = frame->childregs.gprs[6];
                 if (is_compat_task()) {
-                        p->thread.acrs[0] = (unsigned int) regs->gprs[6];
+                        p->thread.acrs[0] = (unsigned int)tls;
                 } else {
-                        p->thread.acrs[0] = (unsigned int)(regs->gprs[6] >> 32);
-                        p->thread.acrs[1] = (unsigned int) regs->gprs[6];
+                        p->thread.acrs[0] = (unsigned int)(tls >> 32);
+                        p->thread.acrs[1] = (unsigned int)tls;
                 }
         }
 #endif /* CONFIG_64BIT */
         return 0;
 }
 
-SYSCALL_DEFINE0(fork)
-{
-        struct pt_regs *regs = task_pt_regs(current);
-        return do_fork(SIGCHLD, regs->gprs[15], regs, 0, NULL, NULL);
-}
-
-SYSCALL_DEFINE4(clone, unsigned long, newsp, unsigned long, clone_flags,
-                int __user *, parent_tidptr, int __user *, child_tidptr)
-{
-        struct pt_regs *regs = task_pt_regs(current);
-
-        if (!newsp)
-                newsp = regs->gprs[15];
-        return do_fork(clone_flags, newsp, regs, 0,
-                       parent_tidptr, child_tidptr);
-}
-
-/*
- * This is trivial, and on the face of it looks like it
- * could equally well be done in user mode.
- *
- * Not so, for quite unobvious reasons - register pressure.
- * In user mode vfork() cannot have a stack frame, and if
- * done by calling the "clone()" system call directly, you
- * do not have enough call-clobbered registers to hold all
- * the information you need.
- */
-SYSCALL_DEFINE0(vfork)
-{
-        struct pt_regs *regs = task_pt_regs(current);
-        return do_fork(CLONE_VFORK | CLONE_VM | SIGCHLD,
-                       regs->gprs[15], regs, 0, NULL, NULL);
-}
-
 asmlinkage void execve_tail(void)
 {
         current->thread.fp_regs.fpc = 0;
diff --git a/arch/s390/kernel/vtime.c b/arch/s390/kernel/vtime.c
index 790334427895..e84b8b68444a 100644
--- a/arch/s390/kernel/vtime.c
+++ b/arch/s390/kernel/vtime.c
@@ -112,7 +112,12 @@ void vtime_task_switch(struct task_struct *prev)
         S390_lowcore.system_timer = ti->system_timer;
 }
 
-void account_process_tick(struct task_struct *tsk, int user_tick)
+/*
+ * In s390, accounting pending user time also implies
+ * accounting system time in order to correctly compute
+ * the stolen time accounting.
+ */
+void vtime_account_user(struct task_struct *tsk)
 {
         if (do_account_vtime(tsk, HARDIRQ_OFFSET))
                 virt_timer_expire();
@@ -127,6 +132,8 @@ void vtime_account(struct task_struct *tsk)
         struct thread_info *ti = task_thread_info(tsk);
         u64 timer, system;
 
+        WARN_ON_ONCE(!irqs_disabled());
+
         timer = S390_lowcore.last_update_timer;
         S390_lowcore.last_update_timer = get_vtimer();
         S390_lowcore.system_timer += timer - S390_lowcore.last_update_timer;
@@ -140,6 +147,10 @@ void vtime_account(struct task_struct *tsk)
 }
 EXPORT_SYMBOL_GPL(vtime_account);
 
+void vtime_account_system(struct task_struct *tsk)
+__attribute__((alias("vtime_account")));
+EXPORT_SYMBOL_GPL(vtime_account_system);
+
 void __kprobes vtime_stop_cpu(void)
 {
         struct s390_idle_data *idle = &__get_cpu_var(s390_idle);
diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c
index ecced9d18986..d91a95568002 100644
--- a/arch/s390/kvm/kvm-s390.c
+++ b/arch/s390/kvm/kvm-s390.c
@@ -608,9 +608,7 @@ static int __vcpu_run(struct kvm_vcpu *vcpu)
                 kvm_s390_deliver_pending_interrupts(vcpu);
 
         vcpu->arch.sie_block->icptcode = 0;
-        local_irq_disable();
         kvm_guest_enter();
-        local_irq_enable();
         VCPU_EVENT(vcpu, 6, "entering sie flags %x",
                    atomic_read(&vcpu->arch.sie_block->cpuflags));
         trace_kvm_s390_sie_enter(vcpu,
@@ -629,9 +627,7 @@ static int __vcpu_run(struct kvm_vcpu *vcpu)
         VCPU_EVENT(vcpu, 6, "exit sie icptcode %d",
                    vcpu->arch.sie_block->icptcode);
         trace_kvm_s390_sie_exit(vcpu, vcpu->arch.sie_block->icptcode);
-        local_irq_disable();
         kvm_guest_exit();
-        local_irq_enable();
 
         memcpy(&vcpu->run->s.regs.gprs[14], &vcpu->arch.sie_block->gg14, 16);
         return rc;