Merge branch 'for-linus' of git://git390.osdl.marist.edu/pub/scm/linux-2.6

* 'for-linus' of git://git390.osdl.marist.edu/pub/scm/linux-2.6:
  [S390] qdio: FCP/SCSI write I/O stagnates on LPAR
  [S390] Fix futex_atomic_cmpxchg_std inline assembly.
  [S390] dcss: Fix Unlikely(x) != y
  [S390] sclp: clean up send/receive naming scheme
  [S390] etr: fix compile error on !SMP
  [S390] qdio: fix qdio_activate timeout handling.
  [S390] Initialize per cpu lowcores on cpu hotplug.
  [S390] find bit corner case.
  [S390] dasd: fix locking in __dasd_device_process_final_queue
  [S390] Make sure enabled wait psw is loaded in default_idle.
  [S390] Let NR_CPUS default to 32/64 on s390/s390x.
  [S390] cio: Do timed recovery on workqueue.
  [S390] cio: Remember to initialize recovery_lock.

5 files changed, 53 insertions, 28 deletions

diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
index 92a4f7b4323a..b21444b681b6 100644
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -100,7 +100,8 @@ config NR_CPUS
         int "Maximum number of CPUs (2-64)"
         range 2 64
         depends on SMP
-        default "32"
+        default "32" if !64BIT
+        default "64" if 64BIT
         help
           This allows you to specify the maximum number of CPUs which this
           kernel will support.  The maximum supported value is 64 and the
diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c
index a6a4729e0e94..1c59ec161cf8 100644
--- a/arch/s390/kernel/process.c
+++ b/arch/s390/kernel/process.c
@@ -114,24 +114,27 @@ extern void s390_handle_mcck(void);
 static void default_idle(void)
 {
         int cpu, rc;
+        int nr_calls = 0;
+        void *hcpu;
 #ifdef CONFIG_SMP
         struct s390_idle_data *idle;
 #endif
 
         /* CPU is going idle. */
         cpu = smp_processor_id();
-
+        hcpu = (void *)(long)cpu;
         local_irq_disable();
         if (need_resched()) {
                 local_irq_enable();
                 return;
         }
 
-        rc = atomic_notifier_call_chain(&idle_chain,
-                                        S390_CPU_IDLE, (void *)(long) cpu);
-        if (rc != NOTIFY_OK && rc != NOTIFY_DONE)
-                BUG();
-        if (rc != NOTIFY_OK) {
+        rc = __atomic_notifier_call_chain(&idle_chain, S390_CPU_IDLE, hcpu, -1,
+                                          &nr_calls);
+        if (rc == NOTIFY_BAD) {
+                nr_calls--;
+                __atomic_notifier_call_chain(&idle_chain, S390_CPU_NOT_IDLE,
+                                             hcpu, nr_calls, NULL);
                 local_irq_enable();
                 return;
         }
diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index 85060659fb12..818bd09c0260 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -626,13 +626,17 @@ static int __cpuinit smp_alloc_lowcore(int cpu)
         if (!lowcore)
                 return -ENOMEM;
         async_stack = __get_free_pages(GFP_KERNEL, ASYNC_ORDER);
-        if (!async_stack)
-                goto out_async_stack;
         panic_stack = __get_free_page(GFP_KERNEL);
-        if (!panic_stack)
-                goto out_panic_stack;
-
-        *lowcore = S390_lowcore;
+        if (!panic_stack || !async_stack)
+                goto out;
+        /*
+         * Only need to copy the first 512 bytes from address 0. But since
+         * the compiler emits a warning if src == NULL for memcpy use copy_page
+         * instead. Copies more than needed but this code is not performance
+         * critical.
+         */
+        copy_page(lowcore, &S390_lowcore);
+        memset((void *)lowcore + 512, 0, sizeof(*lowcore) - 512);
         lowcore->async_stack = async_stack + ASYNC_SIZE;
         lowcore->panic_stack = panic_stack + PAGE_SIZE;
 
@@ -653,9 +657,8 @@ static int __cpuinit smp_alloc_lowcore(int cpu)
 out_save_area:
         free_page(panic_stack);
 #endif
-out_panic_stack:
+out:
         free_pages(async_stack, ASYNC_ORDER);
-out_async_stack:
         free_pages((unsigned long) lowcore, lc_order);
         return -ENOMEM;
 }
@@ -719,8 +722,8 @@ int __cpuinit __cpu_up(unsigned int cpu)
         cpu_lowcore->percpu_offset = __per_cpu_offset[cpu];
         cpu_lowcore->current_task = (unsigned long) idle;
         cpu_lowcore->cpu_data.cpu_nr = cpu;
-        cpu_lowcore->softirq_pending = 0;
-        cpu_lowcore->ext_call_fast = 0;
+        cpu_lowcore->kernel_asce = S390_lowcore.kernel_asce;
+        cpu_lowcore->ipl_device = S390_lowcore.ipl_device;
         eieio();
 
         while (signal_processor(cpu, sigp_restart) == sigp_busy)
@@ -797,23 +800,43 @@ void cpu_die(void)
 
 void __init smp_prepare_cpus(unsigned int max_cpus)
 {
+#ifndef CONFIG_64BIT
+        unsigned long save_area = 0;
+#endif
+        unsigned long async_stack, panic_stack;
+        struct _lowcore *lowcore;
         unsigned int cpu;
+        int lc_order;
 
         smp_detect_cpus();
 
         /* request the 0x1201 emergency signal external interrupt */
         if (register_external_interrupt(0x1201, do_ext_call_interrupt) != 0)
                 panic("Couldn't request external interrupt 0x1201");
-        memset(lowcore_ptr, 0, sizeof(lowcore_ptr));
         print_cpu_info(&S390_lowcore.cpu_data);
-        smp_alloc_lowcore(smp_processor_id());
 
+        /* Reallocate current lowcore, but keep its contents. */
+        lc_order = sizeof(long) == 8 ? 1 : 0;
+        lowcore = (void *) __get_free_pages(GFP_KERNEL | GFP_DMA, lc_order);
+        panic_stack = __get_free_page(GFP_KERNEL);
+        async_stack = __get_free_pages(GFP_KERNEL, ASYNC_ORDER);
 #ifndef CONFIG_64BIT
         if (MACHINE_HAS_IEEE)
-                ctl_set_bit(14, 29); /* enable extended save area */
+                save_area = get_zeroed_page(GFP_KERNEL);
 #endif
-        set_prefix((u32)(unsigned long) lowcore_ptr[smp_processor_id()]);
-
+        local_irq_disable();
+        local_mcck_disable();
+        lowcore_ptr[smp_processor_id()] = lowcore;
+        *lowcore = S390_lowcore;
+        lowcore->panic_stack = panic_stack + PAGE_SIZE;
+        lowcore->async_stack = async_stack + ASYNC_SIZE;
+#ifndef CONFIG_64BIT
+        if (MACHINE_HAS_IEEE)
+                lowcore->extended_save_area_addr = (u32) save_area;
+#endif
+        set_prefix((u32)(unsigned long) lowcore);
+        local_mcck_enable();
+        local_irq_enable();
         for_each_possible_cpu(cpu)
                 if (cpu != smp_processor_id())
                         smp_create_idle(cpu);
diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c
index 3bbac1293be4..76a5dd1b4ce9 100644
--- a/arch/s390/kernel/time.c
+++ b/arch/s390/kernel/time.c
@@ -744,7 +744,6 @@ static void etr_adjust_time(unsigned long long clock, unsigned long long delay)
         }
 }
 
-#ifdef CONFIG_SMP
 static void etr_sync_cpu_start(void *dummy)
 {
         int *in_sync = dummy;
@@ -777,7 +776,6 @@ static void etr_sync_cpu_start(void *dummy)
 static void etr_sync_cpu_end(void *dummy)
 {
 }
-#endif /* CONFIG_SMP */
 
 /*
  * Sync the TOD clock using the port refered to by aibp. This port
diff --git a/arch/s390/lib/uaccess_std.c b/arch/s390/lib/uaccess_std.c
index 28c4500a58d0..d2ffbadb51a7 100644
--- a/arch/s390/lib/uaccess_std.c
+++ b/arch/s390/lib/uaccess_std.c
@@ -293,10 +293,10 @@ int futex_atomic_cmpxchg_std(int __user *uaddr, int oldval, int newval)
 
         asm volatile(
                 "   sacf 256\n"
-                "   cs   %1,%4,0(%5)\n"
-                "0: lr   %0,%1\n"
-                "1: sacf 0\n"
-                EX_TABLE(0b,1b)
+                "0: cs   %1,%4,0(%5)\n"
+                "1: lr   %0,%1\n"
+                "2: sacf 0\n"
+                EX_TABLE(0b,2b) EX_TABLE(1b,2b)
                 : "=d" (ret), "+d" (oldval), "=m" (*uaddr)
                 : "0" (-EFAULT), "d" (newval), "a" (uaddr), "m" (*uaddr)
                 : "cc", "memory" );