Merge branch 'mpi-master' into wip-k-fmlpwip-k-fmlp

Conflicts:
	litmus/sched_cedf.c

1 files changed, 99 insertions, 7 deletions

diff --git a/fs/timerfd.c b/fs/timerfd.c
index b86ab8eff79a..dffeb3795af1 100644
--- a/fs/timerfd.c
+++ b/fs/timerfd.c
@@ -22,16 +22,24 @@
 #include <linux/anon_inodes.h>
 #include <linux/timerfd.h>
 #include <linux/syscalls.h>
+#include <linux/rcupdate.h>
 
 struct timerfd_ctx {
         struct hrtimer tmr;
         ktime_t tintv;
+        ktime_t moffs;
         wait_queue_head_t wqh;
         u64 ticks;
         int expired;
         int clockid;
+        struct rcu_head rcu;
+        struct list_head clist;
+        bool might_cancel;
 };
 
+static LIST_HEAD(cancel_list);
+static DEFINE_SPINLOCK(cancel_lock);
+
 /*
  * This gets called when the timer event triggers. We set the "expired"
  * flag, but we do not re-arm the timer (in case it's necessary,
@@ -51,6 +59,66 @@ static enum hrtimer_restart timerfd_tmrproc(struct hrtimer *htmr)
         return HRTIMER_NORESTART;
 }
 
+/*
+ * Called when the clock was set to cancel the timers in the cancel
+ * list. This will wake up processes waiting on these timers. The
+ * wake-up requires ctx->ticks to be non zero, therefore we increment
+ * it before calling wake_up_locked().
+ */
+void timerfd_clock_was_set(void)
+{
+        ktime_t moffs = ktime_get_monotonic_offset();
+        struct timerfd_ctx *ctx;
+        unsigned long flags;
+
+        rcu_read_lock();
+        list_for_each_entry_rcu(ctx, &cancel_list, clist) {
+                if (!ctx->might_cancel)
+                        continue;
+                spin_lock_irqsave(&ctx->wqh.lock, flags);
+                if (ctx->moffs.tv64 != moffs.tv64) {
+                        ctx->moffs.tv64 = KTIME_MAX;
+                        ctx->ticks++;
+                        wake_up_locked(&ctx->wqh);
+                }
+                spin_unlock_irqrestore(&ctx->wqh.lock, flags);
+        }
+        rcu_read_unlock();
+}
+
+static void timerfd_remove_cancel(struct timerfd_ctx *ctx)
+{
+        if (ctx->might_cancel) {
+                ctx->might_cancel = false;
+                spin_lock(&cancel_lock);
+                list_del_rcu(&ctx->clist);
+                spin_unlock(&cancel_lock);
+        }
+}
+
+static bool timerfd_canceled(struct timerfd_ctx *ctx)
+{
+        if (!ctx->might_cancel || ctx->moffs.tv64 != KTIME_MAX)
+                return false;
+        ctx->moffs = ktime_get_monotonic_offset();
+        return true;
+}
+
+static void timerfd_setup_cancel(struct timerfd_ctx *ctx, int flags)
+{
+        if (ctx->clockid == CLOCK_REALTIME && (flags & TFD_TIMER_ABSTIME) &&
+            (flags & TFD_TIMER_CANCEL_ON_SET)) {
+                if (!ctx->might_cancel) {
+                        ctx->might_cancel = true;
+                        spin_lock(&cancel_lock);
+                        list_add_rcu(&ctx->clist, &cancel_list);
+                        spin_unlock(&cancel_lock);
+                }
+        } else if (ctx->might_cancel) {
+                timerfd_remove_cancel(ctx);
+        }
+}
+
 static ktime_t timerfd_get_remaining(struct timerfd_ctx *ctx)
 {
         ktime_t remaining;
@@ -59,11 +127,12 @@ static ktime_t timerfd_get_remaining(struct timerfd_ctx *ctx)
         return remaining.tv64 < 0 ? ktime_set(0, 0): remaining;
 }
 
-static void timerfd_setup(struct timerfd_ctx *ctx, int flags,
-                          const struct itimerspec *ktmr)
+static int timerfd_setup(struct timerfd_ctx *ctx, int flags,
+                         const struct itimerspec *ktmr)
 {
         enum hrtimer_mode htmode;
         ktime_t texp;
+        int clockid = ctx->clockid;
 
         htmode = (flags & TFD_TIMER_ABSTIME) ?
                 HRTIMER_MODE_ABS: HRTIMER_MODE_REL;
@@ -72,19 +141,24 @@ static void timerfd_setup(struct timerfd_ctx *ctx, int flags,
         ctx->expired = 0;
         ctx->ticks = 0;
         ctx->tintv = timespec_to_ktime(ktmr->it_interval);
-        hrtimer_init(&ctx->tmr, ctx->clockid, htmode);
+        hrtimer_init(&ctx->tmr, clockid, htmode);
         hrtimer_set_expires(&ctx->tmr, texp);
         ctx->tmr.function = timerfd_tmrproc;
-        if (texp.tv64 != 0)
+        if (texp.tv64 != 0) {
                 hrtimer_start(&ctx->tmr, texp, htmode);
+                if (timerfd_canceled(ctx))
+                        return -ECANCELED;
+        }
+        return 0;
 }
 
 static int timerfd_release(struct inode *inode, struct file *file)
 {
         struct timerfd_ctx *ctx = file->private_data;
 
+        timerfd_remove_cancel(ctx);
         hrtimer_cancel(&ctx->tmr);
-        kfree(ctx);
+        kfree_rcu(ctx, rcu);
         return 0;
 }
 
@@ -118,8 +192,21 @@ static ssize_t timerfd_read(struct file *file, char __user *buf, size_t count,
                 res = -EAGAIN;
         else
                 res = wait_event_interruptible_locked_irq(ctx->wqh, ctx->ticks);
+
+        /*
+         * If clock has changed, we do not care about the
+         * ticks and we do not rearm the timer. Userspace must
+         * reevaluate anyway.
+         */
+        if (timerfd_canceled(ctx)) {
+                ctx->ticks = 0;
+                ctx->expired = 0;
+                res = -ECANCELED;
+        }
+
         if (ctx->ticks) {
                 ticks = ctx->ticks;
+
                 if (ctx->expired && ctx->tintv.tv64) {
                         /*
                          * If tintv.tv64 != 0, this is a periodic timer that
@@ -144,6 +231,7 @@ static const struct file_operations timerfd_fops = {
         .release        = timerfd_release,
         .poll           = timerfd_poll,
         .read           = timerfd_read,
+        .llseek         = noop_llseek,
 };
 
 static struct file *timerfd_fget(int fd)
@@ -182,6 +270,7 @@ SYSCALL_DEFINE2(timerfd_create, int, clockid, int, flags)
         init_waitqueue_head(&ctx->wqh);
         ctx->clockid = clockid;
         hrtimer_init(&ctx->tmr, clockid, HRTIMER_MODE_ABS);
+        ctx->moffs = ktime_get_monotonic_offset();
 
         ufd = anon_inode_getfd("[timerfd]", &timerfd_fops, ctx,
                                O_RDWR | (flags & TFD_SHARED_FCNTL_FLAGS));
@@ -198,6 +287,7 @@ SYSCALL_DEFINE4(timerfd_settime, int, ufd, int, flags,
         struct file *file;
         struct timerfd_ctx *ctx;
         struct itimerspec ktmr, kotmr;
+        int ret;
 
         if (copy_from_user(&ktmr, utmr, sizeof(ktmr)))
                 return -EFAULT;
@@ -212,6 +302,8 @@ SYSCALL_DEFINE4(timerfd_settime, int, ufd, int, flags,
                 return PTR_ERR(file);
         ctx = file->private_data;
 
+        timerfd_setup_cancel(ctx, flags);
+
         /*
          * We need to stop the existing timer before reprogramming
          * it to the new values.
@@ -239,14 +331,14 @@ SYSCALL_DEFINE4(timerfd_settime, int, ufd, int, flags,
         /*
          * Re-program the timer to the new value ...
          */
-        timerfd_setup(ctx, flags, &ktmr);
+        ret = timerfd_setup(ctx, flags, &ktmr);
 
         spin_unlock_irq(&ctx->wqh.lock);
         fput(file);
         if (otmr && copy_to_user(otmr, &kotmr, sizeof(kotmr)))
                 return -EFAULT;
 
-        return 0;
+        return ret;
 }
 
 SYSCALL_DEFINE2(timerfd_gettime, int, ufd, struct itimerspec __user *, otmr)