fence: dma-buf cross-device synchronization (v18)

A fence can be attached to a buffer which is being filled or consumed
by hw, to allow userspace to pass the buffer without waiting to another
device.  For example, userspace can call page_flip ioctl to display the
next frame of graphics after kicking the GPU but while the GPU is still
rendering.  The display device sharing the buffer with the GPU would
attach a callback to get notified when the GPU's rendering-complete IRQ
fires, to update the scan-out address of the display, without having to
wake up userspace.

A driver must allocate a fence context for each execution ring that can
run in parallel. The function for this takes an argument with how many
contexts to allocate:
  + fence_context_alloc()

A fence is transient, one-shot deal.  It is allocated and attached
to one or more dma-buf's.  When the one that attached it is done, with
the pending operation, it can signal the fence:
  + fence_signal()

To have a rough approximation whether a fence is fired, call:
  + fence_is_signaled()

The dma-buf-mgr handles tracking, and waiting on, the fences associated
with a dma-buf.

The one pending on the fence can add an async callback:
  + fence_add_callback()

The callback can optionally be cancelled with:
  + fence_remove_callback()

To wait synchronously, optionally with a timeout:
  + fence_wait()
  + fence_wait_timeout()

When emitting a fence, call:
  + trace_fence_emit()

To annotate that a fence is blocking on another fence, call:
  + trace_fence_annotate_wait_on(fence, on_fence)

A default software-only implementation is provided, which can be used
by drivers attaching a fence to a buffer when they have no other means
for hw sync.  But a memory backed fence is also envisioned, because it
is common that GPU's can write to, or poll on some memory location for
synchronization.  For example:

  fence = custom_get_fence(...);
  if ((seqno_fence = to_seqno_fence(fence)) != NULL) {
    dma_buf *fence_buf = seqno_fence->sync_buf;
    get_dma_buf(fence_buf);

    ... tell the hw the memory location to wait ...
    custom_wait_on(fence_buf, seqno_fence->seqno_ofs, fence->seqno);
  } else {
    /* fall-back to sw sync * /
    fence_add_callback(fence, my_cb);
  }

On SoC platforms, if some other hw mechanism is provided for synchronizing
between IP blocks, it could be supported as an alternate implementation
with it's own fence ops in a similar way.

enable_signaling callback is used to provide sw signaling in case a cpu
waiter is requested or no compatible hardware signaling could be used.

The intention is to provide a userspace interface (presumably via eventfd)
later, to be used in conjunction with dma-buf's mmap support for sw access
to buffers (or for userspace apps that would prefer to do their own
synchronization).

v1: Original
v2: After discussion w/ danvet and mlankhorst on #dri-devel, we decided
    that dma-fence didn't need to care about the sw->hw signaling path
    (it can be handled same as sw->sw case), and therefore the fence->ops
    can be simplified and more handled in the core.  So remove the signal,
    add_callback, cancel_callback, and wait ops, and replace with a simple
    enable_signaling() op which can be used to inform a fence supporting
    hw->hw signaling that one or more devices which do not support hw
    signaling are waiting (and therefore it should enable an irq or do
    whatever is necessary in order that the CPU is notified when the
    fence is passed).
v3: Fix locking fail in attach_fence() and get_fence()
v4: Remove tie-in w/ dma-buf..  after discussion w/ danvet and mlankorst
    we decided that we need to be able to attach one fence to N dma-buf's,
    so using the list_head in dma-fence struct would be problematic.
v5: [ Maarten Lankhorst ] Updated for dma-bikeshed-fence and dma-buf-manager.
v6: [ Maarten Lankhorst ] I removed dma_fence_cancel_callback and some comments
    about checking if fence fired or not. This is broken by design.
    waitqueue_active during destruction is now fatal, since the signaller
    should be holding a reference in enable_signalling until it signalled
    the fence. Pass the original dma_fence_cb along, and call __remove_wait
    in the dma_fence_callback handler, so that no cleanup needs to be
    performed.
v7: [ Maarten Lankhorst ] Set cb->func and only enable sw signaling if
    fence wasn't signaled yet, for example for hardware fences that may
    choose to signal blindly.
v8: [ Maarten Lankhorst ] Tons of tiny fixes, moved __dma_fence_init to
    header and fixed include mess. dma-fence.h now includes dma-buf.h
    All members are now initialized, so kmalloc can be used for
    allocating a dma-fence. More documentation added.
v9: Change compiler bitfields to flags, change return type of
    enable_signaling to bool. Rework dma_fence_wait. Added
    dma_fence_is_signaled and dma_fence_wait_timeout.
    s/dma// and change exports to non GPL. Added fence_is_signaled and
    fence_enable_sw_signaling calls, add ability to override default
    wait operation.
v10: remove event_queue, use a custom list, export try_to_wake_up from
    scheduler. Remove fence lock and use a global spinlock instead,
    this should hopefully remove all the locking headaches I was having
    on trying to implement this. enable_signaling is called with this
    lock held.
v11:
    Use atomic ops for flags, lifting the need for some spin_lock_irqsaves.
    However I kept the guarantee that after fence_signal returns, it is
    guaranteed that enable_signaling has either been called to completion,
    or will not be called any more.

    Add contexts and seqno to base fence implementation. This allows you
    to wait for less fences, by testing for seqno + signaled, and then only
    wait on the later fence.

    Add FENCE_TRACE, FENCE_WARN, and FENCE_ERR. This makes debugging easier.
    An CONFIG_DEBUG_FENCE will be added to turn off the FENCE_TRACE
    spam, and another runtime option can turn it off at runtime.
v12:
    Add CONFIG_FENCE_TRACE. Add missing documentation for the fence->context
    and fence->seqno members.
v13:
    Fixup CONFIG_FENCE_TRACE kconfig description.
    Move fence_context_alloc to fence.
    Simplify fence_later.
    Kill priv member to fence_cb.
v14:
    Remove priv argument from fence_add_callback, oops!
v15:
    Remove priv from documentation.
    Explicitly include linux/atomic.h.
v16:
    Add trace events.
    Import changes required by android syncpoints.
v17:
    Use wake_up_state instead of try_to_wake_up. (Colin Cross)
    Fix up commit description for seqno_fence. (Rob Clark)
v18:
    Rename release_fence to fence_release.
    Move to drivers/dma-buf/.
    Rename __fence_is_signaled and __fence_signal to *_locked.
    Rename __fence_init to fence_init.
    Make fence_default_wait return a signed long, and fix wait ops too.

Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com>
Signed-off-by: Thierry Reding <thierry.reding@gmail.com> #use smp_mb__before_atomic()
Acked-by: Sumit Semwal <sumit.semwal@linaro.org>
Acked-by: Daniel Vetter <daniel@ffwll.ch>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

2 files changed, 432 insertions, 1 deletions

diff --git a/drivers/dma-buf/Makefile b/drivers/dma-buf/Makefile
index 4a4f4c9bacd0..d7825bfe630e 100644
--- a/drivers/dma-buf/Makefile
+++ b/drivers/dma-buf/Makefile
@@ -1 +1 @@
-obj-y := dma-buf.o reservation.o
+obj-y := dma-buf.o fence.o reservation.o
diff --git a/drivers/dma-buf/fence.c b/drivers/dma-buf/fence.c
new file mode 100644
index 000000000000..948bf00d955e
--- /dev/null
+++ b/drivers/dma-buf/fence.c
@@ -0,0 +1,431 @@
+/*
+ * Fence mechanism for dma-buf and to allow for asynchronous dma access
+ *
+ * Copyright (C) 2012 Canonical Ltd
+ * Copyright (C) 2012 Texas Instruments
+ *
+ * Authors:
+ * Rob Clark <robdclark@gmail.com>
+ * Maarten Lankhorst <maarten.lankhorst@canonical.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published by
+ * the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ */
+
+#include <linux/slab.h>
+#include <linux/export.h>
+#include <linux/atomic.h>
+#include <linux/fence.h>
+
+#define CREATE_TRACE_POINTS
+#include <trace/events/fence.h>
+
+EXPORT_TRACEPOINT_SYMBOL(fence_annotate_wait_on);
+EXPORT_TRACEPOINT_SYMBOL(fence_emit);
+
+/**
+ * fence context counter: each execution context should have its own
+ * fence context, this allows checking if fences belong to the same
+ * context or not. One device can have multiple separate contexts,
+ * and they're used if some engine can run independently of another.
+ */
+static atomic_t fence_context_counter = ATOMIC_INIT(0);
+
+/**
+ * fence_context_alloc - allocate an array of fence contexts
+ * @num:	[in]	amount of contexts to allocate
+ *
+ * This function will return the first index of the number of fences allocated.
+ * The fence context is used for setting fence->context to a unique number.
+ */
+unsigned fence_context_alloc(unsigned num)
+{
+	BUG_ON(!num);
+	return atomic_add_return(num, &fence_context_counter) - num;
+}
+EXPORT_SYMBOL(fence_context_alloc);
+
+/**
+ * fence_signal_locked - signal completion of a fence
+ * @fence: the fence to signal
+ *
+ * Signal completion for software callbacks on a fence, this will unblock
+ * fence_wait() calls and run all the callbacks added with
+ * fence_add_callback(). Can be called multiple times, but since a fence
+ * can only go from unsignaled to signaled state, it will only be effective
+ * the first time.
+ *
+ * Unlike fence_signal, this function must be called with fence->lock held.
+ */
+int fence_signal_locked(struct fence *fence)
+{
+	struct fence_cb *cur, *tmp;
+	int ret = 0;
+
+	if (WARN_ON(!fence))
+		return -EINVAL;
+
+	if (!ktime_to_ns(fence->timestamp)) {
+		fence->timestamp = ktime_get();
+		smp_mb__before_atomic();
+	}
+
+	if (test_and_set_bit(FENCE_FLAG_SIGNALED_BIT, &fence->flags)) {
+		ret = -EINVAL;
+
+		/*
+		 * we might have raced with the unlocked fence_signal,
+		 * still run through all callbacks
+		 */
+	} else
+		trace_fence_signaled(fence);
+
+	list_for_each_entry_safe(cur, tmp, &fence->cb_list, node) {
+		list_del_init(&cur->node);
+		cur->func(fence, cur);
+	}
+	return ret;
+}
+EXPORT_SYMBOL(fence_signal_locked);
+
+/**
+ * fence_signal - signal completion of a fence
+ * @fence: the fence to signal
+ *
+ * Signal completion for software callbacks on a fence, this will unblock
+ * fence_wait() calls and run all the callbacks added with
+ * fence_add_callback(). Can be called multiple times, but since a fence
+ * can only go from unsignaled to signaled state, it will only be effective
+ * the first time.
+ */
+int fence_signal(struct fence *fence)
+{
+	unsigned long flags;
+
+	if (!fence)
+		return -EINVAL;
+
+	if (!ktime_to_ns(fence->timestamp)) {
+		fence->timestamp = ktime_get();
+		smp_mb__before_atomic();
+	}
+
+	if (test_and_set_bit(FENCE_FLAG_SIGNALED_BIT, &fence->flags))
+		return -EINVAL;
+
+	trace_fence_signaled(fence);
+
+	if (test_bit(FENCE_FLAG_ENABLE_SIGNAL_BIT, &fence->flags)) {
+		struct fence_cb *cur, *tmp;
+
+		spin_lock_irqsave(fence->lock, flags);
+		list_for_each_entry_safe(cur, tmp, &fence->cb_list, node) {
+			list_del_init(&cur->node);
+			cur->func(fence, cur);
+		}
+		spin_unlock_irqrestore(fence->lock, flags);
+	}
+	return 0;
+}
+EXPORT_SYMBOL(fence_signal);
+
+/**
+ * fence_wait_timeout - sleep until the fence gets signaled
+ * or until timeout elapses
+ * @fence:	[in]	the fence to wait on
+ * @intr:	[in]	if true, do an interruptible wait
+ * @timeout:	[in]	timeout value in jiffies, or MAX_SCHEDULE_TIMEOUT
+ *
+ * Returns -ERESTARTSYS if interrupted, 0 if the wait timed out, or the
+ * remaining timeout in jiffies on success. Other error values may be
+ * returned on custom implementations.
+ *
+ * Performs a synchronous wait on this fence. It is assumed the caller
+ * directly or indirectly (buf-mgr between reservation and committing)
+ * holds a reference to the fence, otherwise the fence might be
+ * freed before return, resulting in undefined behavior.
+ */
+signed long
+fence_wait_timeout(struct fence *fence, bool intr, signed long timeout)
+{
+	signed long ret;
+
+	if (WARN_ON(timeout < 0))
+		return -EINVAL;
+
+	trace_fence_wait_start(fence);
+	ret = fence->ops->wait(fence, intr, timeout);
+	trace_fence_wait_end(fence);
+	return ret;
+}
+EXPORT_SYMBOL(fence_wait_timeout);
+
+void fence_release(struct kref *kref)
+{
+	struct fence *fence =
+			container_of(kref, struct fence, refcount);
+
+	trace_fence_destroy(fence);
+
+	BUG_ON(!list_empty(&fence->cb_list));
+
+	if (fence->ops->release)
+		fence->ops->release(fence);
+	else
+		fence_free(fence);
+}
+EXPORT_SYMBOL(fence_release);
+
+void fence_free(struct fence *fence)
+{
+	kfree(fence);
+}
+EXPORT_SYMBOL(fence_free);
+
+/**
+ * fence_enable_sw_signaling - enable signaling on fence
+ * @fence:	[in]	the fence to enable
+ *
+ * this will request for sw signaling to be enabled, to make the fence
+ * complete as soon as possible
+ */
+void fence_enable_sw_signaling(struct fence *fence)
+{
+	unsigned long flags;
+
+	if (!test_and_set_bit(FENCE_FLAG_ENABLE_SIGNAL_BIT, &fence->flags) &&
+	    !test_bit(FENCE_FLAG_SIGNALED_BIT, &fence->flags)) {
+		trace_fence_enable_signal(fence);
+
+		spin_lock_irqsave(fence->lock, flags);
+
+		if (!fence->ops->enable_signaling(fence))
+			fence_signal_locked(fence);
+
+		spin_unlock_irqrestore(fence->lock, flags);
+	}
+}
+EXPORT_SYMBOL(fence_enable_sw_signaling);
+
+/**
+ * fence_add_callback - add a callback to be called when the fence
+ * is signaled
+ * @fence:	[in]	the fence to wait on
+ * @cb:		[in]	the callback to register
+ * @func:	[in]	the function to call
+ *
+ * cb will be initialized by fence_add_callback, no initialization
+ * by the caller is required. Any number of callbacks can be registered
+ * to a fence, but a callback can only be registered to one fence at a time.
+ *
+ * Note that the callback can be called from an atomic context.  If
+ * fence is already signaled, this function will return -ENOENT (and
+ * *not* call the callback)
+ *
+ * Add a software callback to the fence. Same restrictions apply to
+ * refcount as it does to fence_wait, however the caller doesn't need to
+ * keep a refcount to fence afterwards: when software access is enabled,
+ * the creator of the fence is required to keep the fence alive until
+ * after it signals with fence_signal. The callback itself can be called
+ * from irq context.
+ *
+ */
+int fence_add_callback(struct fence *fence, struct fence_cb *cb,
+		       fence_func_t func)
+{
+	unsigned long flags;
+	int ret = 0;
+	bool was_set;
+
+	if (WARN_ON(!fence || !func))
+		return -EINVAL;
+
+	if (test_bit(FENCE_FLAG_SIGNALED_BIT, &fence->flags)) {
+		INIT_LIST_HEAD(&cb->node);
+		return -ENOENT;
+	}
+
+	spin_lock_irqsave(fence->lock, flags);
+
+	was_set = test_and_set_bit(FENCE_FLAG_ENABLE_SIGNAL_BIT, &fence->flags);
+
+	if (test_bit(FENCE_FLAG_SIGNALED_BIT, &fence->flags))
+		ret = -ENOENT;
+	else if (!was_set) {
+		trace_fence_enable_signal(fence);
+
+		if (!fence->ops->enable_signaling(fence)) {
+			fence_signal_locked(fence);
+			ret = -ENOENT;
+		}
+	}
+
+	if (!ret) {
+		cb->func = func;
+		list_add_tail(&cb->node, &fence->cb_list);
+	} else
+		INIT_LIST_HEAD(&cb->node);
+	spin_unlock_irqrestore(fence->lock, flags);
+
+	return ret;
+}
+EXPORT_SYMBOL(fence_add_callback);
+
+/**
+ * fence_remove_callback - remove a callback from the signaling list
+ * @fence:	[in]	the fence to wait on
+ * @cb:		[in]	the callback to remove
+ *
+ * Remove a previously queued callback from the fence. This function returns
+ * true if the callback is succesfully removed, or false if the fence has
+ * already been signaled.
+ *
+ * *WARNING*:
+ * Cancelling a callback should only be done if you really know what you're
+ * doing, since deadlocks and race conditions could occur all too easily. For
+ * this reason, it should only ever be done on hardware lockup recovery,
+ * with a reference held to the fence.
+ */
+bool
+fence_remove_callback(struct fence *fence, struct fence_cb *cb)
+{
+	unsigned long flags;
+	bool ret;
+
+	spin_lock_irqsave(fence->lock, flags);
+
+	ret = !list_empty(&cb->node);
+	if (ret)
+		list_del_init(&cb->node);
+
+	spin_unlock_irqrestore(fence->lock, flags);
+
+	return ret;
+}
+EXPORT_SYMBOL(fence_remove_callback);
+
+struct default_wait_cb {
+	struct fence_cb base;
+	struct task_struct *task;
+};
+
+static void
+fence_default_wait_cb(struct fence *fence, struct fence_cb *cb)
+{
+	struct default_wait_cb *wait =
+		container_of(cb, struct default_wait_cb, base);
+
+	wake_up_state(wait->task, TASK_NORMAL);
+}
+
+/**
+ * fence_default_wait - default sleep until the fence gets signaled
+ * or until timeout elapses
+ * @fence:	[in]	the fence to wait on
+ * @intr:	[in]	if true, do an interruptible wait
+ * @timeout:	[in]	timeout value in jiffies, or MAX_SCHEDULE_TIMEOUT
+ *
+ * Returns -ERESTARTSYS if interrupted, 0 if the wait timed out, or the
+ * remaining timeout in jiffies on success.
+ */
+signed long
+fence_default_wait(struct fence *fence, bool intr, signed long timeout)
+{
+	struct default_wait_cb cb;
+	unsigned long flags;
+	signed long ret = timeout;
+	bool was_set;
+
+	if (test_bit(FENCE_FLAG_SIGNALED_BIT, &fence->flags))
+		return timeout;
+
+	spin_lock_irqsave(fence->lock, flags);
+
+	if (intr && signal_pending(current)) {
+		ret = -ERESTARTSYS;
+		goto out;
+	}
+
+	was_set = test_and_set_bit(FENCE_FLAG_ENABLE_SIGNAL_BIT, &fence->flags);
+
+	if (test_bit(FENCE_FLAG_SIGNALED_BIT, &fence->flags))
+		goto out;
+
+	if (!was_set) {
+		trace_fence_enable_signal(fence);
+
+		if (!fence->ops->enable_signaling(fence)) {
+			fence_signal_locked(fence);
+			goto out;
+		}
+	}
+
+	cb.base.func = fence_default_wait_cb;
+	cb.task = current;
+	list_add(&cb.base.node, &fence->cb_list);
+
+	while (!test_bit(FENCE_FLAG_SIGNALED_BIT, &fence->flags) && ret > 0) {
+		if (intr)
+			__set_current_state(TASK_INTERRUPTIBLE);
+		else
+			__set_current_state(TASK_UNINTERRUPTIBLE);
+		spin_unlock_irqrestore(fence->lock, flags);
+
+		ret = schedule_timeout(ret);
+
+		spin_lock_irqsave(fence->lock, flags);
+		if (ret > 0 && intr && signal_pending(current))
+			ret = -ERESTARTSYS;
+	}
+
+	if (!list_empty(&cb.base.node))
+		list_del(&cb.base.node);
+	__set_current_state(TASK_RUNNING);
+
+out:
+	spin_unlock_irqrestore(fence->lock, flags);
+	return ret;
+}
+EXPORT_SYMBOL(fence_default_wait);
+
+/**
+ * fence_init - Initialize a custom fence.
+ * @fence:	[in]	the fence to initialize
+ * @ops:	[in]	the fence_ops for operations on this fence
+ * @lock:	[in]	the irqsafe spinlock to use for locking this fence
+ * @context:	[in]	the execution context this fence is run on
+ * @seqno:	[in]	a linear increasing sequence number for this context
+ *
+ * Initializes an allocated fence, the caller doesn't have to keep its
+ * refcount after committing with this fence, but it will need to hold a
+ * refcount again if fence_ops.enable_signaling gets called. This can
+ * be used for other implementing other types of fence.
+ *
+ * context and seqno are used for easy comparison between fences, allowing
+ * to check which fence is later by simply using fence_later.
+ */
+void
+fence_init(struct fence *fence, const struct fence_ops *ops,
+	     spinlock_t *lock, unsigned context, unsigned seqno)
+{
+	BUG_ON(!lock);
+	BUG_ON(!ops || !ops->wait || !ops->enable_signaling ||
+	       !ops->get_driver_name || !ops->get_timeline_name);
+
+	kref_init(&fence->refcount);
+	fence->ops = ops;
+	INIT_LIST_HEAD(&fence->cb_list);
+	fence->lock = lock;
+	fence->context = context;
+	fence->seqno = seqno;
+	fence->flags = 0UL;
+
+	trace_fence_init(fence);
+}
+EXPORT_SYMBOL(fence_init);