dma-buf: generalize dma_fence unwrap & merging v3

Introduce a dma_fence_unwrap_merge() macro which allows to unwrap fences
which potentially can be containers as well and then merge them back
together into a flat dma_fence_array.

v2: rename the function, add some more comments about how the wrapper is
    used, move filtering of signaled fences into the unwrap iterator,
    add complex selftest which covers more cases.
v3: fix signaled fence filtering once more

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: https://patchwork.freedesktop.org/patch/msgid/20220518135844.3338-5-christian.koenig@amd.com

1 files changed, 103 insertions, 0 deletions

diff --git a/drivers/dma-buf/dma-fence-unwrap.c b/drivers/dma-buf/dma-fence-unwrap.c
index 711be125428c..502a65ea6d44 100644
--- a/drivers/dma-buf/dma-fence-unwrap.c
+++ b/drivers/dma-buf/dma-fence-unwrap.c
@@ -11,6 +11,7 @@
 #include <linux/dma-fence-array.h>
 #include <linux/dma-fence-chain.h>
 #include <linux/dma-fence-unwrap.h>
+#include <linux/slab.h>
 
 /* Internal helper to start new array iteration, don't use directly */
 static struct dma_fence *
@@ -57,3 +58,105 @@ struct dma_fence *dma_fence_unwrap_next(struct dma_fence_unwrap *cursor)
 	return __dma_fence_unwrap_array(cursor);
 }
 EXPORT_SYMBOL_GPL(dma_fence_unwrap_next);
+
+/* Implementation for the dma_fence_merge() marco, don't use directly */
+struct dma_fence *__dma_fence_unwrap_merge(unsigned int num_fences,
+					   struct dma_fence **fences,
+					   struct dma_fence_unwrap *iter)
+{
+	struct dma_fence_array *result;
+	struct dma_fence *tmp, **array;
+	unsigned int i;
+	size_t count;
+
+	count = 0;
+	for (i = 0; i < num_fences; ++i) {
+		dma_fence_unwrap_for_each(tmp, &iter[i], fences[i])
+			++count;
+	}
+
+	if (count == 0)
+		return dma_fence_get_stub();
+
+	array = kmalloc_array(count, sizeof(*array), GFP_KERNEL);
+	if (!array)
+		return NULL;
+
+	/*
+	 * This trashes the input fence array and uses it as position for the
+	 * following merge loop. This works because the dma_fence_merge()
+	 * wrapper macro is creating this temporary array on the stack together
+	 * with the iterators.
+	 */
+	for (i = 0; i < num_fences; ++i)
+		fences[i] = dma_fence_unwrap_first(fences[i], &iter[i]);
+
+	count = 0;
+	do {
+		unsigned int sel;
+
+restart:
+		tmp = NULL;
+		for (i = 0; i < num_fences; ++i) {
+			struct dma_fence *next;
+
+			while (fences[i] && dma_fence_is_signaled(fences[i]))
+				fences[i] = dma_fence_unwrap_next(&iter[i]);
+
+			next = fences[i];
+			if (!next)
+				continue;
+
+			/*
+			 * We can't guarantee that inpute fences are ordered by
+			 * context, but it is still quite likely when this
+			 * function is used multiple times. So attempt to order
+			 * the fences by context as we pass over them and merge
+			 * fences with the same context.
+			 */
+			if (!tmp || tmp->context > next->context) {
+				tmp = next;
+				sel = i;
+
+			} else if (tmp->context < next->context) {
+				continue;
+
+			} else if (dma_fence_is_later(tmp, next)) {
+				fences[i] = dma_fence_unwrap_next(&iter[i]);
+				goto restart;
+			} else {
+				fences[sel] = dma_fence_unwrap_next(&iter[sel]);
+				goto restart;
+			}
+		}
+
+		if (tmp) {
+			array[count++] = dma_fence_get(tmp);
+			fences[sel] = dma_fence_unwrap_next(&iter[sel]);
+		}
+	} while (tmp);
+
+	if (count == 0) {
+		tmp = dma_fence_get_stub();
+		goto return_tmp;
+	}
+
+	if (count == 1) {
+		tmp = array[0];
+		goto return_tmp;
+	}
+
+	result = dma_fence_array_create(count, array,
+					dma_fence_context_alloc(1),
+					1, false);
+	if (!result) {
+		tmp = NULL;
+		goto return_tmp;
+	}
+	return &result->base;
+
+return_tmp:
+	kfree(array);
+	return tmp;
+}
+EXPORT_SYMBOL_GPL(__dma_fence_unwrap_merge);