summaryrefslogtreecommitdiffstats
path: root/include/rdma
diff options
context:
space:
mode:
authorMatan Barak <matanb@mellanox.com>2017-08-03 16:07:00 +0300
committerDoug Ledford <dledford@redhat.com>2017-08-31 08:35:10 -0400
commit118620d3686b2d624f9a5019f2f14c64cf50d21a (patch)
tree89b4ceb75fe00309fb0ec6f474a01e2b82a9eb64 /include/rdma
parent09e3ebf8c193d3f154c4ffb7cb18995df0243bc6 (diff)
downloadlinux-118620d3686b2d624f9a5019f2f14c64cf50d21a.tar.bz2
IB/core: Add uverbs merge trees functionality
Different drivers support different features and even subset of the common uverbs implementation. Currently, this is handled as bitmask in every driver that represents which kind of methods it supports, but doesn't go down to attributes granularity. Moreover, drivers might want to add their specific types, methods and attributes to let their user-space counter-parts be exposed to some more efficient abstractions. It means that existence of different features is validated syntactically via the parsing infrastructure rather than using a complex in-handler logic. In order to do that, we allow defining features and abstractions as parsing trees. These per-feature parsing tree could be merged to an efficient (perfect-hash based) parsing tree, which is later used by the parsing infrastructure. To sum it up, this makes a parse tree unique for a device and represents only the features this particular device supports. This is done by having a root specification tree per feature. Before a device registers itself as an IB device, it merges all these trees into one parsing tree. This parsing tree is used to parse all user-space commands. A future user-space application could read this parse tree. This tree represents which objects, methods and attributes are supported by this device. This is based on the idea of Jason Gunthorpe <jgunthorpe@obsidianresearch.com> Signed-off-by: Matan Barak <matanb@mellanox.com> Reviewed-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
Diffstat (limited to 'include/rdma')
-rw-r--r--include/rdma/uverbs_ioctl.h40
1 files changed, 39 insertions, 1 deletions
diff --git a/include/rdma/uverbs_ioctl.h b/include/rdma/uverbs_ioctl.h
index 2e8925434d74..cf5b238d2d81 100644
--- a/include/rdma/uverbs_ioctl.h
+++ b/include/rdma/uverbs_ioctl.h
@@ -235,5 +235,43 @@ static inline bool uverbs_attr_is_valid_in_hash(const struct uverbs_attr_bundle_
return test_bit(idx, attrs_hash->valid_bitmap);
}
-#endif
+/* =================================================
+ * Definitions -> Specs infrastructure
+ * =================================================
+ */
+
+/*
+ * uverbs_alloc_spec_tree - Merges different common and driver specific feature
+ * into one parsing tree that every uverbs command will be parsed upon.
+ *
+ * @num_trees: Number of trees in the array @trees.
+ * @trees: Array of pointers to tree root definitions to merge. Each such tree
+ * possibly contains objects, methods and attributes definitions.
+ *
+ * Returns:
+ * uverbs_root_spec *: The root of the merged parsing tree.
+ * On error, we return an error code. Error is checked via IS_ERR.
+ *
+ * The following merges could take place:
+ * a. Two trees representing the same method with different handler
+ * -> We take the handler of the tree that its handler != NULL
+ * and its index in the trees array is greater. The incentive for that
+ * is that developers are expected to first merge common trees and then
+ * merge trees that gives specialized the behaviour.
+ * b. Two trees representing the same object with different
+ * type_attrs (struct uverbs_obj_type):
+ * -> We take the type_attrs of the tree that its type_attr != NULL
+ * and its index in the trees array is greater. This could be used
+ * in order to override the free function, allocation size, etc.
+ * c. Two trees representing the same method attribute (same id but possibly
+ * different attributes):
+ * -> ERROR (-ENOENT), we believe that's not the programmer's intent.
+ *
+ * An object without any methods is considered invalid and will abort the
+ * function with -ENOENT error.
+ */
+struct uverbs_root_spec *uverbs_alloc_spec_tree(unsigned int num_trees,
+ const struct uverbs_object_tree_def **trees);
+void uverbs_free_spec_tree(struct uverbs_root_spec *root);
+#endif