egraph-based midend: draw the rest of the owl (productionized). (#4953)
* egraph-based midend: draw the rest of the owl. * Rename `egg` submodule of cranelift-codegen to `egraph`. * Apply some feedback from @jsharp during code walkthrough. * Remove recursion from find_best_node by doing a single pass. Rather than recursively computing the lowest-cost node for a given eclass and memoizing the answer at each eclass node, we can do a single forward pass; because every eclass node refers only to earlier nodes, this is sufficient. The behavior may slightly differ from the earlier behavior because we cannot short-circuit costs to zero once a node is elaborated; but in practice this should not matter. * Make elaboration non-recursive. Use an explicit stack instead (with `ElabStackEntry` entries, alongside a result stack). * Make elaboration traversal of the domtree non-recursive/stack-safe. * Work analysis logic in Cranelift-side egraph glue into a general analysis framework in cranelift-egraph. * Apply static recursion limit to rule application. * Fix aarch64 wrt dynamic-vector support -- broken rebase. * Topo-sort cranelift-egraph before cranelift-codegen in publish script, like the comment instructs me to! * Fix multi-result call testcase. * Include `cranelift-egraph` in `PUBLISHED_CRATES`. * Fix atomic_rmw: not really a load. * Remove now-unnecessary PartialOrd/Ord derivations. * Address some code-review comments. * Review feedback. * Review feedback. * No overlap in mid-end rules, because we are defining a multi-constructor. * rustfmt * Review feedback. * Review feedback. * Review feedback. * Review feedback. * Remove redundant `mut`. * Add comment noting what rules can do. * Review feedback. * Clarify comment wording. * Update `has_memory_fence_semantics`. * Apply @jameysharp's improved loop-level computation. Co-authored-by: Jamey Sharp <jamey@minilop.net> * Fix suggestion commit. * Fix off-by-one in new loop-nest analysis. * Review feedback. * Review feedback. * Review feedback. * Use `Default`, not `std::default::Default`, as per @fitzgen Co-authored-by: Nick Fitzgerald <fitzgen@gmail.com> * Apply @fitzgen's comment elaboration to a doc-comment. Co-authored-by: Nick Fitzgerald <fitzgen@gmail.com> * Add stat for hitting the rewrite-depth limit. * Some code motion in split prelude to make the diff a little clearer wrt `main`. * Take @jameysharp's suggested `try_into()` usage for blockparam indices. Co-authored-by: Jamey Sharp <jamey@minilop.net> * Take @jameysharp's suggestion to avoid double-match on load op. Co-authored-by: Jamey Sharp <jamey@minilop.net> * Fix suggestion (add import). * Review feedback. * Fix stack_load handling. * Remove redundant can_store case. * Take @jameysharp's suggested improvement to FuncEGraph::build() logic Co-authored-by: Jamey Sharp <jamey@minilop.net> * Tweaks to FuncEGraph::build() on top of suggestion. * Take @jameysharp's suggested clarified condition Co-authored-by: Jamey Sharp <jamey@minilop.net> * Clean up after suggestion (unused variable). * Fix loop analysis. * loop level asserts * Revert constant-space loop analysis -- edge cases were incorrect, so let's go with the simple thing for now. * Take @jameysharp's suggestion re: result_tys Co-authored-by: Jamey Sharp <jamey@minilop.net> * Fix up after suggestion * Take @jameysharp's suggestion to use fold rather than reduce Co-authored-by: Jamey Sharp <jamey@minilop.net> * Fixup after suggestion * Take @jameysharp's suggestion to remove elaborate_eclass_use's return value. * Clarifying comment in terminator insts. Co-authored-by: Jamey Sharp <jamey@minilop.net> Co-authored-by: Nick Fitzgerald <fitzgen@gmail.com>
This commit is contained in:
61
cranelift/codegen/src/prelude_opt.isle
Normal file
61
cranelift/codegen/src/prelude_opt.isle
Normal file
@@ -0,0 +1,61 @@
|
||||
;; Prelude definitions specific to the mid-end.
|
||||
|
||||
;;;;; eclass and enode access ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
|
||||
|
||||
;; An eclass ID.
|
||||
(type Id (primitive Id))
|
||||
|
||||
;; What is the type of an eclass (if a single type)?
|
||||
(decl eclass_type (Type) Id)
|
||||
(extern extractor eclass_type eclass_type)
|
||||
|
||||
;; Helper to wrap an Id-matching pattern and extract type.
|
||||
(decl has_type (Type Id) Id)
|
||||
(extractor (has_type ty id)
|
||||
(and (eclass_type ty)
|
||||
id))
|
||||
|
||||
;; Extract any node(s) for the given eclass ID.
|
||||
(decl multi enodes (Type InstructionImms IdArray) Id)
|
||||
(extern extractor enodes enodes_etor)
|
||||
|
||||
;; Construct a pure node, returning a new (or deduplicated
|
||||
;; already-existing) eclass ID.
|
||||
(decl pure_enode (Type InstructionImms IdArray) Id)
|
||||
(extern constructor pure_enode pure_enode_ctor)
|
||||
|
||||
;; Type of an Id slice (for args).
|
||||
(type IdArray (primitive IdArray))
|
||||
|
||||
(decl id_array_0 () IdArray)
|
||||
(extern constructor id_array_0 id_array_0_ctor)
|
||||
(extern extractor id_array_0 id_array_0_etor)
|
||||
(decl id_array_1 (Id) IdArray)
|
||||
(extern constructor id_array_1 id_array_1_ctor)
|
||||
(extern extractor id_array_1 id_array_1_etor)
|
||||
(decl id_array_2 (Id Id) IdArray)
|
||||
(extern constructor id_array_2 id_array_2_ctor)
|
||||
(extern extractor id_array_2 id_array_2_etor)
|
||||
(decl id_array_3 (Id Id Id) IdArray)
|
||||
(extern constructor id_array_3 id_array_3_ctor)
|
||||
(extern extractor id_array_3 id_array_3_etor)
|
||||
|
||||
;; Extractor to get the min loop-level of an eclass.
|
||||
(decl at_loop_level (u8 Id) Id)
|
||||
(extern extractor infallible at_loop_level at_loop_level)
|
||||
|
||||
;;;;; optimization toplevel ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
|
||||
|
||||
;; The main matcher rule invoked by the toplevel driver.
|
||||
(decl multi simplify (Id) Id)
|
||||
|
||||
;; Mark a node as requiring remat when used in a different block.
|
||||
(decl remat (Id) Id)
|
||||
(extern constructor remat remat)
|
||||
|
||||
;; Mark a node as subsuming whatever else it's rewritten from -- this
|
||||
;; is definitely preferable, not just a possible option. Useful for,
|
||||
;; e.g., constant propagation where we arrive at a definite "final
|
||||
;; answer".
|
||||
(decl subsume (Id) Id)
|
||||
(extern constructor subsume subsume)
|
||||
Reference in New Issue
Block a user