Currently TFRT does not support top-level coroutines, so this functionality will allow to have a single blocking await at the top level until TFRT implements the necessary functionality.
Reviewed By: ezhulenev
Differential Revision: https://reviews.llvm.org/D106730
Interop parallelism requires needs awaiting on results. Blocking awaits are bad for performance. TFRT supports lightweight resumption on threads, and coroutines are an abstraction than can be used to lower the kernels onto TFRT threads.
Reviewed By: ezhulenev
Differential Revision: https://reviews.llvm.org/D106508
* Previously, we were only generating .h.inc files. We foresee the need to also generate implementations and this is a step towards that.
* Discussed in https://llvm.discourse.group/t/generating-cpp-inc-files-for-dialects/3732/2
* Deviates from the discussion above by generating a default constructor in the .cpp.inc file (and adding a tablegen bit that disables this in case if this is user provided).
* Generating the destructor started as a way to flush out the missing includes (produces a link error), but it is a strict improvement on its own that is worth doing (i.e. by emitting key methods in the .cpp file, we root vtables in one translation unit, which is a non-controversial improvement).
Differential Revision: https://reviews.llvm.org/D105070
Depends On D105037
Avoid creating too many tasks when the number of workers is large.
Reviewed By: herhut
Differential Revision: https://reviews.llvm.org/D105126
Depends On D104999
Automatic reference counting based on the liveness analysis can add a lot of reference counting overhead at runtime. If the IR is known to be constrained to few particular "shapes", it's much more efficient to provide a custom reference counting policy that will specify where it is required to update the async value reference count.
Reviewed By: mehdi_amini
Differential Revision: https://reviews.llvm.org/D105037
Depends On D104998
Function calls "transfer ownership" to the callee and it puts additional constraints on the reference counting optimization pass
Reviewed By: mehdi_amini
Differential Revision: https://reviews.llvm.org/D104999
Depends On D104850
Add a test that verifies that canonicalization removes all async overheads if it is statically known that the scf.parallel operation will be computed using a single block.
Reviewed By: herhut
Differential Revision: https://reviews.llvm.org/D104891
Depends On D104780
Recursive work splitting instead of sequential async tasks submission gives ~20%-30% speedup in microbenchmarks.
Algorithm outline:
1. Collapse scf.parallel dimensions into a single dimension
2. Compute the block size for the parallel operations from the 1d problem size
3. Launch parallel tasks
4. Each parallel task reconstructs its own bounds in the original multi-dimensional iteration space
5. Each parallel task computes the original parallel operation body using scf.for loop nest
Reviewed By: herhut
Differential Revision: https://reviews.llvm.org/D104850
Specify the `!async.group` size (the number of tokens that will be added to it) at construction time. `async.await_all` operation can potentially race with `async.execute` operations that keep updating the group, for this reason it is required to know upfront how many tokens will be added to the group.
Reviewed By: ftynse, herhut
Differential Revision: https://reviews.llvm.org/D104780
Depends On D103109
If any of the tokens/values added to the `!async.group` switches to the error state, than the group itself switches to the error state.
Reviewed By: mehdi_amini
Differential Revision: https://reviews.llvm.org/D103203
Depends On D103102
Not yet implemented:
1. Error handling after synchronous await
2. Error handling for async groups
Will be addressed in the followup PRs
Reviewed By: mehdi_amini
Differential Revision: https://reviews.llvm.org/D103109
Support reference counted values implicitly passed (live) only to some of the successors.
Example: if branched to ^bb2 token will leak, unless `drop_ref` operation is properly created
```
^entry:
%token = async.runtime.create : !async.token
cond_br %cond, ^bb1, ^bb2
^bb1:
async.runtime.await %token
async.runtime.drop_ref %token
br ^bb2
^bb2:
return
```
Reviewed By: mehdi_amini
Differential Revision: https://reviews.llvm.org/D103102
Depends On D95311
Previous automatic-ref-counting pass worked with high level async operations (e.g. async.execute), however async values reference counting is a runtime implementation detail.
New pass mostly relies on the save liveness analysis to place drop_ref operations, and does better verification of CFG with different liveIn sets in block successors.
This is almost NFC change. No new reference counting ideas, just a cleanup of the previous version.
Reviewed By: mehdi_amini
Differential Revision: https://reviews.llvm.org/D95390
In particular for Graph Regions, the terminator needs is just a
historical artifact of the generalization of MLIR from CFG region.
Operations like Module don't need a terminator, and before Module
migrated to be an operation with region there wasn't any needed.
To validate the feature, the ModuleOp is migrated to use this trait and
the ModuleTerminator operation is deleted.
This patch is likely to break clients, if you're in this case:
- you may iterate on a ModuleOp with `getBody()->without_terminator()`,
the solution is simple: just remove the ->without_terminator!
- you created a builder with `Builder::atBlockTerminator(module_body)`,
just use `Builder::atBlockEnd(module_body)` instead.
- you were handling ModuleTerminator: it isn't needed anymore.
- for generic code, a `Block::mayNotHaveTerminator()` may be used.
Differential Revision: https://reviews.llvm.org/D98468
This doesn't change APIs, this just cleans up the many in-tree uses of these
names to use the new preferred names. We'll keep the old names around for a
couple weeks to help transitions.
Differential Revision: https://reviews.llvm.org/D99127
This updates the codebase to pass the context when creating an instance of
OwningRewritePatternList, and starts removing extraneous MLIRContext
parameters. There are many many more to be removed.
Differential Revision: https://reviews.llvm.org/D99028
This allows the caller to distinguish between a parse error or an
unmatched keyword. It fixes the redundant error that was emitted by the
caller when the generated parser would fail.
Differential Revision: https://reviews.llvm.org/D98162
The current implementation of Value involves a pointer int pair with several different kinds of owners, i.e. BlockArgumentImpl*, Operation *, TrailingOpResult*. This design arose from the desire to save memory overhead for operations that have a very small number of results (generally 0-2). There are, unfortunately, many problematic aspects of the current implementation that make Values difficult to work with or just inefficient.
Operation result types are stored as a separate array on the Operation. This is very inefficient for many reasons: we use TupleType for multiple results, which can lead to huge amounts of memory usage if multi-result operations change types frequently(they do). It also means that simple methods like Value::getType/Value::setType now require complex logic to get to the desired type.
Value only has one pointer bit free, severely limiting the ability to use it in things like PointerUnion/PointerIntPair. Given that we store the kind of a Value along with the "owner" pointer, we only leave one bit free for users of Value. This creates situations where we end up nesting PointerUnions to be able to use Value in one.
As noted above, most of the methods in Value need to branch on at least 3 different cases which is both inefficient, possibly error prone, and verbose. The current storage of results also creates problems for utilities like ValueRange/TypeRange, which want to efficiently store base pointers to ranges (of which Operation* isn't really useful as one).
This revision greatly simplifies the implementation of Value by the introduction of a new ValueImpl class. This class contains all of the state shared between all of the various derived value classes; i.e. the use list, the type, and the kind. This shared implementation class provides several large benefits:
* Most of the methods on value are now branchless, and often one-liners.
* The "kind" of the value is now stored in ValueImpl instead of Value
This frees up all of Value's pointer bits, allowing for users to take full advantage of PointerUnion/PointerIntPair/etc. It also allows for storing more operation results as "inline", 6 now instead of 2, freeing up 1 word per new inline result.
* Operation result types are now stored in the result, instead of a side array
This drops the size of zero-result operations by 1 word. It also removes the memory crushing use of TupleType for operations results (which could lead up to hundreds of megabytes of "dead" TupleTypes in the context). This also allowed restructured ValueRange, making it simpler and one word smaller.
This revision does come with two conceptual downsides:
* Operation::getResultTypes no longer returns an ArrayRef<Type>
This conceptually makes some usages slower, as the iterator increment is slightly more complex.
* OpResult::getOwner is slightly more expensive, as it now requires a little bit of arithmetic
From profiling, neither of the conceptual downsides have resulted in any perceivable hit to performance. Given the advantages of the new design, most compiles are slightly faster.
Differential Revision: https://reviews.llvm.org/D97804
Depends On D95000
Move async.execute outlining and async -> async.runtime lowering into the separate Async transformation pass
Reviewed By: mehdi_amini
Differential Revision: https://reviews.llvm.org/D95311
[NFC] No new functionality, mostly a cleanup and one more abstraction level between Async and LLVM IR.
Instead of lowering from Async to LLVM coroutines and Async Runtime API in one shot, do it progressively via async.coro and async.runtime operations.
1. Lower from async to async.runtime/coro (e.g. async.execute to function with coro setup and runtime calls)
2. Lower from async.runtime/coro to LLVM intrinsics and runtime API calls
Intermediate coro/runtime operations will allow to run transformations on a higher level IR and do not try to match IR based on the LLVM::CallOp properties.
Although async.coro is very close to LLVM coroutines, it is not exactly the same API, instead it is optimized for usability in async lowering, and misses a lot of details that are present in @llvm.coro intrinsic.
Reviewed By: ftynse
Differential Revision: https://reviews.llvm.org/D94923
This better matches the rest of the infrastructure, is much simpler, and makes it easier to move these types to being declaratively specified.
Differential Revision: https://reviews.llvm.org/D93432
Add an option to pass the number of worker threads to select the number of async regions for parallel for transformation.
```
std::unique_ptr<OperationPass<FuncOp>> createAsyncParallelForPass(int numWorkerThreads);
```
Reviewed By: mehdi_amini
Differential Revision: https://reviews.llvm.org/D92835
Given that OpState already implicit converts to Operator*, this seems reasonable.
The alternative would be to add more functions to OpState which forward to Operation.
Reviewed By: rriddle, ftynse
Differential Revision: https://reviews.llvm.org/D92266
Depends On D89963
**Automatic reference counting algorithm outline:**
1. `ReturnLike` operations forward the reference counted values without
modifying the reference count.
2. Use liveness analysis to find blocks in the CFG where the lifetime of
reference counted values ends, and insert `drop_ref` operations after
the last use of the value.
3. Insert `add_ref` before the `async.execute` operation capturing the
value, and pairing `drop_ref` before the async body region terminator,
to release the captured reference counted value when execution
completes.
4. If the reference counted value is passed only to some of the block
successors, insert `drop_ref` operations in the beginning of the blocks
that do not have reference coutned value uses.
Reviewed By: silvas
Differential Revision: https://reviews.llvm.org/D90716
Depends On D89958
1. Adds `async.group`/`async.awaitall` to group together multiple async tokens/values
2. Rewrite scf.parallel operation into multiple concurrent async.execute operations over non overlapping subranges of the original loop.
Example:
```
scf.for (%i, %j) = (%lbi, %lbj) to (%ubi, %ubj) step (%si, %sj) {
"do_some_compute"(%i, %j): () -> ()
}
```
Converted to:
```
%c0 = constant 0 : index
%c1 = constant 1 : index
// Compute blocks sizes for each induction variable.
%num_blocks_i = ... : index
%num_blocks_j = ... : index
%block_size_i = ... : index
%block_size_j = ... : index
// Create an async group to track async execute ops.
%group = async.create_group
scf.for %bi = %c0 to %num_blocks_i step %c1 {
%block_start_i = ... : index
%block_end_i = ... : index
scf.for %bj = %c0 t0 %num_blocks_j step %c1 {
%block_start_j = ... : index
%block_end_j = ... : index
// Execute the body of original parallel operation for the current
// block.
%token = async.execute {
scf.for %i = %block_start_i to %block_end_i step %si {
scf.for %j = %block_start_j to %block_end_j step %sj {
"do_some_compute"(%i, %j): () -> ()
}
}
}
// Add produced async token to the group.
async.add_to_group %token, %group
}
}
// Await completion of all async.execute operations.
async.await_all %group
```
In this example outer loop launches inner block level loops as separate async
execute operations which will be executed concurrently.
At the end it waits for the completiom of all async execute operations.
Reviewed By: ftynse, mehdi_amini
Differential Revision: https://reviews.llvm.org/D89963
Async execute operation can take async arguments as dependencies.
Change `async.execute` custom parser/printer format to use `%value as %unwrapped: !async.value<!type>` sytax.
Reviewed By: mehdi_amini, herhut
Differential Revision: https://reviews.llvm.org/D88601
Eugene Zhulenev