-
Notifications
You must be signed in to change notification settings - Fork 72
Fix memory consistency non-inline metadata helpers #944
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Open
TaoTao-real
wants to merge
3
commits into
hw-native-sys:main
Choose a base branch
from
TaoTao-real:pr873-vpto-ci-fix-local
base: main
Could not load branches
Branch not found: {{ refName }}
Loading
Could not load tags
Nothing to show
Loading
Are you sure you want to change the base?
Some commits from the old base branch may be removed from the timeline,
and old review comments may become outdated.
Open
Changes from 1 commit
Commits
Show all changes
3 commits
Select commit
Hold shift + click to select a range
File filter
Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
31 changes: 31 additions & 0 deletions
31
test/lit/pto/memory_consistency_noninline_metadata_call.pto
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,31 @@ | ||
| // Copyright (c) 2026 Huawei Technologies Co., Ltd. | ||
| // This program is free software, you can redistribute it and/or modify it under the terms and conditions of | ||
| // CANN Open Software License Agreement Version 2.0 (the "License"). | ||
| // Please refer to the License for details. You may not use this file except in compliance with the License. | ||
| // THIS SOFTWARE IS PROVIDED ON AN "AS IS" BASIS, WITHOUT ANY WARRANTIES OF ANY KIND, EITHER EXPRESS OR IMPLIED, | ||
| // INCLUDING BUT NOT LIMITED TO NON-INFRINGEMENT, MERCHANTABILITY, OR FITNESS FOR A PARTICULAR PURPOSE. | ||
| // See LICENSE in the root of the software repository for the full text of the License. | ||
|
|
||
| // A non-inline helper that only reads communication metadata should not be | ||
| // treated as hidden payload/signal state. | ||
|
|
||
| // RUN: ptoas --pto-arch=a3 %s -o - 2>&1 | FileCheck %s | ||
|
|
||
| module { | ||
| func.func private @metadata_offset(%ctx: !pto.ptr<i32>) -> i32 { | ||
| %c0 = arith.constant 0 : index | ||
| %off = pto.load_scalar %ctx[%c0] : !pto.ptr<i32> -> i32 | ||
| return %off : i32 | ||
| } | ||
|
|
||
| func.func @metadata_helper_call_is_allowed(%ctx: !pto.ptr<i32>) | ||
| attributes {pto.kernel_kind = #pto.kernel_kind<vector>} { | ||
| %c0 = arith.constant 0 : index | ||
| %off = func.call @metadata_offset(%ctx) : (!pto.ptr<i32>) -> i32 | ||
| pto.store_scalar %off, %ctx[%c0] : !pto.ptr<i32>, i32 | ||
| return | ||
| } | ||
| } | ||
|
|
||
| // CHECK-NOT: error: | ||
| // CHECK: AICORE void metadata_helper_call_is_allowed( |
64 changes: 64 additions & 0 deletions
64
test/lit/pto/memory_consistency_noninline_release_call.pto
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,64 @@ | ||
| // Copyright (c) 2026 Huawei Technologies Co., Ltd. | ||
| // This program is free software, you can redistribute it and/or modify it under the terms and conditions of | ||
| // CANN Open Software License Agreement Version 2.0 (the "License"). | ||
| // Please refer to the License for details. You may not use this file except in compliance with the License. | ||
| // THIS SOFTWARE IS PROVIDED ON AN "AS IS" BASIS, WITHOUT ANY WARRANTIES OF ANY KIND, EITHER EXPRESS OR IMPLIED, | ||
| // INCLUDING BUT NOT LIMITED TO NON-INFRINGEMENT, MERCHANTABILITY, OR FITNESS FOR A PARTICULAR PURPOSE. | ||
| // See LICENSE in the root of the software repository for the full text of the License. | ||
|
|
||
| // The callee hides a payload write, but the caller still provides the payload | ||
| // marker explicitly. The memory consistency pass should keep the boundary and | ||
| // infer the required drain from the marker instead of blindly rejecting the | ||
| // call because of the hidden load/store body. | ||
|
|
||
| // RUN: ptoas --pto-arch=a3 %s -o - 2>&1 | FileCheck %s | ||
|
|
||
| module { | ||
| func.func private @producer( | ||
| %tile: !pto.tile_buf<loc=vec, dtype=f32, rows=1, cols=32, v_row=1, v_col=32, blayout=row_major, slayout=none_box, fractal=512, pad=0>, | ||
| %dst_ptr: !pto.ptr<f32>) { | ||
| %c0 = arith.constant 0 : index | ||
| %c1 = arith.constant 1 : index | ||
| %c32 = arith.constant 32 : index | ||
| %dst_view = pto.make_tensor_view %dst_ptr, | ||
| shape = [%c1, %c32], strides = [%c32, %c1] : !pto.tensor_view<?x?xf32> | ||
| %dst = pto.partition_view %dst_view, | ||
| offsets = [%c0, %c0], sizes = [%c1, %c32] | ||
| : !pto.tensor_view<?x?xf32> -> !pto.partition_tensor_view<1x32xf32> | ||
| pto.tstore ins(%tile : !pto.tile_buf<loc=vec, dtype=f32, rows=1, cols=32, v_row=1, v_col=32, blayout=row_major, slayout=none_box, fractal=512, pad=0>) | ||
| outs(%dst : !pto.partition_tensor_view<1x32xf32>) | ||
| return | ||
| } | ||
|
|
||
| func.func @call_hidden_payload_write_with_marker( | ||
| %dst_ptr: !pto.ptr<f32>, | ||
| %signal_ptr: !pto.ptr<i32>) | ||
| attributes {pto.kernel_kind = #pto.kernel_kind<vector>} { | ||
| %c0 = arith.constant 0 : index | ||
| %c1 = arith.constant 1 : index | ||
| %v_i32 = arith.constant 1 : i32 | ||
|
|
||
| %tile = pto.alloc_tile : | ||
| !pto.tile_buf<loc=vec, dtype=f32, rows=1, cols=32, v_row=1, v_col=32, blayout=row_major, slayout=none_box, fractal=512, pad=0> | ||
|
|
||
| func.call @producer(%tile, %dst_ptr) : | ||
| (!pto.tile_buf<loc=vec, dtype=f32, rows=1, cols=32, v_row=1, v_col=32, blayout=row_major, slayout=none_box, fractal=512, pad=0>, | ||
| !pto.ptr<f32>) -> () | ||
|
|
||
| pto.cmo.cacheinvalid %dst_ptr single_cache_line : !pto.ptr<f32> | ||
| pto.fence.barrier_all #pto.fence_scope<gm> | ||
|
|
||
| %sig_view = pto.make_tensor_view %signal_ptr, | ||
| shape = [%c1], strides = [%c1] : !pto.tensor_view<1xi32> | ||
| %sig = pto.partition_view %sig_view, | ||
| offsets = [%c0], sizes = [%c1] | ||
| : !pto.tensor_view<1xi32> -> !pto.partition_tensor_view<1xi32> | ||
| pto.comm.tnotify(%sig, %v_i32 : !pto.partition_tensor_view<1xi32>, i32) | ||
| {notifyOp = #pto<notify_op set>} | ||
| return | ||
| } | ||
| } | ||
|
|
||
| // CHECK: pipe_barrier(PIPE_MTE3) | ||
| // CHECK: dsb( | ||
| // CHECK: pto::comm::TNOTIFY( |
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Narrowing the fail-fast to boundary-ops-only introduces a release/acquire asymmetry that can silently drop an acquire diagnostic.
Before this change, a non-inline callee containing a GM
load_scalar/store_scalarwas rejected here (the oldisMemoryConsistencyRelevantDirectOpreturned true for GM scalar load/store), which forced the helper to be inlined so the acquire walk could see its loads. Now such a callee is accepted, but only the release path was made interprocedural (getReleaseStateForCallsummarizes callee payload producers into the caller). The acquire path was not:annotateSignalAcquire/collectSignalAcquireState/annotateSignalAcquireForBlockhave nofunc.callhandling, and every function is analyzed independently starting from an emptySignalAcquireState.Consequence:
The cacheable load inside
@helperis now reached after an acquire, butdiagnoseAcquireLoadnever fires for it:@helperis analyzed with an empty acquire state, and the caller's acquire state does not cross the call. Previously this program was rejected (forcing inline, after which the load would be diagnosed as needingpto.cmo.cacheinvalid all #pto.address_space<gm>before the cacheable load). So a program that should error now compiles silently -> potential stale GM read after acquire.The motivating metadata helpers (
load_scalaronCommContext) are legitimately exempt, but nothing constrains a now-allowed non-inline callee to metadata-only loads. Is this asymmetry intentional? If so, it would be good to pin it with a regression test and note it in the design doc; otherwise the acquire path likely needs the same interprocedural treatment as the release path. Either way it is currently untested.