[llvm-branch-commits] [flang] [llvm] [MLIR][OpenMP] Add a new AutomapToTargetData conversion pass in FIR (PR #151989)

@@ -0,0 +1,44 @@ +// RUN: fir-opt --omp-automap-to-target-data %s | FileCheck %s +// Test OMP AutomapToTargetData pass. + +module { + fir.global + @_QMtestEarr{omp.declare_target = #omp.declaretarget} target + : !fir.box>> + + func.func @automap()

[llvm-branch-commits] [flang] [llvm] [MLIR][OpenMP] Add a new AutomapToTargetData conversion pass in FIR (PR #151989)

@@ -0,0 +1,44 @@ +// RUN: fir-opt --omp-automap-to-target-data %s | FileCheck %s +// Test OMP AutomapToTargetData pass. + +module { + fir.global + @_QMtestEarr{omp.declare_target = #omp.declaretarget} target + : !fir.box>> + + func.func @automap()

[llvm-branch-commits] [flang] [llvm] [MLIR][OpenMP] Add a new AutomapToTargetData conversion pass in FIR (PR #151989)

@@ -72,6 +77,36 @@ struct EntryBlockArgs { /// \param [in] region - Empty region in which to create the entry block. mlir::Block *genEntryBlock( mlir::OpBuilder &builder, const EntryBlockArgs &args, mlir::Region ®ion); + +// Returns true if the variable has a dynamic size

[llvm-branch-commits] [flang] [llvm] [MLIR][OpenMP] Add a new AutomapToTargetData conversion pass in FIR (PR #151989)

@@ -0,0 +1,130 @@ +//===- AutomapToTargetData.cpp ---===// +// +// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. +// See https://llvm.org/LICENSE.txt for license information. +// SPDX-License-Identifier: Apac

[llvm-branch-commits] [flang] [llvm] [MLIR][OpenMP] Add a new AutomapToTargetData conversion pass in FIR (PR #151989)

@@ -0,0 +1,44 @@ +// RUN: fir-opt --omp-automap-to-target-data %s | FileCheck %s +// Test OMP AutomapToTargetData pass. + +module { + fir.global + @_QMtestEarr{omp.declare_target = #omp.declaretarget} target + : !fir.box>> + + func.func @automap()

[llvm-branch-commits] [flang] [llvm] [MLIR][OpenMP] Add a new AutomapToTargetData conversion pass in FIR (PR #151989)

@@ -72,6 +77,36 @@ struct EntryBlockArgs { /// \param [in] region - Empty region in which to create the entry block. mlir::Block *genEntryBlock( mlir::OpBuilder &builder, const EntryBlockArgs &args, mlir::Region ®ion); + +// Returns true if the variable has a dynamic size

[llvm-branch-commits] [flang] [llvm] [MLIR][OpenMP] Add a new AutomapToTargetData conversion pass in FIR (PR #151989)

https://github.com/skatrak approved this pull request. Thank you again Akash, LGTM minus some NFC suggestions. Please wait for @bhandarkar-pranav's approval before merging. https://github.com/llvm/llvm-project/pull/151989 ___ llvm-branch-commits maili

[llvm-branch-commits] [flang] [llvm] [MLIR][OpenMP] Add a new AutomapToTargetData conversion pass in FIR (PR #151989)

https://github.com/skatrak edited https://github.com/llvm/llvm-project/pull/151989 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits

[llvm-branch-commits] [flang] [llvm] [MLIR][OpenMP] Add a new AutomapToTargetData conversion pass in FIR (PR #151989)

https://github.com/skatrak edited https://github.com/llvm/llvm-project/pull/151989 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits

[llvm-branch-commits] [flang] [llvm] [MLIR][OpenMP] Add a new AutomapToTargetData conversion pass in FIR (PR #151989)

@@ -316,13 +316,13 @@ void createOpenMPFIRPassPipeline(mlir::PassManager &pm, pm.addPass(flangomp::createDoConcurrentConversionPass( opts.doConcurrentMappingKind == DoConcurrentMappingKind::DCMK_Device)); - // The MapsForPrivatizedSymbols pass needs to run before

[llvm-branch-commits] [flang] [llvm] [MLIR][OpenMP] Add a new AutomapToTargetData conversion pass in FIR (PR #151989)

@@ -0,0 +1,171 @@ +//===- AutomapToTargetData.cpp ---===// +// +// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. +// See https://llvm.org/LICENSE.txt for license information. +// SPDX-License-Identifier: Apac

[llvm-branch-commits] [flang] [llvm] [MLIR][OpenMP] Add a new AutomapToTargetData conversion pass in FIR (PR #151989)

@@ -316,13 +316,13 @@ void createOpenMPFIRPassPipeline(mlir::PassManager &pm, pm.addPass(flangomp::createDoConcurrentConversionPass( opts.doConcurrentMappingKind == DoConcurrentMappingKind::DCMK_Device)); - // The MapsForPrivatizedSymbols pass needs to run before

[llvm-branch-commits] [flang] [llvm] [MLIR][OpenMP] Add a new AutomapToTargetData conversion pass in FIR (PR #151989)

@@ -0,0 +1,40 @@ +// RUN: fir-opt --omp-automap-to-target-data %s | FileCheck %s +// Test OMP AutomapToTargetData pass. + +module { + fir.global + @_QMtestEarr{omp.declare_target = #omp.declaretarget} target + : !fir.box>> + + func.func @automap()

[llvm-branch-commits] [flang] [llvm] [MLIR][OpenMP] Add a new AutomapToTargetData conversion pass in FIR (PR #151989)

@@ -0,0 +1,171 @@ +//===- AutomapToTargetData.cpp ---===// +// +// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. +// See https://llvm.org/LICENSE.txt for license information. +// SPDX-License-Identifier: Apac

[llvm-branch-commits] [flang] [llvm] [MLIR][OpenMP] Add a new AutomapToTargetData conversion pass in FIR (PR #151989)

@@ -0,0 +1,171 @@ +//===- AutomapToTargetData.cpp ---===// +// +// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. +// See https://llvm.org/LICENSE.txt for license information. +// SPDX-License-Identifier: Apac

[llvm-branch-commits] [flang] [llvm] [MLIR][OpenMP] Add a new AutomapToTargetData conversion pass in FIR (PR #151989)

@@ -0,0 +1,171 @@ +//===- AutomapToTargetData.cpp ---===// +// +// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. +// See https://llvm.org/LICENSE.txt for license information. +// SPDX-License-Identifier: Apac

[llvm-branch-commits] [flang] [llvm] [MLIR][OpenMP] Add a new AutomapToTargetData conversion pass in FIR (PR #151989)

@@ -0,0 +1,171 @@ +//===- AutomapToTargetData.cpp ---===// +// +// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. +// See https://llvm.org/LICENSE.txt for license information. +// SPDX-License-Identifier: Apac

[llvm-branch-commits] [flang] [llvm] [MLIR][OpenMP] Add a new AutomapToTargetData conversion pass in FIR (PR #151989)

@@ -0,0 +1,171 @@ +//===- AutomapToTargetData.cpp ---===// +// +// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. +// See https://llvm.org/LICENSE.txt for license information. +// SPDX-License-Identifier: Apac

[llvm-branch-commits] [flang] [llvm] [MLIR][OpenMP] Add a new AutomapToTargetData conversion pass in FIR (PR #151989)

https://github.com/skatrak commented: Thank you Akash, a couple of minor comments from me. https://github.com/llvm/llvm-project/pull/151989 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/l

[llvm-branch-commits] [llvm] [OpenMPOpt] Make parallel regions reachable from new DeviceRTL loop functions (PR #150927)

https://github.com/skatrak updated https://github.com/llvm/llvm-project/pull/150927 >From a26b2a03629f3ea8a9852a77db49df73f15115f4 Mon Sep 17 00:00:00 2001 From: Sergio Afonso Date: Fri, 25 Jul 2025 13:52:11 +0100 Subject: [PATCH] [OpenMPOpt] Make parallel regions reachable from new DeviceRTL

[llvm-branch-commits] [llvm] [mlir] [OpenMP][OMPIRBuilder] Support parallel in Generic kernels (PR #150926)

https://github.com/skatrak updated https://github.com/llvm/llvm-project/pull/150926 >From 6a81001ec131371c981e789f3dd402fb277e3c62 Mon Sep 17 00:00:00 2001 From: Sergio Afonso Date: Fri, 4 Jul 2025 16:32:03 +0100 Subject: [PATCH] [OpenMP][OMPIRBuilder] Support parallel in Generic kernels This

[llvm-branch-commits] [llvm] [mlir] [OpenMP][OMPIRBuilder] Use device shared memory for arg structures (PR #150925)

https://github.com/skatrak updated https://github.com/llvm/llvm-project/pull/150925 >From 8b344026f04bd417bb919c866b850d4474639d62 Mon Sep 17 00:00:00 2001 From: Sergio Afonso Date: Thu, 3 Jul 2025 16:47:51 +0100 Subject: [PATCH 1/2] [OpenMP][OMPIRBuilder] Use device shared memory for arg stru

[llvm-branch-commits] [flang] [mlir] [MLIR][OpenMP] Simplify OpenMP device codegen (PR #137201)

https://github.com/skatrak commented: Thank you Kareem for your comments, I got pulled into other work and wasn't able to reply properly until now. https://github.com/llvm/llvm-project/pull/137201 ___ llvm-branch-commits mailing list llvm-branch-commi

[llvm-branch-commits] [flang] [mlir] [MLIR][OpenMP] Simplify OpenMP device codegen (PR #137201)

@@ -3173,19 +3173,14 @@ convertOmpThreadprivate(Operation &opInst, llvm::IRBuilderBase &builder, LLVM::GlobalOp global = addressOfOp.getGlobal(moduleTranslation.symbolTable()); llvm::GlobalValue *globalValue = moduleTranslation.lookupGlobal(global); - - if (!ompBuil

[llvm-branch-commits] [flang] [mlir] [MLIR][OpenMP] Simplify OpenMP device codegen (PR #137201)

@@ -5488,40 +5483,172 @@ convertDeclareTargetAttr(Operation *op, mlir::omp::DeclareTargetAttr attribute, return success(); } -// Returns true if the operation is inside a TargetOp or -// is part of a declare target function. -static bool isTargetDeviceOp(Operation *op) { +n

[llvm-branch-commits] [flang] [mlir] [MLIR][OpenMP] Simplify OpenMP device codegen (PR #137201)

https://github.com/skatrak edited https://github.com/llvm/llvm-project/pull/137201 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits

[llvm-branch-commits] [llvm] [OpenMPOpt] Make parallel regions reachable from new DeviceRTL loop functions (PR #150927)

skatrak wrote: PR stack: - #150922 - #150923 - #150924 - #150925 - #150926 - #150927 ◀️ https://github.com/llvm/llvm-project/pull/150927 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/list

[llvm-branch-commits] [llvm] [OpenMPOpt] Make parallel regions reachable from new DeviceRTL loop functions (PR #150927)

https://github.com/skatrak updated https://github.com/llvm/llvm-project/pull/150927 >From b6c5ff874edf121b3801e773ad53608b01e8ebfd Mon Sep 17 00:00:00 2001 From: Sergio Afonso Date: Fri, 25 Jul 2025 13:52:11 +0100 Subject: [PATCH] [OpenMPOpt] Make parallel regions reachable from new DeviceRTL

[llvm-branch-commits] [llvm] [mlir] [OpenMP][OMPIRBuilder] Support parallel in Generic kernels (PR #150926)

https://github.com/skatrak updated https://github.com/llvm/llvm-project/pull/150926 >From b6e9849e6b00eaa76f3d7ce0747fadf4f53c10d0 Mon Sep 17 00:00:00 2001 From: Sergio Afonso Date: Fri, 4 Jul 2025 16:32:03 +0100 Subject: [PATCH] [OpenMP][OMPIRBuilder] Support parallel in Generic kernels This

[llvm-branch-commits] [llvm] [mlir] [OpenMP][OMPIRBuilder] Use device shared memory for arg structures (PR #150925)

https://github.com/skatrak updated https://github.com/llvm/llvm-project/pull/150925 >From 8b344026f04bd417bb919c866b850d4474639d62 Mon Sep 17 00:00:00 2001 From: Sergio Afonso Date: Thu, 3 Jul 2025 16:47:51 +0100 Subject: [PATCH] [OpenMP][OMPIRBuilder] Use device shared memory for arg structur

[llvm-branch-commits] [mlir] [MLIR][OpenMP] Support allocations of device shared memory (PR #150924)

https://github.com/skatrak created https://github.com/llvm/llvm-project/pull/150924 This patch updates the allocation of some reduction and private variables within target regions to use device shared memory rather than private memory. This is a prerequisite to produce working Generic kernels

[llvm-branch-commits] [mlir] [MLIR][OpenMP] Support allocations of device shared memory (PR #150924)

skatrak wrote: PR stack: - #150922 - #150923 - #150924 ◀️ - #150925 - #150926 - #150927 https://github.com/llvm/llvm-project/pull/150924 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/list

[llvm-branch-commits] [llvm] [mlir] [OpenMP][OMPIRBuilder] Use device shared memory for arg structures (PR #150925)

skatrak wrote: PR stack: - #150922 - #150923 - #150924 - #150925 ◀️ - #150926 - #150927 https://github.com/llvm/llvm-project/pull/150925 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/list

[llvm-branch-commits] [llvm] [mlir] [OpenMP][OMPIRBuilder] Support parallel in Generic kernels (PR #150926)

skatrak wrote: PR stack: - #150922 - #150923 - #150924 - #150925 - #150926 ◀️ - #150927 https://github.com/llvm/llvm-project/pull/150926 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/list

[llvm-branch-commits] [llvm] [OpenMP][OMPIRBuilder] Add device shared memory allocation support (PR #150923)

skatrak wrote: PR stack: - #150922 - #150923 ◀️ - #150924 - #150925 - #150926 - #150927 https://github.com/llvm/llvm-project/pull/150923 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/list

[llvm-branch-commits] [llvm] [OpenMPOpt] Make parallel regions reachable from new DeviceRTL loop functions (PR #150927)

https://github.com/skatrak created https://github.com/llvm/llvm-project/pull/150927 This patch updates the OpenMP optimization pass to know about the new DeviceRTL functions for loop constructs. This change marks these functions as potentially containing parallel regions, which fixes a curren

[llvm-branch-commits] [llvm] [mlir] [OpenMP][OMPIRBuilder] Use device shared memory for arg structures (PR #150925)

https://github.com/skatrak created https://github.com/llvm/llvm-project/pull/150925 Argument structures are created when sections of the LLVM IR corresponding to an OpenMP construct are outlined into their own function. For this, stack allocations are used. This patch modifies this behavior w

[llvm-branch-commits] [llvm] [mlir] [OpenMP][OMPIRBuilder] Support parallel in Generic kernels (PR #150926)

https://github.com/skatrak created https://github.com/llvm/llvm-project/pull/150926 This patch introduces codegen logic to produce a wrapper function argument for the `__kmpc_parallel_51` DeviceRTL function needed to handle arguments passed using device shared memory in Generic mode. >From 87

[llvm-branch-commits] [llvm] [OpenMP][OMPIRBuilder] Add device shared memory allocation support (PR #150923)

https://github.com/skatrak created https://github.com/llvm/llvm-project/pull/150923 This patch adds the `__kmpc_alloc_shared` and `__kmpc_free_shared` DeviceRTL functions to the list of those the OMPIRBuilder is able to create. >From f27aedb004854d46a2c36651d69cfcf23ebac181 Mon Sep 17 00:00:00

[llvm-branch-commits] [flang] [mlir] [flang][OpenMP][NFC] remove globals with mlir::StateStack (PR #144898)

https://github.com/skatrak approved this pull request. Thank you Tom for working on this, I think it's a very nice improvement! LGTM. https://github.com/llvm/llvm-project/pull/144898 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.o

[llvm-branch-commits] [flang] [mlir] [flang][OpenMP][NFC] remove globals with mlir::StateStack (PR #144898)

@@ -2224,10 +2260,13 @@ genSectionsOp(lower::AbstractConverter &converter, lower::SymMap &symTable, lower::pft::Evaluation &eval, mlir::Location loc, const ConstructQueue &queue, ConstructQueue::const_iterator item) { - assert(!sectio

[llvm-branch-commits] [flang] [mlir] [flang][OpenMP][NFC] remove globals with mlir::StateStack (PR #144898)

@@ -1818,16 +1852,17 @@ static void genTargetClauses( llvm::SmallVectorImpl &hasDeviceAddrSyms, llvm::SmallVectorImpl &isDevicePtrSyms, llvm::SmallVectorImpl &mapSyms) { + HostEvalInfo *hostEvalInfo = getHostEvalInfoStackTop(converter); ClauseProcessor cp(conver

[llvm-branch-commits] [flang] [mlir] [flang][OpenMP][NFC] remove globals with mlir::StateStack (PR #144898)

@@ -2224,10 +2260,13 @@ genSectionsOp(lower::AbstractConverter &converter, lower::SymMap &symTable, lower::pft::Evaluation &eval, mlir::Location loc, const ConstructQueue &queue, ConstructQueue::const_iterator item) { - assert(!sectio

[llvm-branch-commits] [flang] [mlir] [flang][OpenMP][NFC] remove globals with mlir::StateStack (PR #144898)

https://github.com/skatrak edited https://github.com/llvm/llvm-project/pull/144898 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits

[llvm-branch-commits] [flang] [mlir] [flang][OpenMP][NFC] remove globals with mlir::StateStack (PR #144898)

@@ -200,9 +201,41 @@ class HostEvalInfo { /// the handling of the outer region by keeping a stack of information /// structures, but it will probably still require some further work to support /// reverse offloading. -static llvm::SmallVector hostEvalInfo; -static llvm::SmallVe

[llvm-branch-commits] [flang] [mlir] [MLIR][OpenMP] Simplify OpenMP device codegen (PR #137201)

https://github.com/skatrak updated https://github.com/llvm/llvm-project/pull/137201 Rate limit · GitHub body { background-color: #f6f8fa; color: #24292e; font-family: -apple-system,BlinkMacSystemFont,Segoe UI,Helvetica,Arial,sans-s

[llvm-branch-commits] [mlir] [MLIR][OpenMP] Assert on map translation functions, NFC (PR #137199)

https://github.com/skatrak updated https://github.com/llvm/llvm-project/pull/137199 >From c6954b3120a87eef7d9cf86f18d4ef342b2e7b25 Mon Sep 17 00:00:00 2001 From: Sergio Afonso Date: Fri, 11 Apr 2025 13:40:14 +0100 Subject: [PATCH] [MLIR][OpenMP] Assert on map translation functions, NFC This pa

[llvm-branch-commits] [flang] [mlir] [mlir][OpenMP] cancel(lation point) taskgroup LLVMIR (PR #137841)

https://github.com/skatrak approved this pull request. Thank you, this LGTM. I just have a non-blocking suggestion. https://github.com/llvm/llvm-project/pull/137841 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.l

[llvm-branch-commits] [flang] [mlir] [mlir][OpenMP] cancel(lation point) taskgroup LLVMIR (PR #137841)

@@ -3075,9 +3115,6 @@ convertOmpCancel(omp::CancelOp op, llvm::IRBuilderBase &builder, llvm::OpenMPIRBuilder::LocationDescription ompLoc(builder); llvm::OpenMPIRBuilder *ompBuilder = moduleTranslation.getOpenMPBuilder(); - if (failed(checkImplementationStatus(*op.getOper

[llvm-branch-commits] [flang] [mlir] [mlir][OpenMP] cancel(lation point) taskgroup LLVMIR (PR #137841)

https://github.com/skatrak edited https://github.com/llvm/llvm-project/pull/137841 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits

[llvm-branch-commits] [llvm] [mlir] [mlir][OpenMP] Convert omp.cancellation_point to LLVMIR (PR #137205)

https://github.com/skatrak approved this pull request. LGTM too, thanks! https://github.com/llvm/llvm-project/pull/137205 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branc

[llvm-branch-commits] [flang] [mlir] [MLIR][OpenMP] Simplify OpenMP device codegen (PR #137201)

https://github.com/skatrak updated https://github.com/llvm/llvm-project/pull/137201 >From 22f22aa0ca2c98dfcc48a70f2f7e0a5b68d7b1d9 Mon Sep 17 00:00:00 2001 From: Sergio Afonso Date: Tue, 22 Apr 2025 12:04:45 +0100 Subject: [PATCH] [MLIR][OpenMP] Simplify OpenMP device codegen After removing ho

[llvm-branch-commits] [flang] [Flang][OpenMP] Minimize host ops remaining in device compilation (PR #137200)

https://github.com/skatrak updated https://github.com/llvm/llvm-project/pull/137200 Rate limit · GitHub body { background-color: #f6f8fa; color: #24292e; font-family: -apple-system,BlinkMacSystemFont,Segoe UI,Helvetica,Arial,sans-s

[llvm-branch-commits] [mlir] [MLIR][OpenMP] Assert on map translation functions, NFC (PR #137199)

@@ -3623,6 +3623,9 @@ static llvm::omp::OpenMPOffloadMappingFlags mapParentWithMembers( LLVM::ModuleTranslation &moduleTranslation, llvm::IRBuilderBase &builder, llvm::OpenMPIRBuilder &ompBuilder, DataLayout &dl, MapInfosTy &combinedInfo, MapInfoData &mapData, uin

[llvm-branch-commits] [flang] [mlir] [MLIR][OpenMP] Simplify OpenMP device codegen (PR #137201)

https://github.com/skatrak updated https://github.com/llvm/llvm-project/pull/137201 >From c86caed6b80bdd4ed4edeb77f65116584e61b83b Mon Sep 17 00:00:00 2001 From: Sergio Afonso Date: Tue, 22 Apr 2025 12:04:45 +0100 Subject: [PATCH] [MLIR][OpenMP] Simplify OpenMP device codegen After removing ho

[llvm-branch-commits] [flang] [Flang][OpenMP] Minimize host ops remaining in device compilation (PR #137200)

https://github.com/skatrak updated https://github.com/llvm/llvm-project/pull/137200 >From 6b734e2277969eeb200a292679c173c713b303f6 Mon Sep 17 00:00:00 2001 From: Sergio Afonso Date: Tue, 15 Apr 2025 16:59:18 +0100 Subject: [PATCH 1/2] [Flang][OpenMP] Minimize host ops remaining in device compi

[llvm-branch-commits] [mlir] [MLIR][OpenMP] Assert on map translation functions, NFC (PR #137199)

https://github.com/skatrak updated https://github.com/llvm/llvm-project/pull/137199 >From 6ea52bf1986b5bba68d4d797590156daa38a4269 Mon Sep 17 00:00:00 2001 From: Sergio Afonso Date: Fri, 11 Apr 2025 13:40:14 +0100 Subject: [PATCH] [MLIR][OpenMP] Assert on map translation functions, NFC This pa

[llvm-branch-commits] [flang] [mlir] [MLIR][OpenMP] Simplify OpenMP device codegen (PR #137201)

skatrak wrote: PR stack: - #137198 - #137199 - #137200 - #137201 https://github.com/llvm/llvm-project/pull/137201 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commit

[llvm-branch-commits] [mlir] [MLIR][OpenMP] Assert on map translation functions, NFC (PR #137199)

skatrak wrote: This PR is based on the following analysis of the current state of OpenMP map translation to LLVM IR. If there are any issues with this PR, they are likely easier to spot on this graph: ![MapInfoData uses](https://github.com/user-attachments/assets/f9f56688-51e5-4ab4-be7c-39b3de

[llvm-branch-commits] [flang] [Flang][OpenMP] Minimize host ops remaining in device compilation (PR #137200)

skatrak wrote: PR stack: - #137198 - #137199 - #137200 - #137201 https://github.com/llvm/llvm-project/pull/137200 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commit

[llvm-branch-commits] [mlir] [MLIR][OpenMP] Assert on map translation functions, NFC (PR #137199)

skatrak wrote: PR stack: - #137198 - #137199 - #137200 - #137201 https://github.com/llvm/llvm-project/pull/137199 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commit

[llvm-branch-commits] [flang] [mlir] [MLIR][OpenMP] Simplify OpenMP device codegen (PR #137201)

https://github.com/skatrak created https://github.com/llvm/llvm-project/pull/137201 After removing host operations from the device MLIR module, it is no longer necessary to provide special codegen logic to prevent these operations from causing compiler crashes or miscompilations. This patch r

[llvm-branch-commits] [flang] [Flang][OpenMP] Minimize host ops remaining in device compilation (PR #137200)

https://github.com/skatrak created https://github.com/llvm/llvm-project/pull/137200 This patch updates the function filtering OpenMP pass intended to remove host functions from the MLIR module created by Flang lowering when targeting an OpenMP target device. Host functions holding target regi

[llvm-branch-commits] [mlir] [MLIR][OpenMP] Assert on map translation functions, NFC (PR #137199)

https://github.com/skatrak created https://github.com/llvm/llvm-project/pull/137199 This patch adds assertions to map-related MLIR to LLVM IR translation functions and utils to explicitly document whether they are intended for host or device compilation only. Over time, map-related handling h

[llvm-branch-commits] [flang] [flang][OpenMP] Extend `do concurrent` mapping to multi-range loops (PR #127634)

@@ -210,6 +223,96 @@ mlir::LogicalResult collectLoopNest(fir::DoLoopOp currentLoop, return mlir::success(); } + +/// Prepares the `fir.do_loop` nest to be easily mapped to OpenMP. In +/// particular, this function would take this input IR: +/// ``` +/// fir.do_loop %i_iv =

[llvm-branch-commits] [flang] [flang][OpenMP] Extend `do concurrent` mapping to multi-range loops (PR #127634)

https://github.com/skatrak approved this pull request. Thank you Kareem, LGTM! https://github.com/llvm/llvm-project/pull/127634 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm

[llvm-branch-commits] [flang] [flang][OpenMP] Extend `do concurrent` mapping to multi-range loops (PR #127634)

@@ -28,64 +28,80 @@ namespace looputils { /// Stores info needed about the induction/iteration variable for each `do /// concurrent` in a loop nest. struct InductionVariableInfo { + InductionVariableInfo(fir::DoLoopOp doLoop) { populateInfo(doLoop); } + /// The operation al

[llvm-branch-commits] [flang] [flang][OpenMP] Extend `do concurrent` mapping to multi-range loops (PR #127634)

https://github.com/skatrak edited https://github.com/llvm/llvm-project/pull/127634 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits

[llvm-branch-commits] [flang] [flang][OpenMP] Map simple `do concurrent` loops to OpenMP host constructs (PR #127633)

https://github.com/skatrak approved this pull request. Thank you, LGTM! https://github.com/llvm/llvm-project/pull/127633 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch

[llvm-branch-commits] [mlir] [flang][OpenMP] Translate OpenMP scopes when compiling for target device (PR #130078)

@@ -5255,6 +5283,51 @@ convertTargetOpsInNest(Operation *op, llvm::IRBuilderBase &builder, return WalkResult::interrupt(); return WalkResult::skip(); } + + // Non-target ops might nest target-related ops, therefore, we + //

[llvm-branch-commits] [flang] [flang][OpenMP] Map simple `do concurrent` loops to OpenMP host constructs (PR #127633)

https://github.com/skatrak edited https://github.com/llvm/llvm-project/pull/127633 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits

[llvm-branch-commits] [flang] [flang][OpenMP] Map simple `do concurrent` loops to OpenMP host constructs (PR #127633)

@@ -24,7 +25,67 @@ namespace flangomp { namespace { namespace looputils { -using LoopNest = llvm::SetVector; +/// Stores info needed about the induction/iteration variable for each `do +/// concurrent` in a loop nest. +struct InductionVariableInfo { + /// The operation alloca

[llvm-branch-commits] [mlir] [flang][OpenMP] Translate OpenMP scopes when compiling for target device (PR #130078)

@@ -5255,6 +5283,51 @@ convertTargetOpsInNest(Operation *op, llvm::IRBuilderBase &builder, return WalkResult::interrupt(); return WalkResult::skip(); } + + // Non-target ops might nest target-related ops, therefore, we + //

[llvm-branch-commits] [mlir] [flang][OpenMP] Translate OpenMP scopes when compiling for target device (PR #130078)

@@ -5315,6 +5320,46 @@ convertTargetOpsInNest(Operation *op, llvm::IRBuilderBase &builder, return WalkResult::interrupt(); return WalkResult::skip(); } + + // Non-target ops might nest target-related ops, therefore, we + //

[llvm-branch-commits] [flang] [flang][OpenMP] Map simple `do concurrent` loops to OpenMP host constructs (PR #127633)

@@ -24,7 +25,50 @@ namespace flangomp { namespace { namespace looputils { -using LoopNest = llvm::SetVector; +/// Stores info needed about the induction/iteration variable for each `do +/// concurrent` in a loop nest. +struct InductionVariableInfo { + /// the operation alloca

[llvm-branch-commits] [flang] [flang][OpenMP] Map simple `do concurrent` loops to OpenMP host constructs (PR #127633)

@@ -0,0 +1,45 @@ +! Tests that if `do concurrent` is not perfectly nested in its parent loop, that +! we skip converting the not-perfectly nested `do concurrent` loop. + +! RUN: %flang_fc1 -emit-hlfir -fopenmp -fdo-concurrent-to-openmp=host %s -o - \ +! RUN: | FileCheck %s + +p

[llvm-branch-commits] [mlir] [MLIR][OpenMP] Minor improvements to BlockArgOpenMPOpInterface, NFC (PR #130789)

https://github.com/skatrak created https://github.com/llvm/llvm-project/pull/130789 This patch introduces a use for the new `getBlockArgsPairs` to avoid having to manually list each applicable clause. Also, the `numClauseBlockArgs()` function is introduced, which simplifies the implementation

[llvm-branch-commits] [flang] [flang][OpenMP] Map simple `do concurrent` loops to OpenMP host constructs (PR #127633)

@@ -152,26 +199,140 @@ class DoConcurrentConversion : public mlir::OpConversionPattern { public: using mlir::OpConversionPattern::OpConversionPattern; - DoConcurrentConversion(mlir::MLIRContext *context, bool mapToDevice) - : OpConversionPattern(context), mapToDevice(

[llvm-branch-commits] [flang] [flang][OpenMP] Map simple `do concurrent` loops to OpenMP host constructs (PR #127633)

@@ -24,7 +25,50 @@ namespace flangomp { namespace { namespace looputils { -using LoopNest = llvm::SetVector; +/// Stores info needed about the induction/iteration variable for each `do +/// concurrent` in a loop nest. +struct InductionVariableInfo { + /// the operation alloca

[llvm-branch-commits] [flang] [flang][OpenMP] Map simple `do concurrent` loops to OpenMP host constructs (PR #127633)

@@ -0,0 +1,45 @@ +! RUN: %flang_fc1 -emit-hlfir -fopenmp -fdo-concurrent-to-openmp=host %s -o - \ +! RUN: | FileCheck %s + +program main +implicit none + +call foo(10) + +contains +subroutine foo(n) +implicit none +integer :: n +

[llvm-branch-commits] [flang] [mlir] [mlir][OpenMP] Pack task private variables into a heap-allocated context struct (PR #125307)

https://github.com/skatrak edited https://github.com/llvm/llvm-project/pull/125307 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits

[llvm-branch-commits] [flang] [mlir] [mlir][OpenMP] Pack task private variables into a heap-allocated context struct (PR #125307)

https://github.com/skatrak approved this pull request. It appears a couple of the changes got somehow lost, but no need for another review from me. LGTM, thanks! https://github.com/llvm/llvm-project/pull/125307 ___ llvm-branch-commits mailing list llv

[llvm-branch-commits] [flang] [Flang][OpenMP] Allow host evaluation of loop bounds for distribute (PR #127822)

https://github.com/skatrak updated https://github.com/llvm/llvm-project/pull/127822 >From bd0589979a0db8bb6c5b7da57ac2054ce690a1f9 Mon Sep 17 00:00:00 2001 From: Sergio Afonso Date: Wed, 19 Feb 2025 15:15:01 + Subject: [PATCH] [Flang][OpenMP] Allow host evaluation of loop bounds for distri

[llvm-branch-commits] [mlir] [MLIR][OpenMP] Support target SPMD (PR #127821)

https://github.com/skatrak updated https://github.com/llvm/llvm-project/pull/127821 >From 03651527c28805cb1d786c5674d4bf5644bd1642 Mon Sep 17 00:00:00 2001 From: Sergio Afonso Date: Wed, 19 Feb 2025 14:41:12 + Subject: [PATCH] [MLIR][OpenMP] Support target SPMD This patch implements MLIR t

[llvm-branch-commits] [llvm] [OpenMPIRBuilder] Split calculation of canonical loop trip count, NFC (PR #127820)

https://github.com/skatrak updated https://github.com/llvm/llvm-project/pull/127820 >From aa6182cc887136d44c8fd180a702f62b381f9b5c Mon Sep 17 00:00:00 2001 From: Sergio Afonso Date: Tue, 18 Feb 2025 14:19:30 + Subject: [PATCH] [OpenMPIRBuilder] Split calculation of canonical loop trip coun

[llvm-branch-commits] [mlir] [MLIR][OpenMP] Host lowering of distribute-parallel-do/for (PR #127819)

https://github.com/skatrak updated https://github.com/llvm/llvm-project/pull/127819 >From e61c7975523361b02d1be6f270edfc4cdc6b4f66 Mon Sep 17 00:00:00 2001 From: Sergio Afonso Date: Tue, 18 Feb 2025 13:07:51 + Subject: [PATCH] [MLIR][OpenMP] Host lowering of distribute-parallel-do/for This

[llvm-branch-commits] [llvm] [OpenMPIRBuilder] Add support for distribute-parallel-for/do constructs (PR #127818)

https://github.com/skatrak updated https://github.com/llvm/llvm-project/pull/127818 >From d164acf8d6a573cfc3e5a8d46f8d242fc3cd190f Mon Sep 17 00:00:00 2001 From: Sergio Afonso Date: Tue, 18 Feb 2025 12:04:53 + Subject: [PATCH] [OpenMPIRBuilder] Add support for distribute-parallel-for/do co

[llvm-branch-commits] [mlir] [MLIR][OpenMP] Host lowering of standalone distribute (PR #127817)

https://github.com/skatrak updated https://github.com/llvm/llvm-project/pull/127817 >From 654c63e0489922e0abac069a95f2433d20f65e53 Mon Sep 17 00:00:00 2001 From: Sergio Afonso Date: Tue, 18 Feb 2025 11:22:43 + Subject: [PATCH] [MLIR][OpenMP] Host lowering of standalone distribute This patc

[llvm-branch-commits] [llvm] [OpenMPIRBuilder] Add support for distribute constructs (PR #127816)

https://github.com/skatrak updated https://github.com/llvm/llvm-project/pull/127816 >From 26638a2456dbc7b423d1f5bbb92650398f76098a Mon Sep 17 00:00:00 2001 From: Dominik Adamski Date: Mon, 17 Feb 2025 14:25:40 + Subject: [PATCH] [OpenMPIRBuilder] Add support for distribute constructs This

[llvm-branch-commits] [flang] [flang][OpenMP] Extend `do concurrent` mapping to multi-range loops (PR #127634)

@@ -102,6 +105,47 @@ mlir::Operation *findLoopIterationVarMemDecl(fir::DoLoopOp doLoop) { return result.getDefiningOp(); } +/// Collects the op(s) responsible for updating a loop's iteration variable with +/// the current iteration number. For example, for the input IR: +/

[llvm-branch-commits] [flang] [flang][OpenMP] Handle "loop-local values" in `do concurrent` nests (PR #127635)

https://github.com/skatrak edited https://github.com/llvm/llvm-project/pull/127635 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits

[llvm-branch-commits] [flang] [flang][OpenMP] Handle "loop-local values" in `do concurrent` nests (PR #127635)

@@ -0,0 +1,62 @@ +! Tests that "loop-local values" are properly handled by localizing them to the +! body of the loop nest. See `collectLoopLocalValues` and `localizeLoopLocalValue` +! for a definition of "loop-local values" and how they are handled. skatrak wrot

[llvm-branch-commits] [flang] [flang][OpenMP] Handle "loop-local values" in `do concurrent` nests (PR #127635)

@@ -361,6 +361,64 @@ void sinkLoopIVArgs(mlir::ConversionPatternRewriter &rewriter, ++idx; } } + +/// Collects values that are local to a loop: "loop-local values". A loop-local +/// value is one that is used exclusively inside the loop but allocated outside +/// of it.

[llvm-branch-commits] [flang] [flang][OpenMP] Handle "loop-local values" in `do concurrent` nests (PR #127635)

@@ -202,6 +202,57 @@ variables: `i` and `j`. These are locally allocated inside the parallel/target OpenMP region similar to what the single-range example in previous section shows. +### Data environment + +By default, variables that are used inside a `do concurrent` loop nes

[llvm-branch-commits] [flang] [flang][OpenMP] Handle "loop-local values" in `do concurrent` nests (PR #127635)

@@ -202,6 +202,57 @@ variables: `i` and `j`. These are locally allocated inside the parallel/target OpenMP region similar to what the single-range example in previous section shows. +### Data environment + +By default, variables that are used inside a `do concurrent` loop nes

[llvm-branch-commits] [flang] [flang][OpenMP] Handle "loop-local values" in `do concurrent` nests (PR #127635)

https://github.com/skatrak approved this pull request. I have a couple of nits, but LGTM otherwise. Thank you! https://github.com/llvm/llvm-project/pull/127635 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.o

[llvm-branch-commits] [flang] [flang][OpenMP] Handle "loop-local values" in `do concurrent` nests (PR #127635)

@@ -361,6 +361,64 @@ void sinkLoopIVArgs(mlir::ConversionPatternRewriter &rewriter, ++idx; } } + +/// Collects values that are local to a loop: "loop-local values". A loop-local +/// value is one that is used exclusively inside the loop but allocated outside +/// of it.

[llvm-branch-commits] [flang] [flang][OpenMP] Map simple `do concurrent` loops to OpenMP host constructs (PR #127633)

@@ -152,26 +231,136 @@ class DoConcurrentConversion : public mlir::OpConversionPattern { public: using mlir::OpConversionPattern::OpConversionPattern; - DoConcurrentConversion(mlir::MLIRContext *context, bool mapToDevice) - : OpConversionPattern(context), mapToDevice(

[llvm-branch-commits] [flang] [flang][OpenMP] Extend `do concurrent` mapping to multi-range loops (PR #127634)

@@ -102,6 +105,47 @@ mlir::Operation *findLoopIterationVarMemDecl(fir::DoLoopOp doLoop) { return result.getDefiningOp(); } +/// Collects the op(s) responsible for updating a loop's iteration variable with +/// the current iteration number. For example, for the input IR: --

[llvm-branch-commits] [flang] [flang][OpenMP] Extend `do concurrent` mapping to multi-range loops (PR #127634)

https://github.com/skatrak edited https://github.com/llvm/llvm-project/pull/127634 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits

  1   2   3   4   5   6   7   >