[PATCH] D146840: [AMDGPU] Replace target feature for global fadd32

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rGa955a3189637: [AMDGPU] Replace target feature for global fadd32 (authored by gandhi21299). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST

[PATCH] D146840: [AMDGPU] Replace target feature for global fadd32

gandhi21299 added a comment. Sounds good, thanks the review @rampitec Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D146840/new/ https://reviews.llvm.org/D146840 ___ cfe-commits mailing list cfe-commits@l

[PATCH] D146840: [AMDGPU] Replace target feature for global fadd32

gandhi21299 updated this revision to Diff 509127. gandhi21299 added a comment. - Adding tests for gfx90a and gfx940 Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D146840/new/ https://reviews.llvm.org/D146840 Files: clang/include/clang/Basic/Built

[PATCH] D146840: [AMDGPU] Replace target feature for global fadd32

gandhi21299 updated this revision to Diff 509067. gandhi21299 added a comment. - removed builtins-amdgcn-gfx908.cl Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D146840/new/ https://reviews.llvm.org/D146840 Files: clang/include/clang/Basic/Builti

[PATCH] D146840: [AMDGPU] Replace target feature for global fadd32

gandhi21299 updated this revision to Diff 509066. gandhi21299 added a comment. - gfx908 does not support return version of the builtin Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D146840/new/ https://reviews.llvm.org/D146840 Files: clang/includ

[PATCH] D146840: [AMDGPU] Replace target feature for global fadd32

gandhi21299 created this revision. gandhi21299 added reviewers: foad, arsenm, rampitec, Joe_Nash. Herald added subscribers: kosarev, StephenFan, kerbowa, tpr, dstuttard, yaxunl, jvesely, kzhuravl. Herald added a project: All. gandhi21299 requested review of this revision. Herald added subscribers:

[PATCH] D92956: Fix range-loop-analysis checks for trivial copyability

gandhi21299 added a comment. Perhaps, a description could be added to this patch before committing. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D92956/new/ https://reviews.llvm.org/D92956 ___ cfe-commit

[PATCH] D92956: Fix range-loop-analysis checks for trivial copyability

gandhi21299 added a comment. Herald added a project: All. @Quuxplusone I can commit this patch on behalf of @fanfuqiang. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D92956/new/ https://reviews.llvm.org/D92956 _

[PATCH] D114553: [HIP] Add atomic load, atomic store and atomic cmpxchng_weak builtin support in HIP-clang

This revision was automatically updated to reflect the committed changes. Closed by commit rGdf0560ca0018: [HIP] Add atomic load, atomic store and atomic cmpxchng_weak builtin support in… (authored by gandhi21299). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews

[PATCH] D114553: [HIP] Add atomic load, atomic store and atomic cmpxchng_weak builtin support in HIP-clang

gandhi21299 added a comment. Thanks for the review, I will merge this patch in. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D114553/new/ https://reviews.llvm.org/D114553 ___ cfe-commits mailing list cfe

[PATCH] D114553: [HIP] Add atomic load, atomic store and atomic cmpxchng_weak builtin support in HIP-clang

gandhi21299 updated this revision to Diff 390377. gandhi21299 added a comment. - replaced integers with memory order macros in SemaCUDA/atomic-ops.cu Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D114553/new/ https://reviews.llvm.org/D114553 Files:

[PATCH] D114553: [HIP] Add atomic load, atomic store and atomic cmpxchng_weak builtin support in HIP-clang

gandhi21299 updated this revision to Diff 390094. gandhi21299 added a comment. applied clang-format Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D114553/new/ https://reviews.llvm.org/D114553 Files: clang/include/clang/AST/Expr.h clang/include/

[PATCH] D114553: [HIP] Add atomic load, atomic store and atomic cmpxchng_weak builtin support in HIP-clang

gandhi21299 updated this revision to Diff 390093. gandhi21299 added a comment. addressed feedback including - correction of function signatures - checking for order combinations - correcting expected store atomic instruction - using predefined macros for memory order Repository: rG LLVM Githu

[PATCH] D114553: [HIP] Add atomic load, atomic store and atomic cmpxchng_weak builtin support in HIP-clang

gandhi21299 added a comment. Passed internal CI Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D114553/new/ https://reviews.llvm.org/D114553 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://

[PATCH] D114553: [HIP] Add atomic load, atomic store and atomic cmpxchng_weak builtin support in HIP-clang

gandhi21299 updated this revision to Diff 389834. gandhi21299 added a comment. - applied clang-format Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D114553/new/ https://reviews.llvm.org/D114553 Files: clang/include/clang/AST/Expr.h clang/includ

[PATCH] D114553: [HIP] Add atomic load, atomic store and atomic cmpxchng_weak builtin support in HIP-clang

gandhi21299 updated this revision to Diff 389833. gandhi21299 added a comment. - added order argument in the builtins and changed the tests accordingly - adding Sema test Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D114553/new/ https://reviews.llv

[PATCH] D114025: [clang][NFC] Inclusive terms: replace some uses of sanity in clang

gandhi21299 added inline comments. Comment at: clang/lib/Sema/SemaChecking.cpp:5536 +// GCC does not enforce these rules for GNU atomics, but we do, because if +// we didn't it would be very confusing. FIXME: For whom? How so? auto IsAllowedValueType = [&](QualType

[PATCH] D113925: [HIP] Add HIP scope atomic operations

gandhi21299 added inline comments. Comment at: clang/lib/Sema/SemaChecking.cpp:5587 // the GNU atomics specification, but we enforce it, because if we didn't it // would be very confusing. FIXME: For whom? How so? Diag(ExprRange.getBegin(), diag::err_atomic_op_ne

[PATCH] D113925: [HIP] Add HIP scope atomic operations

gandhi21299 added inline comments. Comment at: clang/lib/Sema/SemaChecking.cpp:5587 // the GNU atomics specification, but we enforce it, because if we didn't it // would be very confusing. FIXME: For whom? How so? Diag(ExprRange.getBegin(), diag::err_atomic_op_ne

[PATCH] D114553: [HIP] Add atomic load, atomic store and atomic cmpxchng_weak builtin support in HIP-clang

gandhi21299 updated this revision to Diff 389577. gandhi21299 added a comment. removed changes in atomic-ops.cl, CUDA test may be sufficient. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D114553/new/ https://reviews.llvm.org/D114553 Files: clang

[PATCH] D114553: [HIP] Add atomic load, atomic store and atomic cmpxchng_weak builtin support in HIP-clang

gandhi21299 created this revision. gandhi21299 added reviewers: yaxunl, b-sumner, t-tye, tra, rjmccall. gandhi21299 requested review of this revision. Herald added a project: clang. Herald added a subscriber: cfe-commits. Introduce __hip_atomic_load, __hip_atomic_store and __hip_atomic_compare_ex

[PATCH] D92956: Fix range-loop-analysis checks for trivial copyability

gandhi21299 accepted this revision. gandhi21299 added a comment. This revision is now accepted and ready to land. LGTM, thanks for the bug fix! Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D92956/new/ https://reviews.llvm.org/D92956 __

[PATCH] D113925: [HIP] Add HIP scope atomic operations

gandhi21299 added a comment. @yaxunl thanks for the review! My Github account is locked unfortunately so I will have to ask you to push this commit to the main branch. Thank you! Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D113925/new/ https://re

[PATCH] D113925: [HIP] Add HIP scope atomic operations

gandhi21299 updated this revision to Diff 387668. gandhi21299 marked an inline comment as done. gandhi21299 added a comment. - removed debug code - some macro definitions need to be defined for HIP-only compilation Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://review

[PATCH] D113925: [HIP] Add HIP scope atomic operations

gandhi21299 updated this revision to Diff 387396. gandhi21299 added a comment. reapplied clang-format Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D113925/new/ https://reviews.llvm.org/D113925 Files: clang/include/clang/AST/Expr.h clang/includ

[PATCH] D113925: [HIP] Add HIP scope atomic operations

gandhi21299 updated this revision to Diff 387395. gandhi21299 added a comment. clang-formatted code Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D113925/new/ https://reviews.llvm.org/D113925 Files: clang/include/clang/AST/Expr.h clang/include/

[PATCH] D113925: [HIP] Add HIP scope atomic operations

gandhi21299 created this revision. gandhi21299 added reviewers: yaxunl, b-sumner, t-tye, tra, rjmccall. Herald added subscribers: dexonsmith, Anastasia. gandhi21299 requested review of this revision. Herald added a project: clang. Herald added a subscriber: cfe-commits. Add an AtomicScopeModel for

[PATCH] D109707: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols

gandhi21299 added inline comments. Comment at: llvm/lib/Target/AMDGPU/AMDGPUResourceUsageAnalysis.cpp:67 + if (auto *GA = dyn_cast(Op.getGlobal())) +return cast(GA->getOperand(0)); return cast(Op.getGlobal()); arsenm wrote: > gandhi21299 wrote: > > arsenm

[PATCH] D109707: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols

gandhi21299 added inline comments. Comment at: llvm/lib/Target/AMDGPU/AMDGPUResourceUsageAnalysis.cpp:67 + if (auto *GA = dyn_cast(Op.getGlobal())) +return cast(GA->getOperand(0)); return cast(Op.getGlobal()); arsenm wrote: > I thought aliases could inclu

[PATCH] D109707: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols

This revision was automatically updated to reflect the committed changes. Closed by commit rG0567f0333176: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols (authored by gandhi21299). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.

[PATCH] D109707: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols

gandhi21299 added inline comments. Comment at: clang/test/CodeGenCUDA/amdgpu-alias-undef-symbols.cu:3 + +// RUN: %clang -target x86_64-unknown-linux-gnu --offload-arch=gfx906 --cuda-device-only -nogpulib -nogpuinc -x hip -emit-llvm -S -o - %s \ +// RUN: -fgpu-rdc -O3 -mllvm -a

[PATCH] D109707: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols

gandhi21299 added a comment. @thakis can you please check if this solution is sufficient? Thanks for bringing it up Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D109707/new/ https://reviews.llvm.org/D109707 ___

[PATCH] D109707: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols

gandhi21299 updated this revision to Diff 380110. gandhi21299 added a comment. added -target option in the test amdgpu-alias-undef-symbols.cu Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D109707/new/ https://reviews.llvm.org/D109707 Files: clang

[PATCH] D111734: [HIP] Relax conditions for address space cast in builtin args

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rGf92db6d3fff1: [HIP] Relax conditions for address space cast in builtin args (authored by gandhi21299). Repository: rG LLVM Github Monorepo CHANGE

[PATCH] D111734: [HIP] Relax conditions for address space cast in builtin args

gandhi21299 updated this revision to Diff 380098. gandhi21299 added a comment. cleaned up code for readability Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D111734/new/ https://reviews.llvm.org/D111734 Files: clang/lib/Sema/SemaExpr.cpp clang/

[PATCH] D111734: [HIP] Relax conditions for address space cast in builtin args

gandhi21299 added a comment. @tra I see, we sure can rewrite that segment for readability. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D111734/new/ https://reviews.llvm.org/D111734 ___ cfe-commits maili

[PATCH] D111734: [HIP] Relax conditions for address space cast in builtin args

This revision was automatically updated to reflect the committed changes. Closed by commit rG3b48e1170dc6: [HIP] Relax conditions for address space cast in builtin args (authored by gandhi21299). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D111734/n

[PATCH] D111734: [HIP] Relax conditions for address space cast in builtin args

gandhi21299 added a comment. Thanks for the review! Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D111734/new/ https://reviews.llvm.org/D111734 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http

[PATCH] D111734: [HIP] Relax conditions for address space cast in builtin args

gandhi21299 updated this revision to Diff 380067. gandhi21299 marked 3 inline comments as done. gandhi21299 added a comment. corrected sema test, as requested Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D111734/new/ https://reviews.llvm.org/D11173

[PATCH] D109707: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols

This revision was automatically updated to reflect the committed changes. Closed by commit rG03375a3fb33b: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols (authored by gandhi21299). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.

[PATCH] D111734: [HIP] Relax conditions for address space cast in builtin args

gandhi21299 updated this revision to Diff 380048. gandhi21299 added a comment. removed unused diagnostic sema note Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D111734/new/ https://reviews.llvm.org/D111734 Files: clang/lib/Sema/SemaExpr.cpp cl

[PATCH] D111734: [HIP] Relax conditions for address space cast in builtin args

gandhi21299 updated this revision to Diff 380047. gandhi21299 added a comment. installed clang-format, refreshing patch Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D111734/new/ https://reviews.llvm.org/D111734 Files: clang/include/clang/Basic/D

[PATCH] D111734: [HIP] Relax conditions for address space cast in builtin args

gandhi21299 updated this revision to Diff 380046. gandhi21299 added a comment. adding sema test Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D111734/new/ https://reviews.llvm.org/D111734 Files: clang/include/clang/Basic/DiagnosticSemaKinds.td

[PATCH] D109707: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols

gandhi21299 added a comment. ping Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D109707/new/ https://reviews.llvm.org/D109707 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org

[PATCH] D111734: [HIP] Relax conditions for address space cast in builtin args

gandhi21299 added a comment. Passed internal CI, still working on a Sema test. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D111734/new/ https://reviews.llvm.org/D111734 ___ cfe-commits mailing list cfe-

[PATCH] D109707: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols

gandhi21299 added a comment. Passed ePSDB Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D109707/new/ https://reviews.llvm.org/D109707 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.

[PATCH] D109707: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols

gandhi21299 updated this revision to Diff 379530. gandhi21299 added a comment. add a restrictions to what architecture AlwaysInliner should run on, updated the inline-calls.ll test. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D109707/new/ https:/

[PATCH] D111734: [HIP] Relax conditions for address space cast in builtin args

gandhi21299 updated this revision to Diff 379444. gandhi21299 marked an inline comment as done. gandhi21299 added a comment. adding codegen test Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D111734/new/ https://reviews.llvm.org/D111734 Files: cl

[PATCH] D111734: [HIP] Relax conditions for address space cast in builtin args

gandhi21299 updated this revision to Diff 379419. gandhi21299 added a comment. removed irrelevant lines in the test Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D111734/new/ https://reviews.llvm.org/D111734 Files: clang/lib/Sema/SemaExpr.cpp c

[PATCH] D111734: [HIP] Relax conditions for address space cast in builtin args

gandhi21299 created this revision. gandhi21299 added a reviewer: yaxunl. gandhi21299 requested review of this revision. Herald added a project: clang. Herald added a subscriber: cfe-commits. Allow (implicit) address space casting between LLVM-equivalent target address spaces. Repository: rG LL

[PATCH] D109707: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols

gandhi21299 added inline comments. Comment at: llvm/test/CodeGen/AMDGPU/inline-calls.ll:3 ; RUN: llc -march=amdgcn -mcpu=tonga -verify-machineinstrs < %s | FileCheck %s -; RUN: llc -march=r600 -mcpu=redwood -verify-machineinstrs < %s | FileCheck %s @tstellar

[PATCH] D109707: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols

gandhi21299 added a comment. Passed internal CI Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D109707/new/ https://reviews.llvm.org/D109707 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://

[PATCH] D109707: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols

gandhi21299 updated this revision to Diff 378218. gandhi21299 added a comment. added -nogpulib and -nogpuinc flags to amdgpu-alias-undef-symbols.cu Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D109707/new/ https://reviews.llvm.org/D109707 Files:

[PATCH] D109707: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols

gandhi21299 added a comment. ping Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D109707/new/ https://reviews.llvm.org/D109707 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org

[PATCH] D109707: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols

gandhi21299 updated this revision to Diff 377922. gandhi21299 added a comment. refreshing patch Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D109707/new/ https://reviews.llvm.org/D109707 Files: clang/lib/Driver/ToolChains/Clang.cpp clang/test/

[PATCH] D109707: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols

gandhi21299 added a comment. ping Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D109707/new/ https://reviews.llvm.org/D109707 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org

[PATCH] D109707: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols

gandhi21299 updated this revision to Diff 376559. gandhi21299 added a comment. - Since callees may alias to a function pointer, it makes sense for `getCalleeFunction(...)` to return a `Function` which is a cast of the operand of a `GlobalAlias`. Repository: rG LLVM Github Monorepo CHANGES S

[PATCH] D109707: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols

gandhi21299 updated this revision to Diff 376564. gandhi21299 added a comment. - eliminated changes in SIISelLowering Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D109707/new/ https://reviews.llvm.org/D109707 Files: clang/lib/Driver/ToolChains/C

[PATCH] D109707: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols

gandhi21299 added a comment. inline-calls.ll failed on gfx908 due to the change in SIISelLowering.cpp, line 3015. Without the change, there is a failure in AMDGPUResourceAnalysis.cpp, line 65 because Op.getGlobal() is not a Function. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST AC

[PATCH] D109707: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols

gandhi21299 reclaimed this revision. gandhi21299 added a comment. Sorry, that was a mistake. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D109707/new/ https://reviews.llvm.org/D109707 ___ cfe-commits mai

[PATCH] D110772: [clang] Allow pointers from any address space to gfx90a builtins

gandhi21299 updated this revision to Diff 376095. gandhi21299 marked an inline comment as done. gandhi21299 added a comment. - corrected test, address space cast should go under `unsafeAtomicAdd(...)` codegen Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llv

[PATCH] D110772: [clang] Allow pointers from any address space to gfx90a builtins

gandhi21299 created this revision. gandhi21299 added reviewers: yaxunl, rampitec. gandhi21299 requested review of this revision. Herald added a project: clang. Herald added a subscriber: cfe-commits. Explicitly address space cast the pointer if its address space differs from the address space of t

[PATCH] D109707: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols

gandhi21299 added inline comments. Comment at: llvm/lib/Target/AMDGPU/AMDGPUAlwaysInlinePass.cpp:96-97 if (Function* F = dyn_cast(A.getAliasee())) { + if (A.getLinkage() != GlobalValue::InternalLinkage) +continue; A.replaceAllUsesWith(F);

[PATCH] D109707: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols

gandhi21299 updated this revision to Diff 375655. gandhi21299 added a comment. - declare failure when lowering an accessor of a callee which is not a function, in GlobalISel Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D109707/new/ https://reviews

[PATCH] D109707: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols

gandhi21299 added a comment. @yaxunl Should inline-calls.ll be converted into an expected failing test or removed? (to avoid cast failure in AMDGPUResourceAnalysis to break the test) Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D109707/new/ https:

[PATCH] D109707: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols

gandhi21299 updated this revision to Diff 375284. gandhi21299 marked an inline comment as done. gandhi21299 added a comment. - added the `REQUIRES` line as requested by Sam Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D109707/new/ https://reviews.l

[PATCH] D109707: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols

gandhi21299 added a comment. In D109707#3016438 , @gandhi21299 wrote: > - replaced a `cast` with a `dyn_cast` since the return value from > `getCalleeFunction()` is not always a Function > - `RUN on line 2` was causing 2 more scalar registers to be used

[PATCH] D109707: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols

gandhi21299 added a comment. It does not look like function calls are supported yet in AMDGPUCallLowering, is that correct? Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D109707/new/ https://reviews.llvm.org/D109707 ___

[PATCH] D109707: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols

gandhi21299 updated this revision to Diff 375079. gandhi21299 added a comment. - Declare an unhandled call lowering in SelectionDAG when a callee is encountered which cannot be casted into a Function - I am still investigating the effects on GlobalISel side of things, there seems to be a problem

[PATCH] D109707: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols

gandhi21299 added inline comments. Comment at: llvm/lib/Target/AMDGPU/AMDGPUResourceUsageAnalysis.cpp:65 - return cast(Op.getGlobal()); + return dyn_cast(Op.getGlobal()); } arsenm wrote: > I think this is not the right place for this. If we can determine the

[PATCH] D109707: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols

gandhi21299 updated this revision to Diff 374369. gandhi21299 added a comment. - refreshing patch Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D109707/new/ https://reviews.llvm.org/D109707 Files: clang/lib/Driver/ToolChains/Clang.cpp clang/tes

[PATCH] D109707: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols

gandhi21299 updated this revision to Diff 374354. gandhi21299 added a comment. - replaced a `cast` with a `dyn_cast` since the return value from `getCalleeFunction()` is not always a Function - `RUN on line 2` was causing 2 more scalar registers to be used on tonga due to @func_alias not being i

[PATCH] D109707: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols

gandhi21299 updated this revision to Diff 373257. gandhi21299 marked an inline comment as done. gandhi21299 added a comment. - Prevent removing alias if the GlobalAlias does not have internal linkage Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D109

[PATCH] D109707: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols

gandhi21299 added a comment. @yaxunl I think we have two ways to go from here: 1. If appropriate, reset the maximum number of scalar registers allowed in `@kernel3` (inline-calls.ll) to fix the test. 2. Determine a stronger condition for inlining. Repository: rG LLVM Github Monorepo CHANGES

[PATCH] D109707: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols

gandhi21299 added a comment. Internal linkage detection works great for our purposes but it causes a failure in llvm/test/CodeGen/AMDGPU/inline-calls.ll due to `@func_alias` unable to be casted into a `Function`. If we pass through that, the `@kernel3` causes the error: `scalar registers (98) e

[PATCH] D109707: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols

gandhi21299 added a comment. @yaxunl Under what criteria should an alias not be removed? Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D109707/new/ https://reviews.llvm.org/D109707 ___ cfe-commits mailing

[PATCH] D109707: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols

gandhi21299 updated this revision to Diff 372573. gandhi21299 added a comment. - converted the HIP test into a CUDA test Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D109707/new/ https://reviews.llvm.org/D109707 Files: clang/lib/Driver/ToolChain

[PATCH] D109707: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols

gandhi21299 updated this revision to Diff 372551. gandhi21299 added a comment. - added the include header for HIP runtime Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D109707/new/ https://reviews.llvm.org/D109707 Files: clang/lib/Driver/ToolChai

[PATCH] D109707: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols

gandhi21299 added inline comments. Comment at: clang/lib/Driver/ToolChains/Clang.cpp:5069 // where aliases aren't supported. - if (!RawTriple.isOSDarwin() && !RawTriple.isNVPTX() && !RawTriple.isAMDGPU()) CmdArgs.push_back("-mconstructor-aliases"); gandh

[PATCH] D109707: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols

gandhi21299 updated this revision to Diff 372528. gandhi21299 marked an inline comment as done. gandhi21299 added a comment. - set `GlobalOpt` parameter to false by default to disallow alias elimination when the options EarlyInlineAll and EnableFunctionCalls are true and false, respectively. R

[PATCH] D109707: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols

gandhi21299 marked an inline comment as done. gandhi21299 added inline comments. Comment at: clang/lib/Driver/ToolChains/Clang.cpp:5069 // where aliases aren't supported. - if (!RawTriple.isOSDarwin() && !RawTriple.isNVPTX() && !RawTriple.isAMDGPU()) CmdArgs.push_back("-

[PATCH] D109707: [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols

gandhi21299 created this revision. gandhi21299 added reviewers: yaxunl, aeubanks. Herald added subscribers: foad, kerbowa, hiraditya, tpr, nhaehnle, jvesely, arsenm. gandhi21299 requested review of this revision. Herald added projects: clang, LLVM. Herald added subscribers: llvm-commits, cfe-commi

[PATCH] D108150: [Remarks] [AMDGPU] Emit optimization remarks for atomics generating hardware instructions

This revision was automatically updated to reflect the committed changes. Closed by commit rG508b06699a39: [Remarks] [AMDGPU] Emit optimization remarks for atomics generating hardware… (authored by gandhi21299). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.ll

[PATCH] D108150: [Remarks] [AMDGPU] Emit optimization remarks for atomics generating hardware instructions

gandhi21299 marked an inline comment as done. gandhi21299 added a comment. Thanks for the review, I will merge this in as soon as the CI passes. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D108150/new/ https://reviews.llvm.org/D108150 ___

[PATCH] D108150: [Remarks] [AMDGPU] Emit optimization remarks for atomics generating hardware instructions

gandhi21299 updated this revision to Diff 367273. gandhi21299 marked an inline comment as done. gandhi21299 added a comment. - code refactor Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D108150/new/ https://reviews.llvm.org/D108150 Files: clang/

[PATCH] D108150: [Remarks] [AMDGPU] Emit optimization remarks for atomics generating hardware instructions

gandhi21299 added a comment. ping Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D108150/new/ https://reviews.llvm.org/D108150 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org

[PATCH] D108150: [Remarks] [AMDGPU] Emit optimization remarks for atomics generating hardware instructions

gandhi21299 updated this revision to Diff 367042. gandhi21299 added a comment. - corrected logic for ORE in SIISelLowering.cpp Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D108150/new/ https://reviews.llvm.org/D108150 Files: clang/test/CodeGenOp

[PATCH] D108150: [Remarks] [AMDGPU] Emit optimization remarks for atomics generating hardware instructions

gandhi21299 added a comment. My understanding is that since we are reporting unsafe expansion into hw instructions, `fpModeMatchesGlobalFPAtomicMode(RMW)` must be false to match the logic. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D108150/new/

[PATCH] D108150: [Remarks] [AMDGPU] Emit optimization remarks for atomics generating hardware instructions

gandhi21299 added a comment. @rampitec Which part of the logic is wrong? Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D108150/new/ https://reviews.llvm.org/D108150 ___ cfe-commits mailing list cfe-commit

[PATCH] D108150: [Remarks] [AMDGPU] Emit optimization remarks for atomics generating hardware instructions

gandhi21299 updated this revision to Diff 367000. gandhi21299 marked an inline comment as done. gandhi21299 added a comment. - corrected AtomicExpansionKind return value in SIISelLowering as pointed out by reviewer - eliminated previous function declaration which is not defined/used anymore Rep

[PATCH] D108150: [Remarks] [AMDGPU] Emit optimization remarks for atomics generating hardware instructions

gandhi21299 updated this revision to Diff 366958. gandhi21299 added a comment. - requested changes from reviewer Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D108150/new/ https://reviews.llvm.org/D108150 Files: clang/test/CodeGenOpenCL/atomics-c

[PATCH] D108150: [Remarks] [AMDGPU] Emit optimization remarks for atomics generating hardware instructions

gandhi21299 marked 3 inline comments as done. gandhi21299 added inline comments. Comment at: llvm/lib/Target/AMDGPU/SIISelLowering.cpp:12195 + if (!fpModeMatchesGlobalFPAtomicMode(RMW)) +return reportUnsafeHWInst(RMW, AtomicExpansionKind::None); ra

[PATCH] D108150: [Remarks] [AMDGPU] Emit optimization remarks for atomics generating hardware instructions

gandhi21299 added a comment. ping Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D108150/new/ https://reviews.llvm.org/D108150 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org

[PATCH] D108150: [Remarks] [AMDGPU] Emit optimization remarks for atomics generating hardware instructions

gandhi21299 updated this revision to Diff 366765. gandhi21299 marked 2 inline comments as done. gandhi21299 added a comment. - split the OpenCL test into two for brevity - fixed a mistake in SIISelLowering as pointed out by reviewer - added the missing -munsafe-fp-atomics flag Repository: rG L

[PATCH] D108150: [Remarks] [AMDGPU] Emit optimization remarks for atomics generating hardware instructions

gandhi21299 marked 5 inline comments as done. gandhi21299 added inline comments. Comment at: clang/test/CodeGenOpenCL/atomics-remarks-gfx90a.cl:9 +// RUN: %clang_cc1 -cl-std=CL2.0 -O0 -triple=amdgcn-amd-amdhsa -target-cpu gfx90a \ +// RUN: -Rpass=si-lower -munsafe-fp-atomi

[PATCH] D108150: [Remarks] [AMDGPU] Emit optimization remarks for atomics generating hardware instructions

gandhi21299 updated this revision to Diff 366752. gandhi21299 added a comment. - corrected an argument in AtomicExpand pass Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D108150/new/ https://reviews.llvm.org/D108150 Files: clang/test/CodeGenOpenC

[PATCH] D108150: [Remarks] [AMDGPU] Emit optimization remarks for atomics generating hardware instructions

gandhi21299 updated this revision to Diff 366751. gandhi21299 added a comment. - added more tests - addressed feedback from reviewer Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D108150/new/ https://reviews.llvm.org/D108150 Files: clang/test/Cod

[PATCH] D106891: [Remarks] Emit optimization remarks for atomics generating CAS loop

This revision was automatically updated to reflect the committed changes. Closed by commit rGf22ba5187350: [Remarks] Emit optimization remarks for atomics generating CAS loop (authored by gandhi21299). Changed prior to commit: https://reviews.llvm.org/D106891?vs=366683&id=366735#toc Repository

[PATCH] D108150: [Remarks] Emit optimization remarks for atomics generating hardware instructions

gandhi21299 created this revision. gandhi21299 added reviewers: rampitec, arsenm, b-sumner. Herald added subscribers: foad, kerbowa, jfb, hiraditya, Anastasia, nhaehnle, jvesely. gandhi21299 requested review of this revision. Herald added subscribers: llvm-commits, cfe-commits, wdng. Herald added

[PATCH] D106891: [Remarks] Emit optimization remarks for atomics generating CAS loop

gandhi21299 added a comment. Will do, thanks! Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D106891/new/ https://reviews.llvm.org/D106891 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://li

  1   2   3   >