Skip to content

SystemZ Backend: Add support for operations such as FP16_TO_FP and FP_TO_FP16 #50374

New issue
New issue
Closed
#109164
Closed
SystemZ Backend: Add support for operations such as FP16_TO_FP and FP_TO_FP16#50374
#109164
Assignees
JonPsson1
Labels
backend:SystemZbugzillaIssues migrated from bugzilla

Description

@kun-lu20
kun-lu20
opened on Jul 8, 2021
Bugzilla Link 51030
Version unspecified
OS Linux
CC @River707,@ftynse

Extended Description

Hi,

Recently we're running test suite of TensorFlow v2.5.0 on s390x (Ubuntu 18.04).

Test case //tensorflow/compiler/tests:sort_ops_test_cpu fails due to the following error:

LLVM ERROR: Cannot select: 0x3ff14167ca0: f32 = fp16_to_fp 0x3ff14167f10
0x3ff14167f10: i32,ch = load<(dereferenceable load 2 from %ir.4, !alias.scope !​6, !noalias !​4), zext from i16> 0x3ff14197548, 0x3ff141678f8, undef:i64
0x3ff141678f8: i64,ch = load<(load 8 from %ir.3)> 0x3ff14197548, 0x3ff14167890, undef:i64
0x3ff14167890: i64 = add nuw 0x3ff141674e8, Constant:i64<8>
0x3ff141674e8: i64,ch = CopyFromReg 0x3ff14197548, Register:i64 %2
0x3ff14167480: i64 = Register %2
0x3ff14167828: i64 = Constant<8>
0x3ff14167758: i64 = undef
0x3ff14167758: i64 = undef
In function: compare_lt_WCTTAtafbb4__.7

Other test cases such as //tensorflow/python/keras/optimizer_v2:adam_test and //tensorflow/core/kernels/mlir_generated:abs_cpu_f16_f16_gen_test also fail on s390x due to similar reasons. A related issue (tensorflow/tensorflow#44362) has been raised in TensorFlow GitHub issues.

We think the root cause is lack of support for operations such as FP16_TO_FP and FP_TO_FP16 which perform promotions and truncation for half-precision (16 bit) floating numbers in the SystemZ LLVM backend (llvm/lib/Target/SystemZ/SystemZISelLowering.cpp). Could these features be considered to add to SystemZ LLVM backend? Thanks!

Activity

  1. kun-lu20 commented on Sep 21, 2021

    @kun-lu20
    kun-lu20
    on Sep 21, 2021
    Author

    Any updates from the community reg this issue? Thanks!

  2. joker-eph commented on Sep 21, 2021

    @joker-eph
    joker-eph
    on Sep 21, 2021
    Contributor

    Moving out of MLIR: this is a backend issue.

  3. kun-lu20 commented on Oct 8, 2021

    @kun-lu20
    kun-lu20
    on Oct 8, 2021
    Author

    It looks like commit https://reviews.llvm.org/rG8cd8120a7b5d which has been tagged as 13.0.0-rc could solve this issue, since it adds support for arch14 and operations related to FP16 conversion to the SystemZ backend. Could anyone from community help to confirm this? Thanks!

  4. Imagellvmbot
    transferred this issue fromllvm/llvm-bugzilla-archiveon Dec 11, 2021

  5. kun-lu20 commented on Apr 7, 2022

    @kun-lu20
    kun-lu20
    on Apr 7, 2022
    Author

    This issue still persists on TensorFlow v2.8.0 which uses LLVM 15. Looks like specific half-precision (16 bit) operations are still missing in SystemZ LLVM backend.

    Can anyone from community take a look at this issue? Thanks very much!

  6. kun-lu20 commented on Jul 6, 2022

    @kun-lu20
    kun-lu20
    on Jul 6, 2022
    Author

    Recently we've run test cases under //tensorflow/core/kernels/mlir_generated category in TensorFlow v2.9.1 and found that this issue still exists.

    Looks like FP16/F16 related operations are still unsupported in LLVM SystemZ backend for most Z cpu models, which causes these test cases (such as abs_cpu_f16_f16_gen_test and sqrt_cpu_f64_f64_gen_test) to fail when applyFullConversion() or applyPartialConversion() function is invoked. Although this commit has added FP16 support in the new arch14 (z16) model, it seems that arch14 still doesn't have full support for FP16 operations.

    We also found that when building TensorFlow with options -c opt --copt=-O which sets optimization level to 1 and with JIT_Compilation enabled, these test cases would pass and the output .mlir files could be generated successfully.

    We think this could be used as a workaround for now, but to address the root cause, FP16 related operations still need to be added to SystemZ backend.

    Any thoughts or suggestions from the community reg this issue would be greatly appreciated. Thanks!

  7. ImageEugeneZelenko
    added
    backend:SystemZ
    on Jul 6, 2022

  8. beetrees commented on Jun 13, 2024

    @beetrees
    beetrees
    on Jun 13, 2024
    Contributor

    The following function (compiler explorer):

    define half @deref(ptr %p) {
  %x = load half, ptr %p
  ret half %x
}

    currently fails to compile when compiling for s390x-unknown-linux-gnu with the following error:

    LLVM ERROR: Cannot select: 0x89e9c70: f32,ch = load<(load (s16) from %ir.p), anyext from f16> 0x89a9bc8, 0x89e9c00, undef:i64
  0x89e9c00: i64,ch = CopyFromReg 0x89a9bc8, Register:i64 %0
    0x89e9b90: i64 = Register %0
  0x89e9ce0: i64 = undef
In function: deref
PLEASE submit a bug report to https://github.com/llvm/llvm-project/issues/ and include the crash backtrace.
Stack dump:
0.	Program arguments: /opt/compiler-explorer/clang-trunk/bin/llc -o /app/output.s -mtriple=s390x-unknown-linux-gnu <source>
1.	Running pass 'Function Pass Manager' on module '<source>'.
2.	Running pass 'SystemZ DAG->DAG Pattern Instruction Selection' on function '@deref'
 #0 0x00000000037197d8 llvm::sys::PrintStackTrace(llvm::raw_ostream&, int) (/opt/compiler-explorer/clang-trunk/bin/llc+0x37197d8)
 #1 0x000000000371714c SignalHandler(int) Signals.cpp:0:0
 #2 0x00007baf41042520 (/lib/x86_64-linux-gnu/libc.so.6+0x42520)
 #3 0x00007baf410969fc pthread_kill (/lib/x86_64-linux-gnu/libc.so.6+0x969fc)
 #4 0x00007baf41042476 gsignal (/lib/x86_64-linux-gnu/libc.so.6+0x42476)
 #5 0x00007baf410287f3 abort (/lib/x86_64-linux-gnu/libc.so.6+0x287f3)
 #6 0x000000000073359e llvm::UniqueStringSaver::save(llvm::StringRef) (.cold) StringSaver.cpp:0:0
 #7 0x00000000034dca44 llvm::SelectionDAGISel::CannotYetSelect(llvm::SDNode*) (/opt/compiler-explorer/clang-trunk/bin/llc+0x34dca44)
 #8 0x00000000034e3e85 llvm::SelectionDAGISel::SelectCodeCommon(llvm::SDNode*, unsigned char const*, unsigned int) (/opt/compiler-explorer/clang-trunk/bin/llc+0x34e3e85)
 #9 0x00000000019ed9de (anonymous namespace)::SystemZDAGToDAGISel::Select(llvm::SDNode*) SystemZISelDAGToDAG.cpp:0:0
#10 0x00000000034d9f94 llvm::SelectionDAGISel::DoInstructionSelection() (/opt/compiler-explorer/clang-trunk/bin/llc+0x34d9f94)
#11 0x00000000034e92a1 llvm::SelectionDAGISel::CodeGenAndEmitDAG() (/opt/compiler-explorer/clang-trunk/bin/llc+0x34e92a1)
#12 0x00000000034ebed4 llvm::SelectionDAGISel::SelectAllBasicBlocks(llvm::Function const&) (/opt/compiler-explorer/clang-trunk/bin/llc+0x34ebed4)
#13 0x00000000034edd44 llvm::SelectionDAGISel::runOnMachineFunction(llvm::MachineFunction&) (/opt/compiler-explorer/clang-trunk/bin/llc+0x34edd44)
#14 0x00000000019efe0a (anonymous namespace)::SystemZDAGToDAGISel::runOnMachineFunction(llvm::MachineFunction&) SystemZISelDAGToDAG.cpp:0:0
#15 0x00000000034dd861 llvm::SelectionDAGISelLegacy::runOnMachineFunction(llvm::MachineFunction&) (/opt/compiler-explorer/clang-trunk/bin/llc+0x34dd861)
#16 0x000000000282216b llvm::MachineFunctionPass::runOnFunction(llvm::Function&) (.part.0) MachineFunctionPass.cpp:0:0
#17 0x0000000002d58b22 llvm::FPPassManager::runOnFunction(llvm::Function&) (/opt/compiler-explorer/clang-trunk/bin/llc+0x2d58b22)
#18 0x0000000002d58ca1 llvm::FPPassManager::runOnModule(llvm::Module&) (/opt/compiler-explorer/clang-trunk/bin/llc+0x2d58ca1)
#19 0x0000000002d5a950 llvm::legacy::PassManagerImpl::run(llvm::Module&) (/opt/compiler-explorer/clang-trunk/bin/llc+0x2d5a950)
#20 0x000000000084df94 compileModule(char**, llvm::LLVMContext&) llc.cpp:0:0
#21 0x0000000000745af6 main (/opt/compiler-explorer/clang-trunk/bin/llc+0x745af6)
#22 0x00007baf41029d90 (/lib/x86_64-linux-gnu/libc.so.6+0x29d90)
#23 0x00007baf41029e40 __libc_start_main (/lib/x86_64-linux-gnu/libc.so.6+0x29e40)
#24 0x0000000000845bce _start (/opt/compiler-explorer/clang-trunk/bin/llc+0x845bce)
Program terminated with signal: SIGSEGV
Compiler returned: 139

    Other operations involving half also fail with similar errors.

  9. alexrp commented on Jun 28, 2024

    @alexrp
    alexrp
    on Jun 28, 2024
    Member

    FWIW, this is the only remaining blocker I'm aware of for Zig to be able to target s390x:

    ❯ zig cc s390x.c -target s390x-linux-musl
LLVM ERROR: Cannot select: 0x6d24170: i32 = fp_to_fp16 0x6d23a00
  0x6d23a00: f32,ch = CopyFromReg 0x5e82a10, Register:f32 %10
    0x6c03da0: f32 = Register %10
In function: __fixhfsi
  10. Imageuweigand
    added a commit that references this issue on Jul 10, 2024

    core: Limit remaining f16 doctests to x86_64 linux

    0065763
  11. Imagematthiaskrgr
    added a commit that references this issue on Jul 11, 2024

    Rollup merge of rust-lang#127588 - uweigand:s390x-f16-doctests, r=tgr…

    380c787
  12. Imagerust-timer
    added a commit that references this issue on Jul 12, 2024

    Unrolled build for rust-lang#127588

    a535016
  13. Imageandrewrk
    added a commit that references this issue on Aug 12, 2024

    llvm: Disable lowering to f16 on s390x.

    82b0f44
  14. ImageJonPsson1
    self-assigned this
    on Oct 7, 2024

  15. JonPsson1 commented on Oct 7, 2024

    @JonPsson1
    JonPsson1
    on Oct 7, 2024
    Contributor

    Patch in progress here: #109164

  16. ImageJonPsson1
    closed this as completedin #109164on Apr 16, 2025
  17. ImageJonPsson1
    closed this as completedin 6d03f51on Apr 16, 2025
  18. Imageswift-ci
    added a commit that references this issue on Apr 16, 2025

    [SystemZ] Add support for 16-bit floating point. (llvm#109164)

    6d03f51
  19. Imagefolkertdev
    added a commit that references this issue on Jan 17, 2026

    `f16`: remove s390x fixme

    98e02d5
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

Labels

backend:SystemZbugzillaIssues migrated from bugzilla

Type

No type

Projects

No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions