Skip to content

Tracking: full device debug metadata for variables and types #226

New issue
New issue
Open
4 of 4 issues completed
Open
Tracking: full device debug metadata for variables and types#226
4 of 4 issues completed
Labels
IR-loweringLowering between dialects (dialect-mir to LLVM dialect)codegenDevice code-generation pipeline (Rust MIR to IR to PTX)cuda-oxideSIMT programming model: rustc backend, cargo oxide, cuda-device, cuda-host, bookdebug infoDebug information in generated PTX/IR (line/source maps)enhancementNew feature or requestllvm-exportTextual LLVM/NVVM IR exporter (llvm-export crate)trackingUmbrella tracking issue
Milestone
cuda oxide v0.5.0

Description

@nihalpasham
nihalpasham
opened on Jun 15, 2026

Summary

PR #225 adds the first useful device-debug tier: line tables. That gives cuda-gdb source breakpoints, stepping, and basic backtraces.

This issue tracks the larger follow-up: full debug metadata for variables, arguments, and Rust types.

Goal

Make debugger workflows like these possible for device code:

(cuda-gdb) info locals
(cuda-gdb) print idx
(cuda-gdb) print some_struct.field

Likely work

  • carry rustc Body::var_debug_info into the MIR/importer pipeline
  • preserve enough variable/place information through lowering
  • emit DILocalVariable for args and locals
  • emit a useful DIType tree for the first supported Rust type surface
  • emit llvm.dbg.value / equivalent variable location metadata
  • decide how this interacts with optimization, likely including an opt-skip or debug-preserving mode
  • add cuda-gdb validation for local/argument inspection

Non-goal

Line-table source mapping is not tracked here. That is handled by #225.

This issue is about variable/type inspection, not just file/line/column mapping.

Activity

  1. nihalpasham
    added
    enhancementNew feature or request
    debug infoDebug information in generated PTX/IR (line/source maps)
    llvm-exportTextual LLVM/NVVM IR exporter (llvm-export crate)
    codegenDevice code-generation pipeline (Rust MIR to IR to PTX)
    trackingUmbrella tracking issue
    IR-loweringLowering between dialects (dialect-mir to LLVM dialect)
    on Jun 15, 2026

  2. nihalpasham commented on Jun 16, 2026

    @nihalpasham
    nihalpasham
    on Jun 16, 2026
    CollaboratorAuthor

    Opened #233 as the first CUDA_OXIDE_DEBUG=full implementation slice.

    What #233 covers:

    rustc var_debug_info
  -> tagged MIR local alloca
  -> LLVM alloca + DILocalVariable + DIType + dbg.declare
  -> LLVM opt may salvage into #dbg_value
  -> cuda-gdb can inspect supported locals/args where locations survive

    A few choices from the implementation:

    • full debug skips Pliron mem2reg for now, because Pliron does not yet rewrite dbg.declare into dbg.value
    • LLVM opt -O2 stays enabled, since LLVM can salvage many debug locations itself
    • line-table mode strips only PTX .target ..., debug after llc so source breakpoints stay fast
    • full mode keeps PTX .target ..., debug because cuda-gdb needs it for local/argument inspection

    What remains here:

    • richer Rust type trees: ADTs, tuples, arrays, slices, closures
    • fragments/projections like fields, derefs, destructuring
    • precise lexical scopes
    • constants from var_debug_info
    • Pliron-side debug salvage so optimized passes can produce dbg.value directly
  3. nihalpasham
    changed the title [-]Track full device debug metadata for variables and types[/-] [+]Tracking: full device debug metadata for variables and types[/+] on Jun 24, 2026
  4. nihalpasham
    added sub-issues on Aug 21, 2026

  5. nihalpasham commented on Aug 21, 2026

    @nihalpasham
    nihalpasham
    on Aug 21, 2026
    CollaboratorAuthor

    cuda-devtools debugger-assessment follow-ups (validated against main 6c5458fe991bbde32c5bee74d87822aef1b5a691):

    Full variable/type debug children:

    Related line-table follow-ups (outside #226's stated scope; see #224/#225):

    External cuda-devtools dependency:

    The trap() versus breakpoint() API question is expected behavior, not an eighth issue: they intentionally match CUDA C++ __trap() and __brkpt().

  6. nihalpasham
    added this to the cuda oxide v0.5.0 milestone on Oct 25, 2026
  7. nihalpasham
    added
    cuda-oxideSIMT programming model: rustc backend, cargo oxide, cuda-device, cuda-host, book
    on Oct 29, 2026
  8. nihalpasham
    added a commit that references this issue on Nov 4, 2026

    feat(mir-importer): emit host-style { data_ptr, length } DWARF struct…

    d6f1466
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    IR-loweringLowering between dialects (dialect-mir to LLVM dialect)codegenDevice code-generation pipeline (Rust MIR to IR to PTX)cuda-oxideSIMT programming model: rustc backend, cargo oxide, cuda-device, cuda-host, bookdebug infoDebug information in generated PTX/IR (line/source maps)enhancementNew feature or requestllvm-exportTextual LLVM/NVVM IR exporter (llvm-export crate)trackingUmbrella tracking issue

    Type

    No type

    Projects

    No projects

      Milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions