Skip to content

device_async_resource_ref failure in tc example #86

Description

@rootjalex

When I try to run the transitive closure example, compilation fails with an undeclared identifier error:

(srdatalog-env) ajroot@aspen:~/projects/srdatalog-python$ CXX=clang++ CC=clang python examples/run_benchmark.py tc --data ~/projects/datalog/data/vsp_finan/edge.facts
[tc] program: 3 relations, 3 rules (0.0s)
[emit] ./build/jit/TcPlan_DB_2061 (main + 1 batches, 0.0s)
[compile] FAILED after 3.2s
vice_accessible>;
      |       ^
/home/ajroot/projects/srdatalog-python/src/srdatalog/runtime/vendor/rmm/include/rmm/resource_ref.hpp:82:10: error: use of undeclared identifier 'device_async_resource_ref'
   82 |   return device_async_resource_ref{*res};
      |          ^~~~~~~~~~~~~~~~~~~~~~~~~
In file included from /home/ajroot/projects/srdatalog-python/build/jit/TcPlan_DB_2061/main.cpp:1:
In file included from /home/ajroot/projects/srdatalog-python/src/srdatalog/runtime/generalized_datalog/srdatalog.h:43:
In file included from /home/ajroot/projects/srdatalog-python/src/srdatalog/runtime/generalized_datalog/relation_col.h:33:
In file included from /home/ajroot/projects/srdatalog-python/src/srdatalog/runtime/generalized_datalog/gpu/aos_device_array.h:8:
In file included from /home/ajroot/projects/srdatalog-python/src/srdatalog/runtime/generalized_datalog/gpu/./device_array.h:26:
/home/ajroot/projects/srdatalog-python/src/srdatalog/runtime/vendor/rmm/include/rmm/mr/device/per_device_resource.hpp:158:47: error: unknown type name 'device_async_resource_ref'; did you mean 'device_resource_ref'?
  158 |   static std::map<cuda_device_id::value_type, device_async_resource_ref> device_id_to_resource_ref;
      |                                               ^~~~~~~~~~~~~~~~~~~~~~~~~
      |                                               device_resource_ref
/home/ajroot/projects/srdatalog-python/src/srdatalog/runtime/vendor/rmm/include/rmm/resource_ref.hpp:35:7: note: 'device_resource_ref' declared here
   35 | using device_resource_ref = cuda::mr::resource_ref<cuda::mr::device_accessible>;
      |       ^
In file included from /home/ajroot/projects/srdatalog-python/build/jit/TcPlan_DB_2061/main.cpp:1:
In file included from /home/ajroot/projects/srdatalog-python/src/srdatalog/runtime/generalized_datalog/srdatalog.h:43:
In file included from /home/ajroot/projects/srdatalog-python/src/srdatalog/runtime/generalized_datalog/relation_col.h:33:
In file included from /home/ajroot/projects/srdatalog-python/src/srdatalog/runtime/generalized_datalog/gpu/aos_device_array.h:8:
In file included from /home/ajroot/projects/srdatalog-python/src/srdatalog/runtime/generalized_datalog/gpu/./device_array.h:26:
/home/ajroot/projects/srdatalog-python/src/srdatalog/runtime/vendor/rmm/include/rmm/mr/device/per_device_resource.hpp:203:8: error: unknown type name 'device_async_resource_ref'; did you mean 'device_resource_ref'?
  203 | inline device_async_resource_ref set_per_device_resource_ref_unsafe(
      |        ^~~~~~~~~~~~~~~~~~~~~~~~~
      |        device_resource_ref
/home/ajroot/projects/srdatalog-python/src/srdatalog/runtime/vendor/rmm/include/rmm/resource_ref.hpp:35:7: note: 'device_resource_ref' declared here
   35 | using device_resource_ref = cuda::mr::resource_ref<cuda::mr::device_accessible>;
      |       ^
fatal error: too many errors emitted, stopping now [-ferror-limit=]
20 errors generated when compiling for sm_89.
ninja: build stopped: subcommand failed.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions