Merge branch 'nv-legate:branch-24.01' into _Enable_UCX_

.github/ISSUE_TEMPLATE/bug_report.yml

@@ -0,0 +1,97 @@
+name: Bug report
+description: Submit a bug report
+title: "[BUG] "
+labels: TRIAGE
+body:
+  - type: markdown
+    attributes:
+      value: "# Bug report"
+  - type: markdown
+    attributes:
+      value: Thank you for reporting a bug and helping us improve Legate!
+  - type: markdown
+    attributes:
+      value: >
+        Please fill out all of the required information.
+  - type: markdown
+    attributes:
+      value: |
+        ---
+        ## Environment information
+  - type: textarea
+    id: legate_issue
+    attributes:
+      label: Software versions
+      description: >-
+        Run `legate-issue` and paste the output here.
+      placeholder: |
+        Python      :  3.10.11 | packaged by conda-forge | (main, May 10 2023, 18:58:44) [GCC 11.3.0]
+        Platform    :  Linux-5.14.0-1042-oem-x86_64-with-glibc2.31
+        Legion      :  v23.11.00.dev-16-g2499f878
+        Legate      :  23.11.00.dev+17.gb7b50313
+        Cunumeric   :  (ImportError: cannot import name 'LogicalArray' from 'legate.core')
+        Numpy       :  1.24.4
+        Scipy       :  1.10.1
+        Numba       :  (not installed)
+        CTK package : cuda-version-11.8-h70ddcb2_2 (conda-forge)
+        GPU Driver  : 515.65.01
+        GPU Devices :
+          GPU 0: Quadro RTX 8000
+          GPU 1: Quadro RTX 8000
+    validations:
+      required: true
+  - type: input
+    id: jupyter
+    attributes:
+      label: Jupyter notebook / Jupyter Lab version
+      description: >-
+        Please supply if the issue you are reporting is related to Jupyter
+        notebook or Jupyter Lab.
+    validations:
+      required: false
+  - type: markdown
+    attributes:
+      value: |
+        ## Issue details
+  - type: textarea
+    id: expected-behavior
+    attributes:
+      label: Expected behavior
+      description: What did you expect to happen?
+    validations:
+      required: true
+  - type: textarea
+    id: observed-behavior
+    attributes:
+      label: Observed behavior
+      description: What did actually happen?
+    validations:
+      required: true
+  - type: markdown
+    attributes:
+      value: |
+        ## Directions to reproduce
+  - type: textarea
+    id: example
+    attributes:
+      label: Example code or instructions
+      description: >
+        Please provide detailed instructions to reproduce the issue. Ideally this includes a 
+        [Complete, minimal, self-contained example code](https://stackoverflow.com/help/minimal-reproducible-example)
+        given here or as a link to code in another repository. 
+      render: Python
+    validations:
+      required: true
+  - type: markdown
+    attributes:
+      value: |
+        ## Additional information
+  - type: textarea
+    id: traceback-console
+    attributes:
+      label: Stack traceback or browser console output
+      description: >
+        Add any error messages or logs that might be helpful in reproducing and
+        identifying the bug, for example a Python stack traceback.
+    validations:
+      required: false

BUILD.md

@@ -45,13 +45,13 @@ Once you have this environment file, you can install the required packages by
 creating a new conda environment:
 
 ```shell
-conda env create -n legate -f <env-file>.yaml
+mamba env create -n legate -f <env-file>.yaml
 ```
 
 or by updating an existing environment:
 
 ```shell
-conda env update -f <env-file>.yaml
+mamba env update -f <env-file>.yaml
 ```
 
 ## Building through install.py

README.md

@@ -218,15 +218,15 @@ as though they are running on a single processor.
 Legate Core is available [on conda](https://anaconda.org/legate/legate-core):
 
 ```
-conda install -c nvidia -c conda-forge -c legate legate-core
+mamba install -c nvidia -c conda-forge -c legate legate-core
 ```
 
 Only linux-64 packages are available at the moment.
 
 The default package contains GPU support, and is compatible with CUDA >= 12.0
 (CUDA driver version >= r520), and Volta or later GPU architectures. There are
-also CPU-only packages available, and will be automatically selected by `conda`
-when installing on a machine without GPUs.
+also CPU-only packages available, and will be automatically selected when
+installing on a machine without GPUs.
 
 See [BUILD.md](BUILD.md) for instructions on building Legate Core from source.
 

legate/core/store.py

@@ -1233,6 +1233,7 @@ def project(self, dim: int, index: int) -> Store:
             If ``dim`` is not a valid dimension name or ``index`` is
             out of bounds
         """
+        index = int(index)
         dim = dim + self.ndim if dim < 0 else dim
         if dim < 0 or dim >= self.ndim:
             raise ValueError(

legate/driver/launcher.py

@@ -15,7 +15,6 @@
 from __future__ import annotations
 
 import os
-from pathlib import Path
 from typing import TYPE_CHECKING
 
 from .. import install_info
@@ -174,9 +173,6 @@ def _compute_env(self) -> tuple[EnvDict, set[str]]:
                 str(system.legion_paths.legion_jupyter_module)
             )
 
-        # Make sure the base directory for this file is in the python path
-        extra_python_paths.append(str(Path(__file__).parents[1]))
-
         env["PYTHONPATH"] = os.pathsep.join(extra_python_paths)
 
         # If using NCCL prefer parallel launch mode over cooperative groups,

legate/tester/stages/__init__.py

@@ -19,7 +19,6 @@
 from __future__ import annotations
 
 import sys
-from typing import Dict, Type
 
 from .. import FeatureType
 from .test_stage import TestStage
@@ -33,7 +32,7 @@
     raise RuntimeError(f"unsupported platform: {sys.platform}")
 
 #: All the available test stages that can be selected
-STAGES: Dict[FeatureType, Type[TestStage]] = {
+STAGES: dict[FeatureType, type[TestStage]] = {
     "cpus": CPU,
     "cuda": GPU,
     "openmp": OMP,

legate/tester/stages/_osx/gpu.py

@@ -18,13 +18,14 @@
 from typing import TYPE_CHECKING
 
 from ..test_stage import TestStage
-from ..util import UNPIN_ENV, Shard
+from ..util import UNPIN_ENV
 
 if TYPE_CHECKING:
     from ....util.types import ArgList, EnvDict
     from ... import FeatureType
     from ...config import Config
     from ...test_system import TestSystem
+    from ..util import Shard, StageSpec
 
 
 class GPU(TestStage):
@@ -52,3 +53,9 @@ def env(self, config: Config, system: TestSystem) -> EnvDict:
 
     def delay(self, shard: Shard, config: Config, system: TestSystem) -> None:
         time.sleep(config.gpu_delay / 1000)
+
+    def shard_args(self, shard: Shard, config: Config) -> ArgList:
+        raise NotImplementedError()
+
+    def compute_spec(self, config: Config, system: TestSystem) -> StageSpec:
+        raise NotImplementedError()

scripts/util/build-caching.sh

@@ -7,9 +7,9 @@ if [[ -n "$(which sccache)" ]]; then
     CMAKE_CUDA_COMPILER_LAUNCHER="${CMAKE_CUDA_COMPILER_LAUNCHER:-$(which sccache)}";
 elif [[ -n "$(which ccache)" ]]; then
     # Use ccache if installed
-    CMAKE_C_COMPILER_LAUNCHER="${CMAKE_C_COMPILER_LAUNCHER:-$(which cache)}";
-    CMAKE_CXX_COMPILER_LAUNCHER="${CMAKE_CXX_COMPILER_LAUNCHER:-$(which cache)}";
-    CMAKE_CUDA_COMPILER_LAUNCHER="${CMAKE_CUDA_COMPILER_LAUNCHER:-$(which cache)}";
+    CMAKE_C_COMPILER_LAUNCHER="${CMAKE_C_COMPILER_LAUNCHER:-$(which ccache)}";
+    CMAKE_CXX_COMPILER_LAUNCHER="${CMAKE_CXX_COMPILER_LAUNCHER:-$(which ccache)}";
+    CMAKE_CUDA_COMPILER_LAUNCHER="${CMAKE_CUDA_COMPILER_LAUNCHER:-$(which ccache)}";
 fi
 
 export CMAKE_C_COMPILER_LAUNCHER="$CMAKE_C_COMPILER_LAUNCHER"

tests/unit/legate/driver/test_command.py

@@ -1046,10 +1046,13 @@ def test_default_single_rank(self, genobjs: GenObjs) -> None:
 
         assert result == ()
 
-    def test_utility_1_single_rank(self, genobjs: GenObjs) -> None:
+    def test_utility_1_single_rank_no_ucx(self, genobjs: GenObjs) -> None:
         config, system, launcher = genobjs(["--utility", "1"])
 
+        networks_orig = list(install_info.networks)
+        install_info.networks = [x for x in networks_orig if x != "ucx"]
         result = m.cmd_bgwork(config, system, launcher)
+        install_info.networks[:] = networks_orig[:]
 
         assert result == ()
 
@@ -1064,12 +1067,15 @@ def test_utility_1_single_rank_and_ucx(self, genobjs: GenObjs) -> None:
         assert result == ()
 
     @pytest.mark.parametrize("value", ("2", "3", "10"))
-    def test_utiltity_n_single_rank(
+    def test_utiltity_n_single_rank_no_ucx(
         self, genobjs: GenObjs, value: str
     ) -> None:
         config, system, launcher = genobjs(["--utility", value])
 
+        networks_orig = list(install_info.networks)
+        install_info.networks = [x for x in networks_orig if x != "ucx"]
         result = m.cmd_bgwork(config, system, launcher)
+        install_info.networks[:] = networks_orig[:]
 
         assert result == ()
 
@@ -1088,14 +1094,17 @@ def test_utiltity_n_single_rank_and_ucx(
 
     @pytest.mark.parametrize("rank_var", RANK_ENV_VARS)
     @pytest.mark.parametrize("rank", ("0", "1", "2"))
-    def test_default_multi_rank(
+    def test_default_multi_rank_no_ucx(
         self, genobjs: GenObjs, rank: str, rank_var: dict[str, str]
     ) -> None:
         config, system, launcher = genobjs(
             [], multi_rank=(2, 2), rank_env={rank_var: rank}
         )
 
+        networks_orig = list(install_info.networks)
+        install_info.networks = [x for x in networks_orig if x != "ucx"]
         result = m.cmd_bgwork(config, system, launcher)
+        install_info.networks[:] = networks_orig[:]
 
         if "ucx" in install_info.networks:
             assert result == ("-ll:bgwork", "2", "-ll:bgworkpin", "1")
@@ -1120,14 +1129,17 @@ def test_default_multi_rank_and_ucx(
 
     @pytest.mark.parametrize("rank_var", RANK_ENV_VARS)
     @pytest.mark.parametrize("rank", ("0", "1", "2"))
-    def test_utility_1_multi_rank_no_launcher(
+    def test_utility_1_multi_rank_no_launcher_no_ucx(
         self, genobjs: GenObjs, rank: str, rank_var: dict[str, str]
     ) -> None:
         config, system, launcher = genobjs(
             ["--utility", "1"], multi_rank=(2, 2), rank_env={rank_var: rank}
         )
 
+        networks_orig = list(install_info.networks)
+        install_info.networks = [x for x in networks_orig if x != "ucx"]
         result = m.cmd_bgwork(config, system, launcher)
+        install_info.networks[:] = networks_orig[:]
 
         if "ucx" in install_info.networks:
             assert result == ("-ll:bgwork", "2", "-ll:bgworkpin", "1")
@@ -1151,14 +1163,17 @@ def test_utility_1_multi_rank_no_launcher_and_ucx(
         assert result == ("-ll:bgwork", "2", "-ll:bgworkpin", "1")
 
     @pytest.mark.parametrize("launch", ("mpirun", "jsrun", "srun"))
-    def test_utility_1_multi_rank_with_launcher(
+    def test_utility_1_multi_rank_with_launcher_no_ucx(
         self, genobjs: GenObjs, launch: str
     ) -> None:
         config, system, launcher = genobjs(
             ["--utility", "1", "--launcher", launch], multi_rank=(2, 2)
         )
 
+        networks_orig = list(install_info.networks)
+        install_info.networks = [x for x in networks_orig if x != "ucx"]
         result = m.cmd_bgwork(config, system, launcher)
+        install_info.networks[:] = networks_orig[:]
 
         if "ucx" in install_info.networks:
             assert result == ("-ll:bgwork", "2", "-ll:bgworkpin", "1")
@@ -1183,14 +1198,17 @@ def test_utility_1_multi_rank_with_launcher_and_ucx(
     @pytest.mark.parametrize("rank_var", RANK_ENV_VARS)
     @pytest.mark.parametrize("rank", ("0", "1", "2"))
     @pytest.mark.parametrize("value", ("2", "3", "10"))
-    def test_utility_n_multi_rank_no_launcher(
+    def test_utility_n_multi_rank_no_launcher_no_ucx(
         self, genobjs: GenObjs, value: str, rank: str, rank_var: dict[str, str]
     ) -> None:
         config, system, launcher = genobjs(
             ["--utility", value], multi_rank=(2, 2), rank_env={rank_var: rank}
         )
 
+        networks_orig = list(install_info.networks)
+        install_info.networks = [x for x in networks_orig if x != "ucx"]
         result = m.cmd_bgwork(config, system, launcher)
+        install_info.networks[:] = networks_orig[:]
 
         if "ucx" in install_info.networks:
             assert result == ("-ll:bgwork", value, "-ll:bgworkpin", "1")
@@ -1216,14 +1234,17 @@ def test_utility_n_multi_rank_no_launcher_and_ucx(
 
     @pytest.mark.parametrize("launch", ("mpirun", "jsrun", "srun"))
     @pytest.mark.parametrize("value", ("2", "3", "10"))
-    def test_utility_n_multi_rank_with_launcher(
+    def test_utility_n_multi_rank_with_launcher_no_ucx(
         self, genobjs: GenObjs, value: str, launch: str
     ) -> None:
         config, system, launcher = genobjs(
             ["--utility", value, "--launcher", launch], multi_rank=(2, 2)
         )
 
+        networks_orig = list(install_info.networks)
+        install_info.networks = [x for x in networks_orig if x != "ucx"]
         result = m.cmd_bgwork(config, system, launcher)
+        install_info.networks[:] = networks_orig[:]
 
         if "ucx" in install_info.networks:
             assert result == ("-ll:bgwork", value, "-ll:bgworkpin", "1")