Fix spelling

README.md

@@ -22,7 +22,7 @@ register_detected_cuda_toolchains()
 ```
 
 **NOTE**: the use of `register_detected_cuda_toolchains` depends on the environment variable `CUDA_PATH`. You must also
-ensure the host compiler is available. On windows, this means that you will also need to set the environment variable
+ensure the host compiler is available. On Windows, this means that you will also need to set the environment variable
 `BAZEL_VC` properly.
 
 [`detect_cuda_toolkit`](https://github.com/bazel-contrib/rules_cuda/blob/5633f0c0f7/cuda/private/repositories.bzl#L28-L58)
@@ -33,8 +33,7 @@ determains how the toolchains are detected.
 
 - `cuda_library`: Can be used to compile and create static library for CUDA kernel code. The resulting targets can be
   consumed by [C/C++ Rules](https://bazel.build/reference/be/c-cpp#rules).
-- `cuda_objects`: If you don't understand what _device link_ means, you must never use it. This rule produce incomplete
-  object files that can only be consumed by `cuda_library`. It is created for relocatable device code and device link
+- `cuda_objects`: If you don't understand what _device link_ means, you must never use it. This rule produces incomplete object files that can only be consumed by `cuda_library`. It is created for relocatable device code and device link
   time optimization source files.
 
 ### Flags
@@ -45,7 +44,7 @@ Some flags are defined in [cuda/BUILD.bazel](cuda/BUILD.bazel). To use them, for
 bazel build --@rules_cuda//cuda:archs=compute_61:compute_61,sm_61
 ```
 
-In `.bazelrc` file, you can define shortcut alias for the flag, for example:
+In `.bazelrc` file, you can define a shortcut alias for the flag, for example:
 
 ```
 # Convenient flag shortcuts.

cuda/private/actions/dlink.bzl

@@ -109,7 +109,7 @@ def _wrapper_device_link(
     actions = ctx.actions
     pic_suffix = "_pic" if pic else ""
 
-    # Device-link to cubins for each gpu architecture. The stage1 compiled PTX is embeded in the object files.
+    # Device-link to cubins for each gpu architecture. The stage1 compiled PTX is embedded in the object files.
     # We don't need to do any thing about it, presumably.
     register_h = None
     cubins = []

cuda/private/providers.bzl

@@ -23,7 +23,7 @@ cuda_archs = [
 ]
 
 Stage2ArchInfo = provider(
-    """Provides the information of how the stage 2 complation is carried out.
+    """Provides the information of how the stage 2 compilation is carried out.
 
 One and only one of `virtual`, `gpu` and `lto` must be set to True. For example, if `arch` is set to `80` and `virtual` is `True`, then a
 ptx embedding process is carried out for `compute_80`. Multiple `Stage2ArchInfo` can be used for specifying how a stage 1 result is
@@ -66,9 +66,9 @@ if merged a single depset.
         "defines": "A depset of strings. It is used for the compilation during device linking.",
         # direct only:
         "objects": "A depset of objects. Direct artifacts of the rule.",  # but not rdc and pic
-        "pic_objects": "A depset of position indepentent code objects. Direct artifacts of the rule.",  # but not rdc
+        "pic_objects": "A depset of position independent code objects. Direct artifacts of the rule.",  # but not rdc
         "rdc_objects": "A depset of relocatable device code objects. Direct artifacts of the rule.",  # but not pic
-        "rdc_pic_objects": "A depset of relocatable device code and position indepentent code objects. Direct artifacts of the rule.",
+        "rdc_pic_objects": "A depset of relocatable device code and position independent code objects. Direct artifacts of the rule.",
         # transitive archive only (cuda_objects):
         "archive_objects": "A depset of rdc objects. cuda_objects only. Gathered from the transitive dependencies for archiving.",
         "archive_pic_objects": "A depset of rdc pic objects. cuda_objects only. Gathered from the transitive dependencies for archiving.",

cuda/private/repositories.bzl

@@ -48,7 +48,7 @@ def detect_cuda_toolkit(repository_ctx):
         if ptxas_path:
             # ${CUDA_PATH}/bin/ptxas
 
-            # Some distributions instead put CUDA binaries in a seperate path
+            # Some distributions instead put CUDA binaries in a separate path
             # Manually check and redirect there when necessary
             alternative = repository_ctx.path("/usr/lib/nvidia-cuda-toolkit/bin/nvcc")
             if str(ptxas_path) == "/usr/bin/ptxas" and alternative.exists:

cuda/private/rules/cuda_library.bzl

@@ -70,7 +70,7 @@ def _cuda_library_impl(ctx):
         rdc_objects = depset(transitive = [rdc_objects, rdc_dlink_output])
         rdc_pic_objects = depset(transitive = [rdc_pic_objects, rdc_pic_dlink_output])
 
-    # objects to archive: objects directly outputed by this rule and all objects transitively from deps,
+    # objects to archive: objects directly outputted by this rule and all objects transitively from deps,
     # take use_rdc=True and non-pic as an example:
     # rdc_objects: produce with this rule, thus it must be archived in the library produced by this rule
     # archive_rdc_objects: propagate from other `cuda_objects`, so this rule is in charge of archiving them
@@ -155,7 +155,7 @@ cuda_library = rule(
         "alwayslink": attr.bool(default = False),
         "rdc": attr.bool(
             default = False,
-            doc = ("Whether to perform device linking for relocateable device code. " +
+            doc = ("Whether to perform device linking for relocatable device code. " +
                    "Transitive deps that contain device code must all either be cuda_objects or cuda_library(rdc = True)."),
         ),
         "includes": attr.string_list(doc = "List of include dirs to be added to the compile line."),

cuda/private/rules/flags.bzl

@@ -9,7 +9,7 @@ def _cuda_archs_flag_impl(ctx):
 cuda_archs_flag = rule(
     doc = """A build setting for specifying cuda archs to compile for.
 
-To retain the flexiblity of NVCC, the [extended notation](https://docs.nvidia.com/cuda/cuda-compiler-driver-nvcc/index.html#extended-notation) is adopted.
+To retain the flexibility of NVCC, the [extended notation](https://docs.nvidia.com/cuda/cuda-compiler-driver-nvcc/index.html#extended-notation) is adopted.
 
 When passing cuda_archs from commandline, its spec grammar is as follows:
 

cuda/private/toolchain_config_lib.bzl

@@ -243,7 +243,7 @@ def _eval_flag_group_impl(stack, ret, fg, var, eval_iterations, _parse_flag_cach
                 ret[-2].extend(ret[-1])
                 ret.pop()  # The return space is deallocated.
             stack.pop()  # The stack frame is useless anymore,
-            continue  #### and there is no need to procees the current stack frame any further
+            continue  #### and there is no need to process the current stack frame any further
 
         stack[-1][-1] = True  # mark entered = True
 
@@ -275,7 +275,7 @@ def _eval_flag_group_impl(stack, ret, fg, var, eval_iterations, _parse_flag_cach
                 _eval_flags_or_flag_groups(stack, ret, fg, var, recursion_depth, _parse_flag_cache)
 
     if len(stack) != 0:
-        fail("flag_group evaluation imcomplete")
+        fail("flag_group evaluation incomplete")
     return ret
 
 def eval_flag_group(fg, value, max_eval_iterations = 65536, _parse_flag_cache = None):
@@ -402,7 +402,7 @@ def _enable_all_implied(info):
             to_enable.extend([new_name for new_name in reversed(info.selectables_info.implies[name])])
 
     if len(to_enable) != 0:
-        fail("_enable_all_implied imcomplete")
+        fail("_enable_all_implied incomplete")
 
 def _is_implied_by_enabled_activatable(info, name):
     for implied_by in info.selectables_info.implied_by.get(name, []):
@@ -476,7 +476,7 @@ def _check_activatable(info, to_check):
         to_check.extend(reversed(info.selectables_info.implied_by.get(name, [])))
 
     if len(to_check) != 0:
-        fail("_check_activatable imcomplete")
+        fail("_check_activatable incomplete")
 
 def _disable_unsupported_activatables(info):
     enabled = [k for k, v in reversed(info.enabled.items()) if v == True]

tests/toolchain_config_lib/toolchain_config_lib_test.bzl

@@ -83,8 +83,8 @@ def _parse_flag_cache_test_impl(ctx):
     asserts.equals(env, {"v": [0]}, f1.expandables)
     asserts.equals(env, ["v"], cache[flag_str].chunks)
     asserts.equals(env, {"v": [0]}, cache[flag_str].expandables)
-    f1.chunks.append("modifed")
-    f1.expandables["v"].append("modifed")
+    f1.chunks.append("modified")
+    f1.expandables["v"].append("modified")
     asserts.equals(env, ["v"], cache[flag_str].chunks)
     asserts.equals(env, {"v": [0]}, cache[flag_str].expandables)
     f2 = parse_flag(flag_str, cache = cache)