[Doc]: fixing multiple typos in diverse files (#33256)

Signed-off-by: Didier Durand <durand.didier@gmail.com>
Signed-off-by: Didier Durand <2927957+didier-durand@users.noreply.github.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
This commit is contained in:
Didier Durand
2026-01-29 16:52:03 +08:00
committed by GitHub
co-authored by gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
parent abb34ac43a
commit 31b25f6516
14 changed files with 20 additions and 20 deletions
+1 -1
View File
@@ -693,7 +693,7 @@ class OpenAIServingResponses(OpenAIServing):
except ValueError as e:
return self.create_error_response(e)
# NOTE: Implementation of stauts is still WIP, but for now
# NOTE: Implementation of status is still WIP, but for now
# we guarantee that if the status is not "completed", it is accurate.
# "completed" is implemented as the "catch-all" for now.
status: ResponseStatus = "completed"
@@ -30,7 +30,7 @@ class SharedFusedMoE(FusedMoE):
# Disable shared expert overlap if:
# - we are using eplb with non-default backend, because of correctness issues
# - we are using flashinfer with DP, since there nothint to gain
# - we are using flashinfer with DP, since there nothing to gain
# - we are using marlin kernels
backend = self.moe_parallel_config.all2all_backend
self.use_overlapped = (
@@ -22,7 +22,7 @@ class ScaledMMLinearLayerConfig:
@dataclass
class Int8ScaledMMLinearLayerConfig(ScaledMMLinearLayerConfig):
# TODO: Chnage to QuantKey like FP8ScaledMMLinearLayerConfig
# TODO: Change to QuantKey like FP8ScaledMMLinearLayerConfig
is_static_input_scheme: bool
is_channelwise: bool
input_symmetric: bool
@@ -119,7 +119,7 @@ def choose_scaled_mm_linear_kernel(
config (_KernelConfigT): Description of the linear layer
to be implemented.
possible_kernels (dict[PlatformEnum, list[_KernelT]]): A
dictionary of platforms and their list list of possible kernels.
dictionary of platforms and their list of possible kernels.
compute_capability (Optional[int], optional): The compute capability of
the target device, if None uses `current_platform` to get the
compute capability. Defaults to None.
+1 -1
View File
@@ -278,7 +278,7 @@ def find_loaded_library(lib_name: str) -> str | None:
According to according to https://man7.org/linux/man-pages/man5/proc_pid_maps.5.html,
the file `/proc/self/maps` contains the memory maps of the process, which includes the
shared libraries loaded by the process. We can use this file to find the path of the
a loaded library.
loaded library.
""" # noqa
found_line = None
with open("/proc/self/maps") as f:
+1 -1
View File
@@ -1159,7 +1159,7 @@ class Scheduler(SchedulerInterface):
break
# Calculate the number of embeddings to schedule in the current range
# of scheduled encoder placholder tokens.
# of scheduled encoder placeholder tokens.
start_idx_rel = max(0, num_computed_tokens - start_pos)
end_idx_rel = min(
num_encoder_tokens, num_computed_tokens + num_new_tokens - start_pos