Upstream sync 2024 03 18

.buildkite/test-pipeline.yaml

@@ -13,7 +13,7 @@ steps:
 
 - label: Basic Correctness Test
   command: pytest -v -s --forked basic_correctness
-  
+
 - label: Core Test
   command: pytest -v -s core
 
@@ -28,14 +28,14 @@ steps:
   num_gpus: 2 # only support 1 or 2 for now.
 
 - label: Engine Test
-  command: pytest -v -s engine test_sequence.py
+  command: pytest -v -s engine tokenization test_sequence.py
 
 - label: Entrypoints Test
   command: pytest -v -s entrypoints
 
-- label: Kernels Test
-  command: pytest -v -s kernels
-  soft_fail: true
+- label: Kernels Test %N
+  command: pytest -v -s kernels --shard-id=$$BUILDKITE_PARALLEL_JOB --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT
+  parallelism: 4
 
 - label: Models Test
   commands:
@@ -55,8 +55,9 @@ steps:
 - label: Speculative decoding tests
   command: pytest -v -s spec_decode
 
-- label: LoRA Test
-  command: pytest -v -s lora --forked
+- label: LoRA Test %N
+  command: pytest -v -s lora --forked --shard-id=$$BUILDKITE_PARALLEL_JOB --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT
+  parallelism: 4
 
 - label: Metrics Test
   command: pytest -v -s metrics

.buildkite/test-template.j2

@@ -20,6 +20,9 @@ steps:
     agents:
       queue: kubernetes
     soft_fail: {{ step.soft_fail or false }}
+    {% if step.parallelism %}
+    parallelism: {{ step.parallelism }}
+    {% endif %}
     retry:
       automatic:
         - exit_status: -1  # Agent was lost

.github/ISSUE_TEMPLATE/100-documentation.yml

@@ -1,7 +1,7 @@
 name: 📚 Documentation
 description: Report an issue related to https://docs.vllm.ai/
 title: "[Doc]: "
-labels: ["doc"]
+labels: ["documentation"]
 
 body:
 - type: textarea

.github/ISSUE_TEMPLATE/500-feature request.yml

@@ -1,7 +1,7 @@
 name: 🚀 Feature request
 description: Submit a proposal/request for a new vllm feature
 title: "[Feature]: "
-labels: ["feature"]
+labels: ["feature request"]
 
 body:
 - type: markdown

.github/PULL_REQUEST_TEMPLATE.md

@@ -0,0 +1,60 @@
+<details>
+<!-- inside this <details> section, markdown rendering does not work, so we use raw html here. -->
+<summary><b> PR Checklist (Click to expand. Please read before submitting.) </b></summary>
+
+<p>Thank you for your contribution to vLLM! Before submitting the pull request, please ensure the PR meets the following criteria. This helps vLLM maintain the code quality and improve the efficiency of the review process.</p>
+
+<h3>PR Title and Classification</h3>
+<p>Only specific types of PRs will be reviewed. The PR title is prefixed appropriately to indicate the type of change. Please use one of the following:</p>
+<ul>
+    <li><code>[Bugfix]</code> for bug fixes.</li>
+    <li><code>[CI/Build]</code> for build or continuous integration improvements.</li>
+    <li><code>[Doc]</code> for documentation fixes and improvements.</li>
+    <li><code>[Model]</code> for adding a new model or improving an existing model. Model name should appear in the title.</li>
+    <li><code>[Frontend]</code> For changes on the vLLM frontend (e.g., OpenAI API server, <code>LLM</code> class, etc.) </li>
+    <li><code>[Kernel]</code> for changes affecting CUDA kernels or other compute kernels.</li>
+    <li><code>[Core]</code> for changes in the core vLLM logic (e.g., <code>LLMEngine</code>, <code>AsyncLLMEngine</code>, <code>Scheduler</code>, etc.)</li>
+    <li><code>[Hardware][Vendor]</code> for hardware-specific changes. Vendor name should appear in the prefix (e.g., <code>[Hardware][AMD]</code>).</li>
+    <li><code>[Misc]</code> for PRs that do not fit the above categories. Please use this sparingly.</li>
+</ul>
+<p><strong>Note:</strong> If the PR spans more than one category, please include all relevant prefixes.</p>
+
+<h3>Code Quality</h3>
+
+<p>The PR need to meet the following code quality standards:</p>
+
+<ul>
+    <li>We adhere to <a href="https://google.github.io/styleguide/pyguide.html">Google Python style guide</a> and <a href="https://google.github.io/styleguide/cppguide.html">Google C++ style guide</a>.</li>
+    <li>Pass all linter checks. Please use <a href="https://github.com/vllm-project/vllm/blob/main/format.sh"><code>format.sh</code></a> to format your code.</li>
+    <li>The code need to be well-documented to ensure future contributors can easily understand the code.</li>
+    <li>Include sufficient tests to ensure the project to stay correct and robust. This includes both unit tests and integration tests.</li>
+    <li>Please add documentation to <code>docs/source/</code> if the PR modifies the user-facing behaviors of vLLM. It helps vLLM user understand and utilize the new features or changes.</li>
+</ul>
+
+<h3>Notes for Large Changes</h3>
+<p>Please keep the changes as concise as possible. For major architectural changes (>500 LOC excluding kernel/data/config/test), we would expect a GitHub issue (RFC) discussing the technical design and justification. Otherwise, we will tag it with <code>rfc-required</code> and might not go through the PR.</p>
+
+<h3>What to Expect for the Reviews</h3>
+
+<p>The goal of the vLLM team is to be a <i>transparent reviewing machine</i>. We would like to make the review process transparent and efficient and make sure no contributor feel confused or frustrated. However, the vLLM team is small, so we need to prioritize some PRs over others. Here is what you can expect from the review process: </p>
+
+<ul>
+    <li> After the PR is submitted, the PR will be assigned to a reviewer. Every reviewer will pick up the PRs based on their expertise and availability.</li>
+    <li> After the PR is assigned, the reviewer will provide status update every 2-3 days. If the PR is not reviewed within 7 days, please feel free to ping the reviewer or the vLLM team.</li>
+    <li> After the review, the reviewer will put an <code> action-required</code> label on the PR if there are changes required. The contributor should address the comments and ping the reviewer to re-review the PR.</li>
+    <li> Please respond to all comments within a reasonable time frame. If a comment isn't clear or you disagree with a suggestion, feel free to ask for clarification or discuss the suggestion.
+ </li>
+</ul>
+
+<h3>Thank You</h3>
+
+<p> Finally, thank you for taking the time to read these guidelines and for your interest in contributing to vLLM. Your contributions make vLLM a great tool for everyone! </p>
+
+
+</details>
+
+---
+
+Please provide a brief explanation of the motivation behind the PR and the changes it introduces. This helps reviewers understand the context and rationale for the contribution. If possible, please link existing issues this PR will resolve.
+
+

.github/workflows/ruff.yml

@@ -28,7 +28,7 @@ jobs:
         pip install ruff==0.1.5 codespell==2.2.6 tomli==2.0.1
     - name: Analysing the code with ruff
       run: |
-        ruff vllm tests
+        ruff .
     - name: Spelling check with codespell
       run: |
-         codespell --toml pyproject.toml
+        codespell --toml pyproject.toml

CONTRIBUTING.md

@@ -45,31 +45,9 @@ pytest tests/
 If you encounter a bug or have a feature request, please check our issues page first to see if someone else has already reported it.
 If not, please file a new issue, providing as much relevant information as possible.
 
-### Coding Style Guide
+### Pull Requests & Code Reviews
 
-In general, we adhere to [Google Python style guide](https://google.github.io/styleguide/pyguide.html) and [Google C++ style guide](https://google.github.io/styleguide/cppguide.html).
-
-We include a formatting script [`format.sh`](./format.sh) to format the code.
-
-### Pull Requests
-
-When submitting a pull request:
-
-1. Make sure your code has been rebased on top of the latest commit on the main branch.
-2. Ensure code is properly formatted by running [`format.sh`](./format.sh).
-3. Include a detailed description of the changes in the pull request.
-Explain why you made the changes you did.
-If your pull request fixes an open issue, please include a reference to it in the description.
-
-### Code Reviews
-
-All submissions, including submissions by project members, require a code review.
-To make the review process as smooth as possible, please:
-
-1. Keep your changes as concise as possible.
-If your pull request involves multiple unrelated changes, consider splitting it into separate pull requests.
-2. Respond to all comments within a reasonable time frame.
-If a comment isn't clear or you disagree with a suggestion, feel free to ask for clarification or discuss the suggestion.
+Please check the PR checklist in the [PR template](.github/PULL_REQUEST_TEMPLATE.md) for detailed guide for contribution.
 
 ### Thank You
 

benchmarks/backend_request_func.py

@@ -68,7 +68,7 @@ async def async_request_tgi(
                             output.ttft = ttft
                     output.latency = time.perf_counter() - st
 
-                    body = data.decode("utf-8").lstrip("data:")  # noqa
+                    body = remove_prefix(data.decode("utf-8"), "data:")
                     output.generated_text = json.loads(body)["generated_text"]
                     output.success = True
                 else:
@@ -114,7 +114,7 @@ async def async_request_vllm(
                             output.ttft = ttft
                     output.latency = time.perf_counter() - st
 
-                    # When streaming, '\0' is appended to the end of the response.
+                    # When streaming, '\0' is appended to the end of response.
                     body = data.decode("utf-8").strip("\0")
                     output.generated_text = json.loads(
                         body)["text"][0][len(request_func_input.prompt):]
@@ -162,7 +162,7 @@ async def async_request_trt_llm(
                             output.ttft = ttft
                     output.latency = time.perf_counter() - st
 
-                    body = data.decode("utf-8").lstrip("data:")  # noqa
+                    body = remove_prefix(data.decode("utf-8"), "data:")
                     output.generated_text = json.loads(body)["text_output"]
                     output.success = True
 
@@ -196,7 +196,8 @@ async def async_request_deepspeed_mii(
         output = RequestFuncOutput()
         output.prompt_len = request_func_input.prompt_len
 
-        # DeepSpeed-MII doesn't support streaming as of Jan 28 2024, will use 0 as placeholder.
+        # DeepSpeed-MII doesn't support streaming as of Jan 28 2024,
+        # will use 0 as placeholder.
         # https://github.com/microsoft/DeepSpeed-MII/pull/311
         output.ttft = 0
 
@@ -259,7 +260,7 @@ async def async_request_openai_completions(
                         if not chunk:
                             continue
 
-                        chunk = chunk.decode("utf-8").lstrip("data: ")  # noqa
+                        chunk = remove_prefix(chunk.decode("utf-8"), "data: ")
                         if chunk == "[DONE]":
                             latency = time.perf_counter() - st
                         else:
@@ -326,7 +327,7 @@ async def async_request_openai_chat_completions(
                         if not chunk:
                             continue
 
-                        chunk = chunk.decode("utf-8").lstrip("data: ")
+                        chunk = remove_prefix(chunk.decode("utf-8"), "data: ")
                         if chunk == "[DONE]":
                             latency = time.perf_counter() - st
                         else:
@@ -348,6 +349,14 @@ async def async_request_openai_chat_completions(
     return output
 
 
+# Since vllm must support Python 3.8, we can't use str.removeprefix(prefix)
+# introduced in Python 3.9
+def remove_prefix(text: str, prefix: str) -> str:
+    if text.startswith(prefix):
+        return text[len(prefix):]
+    return text
+
+
 ASYNC_REQUEST_FUNCS = {
     "tgi": async_request_tgi,
     "vllm": async_request_vllm,

benchmarks/benchmark_serving.py

@@ -295,7 +295,9 @@ def main(args: argparse.Namespace):
 
         # Save to file
         base_model_id = model_id.split("/")[-1]
-        file_name = f"{backend}-{args.request_rate}qps-{base_model_id}-{current_dt}.json"
+        file_name = (
+            f"{backend}-{args.request_rate}qps-{base_model_id}-{current_dt}.json"
+        )
         with open(file_name, "w") as outfile:
             json.dump(result_json, outfile)
 
@@ -343,7 +345,7 @@ def main(args: argparse.Namespace):
         "--tokenizer",
         type=str,
         help=
-        "Name or path of the tokenizer, if not using the default model tokenizer.",
+        "Name or path of the tokenizer, if not using the default tokenizer.",
     )
     parser.add_argument(
         "--best-of",