-
Notifications
You must be signed in to change notification settings - Fork 29.8k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
src: rewrite task runner in c++ #52609
src: rewrite task runner in c++ #52609
Conversation
Review requested:
|
0b92919
to
558c595
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
lgtm
src/node_task_runner.cc
Outdated
|
||
// Check if input contains any forbidden characters | ||
// If it doesn't, return the input as is. | ||
if (!std::regex_search(input, forbidden_characters)) { |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
std::regex has a bit of problem in terms of platform support and performance when I ported js2c to C++. I'd suggest just write a wrapper operating on the bytes as you find the characters, or you can use std::string::includes
and std::string::search
if the performance of this doesn't matter.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I can replace this for sure but the rest of it is going to be a problem
558c595
to
06f7a29
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Mostly LGTM, two issues (RE and mem leak pending)
06f7a29
to
8ead08d
Compare
8ead08d
to
720350a
Compare
I will review 'soon'. Please be patient. |
This should’ve had the @nodejs/test_runner team tagged. I assume this change is fine, and desirable, but I’d like to hear from them what it might mean for future development of the test runner. Would moving the core (or more than the core) of the test runner into C++ mean that developing new test runner features would become difficult or impossible for the current test runner contributors? Or even if it would, maybe that’s okay since the test runner is close to feature complete by this point? |
This comment was marked as off-topic.
This comment was marked as off-topic.
Sorry, my mistake! |
For |
…perator[] instead of find_field.
It would have been good because then you could have pointed me to such a goal, or I could have found it myself, and there would be no need for a discussion 🙂 For future reference: Lots of discussions on this topic is happening on Twitter in the threads that followed https://x.com/yagiznizipli/status/1785999524143464681?s=46&t=1mQKe1AKaQ-2YwRjxDfrOg |
@lemire Isn't that what the implementation in lib/internal/main/run.js essentially already provides? And npm etc of course already provides it. Question is what is best for the node.js project. Essentially: Taking more of an undici approach or this more tightly integrated approach. Speaking as a co-maintainer of |
I think you may not have examined @anonrig's work fully. It was always motivated by performance and contains significant C++ code (prior to this PR). You are just pointing a part of the implementation that was in JavaScript.
Here is how the functionality has been submitted... The purpose of this pull-request to offer a fast alternative to npm run xxx. With this pull-request, node supports node run test which executes test command inside package.json scripts. Just so we are clear, are we in agreement that |
@anonrig Let me try something. |
@lemire You're making this into something about my personal desires when it's not. I wrote:
I'm confident that @anonrig has understood my point though so any further discussion on this topic from me will happen outside of this issue when/if/where there's a broader discussion on this topic and how it aligns with current or future goals and needs for Node.js as a project. I do hope such a discussion happen and I would gladly be invited to or involved in it. |
I understand both sentiments (contribution should be easy, performance should be good). In this particular feature I think that since this dramatically effects the startup of every process lunched through the task runner performance triumphs. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I expect to have a PR on your PR in a few minutes. But this can be merged as-is in my opinion.
Landed in c5cfdd4 |
PR-URL: nodejs#52609 Reviewed-By: Matteo Collina <[email protected]> Reviewed-By: Daniel Lemire <[email protected]> Reviewed-By: James M Snell <[email protected]> Reviewed-By: Stephen Belanger <[email protected]>
PR-URL: nodejs#52609 Reviewed-By: Matteo Collina <[email protected]> Reviewed-By: Daniel Lemire <[email protected]> Reviewed-By: James M Snell <[email protected]> Reviewed-By: Stephen Belanger <[email protected]>
PR-URL: #52609 Reviewed-By: Matteo Collina <[email protected]> Reviewed-By: Daniel Lemire <[email protected]> Reviewed-By: James M Snell <[email protected]> Reviewed-By: Stephen Belanger <[email protected]>
PR-URL: nodejs#52609 Reviewed-By: Matteo Collina <[email protected]> Reviewed-By: Daniel Lemire <[email protected]> Reviewed-By: James M Snell <[email protected]> Reviewed-By: Stephen Belanger <[email protected]>
PR-URL: nodejs#52609 Reviewed-By: Matteo Collina <[email protected]> Reviewed-By: Daniel Lemire <[email protected]> Reviewed-By: James M Snell <[email protected]> Reviewed-By: Stephen Belanger <[email protected]>
* chore: bump Node.js to v22.9.0 * build: drop base64 dep in GN build nodejs/node#52856 * build,tools: make addons tests work with GN nodejs/node#50737 * fs: add fast api for InternalModuleStat nodejs/node#51344 * src: move package_json_reader cache to c++ nodejs/node#50322 * crypto: disable PKCS#1 padding for privateDecrypt nodejs-private/node-private#525 * src: move more crypto code to ncrypto nodejs/node#54320 * crypto: ensure valid point on elliptic curve in SubtleCrypto.importKey nodejs/node#50234 * src: shift more crypto impl details to ncrypto nodejs/node#54028 * src: switch crypto APIs to use Maybe<void> nodejs/node#54775 * crypto: remove DEFAULT_ENCODING nodejs/node#47182 * deps: update libuv to 1.47.0 nodejs/node#50650 * build: fix conflict gyp configs nodejs/node#53605 * lib,src: drop --experimental-network-imports nodejs/node#53822 * esm: align sync and async load implementations nodejs/node#49152 * esm: remove unnecessary toNamespacedPath calls nodejs/node#53656 * module: detect ESM syntax by trying to recompile as SourceTextModule nodejs/node#52413 * test: adapt debugger tests to V8 11.4 nodejs/node#49639 * lib: update usage of always on Atomics API nodejs/node#49639 * test: adapt test-fs-write to V8 internal changes nodejs/node#49639 * test: adapt to new V8 trusted memory spaces nodejs/node#50115 * deps: update libuv to 1.47.0 nodejs/node#50650 * src: use non-deprecated v8::Uint8Array::kMaxLength nodejs/node#50115 * src: update default V8 platform to override functions with location nodejs/node#51362 * src: add missing TryCatch nodejs/node#51362 * lib,test: handle new Iterator global nodejs/node#51362 * src: use non-deprecated version of CreateSyntheticModule nodejs/node#50115 * src: remove calls to recently deprecated V8 APIs nodejs/node#52996 * src: use new V8 API to define stream accessor nodejs/node#53084 * src: do not use deprecated V8 API nodejs/node#53084 * src: do not use soon-to-be-deprecated V8 API nodejs/node#53174 * src: migrate to new V8 interceptors API nodejs/node#52745 * src: use supported API to get stalled TLA messages nodejs/node#51362 * module: print location of unsettled top-level await in entry points nodejs/node#51999 * test: make snapshot comparison more flexible nodejs/node#54375 * test: do not set concurrency on parallelized runs nodejs/node#52177 * src: move FromNamespacedPath to path.cc nodejs/node#53540 * test: adapt to new V8 trusted memory spaces nodejs/node#50115 * build: add option to enable clang-cl on Windows nodejs/node#52870 * chore: fixup patch indices * chore: add/remove changed files * esm: drop support for import assertions nodejs/node#54890 * build: compile with C++20 support nodejs/node#52838 * deps: update nghttp2 to 1.62.1 nodejs/node#52966 * src: parse inspector profiles with simdjson nodejs/node#51783 * build: add GN build files nodejs/node#47637 * deps,lib,src: add experimental web storage nodejs/node#52435 * build: add missing BoringSSL dep * src: rewrite task runner in c++ nodejs/node#52609 * fixup! build: add GN build files * src: stop using deprecated fields of v8::FastApiCallbackOptions nodejs/node#54077 * fix: shadow variable * build: add back incorrectly removed SetAccessor patch * fixup! fixup! build: add GN build files * crypto: fix integer comparison in crypto for BoringSSL * src,lib: reducing C++ calls of esm legacy main resolve nodejs/node#48325 * src: move more crypto_dh.cc code to ncrypto nodejs/node#54459 * chore: fixup GN files for previous commit * src: move more crypto code to ncrypto nodejs/node#54320 * Fixup Perfetto ifdef guards * fix: missing electron_natives dep * fix: node_use_node_platform = false * fix: include src/node_snapshot_stub.cc in libnode * 5507047: [import-attributes] Remove support for import assertions https://chromium-review.googlesource.com/c/v8/v8/+/5507047 * fix: restore v8-sandbox.h in filenames.json * fix: re-add original-fs generation logic * fix: ngtcp2 openssl dep * test: try removing NAPI_VERSION undef * chore(deps): bump @types/node * src: move more crypto_dh.cc code to ncrypto nodejs/node#54459 * esm: remove unnecessary toNamespacedPath calls nodejs/node#53656 * buffer: fix out of range for toString nodejs/node#54553 * lib: rewrite AsyncLocalStorage without async_hooks nodejs/node#48528 * module: print amount of load time of a cjs module nodejs/node#52213 * test: skip reproducible snapshot test on 32-bit nodejs/node#53592 * fixup! src: move more crypto_dh.cc code to ncrypto * test: adjust emittedUntil return type * chore: remove redundant wpt streams patch * fixup! chore(deps): bump @types/node * fix: gn executable name on Windows * fix: build on Windows * fix: rename conflicting win32 symbols in //third_party/sqlite On Windows otherwise we get: lld-link: error: duplicate symbol: sqlite3_win32_write_debug >>> defined at .\..\..\third_party\electron_node\deps\sqlite\sqlite3.c:47987 >>> obj/third_party/electron_node/deps/sqlite/sqlite/sqlite3.obj >>> defined at obj/third_party/sqlite\chromium_sqlite3/sqlite3_shim.obj lld-link: error: duplicate symbol: sqlite3_win32_sleep >>> defined at .\..\..\third_party\electron_node\deps\sqlite\sqlite3.c:48042 >>> obj/third_party/electron_node/deps/sqlite/sqlite/sqlite3.obj >>> defined at obj/third_party/sqlite\chromium_sqlite3/sqlite3_shim.obj lld-link: error: duplicate symbol: sqlite3_win32_is_nt >>> defined at .\..\..\third_party\electron_node\deps\sqlite\sqlite3.c:48113 >>> obj/third_party/electron_node/deps/sqlite/sqlite/sqlite3.obj >>> defined at obj/third_party/sqlite\chromium_sqlite3/sqlite3_shim.obj lld-link: error: duplicate symbol: sqlite3_win32_utf8_to_unicode >>> defined at .\..\..\third_party\electron_node\deps\sqlite\sqlite3.c:48470 >>> obj/third_party/electron_node/deps/sqlite/sqlite/sqlite3.obj >>> defined at obj/third_party/sqlite\chromium_sqlite3/sqlite3_shim.obj lld-link: error: duplicate symbol: sqlite3_win32_unicode_to_utf8 >>> defined at .\..\..\third_party\electron_node\deps\sqlite\sqlite3.c:48486 >>> obj/third_party/electron_node/deps/sqlite/sqlite/sqlite3.obj >>> defined at obj/third_party/sqlite\chromium_sqlite3/sqlite3_shim.obj lld-link: error: duplicate symbol: sqlite3_win32_mbcs_to_utf8 >>> defined at .\..\..\third_party\electron_node\deps\sqlite\sqlite3.c:48502 >>> obj/third_party/electron_node/deps/sqlite/sqlite/sqlite3.obj >>> defined at obj/third_party/sqlite\chromium_sqlite3/sqlite3_shim.obj lld-link: error: duplicate symbol: sqlite3_win32_mbcs_to_utf8_v2 >>> defined at .\..\..\third_party\electron_node\deps\sqlite\sqlite3.c:48518 >>> obj/third_party/electron_node/deps/sqlite/sqlite/sqlite3.obj >>> defined at obj/third_party/sqlite\chromium_sqlite3/sqlite3_shim.obj lld-link: error: duplicate symbol: sqlite3_win32_utf8_to_mbcs >>> defined at .\..\..\third_party\electron_node\deps\sqlite\sqlite3.c:48534 >>> obj/third_party/electron_node/deps/sqlite/sqlite/sqlite3.obj >>> defined at obj/third_party/sqlite\chromium_sqlite3/sqlite3_shim.obj lld-link: error: duplicate symbol: sqlite3_win32_utf8_to_mbcs_v2 >>> defined at .\..\..\third_party\electron_node\deps\sqlite\sqlite3.c:48550 >>> obj/third_party/electron_node/deps/sqlite/sqlite/sqlite3.obj >>> defined at obj/third_party/sqlite\chromium_sqlite3/sqlite3_shim.obj * docs: remove unnecessary ts-expect-error after types bump * src: move package resolver to c++ nodejs/node#50322 * build: set ASAN detect_container_overflow=0 nodejs/node#55584 * chore: fixup rebase * test: disable failing ASAN test * win: almost fix race detecting ESRCH in uv_kill libuv/libuv#4341
This is a rewrite of the task runner in C++. The benchmark speaks for themselves with a caveat of removing support for
--env-file
and related CLI flags in the task runner. I think the performance can be improved further more since somehow we still interact with V8.While moving the implementation to C++, I've moved the tests for
escapeShell
to C++ as well, since we can't write a unit test running on JS side anymore. (Ref: please take a look atcctest/test_node_task_runner.cc
I'll investigate further after this pull-request to reduce the task runner to around ~5ms.
cc @nodejs/performance