fix: Make TLS initialization thread-safe and idempotent using pthread_once #11090
+44
−4
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
ThomasBrady
requested review from
edsiper,
fujimotos and
koleini
as code owners
WalkthroughIntroduces idempotent pthread_once-based TLS initialization macros and conditional TLS declarations; updates several public TLS-backed symbols to use Changes
Sequence Diagram(s)sequenceDiagram
participant Caller as TLS user (any CU)
participant FLB_API as FLB_TLS_INIT / macro
participant Once as pthread_once
participant InitFn as per-key init function
participant Key as pthread_key_t
rect rgb(240, 248, 255)
Note over Caller,FLB_API: New flow — idempotent per-key init
Caller->>FLB_API: Access TLS (macro ensures init)
FLB_API->>Once: pthread_once(&once_ctl, InitFn)
Once->>InitFn: call (first time only)
InitFn->>Key: pthread_key_create(&key, destructor)
InitFn-->>Once: return
Once-->>FLB_API: return (subsequent no-op)
FLB_API-->>Caller: provide thread-local storage
end
Estimated code review effort🎯 4 (Complex) | ⏱️ ~45 minutes
Suggested labels
Suggested reviewers
Poem
Pre-merge checks and finishing touches✅ Passed checks (2 passed)
✨ Finishing touches
🧪 Generate unit tests (beta)
📜 Recent review detailsConfiguration used: CodeRabbit UI Review profile: CHILL Plan: Pro 📒 Files selected for processing (4)
🚧 Files skipped from review as they are similar to previous changes (2)
🧰 Additional context used🧬 Code graph analysis (2)include/fluent-bit/flb_log.h (1)
include/fluent-bit/flb_thread_storage.h (1)
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (30)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
ThomasBrady
force-pushed
the
thomasbrady/idempotent-tls-init-key
branch
from
df31f5d to
289a534
Compare
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 1
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (4)
include/fluent-bit/flb_log.h(1 hunks)include/fluent-bit/flb_output.h(1 hunks)include/fluent-bit/flb_scheduler.h(1 hunks)include/fluent-bit/flb_thread_storage.h(1 hunks)
🧰 Additional context used
🧬 Code graph analysis (4)
include/fluent-bit/flb_output.h (1)
src/flb_log.c (1)
FLB_TLS_DEFINE(50-56)
include/fluent-bit/flb_thread_storage.h (1)
src/flb_log.c (1)
FLB_TLS_DEFINE(50-56)
include/fluent-bit/flb_log.h (1)
src/flb_log.c (1)
FLB_TLS_DEFINE(50-56)
include/fluent-bit/flb_scheduler.h (1)
src/flb_log.c (1)
FLB_TLS_DEFINE(50-56)
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (29)
- GitHub Check: pr-windows-build / call-build-windows-package (Windows 64bit (Arm64), amd64_arm64, -DCMAKE_SYSTEM_NAME=Windows -DCMA...
- GitHub Check: pr-windows-build / call-build-windows-package (Windows 32bit, x86, x86-windows-static, 3.31.6)
- GitHub Check: pr-windows-build / call-build-windows-package (Windows 64bit, x64, x64-windows-static, 3.31.6)
- GitHub Check: run-ubuntu-unit-tests (-DSANITIZE_UNDEFINED=On, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SIMD=Off, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SIMD=On, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_ARROW=On, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SANITIZE_THREAD=On, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SANITIZE_THREAD=On, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DSANITIZE_ADDRESS=On, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SIMD=On, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SIMD=Off, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SANITIZE_MEMORY=On, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SANITIZE_MEMORY=On, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_COVERAGE=On, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DSANITIZE_ADDRESS=On, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DSANITIZE_UNDEFINED=On, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SMALL=On, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SMALL=On, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_JEMALLOC=Off, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_JEMALLOC=On, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_JEMALLOC=Off, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_JEMALLOC=On, 3.31.6, gcc, g++)
- GitHub Check: pr-compile-without-cxx (3.31.6)
- GitHub Check: pr-compile-system-libs (-DFLB_PREFER_SYSTEM_LIBS=On, 3.31.6, clang, clang++, ubuntu-24.04, clang-14)
- GitHub Check: pr-compile-system-libs (-DFLB_PREFER_SYSTEM_LIBS=On, 3.31.6, gcc, g++, ubuntu-24.04, clang-14)
- GitHub Check: pr-compile-system-libs (-DFLB_PREFER_SYSTEM_LIBS=On, 3.31.6, clang, clang++, ubuntu-22.04, clang-12)
- GitHub Check: pr-compile-centos-7
- GitHub Check: PR - fuzzing test
Comment on lines
+57
to
+62
| #define FLB_TLS_DEFINE(type, name) \ | ||
| pthread_key_t name; \ | ||
| pthread_once_t name##_once = PTHREAD_ONCE_INIT; \ | ||
| void name##_init_func(void) { \ | ||
| pthread_key_create(&name, NULL); \ | ||
| } |
There was a problem hiding this comment.
Handle pthread_key_create failure to avoid undefined TLS state.
With the new pthread_once wrapper we only invoke pthread_key_create() once. If that call returns an error (e.g. EAGAIN, ENOMEM), the key stays uninitialized but the pthread_once guard prevents any retry, so every later FLB_TLS_SET/GET will operate on an invalid key and the failure is silent. Before this change we at least retried on the next FLB_TLS_INIT, so this is a regression in resiliency. Please capture the return code, surface/log it, and keep the caller from proceeding (e.g. by storing the status in a companion variable that FLB_TLS_INIT checks and aborts or reports). That way we fail fast instead of running with a broken TLS slot.
🤖 Prompt for AI Agents
In include/fluent-bit/flb_thread_storage.h around lines 57 to 62, the
pthread_key_create() result is ignored so if it fails the pthread_once guard
prevents retries and TLS key remains uninitialized; modify the macro to capture
the return code from pthread_key_create into a companion int status variable
(e.g. name##_init_rc), set it atomically, and have
FLB_TLS_INIT/FLB_TLS_GET/FLB_TLS_SET check that status and abort/log/return an
error when non-zero; also log the errno/rc when pthread_key_create fails to
surface the root cause and avoid proceeding with an invalid key.
|
@codex review |
|
Codex Review: Didn't find any major issues. Hooray! ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
If Codex has suggestions, it will comment; otherwise it will react with 👍. Codex can also answer questions or update the PR. Try commenting "@codex address that feedback". |
|
@ThomasBrady pls adjust the commit message , it must be prefixed with the file interface name (https://github.com/fluent/fluent-bit/blob/master/CONTRIBUTING.md#commit-changes) 
|
…hread_key_create once for each key This patch ensures pthread_key_create is only called once per TLS key by using pthread_once for thread-safe idempotent initialization. This prevents race conditions and errors when FLB_TLS_INIT is called from multiple locations (different compilation units, hot reloads, etc). Changes: - Modified FLB_TLS_INIT macro to use pthread_once control structure - Updated FLB_TLS_DEFINE to include pthread_once_t and init function - Added FLB_TLS_DECLARE macro for external TLS key declarations - Updated flb_log.h, flb_output.h, and flb_scheduler.h to use new FLB_TLS_DECLARE for proper extern declarations The implementation wraps pthread_key_create in a function called via pthread_once, guaranteeing single initialization regardless of how many times FLB_TLS_INIT is invoked. Signed-off-by: Thomas Brady <thomas.brady@chronosphere.io>
ThomasBrady
force-pushed
the
thomasbrady/idempotent-tls-init-key
branch
from
289a534 to
8f0f528
Compare
Done 👍 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Upstreaming as requested by @cosmo0920.
Background:
flb_reload()callsflb_create()to create a new context for validation.flb_create()callsflb_log_create()which callsFLB_TLS_INIT(flb_worker_ctx). On platforms withoutFLB_HAVE_C_TLSdefined (e.g. Windows),FLB_TLS_INITdirectly callspthread_key_create.FLB_TLS_INITis also called from multiple other locations (flb_lib.c,flb_worker.c, etc.). Each call topthread_key_createwith the same key is undefined behavior per POSIX, causing issues like fluentbit logs not being output after a halted reload.Summary of changes
This patch makes FLB_TLS_INIT(key) idempotent using pthread_once:
Before: Each FLB_TLS_INIT call directly executed pthread_key_create, causing multiple key creations
After: Each FLB_TLS_INIT uses pthread_once to ensure the key is created exactly once, regardless of how many times or from where it's called
flb_log.h,flb_output.h,flb_scheduler.h) to use correct declaration patternThis ensures thread-safe, idempotent TLS initialization across all compilation units and during hot reload, eliminating the undefined behavior.
Enter
[N/A]in the box, if an item is not applicable to your change.Testing
Before we can approve your change; please submit the following in a comment:
If this is a change to packaging of containers or native binaries then please confirm it works for all targets.
ok-package-testlabel to test for all targets (requires maintainer to do).Documentation
Backporting
Fluent Bit is licensed under Apache 2.0, by submitting this pull request I understand that this code will be released under the terms of that license.
Summary by CodeRabbit