https://github.com/oven-sh/bun/pull/25361 needs to be merged before this
PR
## Summary
- Move `last_write_failed` flag from loop-level to per-socket flag for
correctness
## Changes
- Move `last_write_failed` from `loop->data` to `socket->flags`
- More semantically correct since write status is per-socket, not
per-loop
🤖 Generated with [Claude Code](https://claude.com/claude-code)
---------
Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
## Summary
- Fix use-after-free vulnerability during socket adoption by properly
tracking reallocated sockets
- Add safety checks to prevent linking closed sockets to context lists
- Properly track socket state with new `is_closed`, `adopted`, and
`is_tls` flags
## What does this PR do?
This PR improves event loop stability by addressing potential
use-after-free issues that can occur when sockets are reallocated during
adoption (e.g., when upgrading a TCP socket to TLS).
### Key Changes
**Socket State Tracking
([internal.h](packages/bun-usockets/src/internal/internal.h))**
- Added `is_closed` flag to explicitly track when a socket has been
closed
- Added `adopted` flag to mark sockets that were reallocated during
context adoption
- Added `is_tls` flag to track TLS socket state for proper low-priority
queue handling
**Safe Socket Adoption
([context.c](packages/bun-usockets/src/context.c))**
- When `us_poll_resize()` returns a new pointer (reallocation occurred),
the old socket is now:
- Marked as closed (`is_closed = 1`)
- Added to the closed socket cleanup list
- Marked as adopted (`adopted = 1`)
- Has its `prev` pointer set to the new socket for event redirection
- Added guards to
`us_internal_socket_context_link_socket/listen_socket/connecting_socket`
to prevent linking already-closed sockets
**Event Loop Handling ([loop.c](packages/bun-usockets/src/loop.c))**
- After callbacks that can trigger socket adoption (`on_open`,
`on_writable`, `on_data`), the event loop now checks if the socket was
reallocated and redirects to the new socket
- Low-priority socket handling now properly checks `is_closed` state and
uses `is_tls` flag for correct SSL handling
**Poll Resize Safety
([epoll_kqueue.c](packages/bun-usockets/src/eventing/epoll_kqueue.c))**
- Changed `us_poll_resize()` to always allocate new memory with
`us_calloc()` instead of `us_realloc()` to ensure the old pointer
remains valid for cleanup
- Now takes `old_ext_size` parameter to correctly calculate memory sizes
- Re-enabled `us_internal_loop_update_pending_ready_polls()` call in
`us_poll_change()` to ensure pending events are properly redirected
### How did you verify your code works?
Run existing CI and existing socket upgrade tests under asan build
### What does this PR do?
Calls `uncork()` after flushing response headers to ensure data is sent
as soon as possible, improving responsiveness.
This behavior still works correctly even without the explicit `uncork()`
call, due to the deferred uncork logic implemented here:
6e3359dd16/packages/bun-uws/src/Loop.h (L57-L64)
A test already covers this scenario in
`test/js/node/test/parallel/test-http-flush-response-headers.js`.
### How did you verify your code works?
CI
---------
Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
* some fixes
* WIP
* wip wip
* more debug
* closer
* sending a really big payload at once is still broken
* now we need to avoid segfault happening inside onWritable after destroy
* opsie
* cleanup
* more cleanup
* more WIP, closer need to fix cork
* fix cork actually not writing non-optional data
* make onWritable return actually do something
* actually clean the on writable handler
* remove unreachable condition
* we are not looping anymore
* little revert
* fix possible fault
* inform backpressure on chunked encoding
* just queue when tryEnd
* remove unreachable code
* Enable more C++ warnings
* Only enable these ones in release builds
* Update MessagePortChannel.cpp
* Update BunProcess.cpp
---------
Co-authored-by: Jarred Sumner <709451+Jarred-Sumner@users.noreply.github.com>
* Make printing errors faster
* Make Next.js tests less flaky
* Update dev-server.test.ts
* Copy over runner from process pr
* bump next version
* Update javascript.zig
* Set port to 0
* p-queue
* no dont use docker buildx
* Prevent runner from hanging
* Update dev-server.test.ts
* Really fix hanging this time
* Fix bounds check for unix domain socket, support abstract namespace sockets
* Various fixes
* [autofix.ci] apply automated fixes
* Update runner.node.mjs
* Update runner.node.mjs
* windows
* Only care about stdout
* increase concureency + clean up test
* Tweak puppeteer
* Update runner.node.mjs
---------
Co-authored-by: Jarred Sumner <709451+Jarred-Sumner@users.noreply.github.com>
Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
* [server] requestIp and AnyRequestContext
Changed Request.uws_request to the new AnyRequestContext. This
allows grabbing the IP from a Request. Unfinished.
* [server] basic `requestIp` implementation
Currently using uws's requestIpAsText, which always returns a ipv6
string. We should return a `SocketAddress` object to the user instead,
which will contain the formatted address string and what type it is.
We'll have to use requestIpAsBinary and parse that ourselves.
* TypeScript docs, use `bun.String`, return `undefined` instead of `null`
if we can't get the ip.
* binary address formatting
* uws getRemoteAddress binding
* remove dead code
* working
* final touches✨
* I will abide by the results of this poll.
---------
Co-authored-by: Parzival-3141 <29632054+Parzival-3141@users.noreply.github.com>
* fixes SNI on tls sockets and also servername
* 💅
* 💅
* add support for https and wss
* fix bun types
* support Bun.file on ca, key and cert
* 💅
* add setTimeout (makes fastify run)
* fix httpVersion
* fix WebSocketServer and add listen event
* fix ws exports and http listening
* fix default import
* bump uws
* add nodebuffer compatibility
* fix drain and allow more passing tests to run
* fix enqueud messages
* default to arraybuffer
* fix constructor binaryType
* fmt
* fixup
* skip some tests
* skip more
* skip fault tests
* reuse encoder instance
* fix handshake WS Client
* temporary revert handshake fix
* fix handshake
* disable all socket.io test temp
* fixup
* add back socket.io tests
* use node_fs to read cert, ca and key on server.zig
* throw the error returned by NodeFS
* 💅
* added some improvements on server
* undo unintended change
* clean data handler before end calls
* refactor
* make ctx.resp nullable
---------
Co-authored-by: Jarred Sumner <jarred@jarredsumner.com>
