Files
Alexander J. Semenuk 5d5e35fb9b fix: Windows toolchain compatibility (curl 8.21 re-vendor, endless reconfigure loop) (#4355)
## Problem

Windows builds break with a current local toolchain (Scoop LLVM 22.1.8,
CMake 4.4.0, VS 2026), in two independent ways:

1. The build stops at curl's deliberate guard: `#error "no non-blocking
method was found/used/set"` in `third-party/curl/lib/nonblock.c`.
2. From the second configure onward, `cmake --build` re-runs CMake in an
endless loop (observed 42 consecutive reconfigure cycles in a single
build). Likely the same mechanism behind the "endlessly building" VS
2026 note in `docs/setup/dev/vs.md`.

## Root cause

1. `third-party/curl/CMake/CurlTests.c` passes `int *` to
`ioctlsocket()`, whose third parameter is `u_long *`. Clang 22 promotes
`-Wincompatible-pointer-types` to a hard error in C, so the
`HAVE_IOCTLSOCKET_FIONBIO` try_compile silently fails and
`curl_config.h` never defines it. Upstream CI does not see this because
the windows-2022 runner image ships an older LLVM. GCC 14 promotes the
same warning to a hard error, which is very likely the `CurlTests.c.obj`
failure reported from MSYS2 in open-goal/jak-project#3551. Upstream curl
hit the identical problem with GCC 14 and fixed the probe in curl 8.8.0
(curl/curl#13578).
2. The root CMakeLists copies the build tree's `compile_commands.json`
into `<src>/build/` for clangd using `configure_file()`, which registers
its input as a configure dependency. CMake rewrites
`compile_commands.json` late in every generation, after
`CTestTestfile.cmake` and `cmake_install.cmake` (outputs of the same
Ninja regen rule), so once the dependency is registered the rule is
deterministically dirty and every `ninja` invocation re-runs CMake. A
pristine first configure is safe (the file does not exist yet, so the
`if(EXISTS ...)` guard skips the copy), which is why the loop looks
machine- or IDE-specific.

## Fix

1. Per review, re-vendor `third-party/curl` at the `curl-8_21_0` tag
(previously `curl-8_3_0`), which carries the upstream probe fix plus two
years of upstream development; `vendor.yaml` updated to match.
Adjustments the version jump forced:
- curl 8.15 removed the native macOS Secure Transport backend
(`CURL_USE_SECTRANSP`), so macOS now builds curl against OpenSSL like
Linux. The two macOS workflows install Homebrew `openssl@3` and export
`OPENSSL_ROOT_DIR` (keg-only), and the macOS setup docs gained the same
two lines.
- `CURL_BROTLI` / `CURL_ZSTD` switched to AUTO-detection in curl 8.10;
pinned OFF to keep the previous no-compression behavior and avoid
silently linking whatever the CI images happen to have.
- curl's new top-level `BUILD_EXAMPLES` cache option (default ON) leaked
into discord-rpc's identically named option and broke configure at a
nonexistent `examples/send-presence` directory; pinned OFF ahead of the
third-party subdirectories.

The diff is dominated by the mechanical tag-tree swap under
`third-party/curl` (linguist-vendored, collapsed in review). The
hand-written changes are `CMakeLists.txt`, the two macOS workflows,
`docs/setup/system/macos.md`, and `vendor.yaml`.
2. Swap `configure_file()` for `file(COPY ...)`: the same clangd copy
with no configure dependency registered. (`file(COPY_FILE ...
ONLY_IF_DIFFERENT)` would be cleaner still but requires CMake 3.21,
above the declared `cmake_minimum_required(VERSION 3.10)`.)

## Test plan

- [x] Fresh `cmake --preset Release-windows-clang` (LLVM 22, no cache
seeding) completes and logs `Enabled SSL backends: Schannel`; the
FIONBIO probe passes without the previous `#error`
- [x] Full Windows Release build from scratch in the branch worktree
(all 1422 targets)
- [x] goalc-test suite: 1509 passed, 0 failed
- [x] Second consecutive configure with `compile_commands.json` present:
the regen rule in `build.ninja` has no `compile_commands.json` input;
`<src>/build/compile_commands.json` is still refreshed for clangd
- [x] Repeated `ninja` invocations after a full build no longer re-run
CMake
- [x] macOS Intel and ARM CI green (first exercise of the OpenSSL
backend switch)

---

I work off a self-hosted forge, so this GitHub account is quiet; the
configure logs and ninja dirty-node traces from the investigation are
available if anyone wants the raw data.

(AI-assisted)
2026-07-27 19:19:18 -04:00

3.7 KiB
Vendored
Generated

c, SPDX-License-Identifier, Title, Section, Source, See-also, Protocol, Added-in
c SPDX-License-Identifier Title Section Source See-also Protocol Added-in
Copyright (C) Daniel Stenberg, <daniel@haxx.se>, et al. curl CURLOPT_READFUNCTION 3 libcurl
CURLOPT_POST (3)
CURLOPT_READDATA (3)
CURLOPT_SEEKFUNCTION (3)
CURLOPT_UPLOAD (3)
CURLOPT_UPLOAD_BUFFERSIZE (3)
CURLOPT_WRITEFUNCTION (3)
All
7.1

NAME

CURLOPT_READFUNCTION - read callback for data uploads

SYNOPSIS

#include <curl/curl.h>

size_t read_callback(char *buffer, size_t size, size_t nitems, void *userdata);

CURLcode curl_easy_setopt(CURL *handle, CURLOPT_READFUNCTION, read_callback);

DESCRIPTION

Pass a pointer to your callback function, as the prototype shows above.

This callback function gets called by libcurl as soon as it needs to read data in order to send it to the peer - like if you ask it to upload or post data to the server. The data area pointed at by the pointer buffer should be filled up with at most nitems number of bytes by your function. size is always 1.

Set the userdata argument with the CURLOPT_READDATA(3) option.

Your function must return the actual number of bytes that it stored in the data area pointed at by the pointer buffer. Returning 0 signals end-of-file to the library and causes it to stop the current transfer.

If you stop the current transfer by returning 0 "pre-maturely" (i.e before the server expected it, like when you have said you would upload N bytes and you upload less than N bytes), you may experience that the server "hangs" waiting for the rest of the data that is not sent.

The read callback may return CURL_READFUNC_ABORT to stop the current operation immediately, resulting in a CURLE_ABORTED_BY_CALLBACK error code from the transfer.

The callback can return CURL_READFUNC_PAUSE to cause reading from this connection to pause. See curl_easy_pause(3) for further details.

Bugs: when doing TFTP uploads, you must return the exact amount of data that the callback wants, or it is considered the final packet by the server end and the transfer ends there.

If you set this callback pointer to NULL, or do not set it at all, the default internal read function is used. It is doing an fread() on the FILE * userdata set with CURLOPT_READDATA(3).

You can set the total size of the data you are sending by using CURLOPT_INFILESIZE_LARGE(3) or CURLOPT_POSTFIELDSIZE_LARGE(3), depending on the type of transfer. For some transfer types it may be required and it allows for better error checking.

When this option is used in combination with telling libcurl to follow redirects with CURLOPT_FOLLOWLOCATION(3), the data might need to be rewound and sent again. The CURLOPT_SEEKFUNCTION(3) can then be invoked for that rewind operation.

DEFAULT

fread(3)

%PROTOCOLS%

EXAMPLE

size_t read_callback(char *ptr, size_t size, size_t nmemb, void *userdata)
{
  FILE *readhere = (FILE *)userdata;
  curl_off_t nread;

  /* copy as much data as possible into the 'ptr' buffer, but no more than
     'size' * 'nmemb' bytes. */
  size_t retcode = fread(ptr, size, nmemb, readhere);

  nread = (curl_off_t)retcode;

  fprintf(stderr, "*** We read %" CURL_FORMAT_CURL_OFF_T
          " bytes from file\n", nread);
  return retcode;
}

int main(int argc, char **argv)
{
  FILE *file = fopen(argv[1], "rb");
  CURLcode result;

  CURL *curl = curl_easy_init();
  if(curl) {
    /* set callback to use */
    curl_easy_setopt(curl, CURLOPT_READFUNCTION, read_callback);

    /* pass in suitable argument to callback */
    curl_easy_setopt(curl, CURLOPT_READDATA, (void *)file);

    result = curl_easy_perform(curl);
  }
}

HISTORY

CURL_READFUNC_PAUSE return code was added in 7.18.0 and CURL_READFUNC_ABORT was added in 7.12.1.

%AVAILABILITY%

RETURN VALUE

This returns CURLE_OK.