mirror of
https://github.com/open-goal/jak-project
synced 2026-08-20 14:24:45 -04:00
Alexander J. Semenuk
5d5e35fb9b
## 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)
73 lines
2.6 KiB
XML
Vendored
Generated
73 lines
2.6 KiB
XML
Vendored
Generated
<?xml version="1.0" encoding="US-ASCII"?>
|
|
<testcase>
|
|
<info>
|
|
<keywords>
|
|
HTTP
|
|
HTTP GET
|
|
compressed
|
|
</keywords>
|
|
</info>
|
|
# Server-side
|
|
<reply>
|
|
# this deflate chunk has three bytes removed from the beginning and is cut
|
|
# short
|
|
<data crlf="headers">
|
|
HTTP/1.1 200 OK
|
|
Date: Mon, 29 Nov 2004 21:56:53 GMT
|
|
Server: Apache/1.3.31 (Debian GNU/Linux) mod_gzip/1.3.26.1a PHP/4.3.9-1 mod_ssl/2.8.20 OpenSSL/0.9.7d mod_perl/1.29
|
|
Vary: Accept-Encoding
|
|
Content-Type: text/html; charset=ISO-8859-1
|
|
Content-Encoding: deflate
|
|
Content-Length: 1305
|
|
|
|
%hex[%58%db%6e%e3%36%10%7d%37%90%7f%60%fd%d4%02%b6%6e%b6%13%39%70%b4%28%72%d9%04%cd%36%c1%da%05%ba%4f%06%2d%d1%36%1b%49%14%48%ca%b9%3c%f4%db%3b%94%28%89%b1%1c%af%77%83%be%04%48%62%72%e6%9c%c3%e1%0c%49%93%99%7c%7a%4a%62%b4%21%5c%50%96%9e%75%5d%cb%e9%22%92%86%2c%a2%e9%ea%ac%7b%33%bd%eb%fb%fe%68%dc%77%bb%9f%82%ce%e4%97%8b%bb%f3%d9%b7%fb%4b%94%71%f6%0f%09%65%3f%a6%42%02%10%4d%bf%4d%67%97%5f%50%77%2d%65%76%6a%db%4b%4e%c4%3a%21%58%5a%29%91%f6%02%87%0f%24%8d%ec%65%d2%d7%3c%d1%77%ac%a1%15%c9%a8%0b%a2%5b%5a%41%07%a1%ca%a6%da%4d%6f%4e%a3%c0%3d%76%bd%89%6d%18%4a%44%84%25%99%e3%28%22%80%18%8f%fd%be%e3%f7%3d%17%39%c3%53%c7%3d%f5%c6%13%db%f0%1b%84%3c%53%1f%51%e0%39%ce%b0%ef%3a%7d%d7%47%8e%77%ea%c1%cf%40%53%2a%c4%ab%38%52%9c%90%b9%58%33%2e%83%30%e7%71%1d%8e%61%6f%e3%97%79%1c%17%70%84%d3%08%c5%74%d1%a6%16%10%1d%1e%11%a1%96%3a%67%49%52%52%52%82%24%63%b5%00%c7%fc%19%2d%19%47%61%4c%49%2a%fb%82%46%04%fd%f5%f5%16%49%8e%53%b1%84%8a%5a%30%8b%46%c8%50%de%19%0c%a2%02%e1%72%04%a5%5a%a9%70%55%df%25%8d%89%38%ea%e4%42%75%d4%18%e2%39%95%f8%c9%42%37%12%89%3c%cb%40%5f%a0%eb%d9%ec%be%57%fc%9d%f6%d0%15%b4%8f%3a%57%45%fb%e2%e6%7c%d6%43%b3%cb%db%3f%2f%e1%f3%f6%e2%77%80%5d%dd%dc%5e%f6%8a%e1%3f%df%dd%5f%5f%7e%85%36%0c%f0%48%62%88%a9%94%ea%67%4c%c8%9e%6e%e6%d0]hex%
|
|
</data>
|
|
|
|
<datacheck crlf="headers">
|
|
HTTP/1.1 200 OK
|
|
Date: Mon, 29 Nov 2004 21:56:53 GMT
|
|
Server: Apache/1.3.31 (Debian GNU/Linux) mod_gzip/1.3.26.1a PHP/4.3.9-1 mod_ssl/2.8.20 OpenSSL/0.9.7d mod_perl/1.29
|
|
Vary: Accept-Encoding
|
|
Content-Type: text/html; charset=ISO-8859-1
|
|
Content-Encoding: deflate
|
|
Content-Length: 1305
|
|
|
|
</datacheck>
|
|
|
|
</reply>
|
|
|
|
# Client-side
|
|
<client>
|
|
<features>
|
|
libz
|
|
</features>
|
|
<server>
|
|
http
|
|
</server>
|
|
<name>
|
|
HTTP GET deflate compressed content with broken deflate header
|
|
</name>
|
|
<command>
|
|
http://%HOSTIP:%HTTPPORT/%TESTNUMBER --compressed
|
|
</command>
|
|
</client>
|
|
|
|
# Verify data after the test has been "shot"
|
|
<verify>
|
|
<strippart>
|
|
s/^Accept-Encoding: [a-zA-Z, ]*/Accept-Encoding: xxx/
|
|
</strippart>
|
|
<protocol crlf="headers">
|
|
GET /%TESTNUMBER HTTP/1.1
|
|
Host: %HOSTIP:%HTTPPORT
|
|
User-Agent: curl/%VERSION
|
|
Accept: */*
|
|
Accept-Encoding: xxx
|
|
|
|
</protocol>
|
|
<errorcode>
|
|
61
|
|
</errorcode>
|
|
</verify>
|
|
</testcase>
|