mirror of
https://github.com/open-goal/jak-project
synced 2026-08-07 02:06:59 -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)
320 lines
15 KiB
Python
Vendored
Generated
320 lines
15 KiB
Python
Vendored
Generated
#!/usr/bin/env python3
|
|
# -*- coding: utf-8 -*-
|
|
#***************************************************************************
|
|
# _ _ ____ _
|
|
# Project ___| | | | _ \| |
|
|
# / __| | | | |_) | |
|
|
# | (__| |_| | _ <| |___
|
|
# \___|\___/|_| \_\_____|
|
|
#
|
|
# Copyright (C) Daniel Stenberg, <daniel@haxx.se>, et al.
|
|
#
|
|
# This software is licensed as described in the file COPYING, which
|
|
# you should have received as part of this distribution. The terms
|
|
# are also available at https://curl.se/docs/copyright.html.
|
|
#
|
|
# You may opt to use, copy, modify, merge, publish, distribute and/or sell
|
|
# copies of the Software, and permit persons to whom the Software is
|
|
# furnished to do so, under the terms of the COPYING file.
|
|
#
|
|
# This software is distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY
|
|
# KIND, either express or implied.
|
|
#
|
|
# SPDX-License-Identifier: curl
|
|
#
|
|
###########################################################################
|
|
#
|
|
import logging
|
|
import re
|
|
|
|
import pytest
|
|
from testenv import CurlClient, Env
|
|
|
|
log = logging.getLogger(__name__)
|
|
|
|
|
|
class TestBasic:
|
|
|
|
# simple http: GET
|
|
def test_01_01_http_get(self, env: Env, httpd):
|
|
curl = CurlClient(env=env)
|
|
url = f'http://{env.domain1}:{env.http_port}/data.json'
|
|
r = curl.http_get(url=url)
|
|
r.check_response(http_status=200)
|
|
assert r.json['server'] == env.domain1
|
|
|
|
# simple https: GET, any http version
|
|
@pytest.mark.skipif(condition=not Env.have_ssl_curl(), reason="curl without SSL")
|
|
def test_01_02_https_get(self, env: Env, httpd):
|
|
curl = CurlClient(env=env)
|
|
url = f'https://{env.domain1}:{env.https_port}/data.json'
|
|
r = curl.http_get(url=url)
|
|
r.check_response(http_status=200)
|
|
assert r.json['server'] == env.domain1
|
|
|
|
# simple https: GET, h2 wanted and got
|
|
@pytest.mark.skipif(condition=not Env.have_ssl_curl(), reason="curl without SSL")
|
|
@pytest.mark.skipif(condition=not Env.have_h2_curl(), reason="curl without h2")
|
|
def test_01_03_h2_get(self, env: Env, httpd):
|
|
curl = CurlClient(env=env)
|
|
url = f'https://{env.domain1}:{env.https_port}/data.json'
|
|
r = curl.http_get(url=url, extra_args=['--http2'])
|
|
r.check_response(http_status=200, protocol='HTTP/2')
|
|
assert r.json['server'] == env.domain1
|
|
|
|
# simple https: GET, h2 unsupported, fallback to h1
|
|
@pytest.mark.skipif(condition=not Env.have_ssl_curl(), reason="curl without SSL")
|
|
@pytest.mark.skipif(condition=not Env.have_h2_curl(), reason="curl without h2")
|
|
def test_01_04_h2_unsupported(self, env: Env, httpd):
|
|
curl = CurlClient(env=env)
|
|
url = f'https://{env.domain2}:{env.https_port}/data.json'
|
|
r = curl.http_get(url=url, extra_args=['--http2'])
|
|
r.check_response(http_status=200, protocol='HTTP/1.1')
|
|
assert r.json['server'] == env.domain2
|
|
|
|
# simple h3: GET, want h3 and get it
|
|
@pytest.mark.skipif(condition=not Env.have_h3(), reason="h3 not supported")
|
|
def test_01_05_h3_get(self, env: Env, httpd, nghttpx):
|
|
curl = CurlClient(env=env)
|
|
url = f'https://{env.domain1}:{env.h3_port}/data.json'
|
|
r = curl.http_get(url=url, extra_args=['--http3-only'])
|
|
r.check_response(http_status=200, protocol='HTTP/3')
|
|
assert r.json['server'] == env.domain1
|
|
|
|
# simple download, check connect/handshake timings
|
|
@pytest.mark.skipif(condition=not Env.have_ssl_curl(), reason="curl without SSL")
|
|
@pytest.mark.parametrize("proto", Env.http_protos())
|
|
def test_01_06_timings(self, env: Env, httpd, nghttpx, proto):
|
|
curl = CurlClient(env=env)
|
|
url = f'https://{env.authority_for(env.domain1, proto)}/data.json'
|
|
r = curl.http_download(urls=[url], alpn_proto=proto, with_stats=True)
|
|
r.check_stats(http_status=200, count=1,
|
|
remote_port=env.port_for(alpn_proto=proto),
|
|
remote_ip='127.0.0.1')
|
|
# there are cases where time_connect is reported as 0
|
|
assert r.stats[0]['time_connect'] >= 0, f'{r.stats[0]}'
|
|
assert r.stats[0]['time_appconnect'] > 0, f'{r.stats[0]}'
|
|
# ports are reported correctly
|
|
assert r.stats[0]['remote_port'] == env.port_for(proto), f'{r.dump_logs()}'
|
|
assert r.stats[0]['local_port'] > 0, f'{r.dump_logs()}'
|
|
|
|
# simple https: HEAD
|
|
@pytest.mark.parametrize("proto", Env.http_protos())
|
|
@pytest.mark.skipif(condition=not Env.have_ssl_curl(), reason="curl without SSL")
|
|
def test_01_07_head(self, env: Env, httpd, nghttpx, proto):
|
|
curl = CurlClient(env=env)
|
|
url = f'https://{env.authority_for(env.domain1, proto)}/data.json'
|
|
r = curl.http_download(urls=[url], with_stats=True, with_headers=True,
|
|
extra_args=['-I'])
|
|
r.check_stats(http_status=200, count=1, exitcode=0,
|
|
remote_port=env.port_for(alpn_proto=proto),
|
|
remote_ip='127.0.0.1')
|
|
# got the Content-Length: header, but did not download anything
|
|
assert r.responses[0]['header']['content-length'] == '30', f'{r.responses[0]}'
|
|
assert r.stats[0]['size_download'] == 0, f'{r.stats[0]}'
|
|
|
|
# http: GET for HTTP/2, see Upgrade:, 101 switch
|
|
@pytest.mark.skipif(condition=not Env.have_h2_curl(), reason="curl without h2")
|
|
def test_01_08_h2_upgrade(self, env: Env, httpd):
|
|
curl = CurlClient(env=env)
|
|
url = f'http://{env.domain1}:{env.http_port}/data.json'
|
|
r = curl.http_get(url=url, extra_args=['--http2'])
|
|
r.check_exit_code(0)
|
|
assert len(r.responses) == 2, f'{r.responses}'
|
|
assert r.responses[0]['status'] == 101, f'{r.responses[0]}'
|
|
assert r.responses[1]['status'] == 200, f'{r.responses[1]}'
|
|
assert r.responses[1]['protocol'] == 'HTTP/2', f'{r.responses[1]}'
|
|
assert r.json['server'] == env.domain1
|
|
|
|
# http: GET for HTTP/2 with prior knowledge
|
|
@pytest.mark.skipif(condition=not Env.have_h2_curl(), reason="curl without h2")
|
|
def test_01_09_h2_prior_knowledge(self, env: Env, httpd):
|
|
curl = CurlClient(env=env)
|
|
url = f'http://{env.domain1}:{env.http_port}/data.json'
|
|
r = curl.http_get(url=url, extra_args=['--http2-prior-knowledge'])
|
|
r.check_exit_code(0)
|
|
assert len(r.responses) == 1, f'{r.responses}'
|
|
assert r.response['status'] == 200, f'{r.responsw}'
|
|
assert r.response['protocol'] == 'HTTP/2', f'{r.response}'
|
|
assert r.json['server'] == env.domain1
|
|
|
|
# http: strip TE header in HTTP/2 requests
|
|
@pytest.mark.skipif(condition=not Env.have_h2_curl(), reason="curl without h2")
|
|
def test_01_10_te_strip(self, env: Env, httpd):
|
|
curl = CurlClient(env=env)
|
|
url = f'https://{env.authority_for(env.domain1, "h2")}/data.json'
|
|
r = curl.http_get(url=url, extra_args=['--http2', '-H', 'TE: gzip'])
|
|
r.check_exit_code(0)
|
|
assert len(r.responses) == 1, f'{r.responses}'
|
|
assert r.responses[0]['status'] == 200, f'{r.responses[1]}'
|
|
assert r.responses[0]['protocol'] == 'HTTP/2', f'{r.responses[1]}'
|
|
|
|
# http: large response headers
|
|
# send 48KB+ sized response headers to check we handle that correctly
|
|
# larger than 64KB headers expose a bug in Apache HTTP/2 that is not
|
|
# RSTing the stream correctly when its internal limits are exceeded.
|
|
@pytest.mark.parametrize("proto", Env.http_protos())
|
|
def test_01_11_large_resp_headers(self, env: Env, httpd, proto):
|
|
curl = CurlClient(env=env)
|
|
url = f'https://{env.authority_for(env.domain1, proto)}' \
|
|
f'/curltest/tweak?x-hd={48 * 1024}'
|
|
r = curl.http_get(url=url, alpn_proto=proto, extra_args=[])
|
|
r.check_exit_code(0)
|
|
assert len(r.responses) == 1, f'{r.responses}'
|
|
assert r.responses[0]['status'] == 200, f'{r.responses}'
|
|
|
|
# http: response headers larger than what curl buffers for
|
|
@pytest.mark.skipif(condition=not Env.httpd_is_at_least('2.4.64'),
|
|
reason='httpd must be at least 2.4.64')
|
|
@pytest.mark.parametrize("proto", Env.http_h1_h2_protos())
|
|
def test_01_12_xlarge_resp_headers(self, env: Env, httpd, configures_httpd, proto):
|
|
httpd.set_extra_config('base', [
|
|
f'H2MaxHeaderBlockLen {130 * 1024}',
|
|
])
|
|
httpd.reload_if_config_changed()
|
|
curl = CurlClient(env=env)
|
|
url = f'https://{env.authority_for(env.domain1, proto)}' \
|
|
f'/curltest/tweak?x-hd={128 * 1024}'
|
|
r = curl.http_get(url=url, alpn_proto=proto, extra_args=[])
|
|
r.check_exit_code(0)
|
|
assert len(r.responses) == 1, f'{r.responses}'
|
|
assert r.responses[0]['status'] == 200, f'{r.responses}'
|
|
|
|
# http: 1 response header larger than what curl buffers for
|
|
@pytest.mark.skipif(condition=not Env.httpd_is_at_least('2.4.64'),
|
|
reason='httpd must be at least 2.4.64')
|
|
@pytest.mark.parametrize("proto", Env.http_h1_h2_protos())
|
|
def test_01_13_megalarge_resp_headers(self, env: Env, httpd, configures_httpd, proto):
|
|
httpd.set_extra_config('base', [
|
|
'LogLevel http2:trace2',
|
|
f'H2MaxHeaderBlockLen {130 * 1024}',
|
|
])
|
|
httpd.reload_if_config_changed()
|
|
curl = CurlClient(env=env)
|
|
url = f'https://{env.authority_for(env.domain1, proto)}' \
|
|
f'/curltest/tweak?x-hd1={128 * 1024}'
|
|
r = curl.http_get(url=url, alpn_proto=proto, extra_args=[])
|
|
if proto == 'h2':
|
|
r.check_exit_code(16) # CURLE_HTTP2
|
|
else:
|
|
r.check_exit_code(100) # CURLE_TOO_LARGE
|
|
|
|
# http: several response headers, together > 256 KB
|
|
# nghttp2 error -905: Too many CONTINUATION frames following a HEADER frame
|
|
@pytest.mark.skipif(condition=not Env.httpd_is_at_least('2.4.64'),
|
|
reason='httpd must be at least 2.4.64')
|
|
@pytest.mark.parametrize("proto", Env.http_h1_h2_protos())
|
|
def test_01_14_gigalarge_resp_headers(self, env: Env, httpd, configures_httpd, proto):
|
|
httpd.set_extra_config('base', [
|
|
'LogLevel http2:trace2',
|
|
f'H2MaxHeaderBlockLen {1024 * 1024}',
|
|
])
|
|
httpd.reload_if_config_changed()
|
|
curl = CurlClient(env=env)
|
|
url = f'https://{env.authority_for(env.domain1, proto)}' \
|
|
f'/curltest/tweak?x-hd={256 * 1024}'
|
|
r = curl.http_get(url=url, alpn_proto=proto, extra_args=[])
|
|
if proto == 'h2':
|
|
r.check_exit_code(16) # CURLE_HTTP2
|
|
else:
|
|
r.check_exit_code(0) # 1.1 can do
|
|
|
|
# http: one response header > 256 KB
|
|
@pytest.mark.skipif(condition=not Env.httpd_is_at_least('2.4.64'),
|
|
reason='httpd must be at least 2.4.64')
|
|
@pytest.mark.parametrize("proto", Env.http_h1_h2_protos())
|
|
def test_01_15_gigalarge_resp_headers(self, env: Env, httpd, configures_httpd, proto):
|
|
httpd.set_extra_config('base', [
|
|
'LogLevel http2:trace2',
|
|
f'H2MaxHeaderBlockLen {1024 * 1024}',
|
|
])
|
|
httpd.reload_if_config_changed()
|
|
curl = CurlClient(env=env)
|
|
url = f'https://{env.authority_for(env.domain1, proto)}' \
|
|
f'/curltest/tweak?x-hd1={256 * 1024}'
|
|
r = curl.http_get(url=url, alpn_proto=proto, extra_args=[])
|
|
if proto == 'h2':
|
|
r.check_exit_code(16) # CURLE_HTTP2
|
|
else:
|
|
r.check_exit_code(100) # CURLE_TOO_LARGE
|
|
|
|
# http: invalid request headers, GET, issue #16998
|
|
@pytest.mark.parametrize("proto", Env.http_protos())
|
|
def test_01_16_inv_req_get(self, env: Env, httpd, nghttpx, proto):
|
|
curl = CurlClient(env=env)
|
|
url = f'https://{env.authority_for(env.domain1, proto)}/curltest/echo'
|
|
r = curl.http_get(url=url, alpn_proto=proto, extra_args=[
|
|
'-H', "a: a\x0ab"
|
|
])
|
|
# on h1, request is sent, h2/h3 reject
|
|
if proto == 'http/1.1':
|
|
r.check_exit_code(0)
|
|
else:
|
|
r.check_exit_code(43)
|
|
|
|
# http: special handling of TE request header
|
|
@pytest.mark.parametrize("te_in, te_out", [
|
|
pytest.param('trailers', 'trailers', id='trailers'),
|
|
pytest.param('chunked', None, id='chunked'),
|
|
pytest.param('gzip, trailers', 'trailers', id='gzip+trailers'),
|
|
pytest.param('gzip ;q=0.2;x="y,x", trailers', 'trailers', id='gzip+q+x+trailers'),
|
|
pytest.param('gzip ;x="trailers", chunks', None, id='gzip+x+chunks'),
|
|
])
|
|
@pytest.mark.skipif(condition=not Env.have_h2_curl(), reason="curl without h2")
|
|
def test_01_17_TE(self, env: Env, httpd, te_in, te_out):
|
|
proto = 'h2'
|
|
curl = CurlClient(env=env)
|
|
url = f'https://{env.authority_for(env.domain1, proto)}/curltest/echo'
|
|
r = curl.http_download(urls=[url], alpn_proto=proto, with_stats=True,
|
|
with_headers=True,
|
|
extra_args=['-H', f'TE: {te_in}'])
|
|
r.check_response(200)
|
|
if te_out is not None:
|
|
assert r.responses[0]['header']['request-te'] == te_out, f'{r.responses[0]}'
|
|
else:
|
|
assert 'request-te' not in r.responses[0]['header'], f'{r.responses[0]}'
|
|
|
|
# check that an existing https: connection is not reused for http:
|
|
@pytest.mark.skipif(condition=not Env.have_h2_curl(), reason="curl without h2")
|
|
def test_01_18_tls_reuse(self, env: Env, httpd):
|
|
proto = 'h2'
|
|
curl = CurlClient(env=env)
|
|
url1 = f'https://{env.authority_for(env.domain1, proto)}/data.json'
|
|
url2 = f'http://{env.authority_for(env.domain1, proto)}/data.json'
|
|
r = curl.http_download(urls=[url1, url2], alpn_proto=proto, with_stats=True)
|
|
assert len(r.stats) == 2
|
|
assert r.total_connects == 2, f'{r.dump_logs()}'
|
|
|
|
# check that an existing http: connection is not reused for https:
|
|
@pytest.mark.skipif(condition=not Env.have_h2_curl(), reason="curl without h2")
|
|
def test_01_19_plain_reuse(self, env: Env, httpd):
|
|
proto = 'h2'
|
|
curl = CurlClient(env=env)
|
|
url1 = f'http://{env.domain1}:{env.http_port}/data.json'
|
|
url2 = f'https://{env.domain1}:{env.http_port}/data.json'
|
|
r = curl.http_download(urls=[url1, url2], alpn_proto=proto, with_stats=True)
|
|
assert len(r.stats) == 2
|
|
assert r.total_connects == 2, f'{r.dump_logs()}'
|
|
|
|
# use a custom method containing a space
|
|
# check that h2/h3 did send that in the :method pseudo header. #19543
|
|
@pytest.mark.skipif(condition=not Env.curl_is_verbose(), reason="needs verbosecurl")
|
|
@pytest.mark.parametrize("proto", Env.http_protos())
|
|
def test_01_20_method_space(self, env: Env, proto, httpd):
|
|
curl = CurlClient(env=env)
|
|
method = 'IN SANE'
|
|
url = f'https://{env.authority_for(env.domain1, proto)}/curltest/echo'
|
|
r = curl.http_download(urls=[url], alpn_proto=proto, with_stats=True,
|
|
extra_args=['-X', method])
|
|
if proto == 'h2' or proto == 'h3':
|
|
# h2+3 may close the connection for such invalid requests
|
|
re_m = re.compile(r'.*\[:method: ([^\]]+)\].*')
|
|
lines = [line for line in r.trace_lines if re_m.match(line)]
|
|
assert len(lines) == 1, f'{r.dump_logs()}'
|
|
m = re_m.match(lines[0])
|
|
assert m.group(1) == method, f'{r.dump_logs()}'
|
|
else:
|
|
# h1 should give us a real response
|
|
assert len(r.stats) == 1
|
|
r.check_response(http_status=400)
|