Daniel Stenberg [Mon, 24 Feb 2025 10:04:30 +0000 (11:04 +0100)]
contrithanks.sh: update docs/THANKS in place
Now using 'sort' for sorting the names. This has the small side-effect
that it sorts slightly different than the previously used sort function
(emacs).
I think this is a better sort and over all it makes it more convenient
to use the script as it removes a manual step.
Stefan Eissing [Tue, 4 Feb 2025 14:24:00 +0000 (15:24 +0100)]
wolfssl: tls early data support
Enable TLS Early Data for wolfSSL:
- merge WOLFSSL_CTX and WOLFSSL setup from ngtcp2 with the general
implemenation in wolfssl.c
- enable for QUIC via ngtcp2
- give Curl_vquic_tls_init() a `struct alpn_spec` like used for the TCP
case. Adapt gnutls and other users.
- enable pytest test cases for early data with wolfSSL
and while this messes up wolfssl.c anyway, do
- rename all struct/functions with prefix 'wolfssl_' to 'wssl_' to not
pollute that name prefix
- rename `ctx/handle` to `ssl_ctx/ssl`, as used in openssl case
Daniel Stenberg [Sat, 22 Feb 2025 12:05:17 +0000 (13:05 +0100)]
tool_operate: fail SSH transfers without server auth
This now insists on using a server auth option unless --insecure is
provided. As an added bonus, it now also only checks for the knownhosts
file once (if found).
Daniel Stenberg [Fri, 21 Feb 2025 22:48:51 +0000 (23:48 +0100)]
http: make the RTSP version check stricter
- make it only accept version 1.0, as that is the version curl supports
- convert the parser to use strparse
- the status code max is now 999, but it does allow != 3 digits
Stefan Eissing [Sat, 22 Feb 2025 11:46:42 +0000 (12:46 +0100)]
multi: event based rework
Rework the event based handling of transfers and connections to
be "localized" into a single source file with clearer dependencies.
- add multi_ev.c and multi_ev.h
- add docs/internal/MULTI-EV.md to explain the overall workings
- only do event handling book keeping when the socket callback
is set
- add handling for "connection only" event tracking, when internal
easy handles are used that are not really tied to a connection.
Used in connection pool.
- remove transfer member "last_poll" and connections "shutdown_poll"
and keep all that internal to multi_ev.c
- add CURL_TRC_M() for tracing of "multi" related things, including
event handling and connection pool operations. Add new trace
feature "multi" for trace config.
multi traces will show exactly what is going on in regard to
event handling.
- multi: trace transfers "mstate" in every CURL_TRC_M() call
- make internal trace buffer 2048 bytes and end the silliness
with +n here -m there. Adjust test 1652 expectations of resulting
length and input edge cases.
- add trace feature "lib-ids" to perfix libcurl traces with transfer
and connection ids. Useful for debugging libcurl applications.
Viktor Szakats [Fri, 21 Feb 2025 16:30:46 +0000 (17:30 +0100)]
cmake: avoid `-Wnonnull` warning in `HAVE_FSETXATTR_5` detection
Seen in Android 21/35 CI jobs:
```
curl/CMake/CurlTests.c:315:16: warning: null passed to a callee that requires a non-null argument [-Wnonnull]
315 | fsetxattr(0, 0, 0, 0, 0);
| ^
1 warning generated.
```
Ref: https://github.com/curl/curl/actions/runs/13460225795/job/37613494183#step:9:5978
Viktor Szakats [Fri, 21 Feb 2025 15:15:49 +0000 (16:15 +0100)]
GHA/windows: replace GfW with MSYS2 runtime downgrade
We recently switched to a known good version of Git for Windows to avoid
the MSYS2/Cygwin runtime performance regression.
MSYS2 is closer to the source of the MSYS2/Cygwin projects. Its known
good version is newer. Installing the downgrade is faster and safer. It
also allows to restore the scripts to their original iteration, making
the workaround easier to drop once the perf issue is fixed upstream.
Therefore, switch back to using MSYS2, and install the runtime downgrade
before running curl tests.
Also disable `pacman`'s `CheckSpace` for best performance.
Jeremy identified to the root cause of the perf regression in this
Cygwin commit (from 2024-09-17):
https://cygwin.com/git/?p=newlib-cygwin.git;a=commit;h=c7fe29f5cb85242ae2607945762f7e0b9af02513
Co-authored-by: Jeremy Drake
Patch: https://github.com/jeremyd2019/curl/commit/95a404e19ae03ba8d35089e66d9690e3a4f11b7c
Ref: https://github.com/curl/curl/pull/16217#issuecomment-2673158597
Ref: https://github.com/curl/curl/pull/16217#issuecomment-2673461330
Daniel Stenberg [Fri, 21 Feb 2025 13:56:21 +0000 (14:56 +0100)]
asyn-ares: renamed define
It has the same name as the one used in asyn-thread, but for a slightly
different purpose. This not only caused unity build problems, but would
also be confusing and error-prone.
Viktor Szakats [Tue, 18 Feb 2025 01:49:31 +0000 (02:49 +0100)]
configure: silence compiler warnings in feature checks, drop duplicates
Silence compiler warnings (200 of them across the main CI workflows):
```
warning #2193: null argument provided for parameter marked with attribute "nonnull"
warning: Null pointer passed to 1st parameter expecting 'nonnull' [core.NonNullParamChecker]
warning: Null pointer passed to 2nd parameter expecting 'nonnull' [core.NonNullParamChecker]
warning: argument 1 null where non-null expected [-Wnonnull]
warning: argument 2 null where non-null expected [-Wnonnull]
warning: comparison of unsigned expression in '< 0' is always false [-Wtype-limits]
warning: null argument where non-null required (argument 1) [-Wnonnull]
```
Also drop `if ... can be linked` feature checks that were identical to
`if ... is compilable` checks, for:
`closesocket`, `ioctlsocket`, `socket`, `freeaddrinfo`, `getaddrinfo`,
`gethostname`, `getpeername`, `getsockname`,
`CloseSocket` (AmigaOS), `IoctlSocket` (AmigaOS).
Another option is to really do the link checks. But, if they weren't
missed so far, it seems safer to drop than risk a detection failure,
as was the case with AmigaOS functions while working on this PR.
There remain 22 `-Wnonnull` warnings in `gethostbyname_r()`,
`getpeername()` `getsockname()`. Most of the rest is necessary for
detection, or originate from autotools and CMake detection code
templates. Some still fixable, like duplicate libs.
Viktor Szakats [Fri, 10 Jan 2025 10:04:02 +0000 (11:04 +0100)]
build: add Windows CE / CeGCC support, with CI jobs
Make it possible to build curl for Windows CE using the CeGCC toolchain.
With both CMake and autotools, including tests and examples, also in CI.
The build configuration is the default one with Schannel enabled. No
3rd-party dependencies have been tested.
Also revive old code to make Schannel build with Windows CE, including
certificate verification.
Builds have been throughougly tested. But, I've made no functional tests
for this PR. Some parts (esp. file operations, like truncate and seek)
are stubbed out and likely broken as a result. Test servers build, but
they do not work on Windows CE. This patch substitutes `fstat()` calls
with `stat()`, which operate on filenames, not file handles. This may or
may not work and/or may not be secure.
About CeGCC: I used the latest available macOS binary build v0.59.1
r1397 from 2009, in native `mingw32ce` build mode. CeGCC is in effect
MinGW + GCC 4.4.0 + old/classic-mingw Windows headers. It targets
Windows CE v3.0 according to its `_WIN32_WCE` value. It means this PR
restores portions of old/classic-mingw support. It makes the Windows CE
codepath compatible with GCC 4.4.0. It also adds workaround for CMake,
which cannot identify and configure this toolchain out of the box.
Notes:
- CMake doesn't recognize CeGCC/mingw32ce, necessitating tricks as seen
with Amiga and MS-DOS.
- CMake doesn't set `MINGW` for mingw32ce. Set it and `MINGW32CE`
manually as a helper variable, in addition to `WINCE` which CMake sets
based on `CMAKE_SYSTEM_NAME`.
- CMake fails to create an implib for `libcurl.dll`, due to not
recognizing the platform as a Windowsy one. This patch adds the
necessary workaround to make it work.
- headers shipping with CeGCC miss some things curl needs for Schannel
support. Fixed by restoring and renovating code previously deleted
old-mingw code.
- it's sometime non-trivial to figure out if a fallout is WinCE,
mingw32ce, old-mingw, or GCC version-specific.
- WinCE is always Unicode. With exceptions: no `wmain`,
`GetProcAddress()`.
- `_fileno()` is said to convert from `FILE *` to `void *` which is
a Win32 file `HANDLE`. (This patch doesn't use this, but with further
effort it probably could be.)
https://stackoverflow.com/questions/3989545/how-do-i-get-the-file-handle-from-the-fopen-file-structure
- WinCE has no signals, current directory, stdio/CRT file handles, no
`_get_osfhandle()`, no `errno`, no `errno.h`. Some of this stuff is
standard C89, yet missing from this platform. Microsoft expects
Windows CE apps to use Win32 file API and `FILE *` exclusively.
- revived CeGCC here (not tested for this PR):
https://building.enlyze.com/posts/a-new-windows-ce-x86-compiler-in-2024/
On `UNDER_CE` vs. `_WIN32_WCE`: (This patch settled on `UNDER_CE`)
- A custom VS2008 WinCE toolchain does not set any of these.
The compiler binaries don't contain these strings, and has no compiler
option for targeting WinCE, hinting that a vanilla toolchain isn't
setting any of them either.
- `UNDER_CE` is automatically defined by the CeGCC compiler.
https://cegcc.sourceforge.net/docs/details.html
- `UNDER_CE` is similar to `_WIN32`, except it's not set automatically
by all compilers. It's not supposed to have any value, like a version.
(Though e.g. OpenSSL sets it to a version)
- `_WIN32_WCE` is the CE counterpart of the non-CE `_WIN32_WINNT` macro.
That does return the targeted Windows CE version.
- `_WIN32_WCE` is not defined by compilers, and relies on a header
setting it to a default, or the build to set it to the desired target
version. This is also how `_WIN32_WINNT` works.
- `_WIN32_WCE` default is set by `windef.h` in CeGCC.
- `_WIN32_WCE` isn't set to a default by MSVC Windows CE headers (the
ones I checked at least).
- CMake sets `_WIN32_WCE=<ver>`, `UNDER_CE`, `WINCE` for MSVC WinCE.
- `_WIN32_WCE` seems more popular in other projects, including CeGCC
itself. `zlib` is a notable exception amongst curl dependencies,
which uses `UNDER_CE`.
- Since `_WIN32_WCE` needs "certain" headers to have it defined, it's
undefined depending on headers included beforehand.
- `curl/curl.h` re-uses `_WIN32_WCE`'s as a self-guard, relying on
its not-(necessarily)-defined-by-default property:
https://github.com/curl/curl/blob/25b445e4796bcbf9f842de686a8c384b30f6c2a2/include/curl/curl.h#L77
Viktor Szakats [Wed, 19 Feb 2025 11:26:47 +0000 (12:26 +0100)]
schannel: enable ALPN support under WINE 6.0+
ALPN support was announced in 5.6 (2020-04-10). It likely needs a WINE
built against GnuTLS 3.2.0 (2013-05-10) or upper (for macOS, GnuTLS was
made default in WINE 6.0.2). I could confirm ALPN working under 6.0.2
(2021-10-26).
Viktor Szakats [Thu, 30 Jan 2025 15:38:18 +0000 (16:38 +0100)]
cmake: drop `CURL_DISABLE_TESTS` option
curl builds tests with CMake when explicitly building the `testdeps`
target. It's not built by default. It seems overkill to have
a curl-specific variant of this (over CMake's `BUILD_TESTING`)
to disable generating this target.
Its history also doesn't make it obvious why this was necessary,
and there was a long debate how to do it, by the time the original
submitter abandoned CMake. The option also remained uninitialized
and thus undocumented.
MSYS/MSYS2 and Cygwin are the same platform. Adjust code where they were
treated differently.
- drop separate `MSYS` from buildinfo flags. Our code is using the
`CYGWIN` variable and CMake (since v3.21) sets it also for `MSYS`.
- fix test1158 and test1186 to exclude them for all Win32 targets,
instead of just MSYS test envs. To align behavior between MSYS and
Cygwin envs. Required for recent MSYS2 releases which reports itself
as Cygwin, and no longer MSYS, which broke the previous exclusion
logic.
- follow Cygwin bumping its `MAX_PID` value, to avoid PID collisions.
https://cygwin.com/git/?p=newlib-cygwin.git;a=commit;h=363357c023ce01e936bdaedf0f479292a8fa4e0f
Reported-by: Jeremy Drake
Bug: https://github.com/curl/curl/pull/16217#issuecomment-2672768233
Ref: https://www.msys2.org/news/#2025-02-14-moving-msys2-closer-to-cygwin
Closes #16411
Daniel Stenberg [Thu, 20 Feb 2025 15:14:58 +0000 (16:14 +0100)]
http: simplify the check for auth methods
Avoids having to use the correct index into the line. Avoids repeated
use of is_valid_auth_separator.
Require that the following letter is not an alnum instead of checking
explicitly for ch == '\0' || ch == ',' || ISSPACE(ch). After all, the
point is to not erroneously match another auth string using the same
prefix.
Stefan Eissing [Wed, 19 Feb 2025 09:52:34 +0000 (10:52 +0100)]
conn: fix connection reuse when SSL is optional
In curl 8.12 I tried to improve the logic on how we handle connections
that "upgrade" to TLS later, e.g. with a STARTTLS. I found the existing
code hard to read in this regard. But of course, the "improvements" blew
up in my face.
We fixed issues with imap, opo3, smtp in 8.12.1, but ftp was no longer
reusing existing, upgraded control connections as before. This PR adds
checks in our pytest FTP tests that verify reuse is happening as
intended.
I rewrote the logic in url.c again, so that the new test checks now pass.
Reported-by: Zenju on github
Fixes #16384
Closes #16392
Stefan Eissing [Mon, 10 Feb 2025 16:40:11 +0000 (17:40 +0100)]
client writer: handle pause before deocding
Adds a "cw-pause" client writer in the PROTOCOL phase that buffers
output when the client paused the transfer. This prevents content
decoding from blowing the buffer in the "cw-out" writer.
Added test_02_35 that downloads 2 100MB gzip bombs in parallel and
pauses after 1MB of decoded 0's.
This is a solution to issue #16280, with some limitations:
- cw-out still needs buffering of its own, since it can be paused
"in the middle" of a write that started with some KB of gzipped
zeros and exploded into several MB of calls to cw-out.
- cw-pause will then start buffering on its own *after* the write
that caused the pause. cw-pause has no buffer limits, but the
data it buffers is still content-encoded.
Protocols like http/1.1 stop receiving, h2/h3 have window sizes,
so the cw-pause buffer should not grow out of control, at least
for these protocols.
- the current limit on cw-out's buffer is ~75MB (for whatever
historical reason). A potential content-encoding that blows 16KB
(the common h2 chunk size) into > 75MB would still blow the buffer,
making the transfer fail. A gzip of 0's makes 16KB into ~16MB, so
that still works.
A better solution would be to allow CURLE_AGAIN handling in the client
writer chain and make all content encoders handle that. This would stop
explosion of encoding on a pause right away. But this is a large change
of the deocoder operations.
Reported-by: lf- on github
Fixes #16280
Closes #16296
Stefan Eissing [Tue, 28 Jan 2025 13:11:59 +0000 (14:11 +0100)]
http: negotiation and room for alt-svc/https rr to navigate
Add a 'wanted' major HTTP version bitmask next to the 'allowed' bitmask
in HTTP version negotiation. This will try connections as specified in
'wanted', but enabled Alt-Svc and HTTPS-RR to redirect to other major
HTTP versions, if those are 'allowed'.
Changes libcurl internal default to `CURL_HTTP_VERSION_NONE` and removes
the code in curl that sets `CURL_HTTP_VERSION_2TLS` if the command line
does not say anything else.
Stefan Eissing [Wed, 19 Feb 2025 15:49:31 +0000 (16:49 +0100)]
cfilter: remove 'blocking' connect handling
Remove `blocking` argument from cfilter's connect method.
Implement blocking behaviour in Curl_conn_connect() instead for all
filter chains.
Update filters implementations. Several of which did never use the
paramter (QUIC for example). Simplifies connect handling in TLS filters
that no longer need to loop
Fixed a blocking connect call in FTP when waiting on a socket accept()
which only worked because the filter did not implement it.
Daniel Stenberg [Wed, 19 Feb 2025 22:55:31 +0000 (23:55 +0100)]
tool_getparam: clear sensitive arguments better
curl attempts to clear some flags to hide them from snooping neighbors
(on platforms where it works). For example the credentials provided with
-u. Previously it would only do that if there was a space between the
option and the credentials as in "-u joe:s3cr3t" but not when done
without a separating space as in "-ujoe:s3cr3t".
This addresses that previous shortcoming.
Reported-by: kayrus on github
Fixes #16396
Closes #16401
Viktor Szakats [Wed, 19 Feb 2025 16:26:58 +0000 (17:26 +0100)]
build: silence bogus `-Wconversion` warnings with gcc 5.1-5.4
It's fixed in gcc 5.5.0.
Example: https://godbolt.org/z/x6Th8q844
Seen in gcc 5.1.0, 5.4.0 (both 32/64-bit) with dl-mingw:
```
lib/rtsp.c: In function 'rtsp_parse_transport':
lib/rtsp.c:1025:36: error: conversion to 'unsigned char' from 'int' may alter its value [-Werror=conversion]
rtp_channel_mask[idx] |= (unsigned char)(1 << off);
^
lib/mprintf.c: In function 'parsefmt':
lib/mprintf.c:526:31: error: conversion to 'unsigned char' from 'int' may alter its value [-Werror=conversion]
usedinput[width/8] |= (unsigned char)(1 << (width&7));
^
lib/mprintf.c:544:35: error: conversion to 'unsigned char' from 'int' may alter its value [-Werror=conversion]
usedinput[precision/8] |= (unsigned char)(1 << (precision&7));
^
lib/mprintf.c:559:29: error: conversion to 'unsigned char' from 'int' may alter its value [-Werror=conversion]
usedinput[param/8] |= (unsigned char)(1 << (param&7));
^
lib/cfilters.c: In function 'Curl_pollset_change':
lib/cfilters.c:935:25: error: conversion to 'unsigned char' from 'int' may alter its value [-Werror=conversion]
ps->actions[i] |= (unsigned char)add_flags;
^
```
gcc 5.1.0: https://github.com/curl/curl/actions/runs/13413103492/job/37467698381#step:9:21
gcc 5.4.0: https://github.com/curl/curl/actions/runs/13413103492/job/37467694479#step:9:19
Daniel Stenberg [Wed, 19 Feb 2025 07:49:54 +0000 (08:49 +0100)]
strparse: provide access functions
To access the string and the length without having to directly use the
struct field names. Gives more freedom, flexbility and keeps
implementation specifics out of users' code.
Daniel Stenberg [Tue, 18 Feb 2025 22:03:09 +0000 (23:03 +0100)]
cookie: convert to using strparse
- using strparse cleans up the code and makes it easier to read and follow
- remove ? handling never used - since the path is provided without queries nowadays
- simplify sanitize_cookie_path
- avoid the strdup in pathmatch()
Daniel Stenberg [Tue, 18 Feb 2025 15:39:25 +0000 (16:39 +0100)]
ssh: consider sftp quote commands case sensitive
They have always been documented in lowercase. They have never been
claimed to be case insensitive. They mostly map to unix counterparts
that are always lowercase. Switch to case sensitive checks: lowercase.
Daniel Stenberg [Mon, 17 Feb 2025 21:34:21 +0000 (22:34 +0100)]
strparse: speed up the hex parser somewhat
Around 2.3x speed-up parsing many large hexadecimal numbers. The decimal and
octal parser get marginally faster.
Still very readable, compact and easy to follow code.
Tweaks
- combine the max and the overflow check, gains 3ns/num (use a separate
check outside of the loop instead for max < base)
- one less indirection in the pointer, gains 3ns/num
- using the table lookup for hex nums, gains 5ns/num
- unfold the num_digit() macro, gains 3s/num
- use the hexasciitable unconditionally, gains 2ns/num
- use post-increment pointer in the table lookup, gains 1ns/num
- improved valid_digit() using the table for the hex case,
gains 26 ns/num
- use "max char" in valid_digit(), gains 3ns/num
Behavior changes:
- no longer returns STRE_TOO_BIG - only STRE_OVERFLOW
- does not move the char ** on error, which is probably better
Stefan Eissing [Thu, 30 Jan 2025 14:31:16 +0000 (15:31 +0100)]
https-rr: implementation improvements
- fold DoH and async HTTPS-RR handling into common code.
have common cleanups, etc. Have a CURLcode result in async
handling to allow HTTPS RR parsing to fail.
- keep target, ipv4hints, ipv6hints, port and echconfig also
when resolving via cares. We need to know `target` and `port`
when evaluating possible ALPN candidates to not go astray.
- add CURL_TRC_DNS for tracing DNS operations
- replace DoH specific tracing with DNS, use doh as alias
for dns in curl_global_tracea()
Stefan Eissing [Mon, 27 Jan 2025 14:39:13 +0000 (15:39 +0100)]
http: version negotiation
Translate the `data->set.httpwant` which is one of the consts from the
public API (CURL_HTTP_VERSION_*) into a major version mask plus
additional flags for internal handling.
`Curl_http_neg_init()` does the translation and flags setting in http.c,
using new internal consts CURL_HTTP_V1x, CURL_HTTP_V2x and CURL_HTTP_V3x
for the major versions. The flags are
- only_10: when the application explicity asked fro HTTP/1.0
- h2_upgrade: when the application asks for upgrading 1.1 to 2.
- h2_prior_knowledge: when directly talking h2 without ALPN
- accept_09: when a HTTP/0.9 response is acceptable.
The Alt-Svc and HTTPS RR redirections from one ALPN to another obey the
allowed major versions. If a transfer has only h3 enabled, Alt-Svc
redirection to h2 is ignored.
This is the current implementation. It can be debated if Alt-Svc should
be able to override the allowed major versions. Added test_12_06 to
verify the current restriction.
Viktor Szakats [Mon, 17 Feb 2025 21:25:32 +0000 (22:25 +0100)]
GHA/windows: drop no-op `-DCMAKE_BUILD_TYPE=` from MSVC jobs
They use Visual Studio generators, which are multi-target.
The build command does the Release/Debug selection via `--config`.
Also:
- appveyor: drop unnecessary conditional for 3 options.
To sync with GHA.
- appveyor: drop unused `-DCMAKE_INSTALL_PREFIX=`.
To sync with GHA.
- sync cmake option order between GHA and appveyor.
Viktor Szakats [Thu, 9 Jan 2025 10:43:42 +0000 (11:43 +0100)]
cmake: sync OpenSSL(-fork) feature checks with `./configure`
`./configure` uses `AC_CHECK_FUNC` for these checks, with one exception
(`SSL_CTX_set_srp_username`). It's slightly less precise but simpler as
it doesn't need headers and/or macros. Do the same in CMake.
It also allows merging ECH detections across OpenSSL forks in CMake too.
Stefan Eissing [Thu, 13 Feb 2025 14:33:45 +0000 (15:33 +0100)]
ssl session cache: add exportable flag
Give peers and `exportable` flag, set TRUE when sessions for this peer
should not be exported. This evalualtes if the peer uses confidential
information (like srp username/password), a client certificate OR if the
"ssl_peer_key" contains relative paths.
When SSL is configured with paths for relevant components, like CA trust
anchors, an attempt is made to make this path absolute. When that does
not work or the infrstructure is not available, the peer key is marked
as *local*.
Exporting sessions based on relative paths may lead to confusion when
later imported in another execution context.
Stefan Eissing [Sun, 16 Feb 2025 14:19:20 +0000 (15:19 +0100)]
hash: use single linked list for entries
Curl's double linked list is proven code, but it comes with some
additional memory overhead. Since hash's internal list of elements needs
only forward traversals, it seems worthwhile to use a single linked list
internally.
This saves 3 pointers per entry plus 3 pointers per slot.
Daniel Stenberg [Mon, 17 Feb 2025 10:24:49 +0000 (11:24 +0100)]
lib: simplify more white space loops
Since the ISBLANK() and ISSPACE() macros check for specific matches,
there is no point in using while(*ptr && ISSPACE(*ptr)) etc, as the
'*ptr' check is then superfluous.
Jay Satiro [Sun, 16 Feb 2025 07:49:43 +0000 (02:49 -0500)]
scripts/managen: fix parsing of markdown code sections
- Terminate a code section before parsing a heading line.
Prior to this change when a code line (eg " code") was followed
by a heading line (eg "## heading") the code section in the output
was terminated after converting the header instead of before. That led
to some weird formatting outputs depending on the nroff or roffit etc.
Viktor Szakats [Fri, 7 Feb 2025 12:44:39 +0000 (13:44 +0100)]
cmake: misc tidy-ups
- replace `add_compile_options()`, `add_definitions()` with directory
properties. To harmonize this across all scripts. The new commands are
verbose, but describe better how they work. The syntax is also closer
to setting target properties, helps grepping.
- prefer `CMAKE_INSTALL_PREFIX` over `--prefix` (in tests, CI).
Viktor Szakats [Tue, 11 Feb 2025 01:46:29 +0000 (02:46 +0100)]
build: fix compiler warnings in feature detections
Fix or silence compiler warnings happening in feature detections
to reduce log noise. Warnings may also get promoted to errors in certain
cases, causing missed detections.
It reduces the number of warnings by 4500+ across the linux, linux-old,
macos, non-native and windows GHA workflows (~142 jobs).
Also move picky warning logic for MSVC/Borland to
`CMake/PickyWarnings.cmake. To make them listed in the picky-warnings
log output, and to also apply to feature detections to make them compile
under the same conditions as source code. The hope is to help catching
issues faster. It also improves code quality of feature tests.
Fixed/silenced:
```
warning #177: variable "dummy" was declared but never referenced
warning #177: variable "flag" was declared but never referenced
warning #177: variable "res" was declared but never referenced
warning #592: variable "s" is used before its value is set
warning #1011: missing return statement at end of non-void function "main"
warning #1786: function "SSL_CTX_set_srp_password" (declared at line 1888 of "/usr/include/openssl/ssl.h") was declared deprecated ("Since OpenSSL 3.0")
warning #1786: function "SSL_CTX_set_srp_username" (declared at line 1887 of "/usr/include/openssl/ssl.h") was declared deprecated ("Since OpenSSL 3.0")
warning #2332: a value of type "const char *" cannot be assigned to an entity of type "char *" (dropping qualifiers)
warning: 'SSL_CTX_set_srp_password' is deprecated [-Wdeprecated-declarations]
warning: 'SSL_CTX_set_srp_password' is deprecated: Since OpenSSL 3.0 [-Wdeprecated-declarations]
warning: 'SSL_CTX_set_srp_username' is deprecated [-Wdeprecated-declarations]
warning: 'SSL_CTX_set_srp_username' is deprecated: Since OpenSSL 3.0 [-Wdeprecated-declarations]
warning: 'b' is used uninitialized [-Wuninitialized]
warning: 'gethostname' redeclared without dllimport attribute: previous dllimport ignored [-Wattributes]
warning: Undefined or garbage value returned to caller [core.uninitialized.UndefReturn]
warning: Value stored to 'i' is never read [deadcode.DeadStores]
warning: assigning to 'char *' from 'const char *' discards qualifiers [-Wincompatible-pointer-types-discards-qualifiers]
warning: assignment discards 'const' qualifier from pointer target type [-Wdiscarded-qualifiers]
warning: control reaches end of non-void function [-Wreturn-type]
warning: empty expression statement has no effect; remove unnecessary ';' to silence this warning [-Wextra-semi-stmt]
warning: excess elements in struct initializer
warning: initialization discards 'const' qualifier from pointer target type [-Wdiscarded-qualifiers]
warning: macro "_FILE_OFFSET_BITS" is not used [-Wunused-macros]
warning: macro "_REENTRANT" is not used [-Wunused-macros]
warning: missing braces around initializer [-Wmissing-braces]
warning: no previous extern declaration for non-static variable 'off_t_is_large' [-Wmissing-variable-declarations]
warning: no previous prototype for 'check' [-Wmissing-prototypes]
warning: no previous prototype for function 'check' [-Wmissing-prototypes]
warning: null argument where non-null required (argument 2) [-Wnonnull]
warning: passing 'const char[1]' to parameter of type 'char *' discards qualifiers [-Wincompatible-pointer-types-discards-qualifiers]
warning: passing argument 2 of 'SSL_CTX_set_srp_password' discards 'const' qualifier from pointer target type [-Wdiscarded-qualifiers]
warning: passing argument 2 of 'SSL_CTX_set_srp_username' discards 'const' qualifier from pointer target type [-Wdiscarded-qualifiers]
warning: unused parameter 'c' [-Wunused-parameter]
warning: unused parameter 'f' [-Wunused-parameter]
warning: unused variable 'data' [-Wunused-variable]
warning: unused variable 'dummy' [-Wunused-variable]
warning: unused variable 'flag' [-Wunused-variable]
warning: unused variable 'res' [-Wunused-variable]
warning: unused variable 's' [-Wunused-variable]
warning: variable 's' set but not used [-Wunused-but-set-variable]
warning: variable 'ts' set but not used [-Wunused-but-set-variable]
```
Viktor Szakats [Fri, 27 Dec 2024 23:27:26 +0000 (00:27 +0100)]
cmake: add pre-fill for Unix, enable in GHA/macos, verify pre-fills
TL;DR: Save 10 minutes of CI time for GHA/macos jobs using pre-fills and
add pre-fill verification for Apple and Windows. Also restores Xcode job
and saves 1.5-10 minutes configuring iOS jobs.
Pre-filling feature detection results can bring down the CMake configure
step to ~5 seconds on most GHA runners, ~10 seconds in slow envs like
Cygwin/MSYS2.
The potential savings per job are:
- 5-40 (average 19) seconds on GHA/macos (33 jobs)
- ~10 seconds on GHA for iOS GNU Makefile (1 job)
- 1.5-10 minutes on GHA for iOS Xcode generator (1 job)
- 10 seconds on GHA/linux with native Ubuntu (12 jobs)
- 40 seconds for Cygwin/MSYS2 (2 jobs)
- 5-10 seconds for virtualized BSDs, native CPU (3 jobs)
- ~60 seconds for virtualized BSDs, emulated CPU (1 job)
On native Windows pre-filling has been in place for a long time and
saving 8 minutes (VS2019-VS2015) to 1.5-2 minutes (VS2022), 3 minutes
(VS2022 UWP), and 30-60 seconds (MinGW), per CI job.
The downside is that detection results need to be manually collected and
filtered to those that universally apply to all platforms that they are
enabled on. Another downside is that by using a cache, we're not running
the actual detections, and thus won't catch regressions in them. It
means we must make sure that the cache is solid and matches with actual
detections results. An upside is that it gives a rough overview of which
features are available on which platforms. Another upside is pre-filled
values do work for feature detections skipped for cross-builds, e.g.
`HAVE_WRITABLE_ARGV`.
This PR adds a pre-fill cache that supports all Unixes (except OmniOS)
used in CI, and makes it usable with an internal option. It also enables
it for GHA/macos CI jobs, where the maximum savings are. And also for
the two iOS [1] and two Cygwin/MSYS2 jobs. The latters don't have
pre-fill checks and we can drop them if they turn into a hassle.
Saving:
- 10 minutes of CI time per GHA/macos workflow run. [2]
- ~80 seconds per GHA/windows workflow run with Cygwin/MSYS2.
(offsetting the cost of pre-fill verifications)
- 1.5-10 minutes per GHA/non-native runs with iOS jobs. [3]
You can enable pre-fill locally with `-D_CURL_PREFILL=ON`. It's
experimental, and if you experience a problem, file a PR or an Issue.
This PR also adds a pre-fill checker for macOS and MinGW/MSVC Windows
GHA jobs to catch if the cache diverges from real detections. It also
adds this logic to AppVeyor, but doesn't enable it due to the perf
penalty of 2 minutes mininum.
The pre-fill checker works by configuring out-of-tree with and without
pre-fill, then diffing their `lib/curl_config.h` outputs.
Exceptions are 3 detection results exposed indirectly [4], and missing
to expose 2, of which one is the C89 header `stddef.h`. While we assume
the C99 `stdint.h` available outside iOS. We can expose them in the
future, if necessary.
The pre-fill checks cost in total:
- ~20 seconds for macOS
- ~40 seconds for MinGW on GHA
- ~80 seconds for MSVC on GHA (UWP would be 2x this)
An extra time saving potential is caching type sizes. They are
well-known, and seldom change, esp. in CI. GHA/Windows jobs spend 8-17
seconds per job on these ~12 feature checks. ~5s on Cygwin/MSYS2. Couple
of seconds on other platforms. (This PR doesn't make this optimization.)
Another opportunity is doing the same for autotools, which typically
spends more time in the configuration step than cmake.
[4] detection results exposed indirectly in `curl_config.h`:
- `HAVE_FILE_OFFSET_BITS` via `_FILE_OFFSET_BITS`
- `HAVE_GETHOSTBYNAME_R_*_REENTRANT` via `NEED_REENTRANT`
- `HAVE_SOCKADDR_IN6_SIN6_ADDR` via `USE_IPV6`
Jay Satiro [Sat, 15 Feb 2025 18:32:34 +0000 (13:32 -0500)]
curl_msh3: remove verify bypass from DEBUGBUILDs
- Remove the workaround that disabled peer verification in DEBUGBUILDs
when CA certs were provided.
The workaround was part of a TODO that disabled verification in
DEBUGBUILDs with a CAfile/path because apparently there's no way to set
those options in msh3 and that caused some tests to fail. Instead the
tests should fail and this problem should not be covered up.