gbdev/rgbds
1
0
Fork 0
mirror of https://github.com/gbdev/rgbds.git synced 2026-09-23 14:17:06 +00:00
Code Issues Packages Projects Releases Wiki Activity
3,707 Commits 3 Branches 53 Tags
d43ef537456d56fab0b64d5fbbfceecc617aa470
Commit Graph
100 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
ISSOtm d43ef53745 Remove unreachable bank range check 
The only case where it is not generated by our own code
(which we ought to be `assume`ing is correct) is when the bank
is used-specified, but that is caught by an earlier check.
 2026-09-18 21:19:37 +02:00
ISSOtm ef709fade6 Refactor and fix sect packing's bank iteration 
Moving both pieces of related code together, as it were.
This also ends up changing the behaviour of scrambling,
where sections could “leak” out of the scrambling's bank pool
and resume being placed normally.

We have discussed it offline, and decided that this was a bug.
The test suite has been updated accordingly, which even gives
the occasion to move this test from a special case to the main
generic loop!
 2026-09-18 21:19:37 +02:00
ISSOtm 8685bcbfbf Refactor all scrambling-related dispatch into a single func 
Yay for (poor man's) pattern matching! 🦀
 2026-09-18 21:19:37 +02:00
ISSOtm 1ca4282e31 Move a function closer to its usage point 
Trying to keep code grouped together well.
 2026-09-18 21:19:37 +02:00
Eldred Habert afe1f01d00 Fix linker script align 16 in floating address context (#2115) 
Turns out we were just using the wrong var lmao
 2026-09-17 20:46:55 -04:00
ISSOtm f234796428 Avoid allocating for our linkdef names (#2111) 
This reverts a change introduced in fd78a9ae8, though it wasn't that commit's main point
so I'm feeling okay with undoing that.

This feels like an overkill change, using a static string is good enough for this
since we never modify this. I have considered using `string_view` instead, to have
the best of both worlds, but that's not NUL-terminated so our print functions
get a little grumpy.
 2026-09-17 13:58:30 -04:00
ISSOtm 0b9f8ab523 Use Section::typeInfo() more 2026-09-17 13:19:26 -04:00
Eldred Habert 3b250d5386 Emit auxiliary files even if asserts fail (#2104) 
Closes #2100
 2026-09-17 17:47:53 +02:00
ISSOtm 6f69867d04 Make more RGBLINK tests use at-files 
Less special handling = good

Possibly this can be done to more tests, but this is what I found.
 2026-09-17 10:25:47 -04:00
ISSOtm 960f82126f Fix to actually allow one-past-the-end zero-sized sections in linker scripts 2026-09-17 10:19:47 -04:00
ISSOtm b9b5c9a2b4 Avoid reporting redundant errors on too-large fixed sect addr 
This would happen for any such section of non-zero size.
 2026-09-17 10:17:26 -04:00
ISSOtm e39dbe1752 Remove no-op in link test script 
No, `printf` does not do that.
This works because making it read from stdin suppresses
any words after the first one.
 2026-09-17 09:35:32 -04:00
ISSOtm 9f175f7923 Use SectionTypeInfo struct member functions for section type info 2026-09-17 09:33:21 -04:00
ISSOtm 6587941512 Allow scrambling SRAM banks up to the full 8-bit range 2026-09-15 23:06:19 -04:00
Eldred Habert 0fe43bf0cb Use CMake-native methods to do our static-build macOS trickery (#2083) 
This should remain equivalent for our current setups,
but play nicer with anyone wanting to experiment further.

It *may* break for someone building on an Intel Mac,
but this is a libpng limitation and thus out of our control.
 2026-09-14 09:55:34 -04:00
Eldred Habert c8e8a2f522 Avoid duplicating warning flags when linking with the Makefile (#2085) 
The warning flag list with `make develop` is quite long,
and this duplication is making those command lines particularly indigestible.
 2026-09-13 20:35:54 -04:00
Eldred Habert de1e1ff688 Actually silence -Wformat et al. on MinGW (#2084) 
Thus far, we have just ensured they wouldn't fail the build,
but they still pollute the build log, so we should silence them entirely.
 2026-09-13 20:25:13 -04:00
ISSOtm 9d69a08909 Determine the number of fetch/clone jobs ourselves 
Apparently our CI has some versions of Git too old to support zero to mean that.
 2026-09-11 16:31:19 -04:00
ISSOtm 970bb84450 Fix updating ext-test repos from existing copies 
The current setup fails with a "cannot read tree" error.

Also added flags to parallelise each fetch, which is relevant when submodules
are involved. (So, pretty much just starter-kit, haha.)
It would be nice to also parallelise the various fetches, but that would hit
the same remote multiple times, and could lead to rate-limiting...
 2026-09-11 16:17:26 -04:00
ISSOtm a0ab1802ff Avoid fetching external repos on 32-bit Cygwin 
The comment says we aren't doing this, but we are
 2026-09-03 13:39:24 +02:00
ISSOtm bf3ab7cb18 Explicitly specify the Python version to be installed 
Turns out `python3-imaging` only redirects to Python 3.**6**'s version,
not the latest like the metapackage does.
 2026-09-03 13:39:24 +02:00
ISSOtm d9003f633a Allow the CMake user to provide their own version string 
This will be useful for downstream packagers, as well as rgbds-live
 2026-08-15 20:25:35 +02:00
ISSOtm cda77721d4 Avoid assuming that RGBDS is the root CMakeLists 
This breaks rgbds-live building us as a subproject, for example
 2026-08-15 20:05:17 +02:00
Eldred Habert 3b1606c63c Fix typo'd APT option in release workflow 
Fixing up bcff14b596
 2026-07-20 21:53:43 +02:00
ISSOtm 3eb8c7e8f2 Let tests know they are being run under Cygwin 2026-07-05 20:58:50 -04:00
ISSOtm 4a77c42ed7 Remove redundant labels from tests 
It's possible to filter tests by name (`-R`)
 2026-07-02 10:39:59 -04:00
ISSOtm b12eac356e Try to automatically determine the test OS name 2026-07-02 10:39:59 -04:00
ISSOtm 82702d8149 Point out an ordering dep isn't a success one 2026-07-02 10:39:59 -04:00
ISSOtm 1f9cc9651a Avoid fetching some repos twice 
Also enables a nice bit of logic cleanup
 2026-07-02 10:39:59 -04:00
ISSOtm 04b9c7fa3d Move external tests files to tests/external/ 2026-07-02 10:39:59 -04:00
ISSOtm 5e8a6cde20 Remove the need to explicitly list external tests 
I like having a single source of truth!
 2026-07-02 10:39:59 -04:00
ISSOtm d59081a918 Shorten external test prop names slightly 
Makes the scripts a bit easier to read, I think
 2026-07-02 10:39:59 -04:00
ISSOtm e3fc46f933 Remove static compile flags 
We support CMake now, which can generate a more accurate `compile_commands.json` database
(and our `develop` preset automatically enables them)
 2026-06-30 16:52:31 +02:00
ISSOtm 0ed8c2f8ee Docu-comment an implicit correctness assumption 2026-05-14 14:47:23 +02:00
ISSOtm d5186b088b Link against (non-debug) MSVC runtime statically 2026-05-14 14:47:23 +02:00
ISSOtm fee177d0b9 Link zlib and libpng statically by default 
Only when we are building them ourselves, that is.
 2026-05-14 14:47:23 +02:00
ISSOtm 1af8bdda16 Have CMake put RGBDS executables in our own dir 
Not the top-level if we are part of a larger project!
 2026-05-06 18:49:19 +02:00
ISSOtm fa22b78e7d Use type instead of which in shell scripts 
The latter is not a Bash built-in,
and some environments turn out to produce non-conforming
return codes.

`type` is provided by Bash itself,
so we can rely on it better.
 2026-05-05 16:03:33 +02:00
ISSOtm 37d39f91f6 Build using GCC 9 in CI 
Checking that we don't regress this later.
Ubuntu doesn't package GCC 8, and we don't support it either,
so that is a good compromise for now.
 2026-04-26 23:30:59 +02:00
ISSOtm d5ce5329ea Fix compilation with GCC 9 2026-04-26 23:30:59 +02:00
ISSOtm 6bcd79b997 Revert "Derive operator!= from operator== (#1660)" 
This reverts commit 993879a2ed.
 2026-04-26 23:30:59 +02:00
ISSOtm decc5f7153 Let parallelism be determined automatically 
Ninja is parallel by default;
however, it is not used on FreeBSD, so we still have to specify it there.
 2026-04-20 15:32:54 +02:00
ISSOtm a3c953ea46 Pass C++ compiler to CMake via env var 
This is a little terser
 2026-04-20 15:32:54 +02:00
ISSOtm d43049c84d Imply CMake source dir 
Shortening our command lines slightly
 2026-04-20 15:32:54 +02:00
ISSOtm b6ff54acc3 Avoid libpng compilation on Windows spamming the Microsoft banner 2026-04-20 15:32:54 +02:00
ISSOtm 71fc449452 Use Ninja on Unix platforms 
It has nicer UX, only showing full command lines when they fail.
It could also be slightly faster, though I've seen no measurable improvement.
 2026-04-20 15:32:54 +02:00
ISSOtm 59dd1fbbf8 Correct naming of MSBuild CMake presets 
You can build with MSVC without MSBuild
 2026-04-20 15:32:54 +02:00
ISSOtm 2502d130eb Use Ninja on Windows 
MSBuild is slow, and its output is hard to parse visually
(it's all interwoven, like Make without `--output-sync`).
Since we use CMake anyway, we don't care about the build system.
 2026-04-20 15:32:54 +02:00
ISSOtm a78a268cb4 Pass convenience options to Make invoked via CMake 
Can't pass them to plain Make invocations,
because we try to be compatible with macOS' old Make.
 2026-04-20 15:32:54 +02:00
ISSOtm aeedfb35a6 Set up Cygwin before cloning our repo 
Otherwise different Gits set up and tear down the repo,
leading to issues.
 2026-04-16 03:16:56 +02:00
ISSOtm 63609871f4 Cache downloaded dependencies for MinGW build also 
Forgot about you!
 2026-04-16 02:56:14 +02:00
ISSOtm 393eaa9dc1 Fix a Fix typo ;) 2026-04-16 02:54:40 +02:00
ISSOtm 185a5c5369 Add necessary flags to apt install for checkformat 2026-04-16 01:58:42 +02:00
ISSOtm 4ea15ed85a Touch up checkformat.bash 
The logic can be slightly simplified,
and the script doesn't require Bash after all.
 2026-04-16 01:05:03 +02:00
ISSOtm b218fd65a0 Install clang-format on ubuntu-slim runner 
Unlike the “fat” runners, that's not installed by default.
Why that causes `find` to nonetheless exit normally, I have no idea.
 2026-04-16 01:02:30 +02:00
ISSOtm 81edc27b6e Fix two warnings on MinGW 2026-04-16 00:57:33 +02:00
ISSOtm dc52487d31 Make -Wformat non-fatal on MinGW 
See the comment for a rationale.
 2026-04-16 00:57:33 +02:00
ISSOtm 6b994b1737 Fix some warnings raised by MSVC 2026-04-16 00:57:33 +02:00
ISSOtm 0849eb801e Decouple -Werror from MORE_WARNINGS 2026-04-16 00:57:33 +02:00
ISSOtm e6916360f4 Set some flags only for our own code 2026-04-16 00:57:33 +02:00
ISSOtm a802bcb320 Switch MinGW building to CMake 
Automatically grabs and compiles libpng, and avoids having hacky lines in our Makefile
(the compiler specification *should* be orthogonal to the build target!)

Also move the MinGW package install to the `install_deps.sh` script,
to move logic off of the YAML.
 2026-04-16 00:57:33 +02:00
ISSOtm 846aa975f1 Stop trying to install md5sha1sum 
macOS already provides it and Homebrew doesn't override it,
so that's a no-op anyway.
 2026-04-16 00:25:58 +02:00
ISSOtm b65649ca73 Tidy up system info 2026-04-16 00:25:58 +02:00
ISSOtm 62ddabb570 Avoid hardcoding Brew's prefix 2026-04-16 00:25:58 +02:00
ISSOtm d48d6fe5d9 Bump our macOS CI runners 2026-04-16 00:25:58 +02:00
Eldred Habert ac0f93ffda Delete GCC static analyzer workflow (#1950) 
As for GCC 15, it is stated not to be suitable for C++ code,
and we have gotten annoyingly many false positives out of it,
so let's just disable it (as per
https://github.com/gbdev/rgbds/pull/1950#issuecomment-4243959126).
 2026-04-14 22:04:20 +02:00
ISSOtm a1712cb6d9 Run patch format check on ubuntu-slim 2026-04-14 22:03:29 +02:00
Eldred Habert 2968bee86f Run macos-static tests using CTest (#1947) 
We can, so why not? It has benefits.
 2026-04-14 01:25:47 +02:00
ISSOtm dec7a74166 Fix macOS static build 
- Turns out the variable wasn't being set, and thus the flags weren't being passed
- Passing the flags via those variables is basically a bad idea, and it
  turns out to break libpng's config script
- The flags need to be passed during link as well, otherwise the binary
  isn't linked as a fat one
 2026-04-14 00:01:14 +02:00
Eldred Habert 9c4e53d1c4 Transition CI from pkg-config to pkgconf (#1942) 
Upstream has changed the providing implementation,
so that package will eventually be removed.
 2026-04-13 19:46:05 +02:00
Eldred HabertandRangi 463edc3c01 Enable color output in our Linux and macOS CI tests (#1939) 
Co-authored-by: Rangi <[email protected]>
 2026-04-13 19:31:01 +02:00
Eldred Habert 65ec1af1e8 Switch macOS static build to CMake (#1938) 
This moves the tortuous flags out of our CI scripts,
and allows us to benefit from the libpng download caching.
 2026-04-12 22:26:15 -04:00
Eldred Habert bc3572ec37 Stop trying to explicitly install pkg-config in CI (#1940) 
Turns out it's already installed by default,
but under a different name that's causing extra ops
for no real gain.
 2026-04-12 22:08:38 -04:00
Eldred Habert 78281a4aaa Break down CTest cases into each test suite (#1931) 
This allows parallelising the individual programs'
test suites, which can save a few minutes!
 2026-04-12 21:29:12 -04:00
ISSOtm 6e51ab6f55 Remove a duplicate warning flag 
We always pass this one
 2026-04-13 01:52:25 +02:00
ISSOtm 52aaf7e3b3 Fix MSVC dependent option syntax 
This is not how CMake works lol
 2026-04-13 01:37:38 +02:00
ISSOtm 09db4f25dd Bump the maximum version of CMake tested with 
Future-proofing!
This is both the version on my own machine,
and what's distributed by Homebrew (thus on GitHub's runners).
 2026-04-13 00:33:31 +02:00
Eldred Habert fed5db7743 Move documentation outside of the root directory (#1936) 2026-04-12 18:09:40 -04:00
ISSOtm c6dbec7830 Stop trying to run version test on Cygwin and FreeBSD 
Fixes #1934
 2026-04-12 23:40:39 +02:00
Eldred Habert 017b0119e6 Overhaul checkdiff CI (#1930) 2026-04-12 16:06:28 -04:00
Eldred Habert 0e6e91d048 Avoid copying RGBFIX binary when testing it (#1933) 
This is just cleaner in general,
and also avoids separating it from any DLLs it may be needing.
 2026-04-12 13:23:51 -04:00
Eldred Habert 123201ecc6 Avoid repeatedly specifying bash as the command shell in CI (#1926) 2026-04-10 22:04:16 -04:00
Eldred Habert 422e8eec20 Generate Bison output in CMake build tree (#1929) 
Again, moving away from in-tree builds
 2026-04-10 18:46:57 -04:00
ISSOtm 4577ff16bc Generate Bison output in CMake build tree 
Again, moving away from in-tree builds
 2026-04-10 22:56:59 +02:00
ISSOtm b430d5a8b8 Note why we are not using a CMake list to pass Bison flags 2026-04-10 22:33:54 +02:00
ISSOtmandvulcandth a5be1d886e Copy DLLs to appropriate dir for install-less Windows testing 
Co-authored-by: vulcandth <[email protected]>
 2026-04-10 18:01:16 +02:00
ISSOtm ace2f3e613 Document why lib src cache is cross-OS 
See https://github.com/gbdev/rgbds/pull/1919#discussion_r3048475606
 2026-04-08 12:25:43 +02:00
ISSOtm 8d1a60d6b5 Switch release process to a lightweight runner 
Easier to provision, less concurrency overhead for us
 2026-04-08 12:25:43 +02:00
ISSOtm 11ac22d8f5 Switch from an Action to gh release 
Fewer supply chain risks whooooo
 2026-04-08 12:25:43 +02:00
ISSOtm 812e1f6d4b Remove dead code checking for version tag 
It was also vulnerable to template injection attacks.
 2026-04-08 12:25:43 +02:00
ISSOtm 8dc052475b Avoid caching in our release workflow 
This is a malware distrib vector,
and we run this workflow sporadically enough
that we can afford to do so uncached.
 2026-04-08 12:25:43 +02:00
ISSOtm 133eaa4e92 Build Windows release in, well, Release mode 2026-04-08 12:25:43 +02:00
ISSOtm 57f1894fbd Fix installing Windows build deps on release 2026-04-08 12:25:43 +02:00
ISSOtm 021fe01eb0 Avoid CMake looking at test dir if testing is disabled 
This saves building the two test support programs.
 2026-04-08 12:25:43 +02:00
ISSOtm bcff14b596 Clean up docs workflows 
Usethe `ubuntu-slim` runner to (hopefully) get faster setup times
 2026-04-07 21:48:26 +02:00
Eldred Habert 44b6a2c453 Add a key that turns out to be required 
The documentation is super flaky tbh
 2026-03-31 02:14:24 +02:00
Eldred Habert 3fc3130364 Fix some YAML syntax 
I prefer TOML and JSON anyway...
 2026-03-31 02:12:29 +02:00
Eldred Habert 5090395ca7 Enable GH Actions to be automatically bumped 
Turns out those get new major releases somewhat often, and we get deprecation warnings and brownouts and all that.
Automatic bumps should help us avoid that, though I've set a *very* large scan interval to shield us from churn.
 2026-03-31 02:03:37 +02:00
ISSOtm 8025c2ae66 Require uploaded artifacts to be non-empty 
We can get silent failures otherwise
 2026-03-28 11:31:49 +01:00
ISSOtm f7831ce8da Simplify CTest pre-run test command 2026-03-28 01:21:08 +01:00