Commit Graph
2593 Commits
Author SHA1 Message Date
Rangi b99069e25d Fix reversing an image with an empty -p/--palette file (#2227) 2026-10-22 14:41:11 -04:00
Rangi 44b55dce69 Fix parsing of SDCC .rel objects' "b_" magic bank symbols (#2226)
This was a regression from 48b2e94aa3
swapping where `+ 1` applies when converting `strcmp` to `ends_with`.
2026-10-22 13:42:09 -04:00
Rangi ada6be082e Add error check for empty OPT string 2026-10-21 17:25:21 -04:00
Rangi 129b4dabee Fix an off-by-one overflow bug in SDCC object parsing (#2222) 2026-10-21 15:59:00 -04:00
Rangi 85f96d671f Allow empty one-past-the-end sections (#2221) 2026-10-21 15:58:30 -04:00
Rangi 72c9171216 Output UINT16_MAX for unused grayscale palette slots (#2219) 2026-10-21 10:01:41 -04:00
Rangi 4864f37662 Do not allow -c embedded for non-indexed PNGs even with PLTE chunks 2026-10-20 11:17:26 -04:00
ISSOtm d766efe95d Use a vector for emitting section patches
Turns out this makes errors be emitted in source order,
which is far more intuitive and pleasant-looking!
2026-10-17 14:39:51 -04:00
ISSOtm 6883dd903e Omit a type name where possible
This also allows omitting a header
(though it gets included implicitly by `<queue>` instead,
but that's fine).

This also makes our code slightly more amenable to change,
of course.
2026-10-17 14:15:54 -04:00
Rangi 170b341315 Refactor unassignedSections into a flat list sorted by constraints (#2218)
Also removes the now-redundant `nbSectionsToAssign` counter
2026-10-17 14:12:58 -04:00
Rangi d1fb662501 Fix rgbgfx -c/--colors to read GBC palettes with less than 4 colors (#2216) 2026-10-16 01:47:23 +02:00
Rangi 451cf69f7d Fix a typo 2026-10-15 10:21:42 -04:00
Rangi c5fe5f0762 Use else if in checkNBit for symmetry with checkPatchSize 2026-10-15 10:21:27 -04:00
Rangi 292f39deb3 Ensure that STRCMP is either -1, 0, or 1 2026-10-15 10:20:24 -04:00
Rangi 586bd2eec8 Non-constant expressions do not fail a static_assert (#2215) 2026-10-14 19:24:47 -04:00
Eldred Habert d663ab59e4 Enable more sanitizers (#2154) 2026-10-14 18:00:14 -04:00
ISSOtm 6818759a8c Tiny refactor 2026-10-11 18:56:32 +02:00
Rangi 6df6868f0d Check assert message requirement whether or not they are deferred 2026-10-10 22:36:19 -04:00
ISSOtm 82ee97b088 Check for builtin symbols directly, and document why 2026-10-11 03:44:39 +02:00
ISSOtm bf894bce6f Encapsulate Symbol and Section ID checks into isReferenced methods 2026-10-11 03:44:39 +02:00
Rangi af192a95b1 Do not output unreferenced unexported symbols in object files 2026-10-11 03:44:39 +02:00
Rangi 9d6dab13d8 Sort linked sections by alignment before size (#2201)
Note that Libbet has no canonical hash and is sensitive to floating section placement
2026-10-11 03:40:15 +02:00
Rangi d73a05aa78 Fix backtraces for REPT nodes with quiet non-REPT ancestors (#2196) 2026-10-03 22:16:58 +02:00
Rangi 9095c5b320 Do not reuse Patch struct for outputting assertions (#2184) 2026-10-03 19:24:49 +02:00
Rangi 6c76606839 Remove unused RawTiles class
This has been left over ever since its introduction in 8c62e80c1 (#981).
2026-10-02 10:28:27 -04:00
Rangi c0a16e4bad Move File-opening of PNG image into Png constructor (#2161)
Also rename `File::c_str` to `display_name` for clarity
2026-09-27 13:19:05 -04:00
Rangi42 f37e289fde Fix off-by-one error in transparent palette count 2026-09-27 09:07:33 -04:00
Rangi fcd4031034 Combine sect_BinaryFile and sect_BinaryFileSlice (#2169) 2026-09-26 22:05:39 -04:00
Rangi 321ad497fa Fix RGBFIX --output with a pre-existing or same-as-input file (#2145) 2026-09-27 01:31:53 +02:00
RangiandISSOtm b732f59472 Improve handling of platform-specific build options (#2162)
* Opt into POSIX functions with `_POSIX_C_SOURCE` before any system
  headers are `#include`d
* Allow user to override `make develop`'s `WARNFLAGS` and `CXXFLAGS`
* Use `make develop` on 32-bit Cygwin with sanitizers disabled
* Do not unnecessarily redefine `fseek` and `ftell`
* Disable a false-positive `-Wno-null-dereference` on 32-bit Cygwin

Co-authored-by: ISSOtm <[email protected]>
2026-09-26 16:31:25 -04:00
Rangi fd7715dcb0 Fix inconsistent handling of transparent colors in the middle of palettes 2026-09-26 19:20:18 +02:00
Rangi e6e5133bcf RGBGFX honors -c/--colors in -r/--reverse mode (#2144)
f4463b17 (#1254) was an incomplete fix for #1166
2026-09-26 19:13:34 +02:00
Rangi 03c71ced93 Fix FLOATING in linker scripts conflicting with section constraints (#2148) 2026-09-26 07:54:48 -04:00
Rangi 646cb05faf Fix BANK(@) discrepancy within LOAD blocks (#2149)
RGBASM and RGBLINK disagreed on which section to consider `@` to be in;
we have decided that the latter is correct.

The `assume` change is known to be suspicious; for more discussion, see
https://github.com/gbdev/rgbds/pull/2149#discussion_r4106902830
2026-09-25 19:18:52 +02:00
Rangi eca8c1f3e7 Do not deprecate implicit transparent color 0 2026-09-22 12:03:33 -04:00
Eldred Habert 63a06f02c3 Fix constrained empty sections escaping their regions (#2137) 2026-09-21 15:54:10 -04:00
Rangi ace8db1b73 Fix RGBGFX -t/--trim-end combined with -B/--background-color for non-deduplicated output (#2130) 2026-09-20 15:05:06 -04:00
Rangi 408e861ecc Correct warning message about different RGBGFX -L and -r widths 2026-09-20 14:32:04 -04:00
ISSOtm e1ed304e6b Apply the same enabling logic for UBSan as for ASan 2026-09-19 23:30:35 +02:00
ISSOtm bb0fc08980 Enable our default ASan flags on macOS
Except for the leak detection (Clang complains it's not supported there),
for which I have corrected the comment and its location.
2026-09-19 23:30:35 +02:00
Rangi b7f6c641f1 Define more debug options for make develop (#2124) 2026-09-19 22:07:10 +02:00
Rangi c151450d9f Fix underflowing SECTION FRAGMENT fixed address in RGBLINK (#2122) 2026-09-18 22:00:27 -04:00
Rangi 76bba2809b Fix incorrect EMPTY space in .map files (#2119) 2026-09-19 03:17:15 +02:00
Eldred Habert 44328d522b Avoid using snprintf where GCC can check sizes (#2121)
Interestingly, Clang's `-Wformat-overflow` seems to only warn
about buffers so small they will *always* underflow,
whereas GCC tries to infer whether they can overflow *at all*.
(I'm guessing they lean towards false-negatives and false-positives resp.)

Anyway, calling `sprintf` here remains safe, since GCC (notably, via CI)
checks our work, and it simplifies the code ever so slightly while
also providing an (admittedly negligible) performance improvement.

Note that there may be more locations where we could use `sprintf`,
but a cursory glance at our other uses of `snprintf` didn't seem fruitful.
(One way to test is to set the target buffer size to 1, and see if a
warning pops up for such a trivially-wrong size. If not, you can be certain
the compiler won't be able to help you with a more realistic size.)
2026-09-19 03:15:07 +02:00
Rangi 142eae4558 Fix repeated "And N more" in overlay error message (#2118) 2026-09-18 17:35:30 -04:00
Rangi 89403ca23b Fix RGBGFX to support -C/--color-curve in -r/--reverse mode (#2110)
Clarifies error message when RGBGFX reverse mode has `-c` mismatching `-p`.
2026-09-18 17:35:22 -04:00
ISSOtm aab22852d3 Factor out free space manipulation 2026-09-18 21:19:37 +02:00
ISSOtm 0e41ace1e3 Refactor section placement algorithm
Massively simplified, by not trying to shoehorn differently-behaving
cases into the same one loop!

This may have introduced bugs (the test suite has caught three different
oversights, two of which via external projects, in fact!), but this also
makes the logic clearer and more streamlined, so that we are also less
likely to have any latent or future bugs.

In particular, previously, the first iteration of the loop could attempt
placement at an address not matching the section's constraints,
which made advancing to the next target address unnecessarily complicated
(https://github.com/gbdev/rgbds/pull/2064#discussion_r3985466596),
among other weirdness. The code ended up being defensive, and thus the
overall logic was murky.

I'm also expecting that this should provide a performance improvement due
to being essentially a form of loop-invariant code motion (and very likely
one a compiler couldn't have performed automatically), though I haven't
measured.
2026-09-18 21:19:37 +02:00
ISSOtm ce5c8ce8ac Respect zero-size sections' align offsets
Fixes #2095

This code pre-dates the introduction of alignment offsets,
so it was never updated to take it into account.

That said, some of the code that I have factored out is a little awkward because
it will be used in an upcoming refactoring of the whole assignment loop,
since that turns out to be more byzantine than it ought to be due to
trying to handle a too disparate set of cases in a single loop.
2026-09-18 21:19:37 +02:00
ISSOtm c49828f7c9 Fix memory region alignment checks
Turns out the checks could fail spuriously if the starting alignment is non-zero,
e.g. `align 8,$C0` in HRAM. (Fixes #2113.)
This was tripped up by a test added to exercise the linker's section placement algorithm
on a similar aligment-related issue. :D

Note that this may be caused by the code in question pre-dating alignment offsets,
and we never checked. That said, these checks only work due to the regions' starting
addresses being no more aligned than their size (..if that makes sense?), so allowing
custom memory regions (#524) could violate that assumption and make those checks incorrect.
2026-09-18 21:19:37 +02:00