Commit Graph

1188 Commits

Author SHA1 Message Date
ISSOtm
2b0c34ecb5 Fix a few code style errors 2020-04-08 00:44:41 +02:00
ISSOtm
b0ec8468e6 Allow specifying offset in addition to alignment 2020-04-08 00:40:41 +02:00
ISSOtm
e82ad21704 Use a single byte for alignment 2020-04-07 21:19:09 +02:00
ISSOtm
e098bf47ba Allow references to be overridden by constant symbols
RGBLINK is capable of handling it now.
Though it'd be ideal for RGBASM to directly catch it.

Fixes #496.
2020-04-07 20:57:20 +02:00
ISSOtm
9f82fa4cf7 Fix BANK(@) outside sections causing crashes 2020-04-07 15:51:17 +02:00
ISSOtm
927c65e863 Fix incorrect PC in LOAD blocks at link time 2020-04-07 14:44:51 +02:00
ISSOtm
5b6c1569a4 Make failure to open file a fatal error 2020-04-07 11:36:44 +02:00
ISSOtm
bdad1499fe Merge branch 'release' 2020-04-03 11:58:53 +02:00
Eldred Habert
1f5ca39559 Merge pull request #494 from ISSOtm/docs
Overhaul man pages
2020-04-03 11:56:41 +02:00
ISSOtm
bcfeb49d6b Allow labels to be passed to DEF 2020-04-02 16:54:41 +02:00
ISSOtm
a3d8836671 Prevent assertions outside sections from crashing 2020-03-29 12:18:24 +02:00
ISSOtm
db1eb8fbcb Revert "Prevent RGBASM from outputting corrupted files"
This reverts commit 06fe27c516.

According to Microsoft's documentation
https://docs.microsoft.com/en-us/cpp/c-runtime-library/reference/tmpfile?view=vs-2019
`tmpfile` attempts to create the temporary file at the **root** folder
This seems to assume that the user has admin rights; might be a compat
thing, idk, but it breaks on people's computers.
(CI didn't catch it, annoyingly.)

Reverting to make RGBASM usable on most Windows computers.
(Sanely-configured ones, at least.)
Another solution to #446 needs to be figured out, yay...
2020-03-27 23:36:11 +01:00
ISSOtm
6579120d9e Simplify symbol-writing logic 2020-03-27 12:32:14 +01:00
ISSOtm
84cd9f2db9 Fix segfaults when using PC outside a section 2020-03-27 12:30:09 +01:00
ISSOtm
06fe27c516 Prevent RGBASM from outputting corrupted files
Properly fixes #451
2020-03-27 11:19:02 +01:00
ISSOtm
5039af6af1 Document new "unionized section" feature 2020-03-26 23:50:30 +01:00
ISSOtm
9113647d41 Fix incorrect macro used for x-ref 2020-03-26 23:11:01 +01:00
ISSOtm
713eef9ef2 Update mandoc style sheet 2020-03-26 23:11:01 +01:00
ISSOtm
7635a2fc74 Reword and give an example of symbol interpolation 2020-03-26 23:11:01 +01:00
ISSOtm
7955447ca4 Overhaul man pages 2020-03-26 23:11:01 +01:00
ISSOtm
187f88aa50 Only copy necessary characters for bare labels 2020-03-26 23:10:51 +01:00
ISSOtm
caddd61f17 Remove "column 1" restriction for local labels
Because a local symbol is necessarily a label, the disambiguation
is not necessary. This is a first step towards fixing #457.
2020-03-26 22:36:40 +01:00
ISSOtm
29629245a4 Use new hashmap implementation for symbols 2020-03-24 12:50:53 +01:00
ISSOtm
7deb8d3e8a Allow purging exported symbols 2020-03-24 10:52:02 +01:00
ISSOtm
cb0a882a31 Order warnings alphabetically 2020-03-23 16:03:36 +01:00
ISSOtm
df2c0dc2f9 Check for unsatisfiable alignment constraints
Fixes #493
2020-03-22 11:54:57 +01:00
ISSOtm
062fa5392b Still override alignment even if not satisfied 2020-03-22 11:52:31 +01:00
ISSOtm
4877bb783c Add more tests for unionized sections + fix bugs
Implementing those tests found a few bugs... oops
2020-03-22 11:14:04 +01:00
ISSOtm
e123b6dec7 Implement unionized sections in RGBLINK 2020-03-22 11:13:39 +01:00
ISSOtm
cb52ae0f26 Implement unionized sections in RGBASM
This touched a lot more code than initially expected, for two reasons.

First, this broke a big RGBASM assumption: that sections are always being
written to at their end. This plus other problems required touching
basically the entirety of `section.c`.

Second, I tried different solutions to solve the above problem, and along
the way I cleaned up many things around. (I believe that keeping this to
"cleanup" commits yields subpar results, and since it's boring they get
postponed anyways.)

RGBLINK support still needs to be added, but this will come next.
2020-03-22 11:06:17 +01:00
ISSOtm
46a402f7d7 Prevent passing assertions that RGBASM passed to RGBLINK 2020-03-22 10:46:37 +01:00
Eldred Habert
fe824e0068 Merge pull request #490 from ISSOtm/const
Implement `ISCONST`, reporting compile-time constness
2020-03-21 23:20:01 +01:00
ISSOtm
fb58166e5d Add assertions
Closes #292
2020-03-21 23:00:38 +01:00
ISSOtm
8191e5eb27 Define version symbols
Major blunder. That warrants a new release on its own...
2020-03-21 19:26:40 +01:00
ISSOtm
03967bd623 Prevent purging referenced symbols
This is an immediate fix for #492, although #342 is needed to implement the
desired functionality.
2020-03-21 15:42:52 +01:00
ISSOtm
af22cf182b Simplify out_PCRelByte by using PC 2020-03-20 20:22:19 +01:00
ISSOtm
5a0fcda4c8 Prevent POPS within LOAD blocks 2020-03-20 18:57:37 +01:00
ISSOtm
deb91f679d Reset symbol scope on every section change 2020-03-20 17:59:47 +01:00
ISSOtm
fb5e768142 Prevent using LOAD blocks outide code sections 2020-03-20 16:37:10 +01:00
ISSOtm
ecf44c784c Reject including directories 2020-03-20 01:24:53 +01:00
ISSOtm
5bca1172ff Fix section continuation when only one bank exists 2020-03-19 23:07:07 +01:00
ISSOtm
136446fccb Improve checking of RST and LDH values at assembly time 2020-03-15 15:23:39 +01:00
ISSOtm
7ddbe44b21 Use hashmap system for charmap 2020-03-15 15:23:32 +01:00
ISSOtm
af60e7f74a Allow charmap creation even if its base doesn't exist 2020-03-15 00:40:45 +01:00
ISSOtm
2f16e82cf7 Improve PC offset management
Basically make it always point to the instruction's first byte.
This was the behavior all whom I asked to intuitively expected.
2020-03-15 00:18:10 +01:00
ISSOtm
7b54312d97 Deprecate OPT z in favor of OPT p
Fixes #298
2020-03-14 17:09:15 +01:00
ISSOtm
5b98beec8b Use left recursion instead of right
> Any kind of sequence can be defined using either left recursion or right
> recursion, but you should always use left recursion, because it can
> parse a sequence of any number of elements with bounded stack space.
https://www.gnu.org/software/bison/manual/html_node/Recursion.html
2020-03-14 16:22:33 +01:00
ISSOtm
6662c86d5e Define 3-bit value when invalid 2020-03-14 16:17:30 +01:00
ISSOtm
2f466c2939 Revamp macro arg system
This should significantly improve performance: on pokecrystal builds, perf
reported as much CPU time spent on `yyparse` as on `sym_UseNewMacroArgs`
Measurements show ~6 seconds of improvement on that codebase.

This also fixes #321, as a bonus, due to saner management!
2020-03-14 16:13:40 +01:00
ISSOtm
a259f53b52 Rename macro functions with proper prefix 2020-03-13 23:20:27 +01:00