1
0
mirror of https://github.com/cc65/cc65.git synced 2024-12-23 19:29:37 +00:00
Commit Graph

9530 Commits

Author SHA1 Message Date
Greg King
c00c89c2b7 Directly clone the document repo. 2022-07-21 03:24:17 +02:00
Greg King
4d045a68e3 Put doc.git next to cc65. 2022-07-21 03:24:17 +02:00
Greg King
964c4a220e Find out where doc.git is. 2022-07-21 03:24:17 +02:00
Greg King
e004e92f21 Fixed the action/checkout steps.
I misunderstood that action's docs.
2022-07-21 03:24:17 +02:00
Greg King
2290e70e0c Experiment: checkout the second repo first.
Maybe that will give it the github.token that it needs.
2022-07-21 03:24:17 +02:00
Greg King
4c74d4b851 Trying a verbose Git push.
If I'm lucky, it will tell me more.
2022-07-21 03:24:17 +02:00
Greg King
cb20b2e968 Tried to authorize a push by setting the access token explicitly. 2022-07-21 03:24:17 +02:00
Greg King
3232ce1bad Fixed a shell syntax bug. Try again. 2022-07-21 03:24:17 +02:00
Greg King
735f0cda25 Added Github action steps that update the online documentation repo. 2022-07-21 03:24:16 +02:00
mrdudz
e8e1889b0b handle loadaddr the same as with the other CBM targets 2022-07-21 03:22:52 +02:00
rofl0r
9cb8bf9500 cl65: fix regression in --print-target-path
since the "empty prefix means run from current dir" hack was removed,
cl65 --print-target-path no longer prints the path relative to the
binary, but the hardcoded library path which points to prefix, because
the code adds the hardcoded path first to the pathsearch, but then actually
only returns the first entry rather than looking whether the path exists.

closes #1754
2022-07-21 03:22:52 +02:00
Stephan Mühlstrasser
8ec1f76b9c Fix switched meaning of '\n' and '\r'
When conio was implemented for the osic1p target, the cbm
target was used as a blueprint. But it was overlooked that the
cbm target encodes CR as 0x0A and LF as 0x0D, while the
osic1p target uses the ASCII encoding of CR and LF. Therefore
conio output of '\n' moved the active position to the start of the
line and '\r' moved the active position to the next line.

This change implements the correct semantics of '\n' and '\r'
in conio for the osic1p target.

Fixes #1747.
2022-07-21 03:22:52 +02:00
Greg King
f69b993ee6 Fixed a multi-line comment.
The left border is thicker in cc65 than it is in VICE.
2022-07-21 03:22:52 +02:00
mrdudz
85fb6ba944 add check for +x flag 2022-07-21 03:22:52 +02:00
mrdudz
8e787c0c0a remove +x flag 2022-07-21 03:22:52 +02:00
compyx
4595d08c6e Fix set but unused variable 2022-07-21 03:22:52 +02:00
mrdudz
3717f36c52 rename the actual checkstyle target too :) 2022-07-21 03:22:52 +02:00
mrdudz
0e4d3aa7c5 rename the "check" target to "checkstyle".
also created a new "check" target which runs all tests, as it is common with GNU programs
2022-07-21 03:22:52 +02:00
mrdudz
0113d1328b don't try to build the checkversion sample for targets that do not have printf 2022-07-21 03:22:52 +02:00
mrdudz
976d07b98f clean and mostlyclean should also clean the test and targettest directories 2022-07-21 03:22:52 +02:00
Bob Andrews
08bad5b93b Update README.md 2022-07-21 03:22:52 +02:00
polluks2
133d703964 smarter manufacturer
some kind of promotion
2022-07-21 03:22:52 +02:00
mrdudz
1334a10101 sample program that demonstrates how to compare the value of __CC65__ for
any version of the compiler
2022-07-21 03:22:52 +02:00
mrdudz
b9779927e7 document changes to __CC65__ 2022-07-21 03:22:52 +02:00
mrdudz
354dcd6101 define __CC65__ as ((VER_MAJOR * 0x100) + VER_MINOR), fixes issue #1667 2022-07-21 03:22:52 +02:00
mrdudz
15047d9754 simple test related to #1678 2022-07-21 03:22:52 +02:00
mrdudz
3943dc1216 Fix endless loop on expanding a recursive macro, fixed issue #1678, patch by kugelfuhr 2022-07-21 03:22:52 +02:00
mrdudz
b8d0bd7f95 Add a directory for tests that exercise the supported C standard differences.
As a first test added the snippet given in issue #1670
2022-07-21 03:22:52 +02:00
mrdudz
4579d584b1 fix some header files to conform with C99 hopefully, should fix issue #1670 2022-07-21 03:22:52 +02:00
mrdudz
db2dfc9226 must include stdarg.h to use varargs 2022-07-21 03:22:52 +02:00
Bob Andrews
738d6dba88 Update Contributing.md 2022-07-21 03:22:52 +02:00
Bob Andrews
cb5ced7c9d Update Contributing.md 2022-07-21 03:22:52 +02:00
mrdudz
4285a30236 At the end of the optimizer run, remove jump cascades again. fixes the
regression reported in #1652
2022-07-21 03:22:52 +02:00
mrdudz
8ec41083e9 add extra optimizer step that replaces BRA by JMP if the target is external 2022-07-21 03:22:52 +02:00
mrdudz
4ddbe5f838 test related to issue #1652, shows regression broken by #1231 and the
case that #1231 improved.
2022-07-21 03:22:52 +02:00
mrdudz
ab7079df00 add sourcelines to the generated asm for easier inspection,
redirect output of sim65, including cycle count, into a seperate .out file per program
2022-07-21 03:22:52 +02:00
Bob Andrews
bfe9ddeb6a Update Contributing.md 2022-07-21 03:22:52 +02:00
polluks2
f8a940bce4 Some clean-up
One exit() too much
file is not optional
2022-07-21 03:22:52 +02:00
polluks2
395e072c3b typo 2022-07-21 03:22:52 +02:00
Bob Andrews
d365b8a4ee Adjust comments 2022-07-21 03:22:52 +02:00
rofl0r
5867d3a020 change GetProgPath() to return full & resolved binary path
GetProgPath() now resolves the path derived from argv[0] always
via realpath(3p) to its real location in the filesystem, and returns
the path including binary name, effectively making it work like
windows' GetModuleFileName(), so we can re-use the existing code
to strip the trailing binary name.
since symlinks are now always resolved, we no longer need the
special case code for linux to use /proc/self/exe for this purpose.
2022-07-21 03:22:52 +02:00
Bob Andrews
546e82965d Try reading from /proc/self first on linux, this is needed to make the edgy "make avail" work 2022-07-21 03:22:52 +02:00
rofl0r
6f74a2e9c7 rename AddSubSearchPathFromWinBin to AddSubSearchPathFromBin 2022-07-21 03:22:52 +02:00
rofl0r
6a5375ace4 implement AddSubSearchPathFromWinBin() counterpart for unix
this looks up paths relative to the binary used to start the
specific application.
2022-07-21 03:22:52 +02:00
rofl0r
6eeaf7485c build: allow empty prefix
there was some concern that this will break windows' way of doing
file lookups relatively from the binary, rather than via hardcoded
locations, but so far each occurence adding e.g. "CA65_INC" to
the pathsearch is already shielded with an #ifndef _WIN32.

addressing #1726
2022-07-21 03:22:52 +02:00
jede
e9bf0dd64f fix space 2022-07-21 03:22:52 +02:00
jede
16d896b73f fix mkdir 2022-07-21 03:22:52 +02:00
jede
e1fbb8465b fix spaces 2022-07-21 03:22:52 +02:00
jede
e1b8508798 fix typo 2022-07-21 03:22:52 +02:00
jede
244da9c911 missing fd_management and cleaning 2022-07-21 03:22:52 +02:00