greg-king5 [Sat, 26 Nov 2016 22:30:38 +0000 (17:30 -0500)]
Merge pull request #358 from EtchedPixels/master
* scanner: Correct handling of \0101
The C language has this oddity that octal constants are 3 bytes so the sequence
"\0101" is two bytes and well defined by the langage. cc65 currently misparses
this as a 1 byte octal code. Add a count to fix this.
Signed-off-by: Alan Cox <etchedpixels@gmail.com>
* cc65: remove un-needed logic from octal parsing
We no longer need the extra error handling logic for octal parsing so simplify
it as requested by Greg King.
Alan Cox [Sat, 19 Nov 2016 13:02:19 +0000 (13:02 +0000)]
scanner: Correct handling of \0101
The C language has this oddity that octal constants are 3 bytes so the sequence
"\0101" is two bytes and well defined by the langage. cc65 currently misparses
this as a 1 byte octal code. Add a count to fix this.
Sven Oliver Moll [Tue, 13 Sep 2016 09:54:56 +0000 (11:54 +0200)]
removed 'make clean' invoked from test/Makefile for test/assembler/Makefile, as all artifacts are now created in testwrk and will be cleaned up out of directory
Stefan Dorndorf, author of XDOS, pointed out that retrieving the
default device by looking at an undocumented memory location won't
work in future XDOS versions.
He also showed a way to get the default device in a compatible
manner.
This change implements his method and adds a version check (XDOS
versions below 2.4 don't support this -- for them the behaviour
will be the same as, for example, AtariDOS: no notion of a default
drive).
Brad Smith [Tue, 12 Jul 2016 00:48:47 +0000 (20:48 -0400)]
Fix broken rand() implementation. The high 8 bits were unused, reducing it to a 24-bit implementation (while still doing all the work for a 32-bit one). The best entropy is in the unused high byte, returning these bits in A instead of bits 8-15, which had considerably lower entropy (i.e. rand() & 255 was effectively a 16-bit LCG).
Greg King [Sun, 3 Jul 2016 11:07:09 +0000 (07:07 -0400)]
Fixed how ca65 handles some debug info from cc65.
ca65 used to claim that an assembler error/warning was found on a C code line; and, that an Assembly line is only indirectly related to it.
Now, ca65 says that the Assembly line has the problem; and, that the Assembly line was produced from the C line.
Greg King [Sun, 19 Jun 2016 21:38:37 +0000 (17:38 -0400)]
Made the overlay demo programs compile for CBM targets again.
<conio.h> includes target-specific headers; so, we didn't bother to include <cbm.h> where it is needed. But, '#include <conio.h>' was removed from some files; so now, we must include <cbm.h> explicitly.
Oliver Schmidt [Sun, 19 Jun 2016 16:55:00 +0000 (18:55 +0200)]
Improved doc and samples default target.
The 'all' target deliberately doesn't build the doc nor the samples. But that doesn't mean that the Makefiles in the 'doc' and 'samples' directories must default to the (empty) 'all' target.
Oliver Schmidt [Sun, 19 Jun 2016 13:03:20 +0000 (15:03 +0200)]
Added CONIO cursor support.
For quite some time I deliberately didn't add cursor support to the Apple II CONIO imöplementation. I consider it inappropriate to increase the size of cgetc() unduly for a rather seldom used feature.
There's no hardware cursor on the Apple II so displaying a cursor during keyboard input means reading the character stored at the cursor location, writing the cursor character, reading the keyboard and finally writing back the character read initially.
The naive approach is to reuse the part of cputc() that determines the memory location of the character at the cursor position in order to read the character stored there. However that means to add at least one additional JSR / RTS pair to cputc() adding 4 bytes and 12 cycles :-( Apart from that this approach means still a "too" large cgetc().
The approach implemented instead is to include all functionality required by cgetc() into cputc() - which is to read the current character before writing a new one. This may seem surprising at first glance but an LDA(),Y / TAX sequence adds only 3 bytes and 7 cycles so it cheaper than the JSR / RTS pair and allows to brings down the code increase in cgetc() down to a reasonable value.
However so far the internal cputc() code in question saved the X register. Now it uses the X register to return the old character present before writing the new character for cgetc(). This requires some rather small adjustments in other functions using that internal cputc() code.
The final part of exec() called 'excexit' and only then restored the
stack pointer to its value at program entry. 'excexit' does all
cleanup (the same as '_exit()'), which means that on the atarixl
target the ROM is banked in again. On big programs the 'SP_save'
variable might reside at a high memory address which is no longer
accessible after the ROM has been banked in.
The change just moves the restoration of the stack pointer before
the call to 'excexit'.
Another change lets exec.s compile if UCASE_FILENAME is not defined.
And some other small cleanups, also in open.s.
- Adds new ENOEXEC error code, also used by Apple2 targets.
- Maximum command line length is 40, incl. program name. This is
an XDOS restriction.
- testcode/lib/tinyshell.c has been extended to be able to run
programs.
projects / cc65 / log