d0p1/ack
1
0
Fork 0
Code Issues Pull requests Packages Projects Releases Wiki Activity
8251 commits 2 branches 0 tags 43 MiB
Commit graph

332 commits

Author SHA1 Message Date
David Given 97e6d8f66a Update README. 2019-06-16 20:10:13 +02:00
David Given 2b013c34dc Some byte shaving; lseek returns the offset. 2019-06-16 20:04:07 +02:00
David Given 50dca8b954 First at-least-slightly working version of the CP/M read/write stuff. Not as 
bad as I expected, but far too big.
 2019-06-16 19:04:17 +02:00
David Given 3131dc9915 Partially working port of stdio to CP/M. I'm not sure this will work; it's 
getting way too complicated (stdio is horribly subtle). I think I need to
rethink things.
 2019-06-15 22:22:01 +02:00
David Given 9109d7af7f First stage in modularising FILE*. Refactor so that printf/scanf don't rely on 
FILE* innards; allow plats to replace the entire emulated FILE* system.
 2019-06-15 13:07:10 +02:00
David Given 784fc67596 Alignment issues mean we can't access 16-bit CP/M structure elements directly. 2019-06-11 22:33:25 +02:00
David Given 0607529df2 Miscellaneous bugfixed and renamings. 2019-06-11 21:33:57 +02:00
David Given a804375560 Miscellaneous byte shaving; you can now choose whether or not you want the CCP 
overwritten or not, and cpm_exit() does the right thing.
 2019-06-11 20:32:00 +02:00
David Given 3feb79ad0c Simplify the .trp API to make it a bit smaller. 2019-06-11 20:02:03 +02:00
David Given 4e90de00e9 Don't pull in all the trap handlers every time. 2019-06-11 19:47:42 +02:00
David Given f58d7e7d30 Add missing file. 2019-06-11 19:40:46 +02:00
David Given 48a9c48f56 Generate traps procedurally. For some reason they always keep getting pulled 
in, though...
 2019-06-11 00:12:00 +02:00
David Given ec2ea1feff Add missing file. 2019-06-10 23:57:43 +02:00
David Given d0967e683b Extend the CP/M libsys with a full set of (hopefully correct) 2.2 BDOS calls. 2019-06-10 23:54:23 +02:00
David Given 3f938d651b EM requires 2-alignment in structures, sadly. 2019-06-10 18:33:04 +02:00
George Koehler bec236c108 Include more headers to declare functions. 
This causes clang to give fewer warnings of implicit declarations of
functions.

In mach/pdp/cv/cv.c, rename wr_int2() to cv_int2() because it
conflicts with wr_int2() in <object.h>.

In util/ack, rename F_OK to F_TRANSFORM because it conflicts with F_OK
for access() in <unistd.h>.
 2019-03-22 15:59:35 -04:00
George Koehler 84f65f7ce3 Raise _XOPEN_SOURCE from 500 to 700 
In OpenBSD, _XOPEN_SOURCE 500 forces _POSIX_C_SOURCE down to 199506,
which hides the declarations of openat() and fdopendir() and causes
compiler warnings.

Don't set _POSIX_C_SOURCE, because _XOPEN_SOURCE may set
_POSIX_C_SOURCE to a different value.
 2019-03-22 14:35:07 -04:00
David Given 79a38ecc08 Instead of using parameterised rsts for stack access, add a huge swathe of 
automatically built helper tools. Star Trek goes up from 40243 to 40779 bytes,
but should be a lot faster.
 2019-02-13 22:45:22 +01:00
David Given 073451dd44
Merge pull request #154 from ccodere/buildfix 
Bugfix of compilation on gcc 4.8
 2019-02-12 21:51:46 +01:00
carl 664338aa49 Fix compilation issue on C90 compilers (gcc 4.8 was failing to compile with default flags) - updated following comment. 2019-02-13 00:17:45 +08:00
carl c63f527dde Fix compilation issue on C90 compilers (gcc 4.8 was failing to compile with default flags). 2019-02-12 00:46:09 +08:00
David Given 34e9573929 Merge from default. 2019-02-10 23:40:07 +01:00
David Given 666d9a3edb Add stray header. 2019-02-10 12:57:00 +01:00
David Given aa6aefdaa0 Add missing header. 2019-02-10 00:45:09 +01:00
David Given 810afbb5b9 Don't run the floating point tests on i80. 2019-02-09 14:24:31 +01:00
David Given bcc9c16bf7 More z80ex removal. 2019-02-09 13:42:09 +01:00
David Given 343cdf7983 Remove stray z80ex reference. 2019-02-09 13:30:29 +01:00
David Given 19f5ee3399 CP/M sbrk now handles overflow correctly. 2019-02-08 21:43:52 +01:00
David Given 22032575c5 Add a 'disassemble' command to the CP/M emulator. 2019-02-08 20:54:44 +01:00
David Given b47d59aa4a Merge from default. 2019-02-07 23:45:47 +01:00
David Given 8876ddcad2 Build the CP/M tests (which got turned off accidentally). Add a disassembler to 
the emulator debugger.
 2019-02-07 23:36:45 +01:00
David Given 7473601172 Add the cpm emulator from Cowgol, modified to work with an 8080 emulator rather 
than libz80ex; enable tests for cpm.
 2019-02-07 22:39:57 +01:00
David Given 5ac280026e Experimentally use rst 1 to optimise frame variable loads. Reduces Star Trek 
from 43892 to 43205 bytes.
 2019-02-07 00:06:32 +01:00
David Given 365c89c9e7 Updated READMEs with the MIPS platform. 2018-09-18 23:49:31 +02:00
David Given 9333b323a5 Make the MIPS boot code produce correct code. 2018-09-18 23:39:04 +02:00
David Given ec0891469f Array descriptor ranges are inclusive; adjust aar and the tests to check for 
this. All the tests except rotate build now!
 2018-09-17 19:34:38 +01:00
David Given b81e060372 Add huge numbers of failing tests. 2018-09-10 23:40:03 +02:00
David Given 776e93221f Thinko fix. 2018-09-09 18:57:44 +02:00
David Given c769fbec93 Use the correct ELF processor flags; align properly. 2018-09-09 18:57:35 +02:00
David Given f3bc9ef92a Call proper Linux system calls proper-like. 2018-09-09 18:57:12 +02:00
David Given ec46643124 Allow the ELF processor flags to be set. Generate MIPS ELF executables, not 
PowerPC ones.
 2018-09-09 14:29:34 +02:00
David Given 26fe3f7530 Added library skeletons. 2018-09-05 00:07:07 +02:00
David Given 98afb42095 Added a boot.o based on the PowerPC one. 2018-09-05 00:00:13 +02:00
David Given 1d6ecddcf4 The MIPS backend is still full of holes, and cut-and-pasted PowerPC code, but 
is beginning to look like an actual code generator.
 2018-09-02 18:57:25 +02:00
David Given 4741ed8e14 Add a completely non-tested table-based MIPS assembler. 2018-09-01 19:35:31 +02:00
David Given ab7002c0ee All the Linux include directories are the same, so common them all out into a 
single one, like OSX.
 2018-06-24 19:57:43 +02:00
David Given f45b357740 Remove the dependency on the glibc-only byteswap.h. 2018-06-24 00:01:06 +02:00
David Given 1fc20e90c0 Remove c99-isms. 2018-06-23 23:56:34 +02:00
David Given 64ae1d7583 Move time() into sys; change the name of the variable which controls it. 2018-06-23 19:40:50 +02:00
David Given 538eefb573 Get em22 working. Remove the leading-underscore system calls from its libsys. 2018-06-23 15:57:57 +02:00
David Given 74d166ef9a Get the pdp7 plat working; remove the leading-underscore syscalls. 2018-06-23 15:18:26 +02:00
David Given 3e40374297 Made the rpi platform work with the new libc layout. 2018-06-23 13:05:27 +02:00
David Given dbb2062d03 Made pc86 work with the new libc. 2018-06-23 12:59:40 +02:00
David Given 680b4071b7 Made the two OSX platforms work with the new libc layout. 2018-06-23 12:55:59 +02:00
David Given b727b7affd Make linuxppc work. 2018-06-23 12:41:54 +02:00
David Given b267b56ad9 Made linux68k work; commoned up all the generic linux includes. 2018-06-23 12:35:17 +02:00
David Given 66815ff987 Make linux386 work with the new libc layout. 2018-06-23 12:13:33 +02:00
David Given 32c881474e Create a basic and probably wrong common unistd.h, which replaces the plat one. 
Made this work with cpm (but nothing else yet).
 2018-06-23 11:46:55 +02:00
David Given 94ffa3ba88 libc floating point and stdio floating point can now be turned on and off using 
ACKCONF variables.
 2018-06-23 00:04:14 +02:00
David Given 9bf1d548fc Rearrange the libc config files, ready for adding new options. 2018-06-22 23:43:15 +02:00
David Given 60b7d8de6e
Merge pull request #96 from kernigh/kernigh-emu-1 
Fixes to build emulators, ass, int on OpenBSD
 2018-06-20 22:30:01 +02:00
David Given 698613cd7d Turns out that the compiler needs access to the libpc headers too; refactor 
into a separate target to avoid build loops and make it work.
 2018-06-18 22:17:38 +02:00
David Given 11854f4dfe Ansify and warning fix. 2018-06-17 22:30:27 +02:00
David Given 13e195d19f Warning fixes. 2018-06-17 15:42:26 +02:00
David Given ddca7276c0
Merge pull request #102 from davidgiven/dtrg-powerpc 
Add a PowerPC simulator for running the tests.
 2018-06-17 11:45:26 +02:00
David Given 89e8956bb2 Fix edge case in rlwnm when the mask is 32 bits wide. All tests now pass. 2018-06-17 11:00:01 +02:00
David Given 7ec9f54679 Turns out that andi and andis only have . forms. Fixed; another test passes. 2018-06-17 10:43:39 +02:00
David Given 78eaf836be Turns out I was returning values from syscalls in the wrong register; fixed. 
More tests pass.
 2018-06-17 10:22:20 +02:00
David Given ab660a44e9 Lots of floating point, bugfixes, and system calls. Most of the tests pass now. 2018-06-17 09:24:01 +02:00
David Given 67efbb5f7f Do a bit of floating point stuff; added brk(); fixed a horrible bug where 
stores with 16-bit displacements were storing the register number and not the
contents of the register.
 2018-06-16 22:55:23 +02:00
David Given 5d7cdd2c67 Fix a setcr0 bug which was trashing the condition register; some system 
calls are now implemented. A few tests more-or-less pass (but crash on
exit).
 2018-06-16 08:35:36 +02:00
David Given 39a49265ce
Merge pull request #100 from kernigh/kernigh-pdp 
Two fixes for ack -mpdpv7
 2018-06-15 15:40:24 +09:00
George Koehler 5db312a1c0 Prevent crash in isatty() 
The existing code allocated 2 bytes (char*), but gtty() needs 6 bytes
(struct sgttyb), so isatty() smashed the stack and corrupted its
return address, probably causing SIGBUS or SIGSEGV.

Fix by switching to TIOCGETD, which needs 2 bytes.  TIOCGETD isn't in
the manual for tty(4), but does appear in
https://minnie.tuhs.org//cgi-bin/utree.pl?file=V7/usr/sys/dev/tty.c

This fixes hilo_c.pdpv7 and hilo_mod.pdpv7 in simh-pdp11.
 2018-06-15 00:48:29 -04:00
George Koehler 5de5611c24 Use 2-byte alignment in pdpv7 to prevent SIGBUS. 
Change the alignment in C structs (wa, pa, sa, and so on) from 1 to 2
bytes.  This prevents the SIGBUS when PDP-11 Unix V7 catches the
misalignment.  This fixes hilo_p.pdpv7 in simh-pdp11.

Change ALIGN to document that sections have 2-byte alignment.  This
change should have no effect, because the sections only contain 2-byte
values.
 2018-06-15 00:17:12 -04:00
David Given 5111556d14 Got the emulator to the point where we're hitting the first major emulation bug 
(the calloc test hangs).
 2018-06-14 05:53:39 -07:00
David Given 853af4c4dd Some code runs (up to needing setcr0). 2018-06-14 04:41:21 -07:00
David Given b71145f79d Implement a non-functional skeleton of the PowerPC emulator (with, hopefully, 
most of the instruction decode done).
 2018-06-14 03:07:36 -07:00
David Given 31c2ba3f94 File cleanup. 2018-06-13 21:47:08 +09:00
David Given 7eaa235fd1 Bodge the ancient em monitor library into building as a libsys for the em22 
plat. It's completely untested, but it builds and makes e.out executables.
 2018-06-13 21:44:25 +09:00
David Given 15161ebb35 Remember to always link the libsys and libend. 2018-06-13 20:57:31 +09:00
David Given 6ae38887a7 It appears that the parameter to lol technically has to be word aligned; having 
a non-word aligned parameter is illegal (but most of the toolchain accepts it).
So, word align data structures for em22.
 2018-06-12 20:54:15 +09:00
George Koehler 6f9e418f94 Remove extra #include <stdio.h> to fix uint 
From my request at https://github.com/kstenerud/Musashi/pull/31

There was a conflict between Musahi's uint and OpenBSD's uint:

    $ cc -o m68kmake m68kmake.c
    $ ./m68kmake
    $ cc -DM68K_COMPILE_FOR_MAME=0 -c m68kcpu.c
    In file included from m68kcpu.c:565:
    In file included from /usr/include/stdio.h:46:
    /usr/include/sys/types.h:58:22: error: cannot combine with previous 'int'
          declaration specifier
    typedef unsigned int    uint;           /* Sys V compatibility */
                            ^
    ./m68kcpu.h:75:25: note: expanded from macro 'uint'
    #define uint   unsigned int
                            ^
    1 error generated.
 2018-06-10 16:24:35 -04:00
George Koehler 5404dace3a Stop using <X11/Xfuncproto.h> from x86emu. 
The code included <X11/Xfuncproto.h> only for _X_ATTRIBUTE_PRINTF,
which tells some compilers to check the printf format string.  Remove
so ACK doesn't need any X11 headers.

Some machines don't have the X11 headers, or they need a flag like
-I/usr/X11R[67]/include to find them.
 2018-06-10 16:00:23 -04:00
David Given d623440c77 Add the core of a simple em22 platform. Unfortunately it doesn't work; the old 
em libmon vanished decades ago (or never existed), and also ass appears to have
a different idea of what the em opcodes are to everything else and gets
confused.
 2018-06-10 20:25:48 +09:00
David Given 19cd42124f Rearrange the tests into sets; allow plats to skip sets; pc86 and linux68k now 
skip the tests which use floating point instructions (as their emulators don't
support them).
 2018-06-08 15:59:04 +09:00
David Given 1d4ddb939d Merge from trunk. 2018-06-08 00:05:50 +09:00
David Given b9c0e1d964 Dummied out some more system calls --- more tests pass. 2018-06-07 13:12:15 +09:00
David Given 2682c125e0 More real ELF loading and brk handling; more tests pass. 2018-06-07 13:03:36 +09:00
David Given 396795105f Beat the 68k simulator into enough shape that at least *some* of the tests pass. 2018-06-06 22:39:43 +09:00
David Given 4b26f75d23 Hang on, the 68k is big-endian --- I'm extracting words (and bytes) from longs 
incorrectly.
 2018-06-06 12:49:21 +09:00
David Given 2cf39be752 Get the emulator to the point where it'll start to run code; it then fails, 
because the version of Musashi I imported doesn't support floating point...
 2018-06-05 23:17:22 +09:00
David Given 5034ed1c39 Crudely beat the standard Masushi simulator into working in the ACK framework. 2018-06-05 17:45:00 +09:00
David Given c6b9446722 Check in the x86emu-based emulator. Sadly, just like the 8086tiny-based 
emulator, FPU instructions aren't supported.
 2018-06-04 16:19:45 +09:00
David Given 0eb89d5d95 Add a clean copy of the Musashi emulator. 2018-06-05 17:04:47 +09:00
David Given 9132641100 Add the pc86emu skeleton. 2018-06-03 08:24:21 -07:00
David Given eb96ea5d28 Add a clean copy of the x86emu library. 2018-06-03 08:06:23 -07:00
George Koehler 019c7f9da7 Update cpm/README for console changes, ack -fp. 2018-05-07 16:08:44 -04:00
George Koehler 154509038d Enable the line editor in read() for CP/M. 
This changes the BDOS call from CPM_BDOS_CONSOLE_INPUT to
CPN_BDOS_READ_CONSOLE_BUFFER.  This allows commands like ^H to delete
characters and ^C to exit to CCP.  This is more like how Unix read(2)
uses canonical mode of termios to read a line.

This change has a disadvantage: the user buffer to read(2) must now be
large enough for an entire line.  This is because CP/M, unlike Unix,
lacks a kernel buffer to hold the rest of the line.  If you use a
buffered input library like stdio to call read(2), then it works; but
if you try to read part of a line or a single character, then it
doesn't work.
 2018-05-04 18:21:01 -04:00
George Koehler 4a3b7be795 Use ack -mcpm -fp for mandelbrot.c, startrek.c 
Add a variable %{ackldflags} so I can pass `-fp`.  This change seems
to cause the build to relink every ackprogram, because the link now
needs to use %{ackldflags} even if the flags are empty.

mandelbrot_c_cpm runs in YAZE-AG; startrek_c_cpm doesn't run because
it doesn't fit in the 16-bit address space.
 2018-04-28 13:18:27 -04:00
George Koehler 93e01eb5d1 Teach i80 ncg to use libfp. Enable ack -mcpm -fp. 
Old .o files stop working if they use floating point.  One must
recompile those files.  Old files don't call libfp in the correct way,
and may use symbols that I removed from libem.  I don't keep old
symbols in libem/flp.s, because a program that pulls both libfp and
flp.s would get "multiply defined" errors in the linker.

I teach mach/i80/ncg/table to use libfp by copying or adapting the
patterns from mach/i86/ncg/table.  I did not test all the patterns,
but I did use `ack -mcpm -fp -O4` to compile examples/mandelbrot.c,
then I ran it in the emulator YAZE-AG.  It worked, but it was slow.
 2018-04-25 16:09:56 -04:00