Brett Gordon [Thu, 25 May 2017 17:02:05 +0000 (13:02 -0400)]
exec(): take PROGLOAD from the binary header
Alan Cox [Sun, 21 May 2017 21:50:28 +0000 (22:50 +0100)]
ed: fix bugs and warnings
With this fixed it seems to be stable on M68K.
Alan Cox [Sun, 21 May 2017 21:20:18 +0000 (22:20 +0100)]
v68-banked: updated README
Bring a bit more into line with the code status
Alan Cox [Sun, 21 May 2017 21:18:05 +0000 (22:18 +0100)]
v68-banked: fix swap handling
This depends upon the blkdev fixes
Alan Cox [Sun, 21 May 2017 21:16:41 +0000 (22:16 +0100)]
blkdev: allow for 16bit block counts
On a 32bit system we can end up writing more than 256 blocks in a run when
we swap so blkdev needs to use 16bits for the block counter.
(We could make this a typedef for 8 v 16bit but it's not clear it matters
enough)
Alan Cox [Sun, 21 May 2017 20:35:16 +0000 (21:35 +0100)]
v68-banked: Get us to the point things work if we don't swap.
If we do swap it breaks. Not yet sure why.
Alan Cox [Sat, 20 May 2017 22:59:32 +0000 (23:59 +0100)]
v68-banked: Fix various silly things
This gets us to the point we hit fork. Our fork() path then fails for
reasons still under investigation.
Alan Cox [Sat, 20 May 2017 18:14:56 +0000 (19:14 +0100)]
v68: do some updating on the in progress banked memory model
Alan Cox [Sat, 20 May 2017 18:14:29 +0000 (19:14 +0100)]
v68: don't rewrite reset vectors or scribble off end
Alan Cox [Sat, 20 May 2017 18:12:54 +0000 (19:12 +0100)]
68000: introduce defines for the low byte of the page
We need this for some of the platforms so we can do a straight mov.b
from offset(a5) to the I/O register
Alan Cox [Sat, 20 May 2017 18:10:44 +0000 (19:10 +0100)]
v68: fix old tool reference in other builds
Alan Cox [Sat, 20 May 2017 16:37:23 +0000 (17:37 +0100)]
v68: more README updates
Alan Cox [Sat, 20 May 2017 15:58:26 +0000 (16:58 +0100)]
v68: update documentation on state
Alan Cox [Sat, 20 May 2017 15:56:48 +0000 (16:56 +0100)]
68000: fix time_t split definition
With this corrected things like "date" now work properly on M68K
Alan Cox [Sat, 20 May 2017 15:49:22 +0000 (16:49 +0100)]
qrun: fix a nasty memory scribble
Also clean up the graphics check
Alan Cox [Sat, 20 May 2017 15:45:04 +0000 (16:45 +0100)]
Makefile: adjust make clean
Make the default Makefile also clear out .o files
Alan Cox [Sat, 20 May 2017 15:44:06 +0000 (16:44 +0100)]
v68: remove last accidental use of old toolchain
EtchedPixels [Mon, 15 May 2017 11:51:58 +0000 (12:51 +0100)]
Merge pull request #502 from beretta42/dasm09
dasm09: a 6809/6309 disassembler
Brett Gordon [Mon, 8 May 2017 01:32:12 +0000 (21:32 -0400)]
dasm09: actually handle errors.
Brett Gordon [Mon, 8 May 2017 01:20:09 +0000 (21:20 -0400)]
Add test code
Brett Gordon [Sat, 6 May 2017 02:12:54 +0000 (22:12 -0400)]
coco3: build dasm09 disassembler
Brett Gordon [Sat, 6 May 2017 02:00:31 +0000 (22:00 -0400)]
dasm09: add package install, man page, mod docs
Brett Gordon [Sat, 6 May 2017 01:25:50 +0000 (21:25 -0400)]
dasm09: Use Clib's buffering to allow large input images.
Brett Gordon [Sat, 6 May 2017 00:40:21 +0000 (20:40 -0400)]
dasm09 built for fuzix. 16k buffer.
EtchedPixels [Wed, 26 Apr 2017 12:15:21 +0000 (13:15 +0100)]
Merge pull request #500 from fgeorgatos/patch-1
reformatting one-liner comments, for consistency
Fotis Georgatos [Wed, 26 Apr 2017 09:23:47 +0000 (10:23 +0100)]
reformatting one-liner comments, for consistency
Alan Cox [Sat, 22 Apr 2017 18:17:23 +0000 (19:17 +0100)]
init: #facepalm
Alan Cox [Sat, 22 Apr 2017 18:07:35 +0000 (19:07 +0100)]
trs80: correct devhd return value
Alan Cox [Sat, 22 Apr 2017 08:26:21 +0000 (09:26 +0100)]
devices: Fix an off by one error in the maximum device checks
If we have 1 major then the highest permitted device is 0,255 not 1, 255.
This only ever showed up if you either mknod a device once bigger than that
supported or attempt to boot a device one higher than the limit.
Alan Cox [Sat, 22 Apr 2017 00:11:40 +0000 (01:11 +0100)]
mini_ide: correct return for raw disk writes
Alan Cox [Fri, 21 Apr 2017 23:06:38 +0000 (00:06 +0100)]
coco2cart: fix small build bug
Alan Cox [Fri, 21 Apr 2017 23:05:32 +0000 (00:05 +0100)]
coco2cart: assorted fixes
- Fix bugs in the sdc driver
- Build both sdc and ide versions
- Make swap auto-switch according to version
- Enable rawmode binary loading
Alan Cox [Fri, 21 Apr 2017 15:51:44 +0000 (16:51 +0100)]
coco2cart: trivial improvements
Alan Cox [Fri, 21 Apr 2017 15:43:29 +0000 (16:43 +0100)]
sdc.s: Fix merge problem
Alan Cox [Fri, 21 Apr 2017 11:55:12 +0000 (12:55 +0100)]
start: It's 2017
Alan Cox [Fri, 21 Apr 2017 11:54:19 +0000 (12:54 +0100)]
trs80: turn on discard movement
Bring the trs80 start up code into line with the newer binman arrangement
Alan Cox [Fri, 21 Apr 2017 11:53:39 +0000 (12:53 +0100)]
trs80: update README
Offset changes for corrected disk image and the makefile now builds the
boot loader with sdasz80 automatically
Alan Cox [Fri, 21 Apr 2017 11:53:20 +0000 (12:53 +0100)]
makejv3: correct sector offsets to match real TRS80 images
Alan Cox [Fri, 21 Apr 2017 10:37:14 +0000 (11:37 +0100)]
makejv3: fix warning
Alan Cox [Thu, 20 Apr 2017 14:25:09 +0000 (15:25 +0100)]
mtx: fix keyboard handling bug
Alan Cox [Wed, 19 Apr 2017 23:18:45 +0000 (00:18 +0100)]
mtx: use build-mini-filesystem so it fits
Alan Cox [Wed, 19 Apr 2017 23:16:48 +0000 (00:16 +0100)]
mtx/devfd: Fix multiple bugs in the fd driver code
Alan Cox [Wed, 19 Apr 2017 23:13:42 +0000 (00:13 +0100)]
syscall_exec16: handle disk read failure on exec properly
We now kill the process so we don't execute a misloaded image
Alan Cox [Wed, 19 Apr 2017 14:25:23 +0000 (15:25 +0100)]
v68: Remove debug code
Alan Cox [Wed, 19 Apr 2017 11:49:36 +0000 (12:49 +0100)]
README.md: add fuzix.org URL
Alan Cox [Tue, 18 Apr 2017 21:59:13 +0000 (22:59 +0100)]
utils: fix 6809 Make clean
Alan Cox [Tue, 18 Apr 2017 20:56:20 +0000 (21:56 +0100)]
levee: Add 68K build
We should tune the defines but we'll get to that later
Alan Cox [Tue, 18 Apr 2017 13:54:57 +0000 (14:54 +0100)]
68000: more minor fixes for softmmu
Next stop is making fork() work I think
Alan Cox [Tue, 18 Apr 2017 13:54:44 +0000 (14:54 +0100)]
Merge branch 'master' of https://github.com/EtchedPixels/FUZIX
EtchedPixels [Fri, 24 Mar 2017 19:52:01 +0000 (19:52 +0000)]
Merge pull request #491 from beretta42/netdev
libs: missed a header.
EtchedPixels [Fri, 24 Mar 2017 19:51:23 +0000 (19:51 +0000)]
Merge pull request #493 from beretta42/swap9
Make coco3 platform swap (finally)
Brett Gordon [Sat, 4 Mar 2017 14:11:03 +0000 (09:11 -0500)]
coco3: remove debugging printf
Brett Gordon [Tue, 28 Feb 2017 19:30:29 +0000 (14:30 -0500)]
coco3: Do bank16 swapping.
Brett Gordon [Tue, 28 Feb 2017 20:03:06 +0000 (15:03 -0500)]
bank16k: add swapping
Brett Gordon [Sat, 25 Feb 2017 21:20:34 +0000 (16:20 -0500)]
libs: missed a header.
EtchedPixels [Wed, 22 Feb 2017 12:52:50 +0000 (12:52 +0000)]
Merge pull request #490 from beretta42/netno2_2
Allow for networking w/o level 2
Brett Gordon [Tue, 21 Feb 2017 19:11:05 +0000 (14:11 -0500)]
coco3: regress to level 1.
Brett Gordon [Tue, 21 Feb 2017 19:10:25 +0000 (14:10 -0500)]
net: add conditional for level 2 specific code
Brett Gordon [Tue, 21 Feb 2017 19:06:05 +0000 (14:06 -0500)]
kernel: allow for net calls without level 2
EtchedPixels [Sun, 12 Feb 2017 14:21:50 +0000 (14:21 +0000)]
Merge pull request #488 from beretta42/coconic
coconic: change to static address on init, enable 16 bit xfers
EtchedPixels [Sun, 12 Feb 2017 13:58:42 +0000 (13:58 +0000)]
Merge pull request #489 from beretta42/bto
Add a timeout to dw-over-becker, simplify.
EtchedPixels [Sun, 12 Feb 2017 13:56:28 +0000 (13:56 +0000)]
Merge pull request #487 from beretta42/ntpdate
netd: proggie to set system time from ntp server
Brett Gordon [Thu, 9 Feb 2017 22:25:33 +0000 (17:25 -0500)]
ntpdate: second pass.
Brett Gordon [Thu, 9 Feb 2017 17:40:28 +0000 (12:40 -0500)]
a start.
Alan Cox [Mon, 6 Feb 2017 23:34:39 +0000 (23:34 +0000)]
68000-softmmu: first guess at fork()
Alan Cox [Sat, 4 Feb 2017 23:50:17 +0000 (23:50 +0000)]
README.md: minor updates
Alan Cox [Sat, 4 Feb 2017 23:31:57 +0000 (23:31 +0000)]
syscall_68000: generate correct tool references
Alan Cox [Sat, 4 Feb 2017 23:31:21 +0000 (23:31 +0000)]
fuzix68000: get the right compiler for the stubs
Alan Cox [Sat, 4 Feb 2017 23:30:23 +0000 (23:30 +0000)]
Makefile: add flat_mem work
Alan Cox [Sat, 4 Feb 2017 23:29:46 +0000 (23:29 +0000)]
v68-softmmu: commit WIP code
Alan Cox [Sat, 4 Feb 2017 23:29:38 +0000 (23:29 +0000)]
v68: note bits to sort out
Alan Cox [Sat, 4 Feb 2017 23:26:37 +0000 (23:26 +0000)]
flat_mem: thoughts on swap
Alan Cox [Sat, 4 Feb 2017 23:16:50 +0000 (23:16 +0000)]
flat_mem: further debug
We now initialize nicely, start init and get to fork, which needs wiring into
the codebase
Alan Cox [Sat, 4 Feb 2017 23:11:34 +0000 (23:11 +0000)]
lowlevel-68000: debug the new syscall ABI
Dumb bug removed and it now seems to work rather better
Alan Cox [Sat, 4 Feb 2017 23:11:12 +0000 (23:11 +0000)]
flat_mem: note sizing work to do
Alan Cox [Sat, 4 Feb 2017 20:59:10 +0000 (20:59 +0000)]
build-filesystem-ng: Fix 'experimental push on scalar is now forbidden'
At least it seems to, I don't speak perl, someone should check !
Alan Cox [Fri, 3 Feb 2017 22:32:23 +0000 (22:32 +0000)]
flat_mem: initial pass over imported code and glue
Possibly we should rewrite this for clarity but let's start by seeing if it
is workable.
Alan Cox [Thu, 2 Feb 2017 14:32:08 +0000 (14:32 +0000)]
simple: add a TODO note
Alan Cox [Thu, 2 Feb 2017 14:31:56 +0000 (14:31 +0000)]
kernel: update TODO
Alan Cox [Thu, 2 Feb 2017 14:31:40 +0000 (14:31 +0000)]
platform_param: correct signed/unsigned discrepancy
EtchedPixels [Mon, 30 Jan 2017 17:40:58 +0000 (17:40 +0000)]
Merge pull request #484 from BartHM/patch-1
Update trsload80.s to work on real hardware
Bart [Mon, 30 Jan 2017 16:06:02 +0000 (17:06 +0100)]
Update trsload80.s to work on real hardware
In order to run Fuzix on a real TRS-80 Model 4 some corrections have to be made to the bootloader:
- The correct mode was not selected, Video Page #1 needs to be selected. The emulator SDLTRS does not behave correctly; on a real Model 4 Video Page #0 starts on FC00 with a distorted screen as result. So: OUT (0x84),#0x86 is working correctly.
- Stackpointer out of the way on 0xEE00, doesn't overwrite code.
- The ctc6845 registers are not loaded correctly, they must be read in reverse order. So INC HL must be DEC HL and the label must be put at the end of the table.
- The Clear Screen routine was not implemented correctly.
- On a real Model 4 a formatted floppy must be used to boot. This will be a SDDD disk, 40 tracks, formatted for TRS-DOS 6 or LS-DOS 6. These formatted disks have 18 sectors per track, starting on sector 0 and ending on 17. So an extra DEC A is used and a test against sector 18 instead of 19.
With these changes the bootloader is still 256 bytes and now works on SDLTRS and on a real Model 4 128K. A real or virtual formatted floppy is necessary, so the instructions need an update as well.
EtchedPixels [Mon, 30 Jan 2017 14:24:12 +0000 (14:24 +0000)]
Merge pull request #483 from beretta42/ping
netd: route locally destined eth frames back to ourselfs.
We should do this by IP header inspection, but this gives us the framework.
Brett Gordon [Mon, 30 Jan 2017 13:53:09 +0000 (08:53 -0500)]
netd: route locally destined eth frames back to ourselfs.
EtchedPixels [Sat, 28 Jan 2017 14:18:58 +0000 (14:18 +0000)]
Merge pull request #480 from willsowerbutts/ds1302
ds1302 improvements
EtchedPixels [Sat, 28 Jan 2017 14:18:00 +0000 (14:18 +0000)]
Merge pull request #479 from willsowerbutts/floppy
Kernel: devfd: Optimisations for reliable operation on 6MHz Z80
EtchedPixels [Sat, 28 Jan 2017 14:05:32 +0000 (14:05 +0000)]
Merge pull request #482 from beretta42/ping
Add basic raw socket support to uip + ping client to test
EtchedPixels [Sat, 28 Jan 2017 14:04:59 +0000 (14:04 +0000)]
Merge pull request #481 from willsowerbutts/optimisations
Optimisations
Brett Gordon [Thu, 26 Jan 2017 17:32:40 +0000 (12:32 -0500)]
coconic: change to static address on init, enable 16 bit xfers
Will Sowerbutts [Wed, 25 Jan 2017 14:53:31 +0000 (14:53 +0000)]
Kernel: n8vem-mark4: Fix timer interrupt rate
Will Sowerbutts [Wed, 25 Jan 2017 14:36:39 +0000 (14:36 +0000)]
Kernel: p112: Use mnemonic to disable floppy motors, do not enable DMA
Will Sowerbutts [Wed, 25 Jan 2017 14:36:50 +0000 (14:36 +0000)]
Kernel: p112: Fix timer interrupt rate
Brett Gordon [Tue, 24 Jan 2017 16:26:35 +0000 (11:26 -0500)]
netd: add some todo's
Brett Gordon [Tue, 24 Jan 2017 16:21:46 +0000 (11:21 -0500)]
netd: mark unused raw sockets with protocol no. 255 rather than 0
Brett Gordon [Tue, 24 Jan 2017 16:06:08 +0000 (11:06 -0500)]
netd: install ping
Brett Gordon [Tue, 24 Jan 2017 16:02:33 +0000 (11:02 -0500)]
netd: add ping client (to test raw socks)
Brett Gordon [Tue, 24 Jan 2017 15:59:19 +0000 (10:59 -0500)]
netd: add ping client (to test raw socks)
Brett Gordon [Tue, 24 Jan 2017 15:14:33 +0000 (10:14 -0500)]
netd: send dup packets to each match raw socket
Will Sowerbutts [Mon, 23 Jan 2017 23:07:48 +0000 (23:07 +0000)]
dev_fd: Only spin up motors for drives that are in use
Previously we spun up all floppy motors at once. The current required to
do this places a lot of strain on power supplies. This method is a bit
kinder to them.
Brett Gordon [Mon, 23 Jan 2017 15:02:59 +0000 (10:02 -0500)]
netd: make uip send IP header too.