FUZIX.git
5 years agotiny68k: 68010 hooks
Alan Cox [Mon, 4 Feb 2019 00:38:17 +0000 (00:38 +0000)]
tiny68k: 68010 hooks

5 years ago68000: Basic 68010 support
Alan Cox [Mon, 4 Feb 2019 00:37:59 +0000 (00:37 +0000)]
68000: Basic 68010 support

5 years agotiny68k: fix pagesize
Alan Cox [Mon, 4 Feb 2019 00:36:33 +0000 (00:36 +0000)]
tiny68k: fix pagesize

5 years agomakeversion: redefine as banks per 64K, which means 8bit is unchanged
Alan Cox [Sun, 3 Feb 2019 21:24:34 +0000 (21:24 +0000)]
makeversion: redefine as banks per 64K, which means 8bit is unchanged

5 years ago68000flat: don't push unneeded pid return word onto switch stacks
Alan Cox [Sun, 3 Feb 2019 21:21:58 +0000 (21:21 +0000)]
68000flat: don't push unneeded pid return word onto switch stacks

5 years agoMerge branch 'master' of https://github.com/EtchedPixels/FUZIX
Alan Cox [Fri, 1 Feb 2019 14:27:48 +0000 (14:27 +0000)]
Merge branch 'master' of https://github.com/EtchedPixels/FUZIX

5 years agotravis: switch from building z80pack to n8vem-mark4 as a test
Alan Cox [Fri, 1 Feb 2019 14:26:56 +0000 (14:26 +0000)]
travis: switch from building z80pack to n8vem-mark4 as a test

That we we cover Z180.

5 years agoMerge pull request #688 from dbotw/setdate
EtchedPixels [Fri, 1 Feb 2019 12:12:53 +0000 (12:12 +0000)]
Merge pull request #688 from dbotw/setdate

setdate: fix range check of hours and minutes

5 years agoz80pack: move common up to F400 ; more buffers
Alan Cox [Fri, 1 Feb 2019 12:12:03 +0000 (12:12 +0000)]
z80pack: move common up to F400 ; more buffers

5 years agosetdate: fix range check of hours and minutes
Jens Guenther [Thu, 31 Jan 2019 18:23:36 +0000 (18:23 +0000)]
setdate: fix range check of hours and minutes

5 years agobuddy: Throw out the current stuff in favour of a small tight model
Alan Cox [Tue, 29 Jan 2019 23:03:50 +0000 (23:03 +0000)]
buddy: Throw out the current stuff in favour of a small tight model

Not yet tested

5 years agomalloc: lay basis for new buddy allocator that can also pair with flat.c
Alan Cox [Tue, 29 Jan 2019 23:02:55 +0000 (23:02 +0000)]
malloc: lay basis for new buddy allocator that can also pair with flat.c

5 years agotty: fix a silly braces bug
Alan Cox [Tue, 29 Jan 2019 23:02:01 +0000 (23:02 +0000)]
tty: fix a silly braces bug

5 years agov68: tidy up
Alan Cox [Tue, 29 Jan 2019 23:01:46 +0000 (23:01 +0000)]
v68: tidy up

5 years agotiny68k: clean up minor things
Alan Cox [Tue, 29 Jan 2019 23:01:05 +0000 (23:01 +0000)]
tiny68k: clean up minor things

5 years agolevel2: remove dead variable
Alan Cox [Tue, 29 Jan 2019 22:59:49 +0000 (22:59 +0000)]
level2: remove dead variable

5 years agom68000: document Minix trap usage
Alan Cox [Tue, 29 Jan 2019 22:58:51 +0000 (22:58 +0000)]
m68000: document Minix trap usage

5 years agosc108: fix signal handling bug
Alan Cox [Mon, 28 Jan 2019 00:54:10 +0000 (00:54 +0000)]
sc108: fix signal handling bug

5 years agoMakefile: update tiny68k status
Alan Cox [Mon, 28 Jan 2019 00:09:24 +0000 (00:09 +0000)]
Makefile: update tiny68k status

5 years agoreboot: add suspend mode
Alan Cox [Mon, 28 Jan 2019 00:08:50 +0000 (00:08 +0000)]
reboot: add suspend mode

5 years agosc114: fix bug reported by herrw@pluspora
Alan Cox [Sun, 27 Jan 2019 22:13:16 +0000 (22:13 +0000)]
sc114: fix bug reported by herrw@pluspora

5 years agoz80pack32: now obsolete
Alan Cox [Sun, 27 Jan 2019 22:00:39 +0000 (22:00 +0000)]
z80pack32: now obsolete

5 years agorc2014tiny: use ds1302 rtc detect
Alan Cox [Sun, 27 Jan 2019 21:59:55 +0000 (21:59 +0000)]
rc2014tiny: use ds1302 rtc detect

5 years agorc2014: use and report rtc presence
Alan Cox [Sun, 27 Jan 2019 21:59:35 +0000 (21:59 +0000)]
rc2014: use and report rtc presence

5 years agoudata: Fix the remaining legacy udata sizes
Alan Cox [Sun, 27 Jan 2019 21:58:36 +0000 (21:58 +0000)]
udata: Fix the remaining legacy udata sizes

In some cases this should also fix a few bugs

5 years agoubee: fix the udata size bug
Alan Cox [Sun, 27 Jan 2019 21:46:41 +0000 (21:46 +0000)]
ubee: fix the udata size bug

5 years agosc108: use the probing ds1302 support
Alan Cox [Sun, 27 Jan 2019 21:30:26 +0000 (21:30 +0000)]
sc108: use the probing ds1302 support

5 years agovarious: Sort out more U_DATA_TOTALSIZE cases
Alan Cox [Sun, 27 Jan 2019 21:28:01 +0000 (21:28 +0000)]
various: Sort out more U_DATA_TOTALSIZE cases

This changed historically and there is some usage confusion to clean up

5 years agosc108: we have a proper bootloader now (yay)
Alan Cox [Sun, 27 Jan 2019 21:27:03 +0000 (21:27 +0000)]
sc108: we have a proper bootloader now (yay)

5 years agorc2014: fix dead comment
Alan Cox [Sun, 27 Jan 2019 21:26:52 +0000 (21:26 +0000)]
rc2014: fix dead comment

5 years agods1302: add detection logic
Alan Cox [Sun, 27 Jan 2019 21:26:32 +0000 (21:26 +0000)]
ds1302: add detection logic

5 years agosc108: documentation update
Alan Cox [Sun, 27 Jan 2019 21:26:11 +0000 (21:26 +0000)]
sc108: documentation update

5 years agosc108: 512 bytes for the udata save
Alan Cox [Sun, 27 Jan 2019 21:25:30 +0000 (21:25 +0000)]
sc108: 512 bytes for the udata save

Nasty because the wrong size scribbles over a buffer beyond tmpbuf if it
is wrong.

5 years agosc108: fix boot loader update other code
Alan Cox [Sun, 27 Jan 2019 19:32:33 +0000 (19:32 +0000)]
sc108: fix boot loader update other code

Still has a weird crash to dig into

5 years agosc108: sync with CTC and tty work for RC2014
Alan Cox [Sun, 27 Jan 2019 18:27:28 +0000 (18:27 +0000)]
sc108: sync with CTC and tty work for RC2014

5 years agorc2014-tiny: resync the tty and CTC work
Alan Cox [Sun, 27 Jan 2019 18:11:59 +0000 (18:11 +0000)]
rc2014-tiny: resync the tty and CTC work

5 years agorc2014: Support for CTC, fix a load of serial related bugs
Alan Cox [Sun, 27 Jan 2019 17:55:31 +0000 (17:55 +0000)]
rc2014: Support for CTC, fix a load of serial related bugs

Also add support for a second SIO2 card

5 years agorc2014-sbc64: tidy
Alan Cox [Sun, 27 Jan 2019 17:54:01 +0000 (17:54 +0000)]
rc2014-sbc64: tidy

5 years agorc2014-sbc64: Final fixes for SIO cards
Alan Cox [Sun, 27 Jan 2019 17:08:07 +0000 (17:08 +0000)]
rc2014-sbc64: Final fixes for SIO cards

5 years agorc2014-sbc64: Further serial work
Alan Cox [Sun, 27 Jan 2019 16:23:46 +0000 (16:23 +0000)]
rc2014-sbc64: Further serial work

- Fix SIO probing
- Support second SIO card at 0x84
- Tidy up SIO code in general

SIO isn't yet working

5 years agotiny68k: so I remember to tackle the 2MB board
Alan Cox [Sun, 27 Jan 2019 15:40:35 +0000 (15:40 +0000)]
tiny68k: so I remember to tackle the 2MB board

5 years agorc2014-sbc64: Further work
Alan Cox [Sun, 27 Jan 2019 15:34:55 +0000 (15:34 +0000)]
rc2014-sbc64: Further work

- Disable interrupts during a bitbang serial then it works rather better
- Tie SIO to a presence check (not yet probed for)
- Autodetect CTC
- CTC programming and default to CTC if possible
- Fix interrupt vector setup
- Assorted cleanups
- First cut at suspend/resume properly

5 years agoMakefile: add the new platform
Alan Cox [Sat, 26 Jan 2019 23:33:47 +0000 (23:33 +0000)]
Makefile: add the new platform

5 years agoz80sbc64: another day another platform
Alan Cox [Sat, 26 Jan 2019 23:32:03 +0000 (23:32 +0000)]
z80sbc64: another day another platform

5 years agoprocess: Fix nready panic and hangs when using rts/cts
Alan Cox [Sat, 26 Jan 2019 23:03:29 +0000 (23:03 +0000)]
process: Fix nready panic and hangs when using rts/cts

Actually what happens is that spend long enough waiting to write the byte
that we hit the poll timeout. In the usual case it doesn't matter but if

1. You hit a key
2. We are idle
3. We are doing idle key polling
4. The echo is blocked for a long long time

We then try and sleep the current process, only we are in idle so there isn't
one. Idle is really like being in an interrupt.

Now arguably we should fix tty output to retry the echo write until it gets
kicked for timing offences and then give up but tht means auditing all the
ininterrupt uses and also encourages the existing bad behaviour of not running
a small tx buffer.

So if it annoys you fix your tty output code 8)

5 years agokeycode: We need an UNDO key symbol
Alan Cox [Sat, 26 Jan 2019 22:26:15 +0000 (22:26 +0000)]
keycode: We need an UNDO key symbol

5 years ago68000: allow configuration of di/ei masks
Alan Cox [Sat, 26 Jan 2019 22:25:30 +0000 (22:25 +0000)]
68000: allow configuration of di/ei masks

We need this as the Atari ST for example expects to run at IPL 3 otherwise
you get bombed with horizontal timing interrupts

5 years agobankfixed: add a DEBUG undef
Alan Cox [Sat, 26 Jan 2019 22:25:09 +0000 (22:25 +0000)]
bankfixed: add a DEBUG undef

5 years agoatarist: filling in the blanks
Alan Cox [Sat, 26 Jan 2019 22:24:50 +0000 (22:24 +0000)]
atarist: filling in the blanks

Still a work in progress with much to complete

5 years agoatarist: quick pass over the old draft code to match current state
Alan Cox [Wed, 23 Jan 2019 19:23:00 +0000 (19:23 +0000)]
atarist: quick pass over the old draft code to match current state

5 years agotiny68k: build with new core code
Alan Cox [Wed, 23 Jan 2019 19:22:40 +0000 (19:22 +0000)]
tiny68k: build with new core code

5 years agov68: resync with moves of code to core
Alan Cox [Wed, 23 Jan 2019 19:21:59 +0000 (19:21 +0000)]
v68: resync with moves of code to core

5 years agotiny68k: remove code now in lowlevel 68K support
Alan Cox [Wed, 23 Jan 2019 19:20:52 +0000 (19:20 +0000)]
tiny68k: remove code now in lowlevel 68K support

5 years ago68000: move common code for fork/switching on 68000 flat model to lib
Alan Cox [Wed, 23 Jan 2019 14:29:08 +0000 (14:29 +0000)]
68000: move common code for fork/switching on 68000 flat model to lib

5 years ago68000: move generic asm code from tiny68K into the core code
Alan Cox [Wed, 23 Jan 2019 14:25:21 +0000 (14:25 +0000)]
68000: move generic asm code from tiny68K into the core code

This will break the v68 platforms (again) but they can just be cleaned
up afterwards.

5 years agotiny68k: remove some missed debug messages
Alan Cox [Wed, 23 Jan 2019 14:18:06 +0000 (14:18 +0000)]
tiny68k: remove some missed debug messages

5 years ago68000: dynamic udata, use actual available memory space for allocs
Alan Cox [Wed, 23 Jan 2019 14:17:01 +0000 (14:17 +0000)]
68000: dynamic udata, use actual available memory space for allocs

5 years agotiny68k: use defined for size of udata block to set stack
Alan Cox [Wed, 23 Jan 2019 14:16:20 +0000 (14:16 +0000)]
tiny68k: use defined for size of udata block to set stack

5 years ago68000: define udata block size for asm
Alan Cox [Wed, 23 Jan 2019 14:15:59 +0000 (14:15 +0000)]
68000: define udata block size for asm

5 years agokdata: correct layout for level 2 and no net
Alan Cox [Wed, 23 Jan 2019 14:15:47 +0000 (14:15 +0000)]
kdata: correct layout for level 2 and no net

5 years agosyscall_proc: fix fork debugging messages on 32bit
Alan Cox [Tue, 22 Jan 2019 23:52:40 +0000 (23:52 +0000)]
syscall_proc: fix fork debugging messages on 32bit

5 years agosyscall_exec32: ifdef out most of core dump for now
Alan Cox [Tue, 22 Jan 2019 23:52:00 +0000 (23:52 +0000)]
syscall_exec32: ifdef out most of core dump for now

We need to do this very differently to 8bit, probably with a callback into
the mapper

5 years agomakeproc: fix comment and also fix it to use the right udata
Alan Cox [Tue, 22 Jan 2019 23:51:30 +0000 (23:51 +0000)]
makeproc: fix comment and also fix it to use the right udata

5 years agotiny68k: and plumb together all the new stuff
Alan Cox [Tue, 22 Jan 2019 23:51:04 +0000 (23:51 +0000)]
tiny68k: and plumb together all the new stuff

5 years agokernel: header tweaks
Alan Cox [Tue, 22 Jan 2019 23:49:09 +0000 (23:49 +0000)]
kernel: header tweaks

We add a new method to set p->p_udata on platform basis
We allow the page switch code to be passed a death argument so that flat and
similar models can do a one way copy if they are copying store to direct
mapped and the existing direct map is dead

(It's a common case when you fork and execve. The child has to put the parent
back but there's no point saving the child as it's going to free and realloc
a mapping)

5 years agotiny68k: status update
Alan Cox [Tue, 22 Jan 2019 23:48:44 +0000 (23:48 +0000)]
tiny68k: status update

5 years agotiny68k: use the flat model
Alan Cox [Tue, 22 Jan 2019 23:48:26 +0000 (23:48 +0000)]
tiny68k: use the flat model

5 years agoflat: debug the flat memory model
Alan Cox [Tue, 22 Jan 2019 23:47:50 +0000 (23:47 +0000)]
flat: debug the flat memory model

It now seems to work ok

5 years agolowlevel-68000: expose copy_blocks
Alan Cox [Tue, 22 Jan 2019 23:46:56 +0000 (23:46 +0000)]
lowlevel-68000: expose copy_blocks

Doesn't yet work. Debug in progress

5 years agomalloc: Fix debug messages, and add used/available data
Alan Cox [Tue, 22 Jan 2019 23:46:20 +0000 (23:46 +0000)]
malloc: Fix debug messages, and add used/available data

5 years agotiny68k: introduce memcpy32
Alan Cox [Tue, 22 Jan 2019 23:45:26 +0000 (23:45 +0000)]
tiny68k: introduce memcpy32

5 years agoMakefile: add tiny68k, tweak trs80 doc
Alan Cox [Mon, 21 Jan 2019 22:38:47 +0000 (22:38 +0000)]
Makefile: add tiny68k, tweak trs80 doc

5 years agov68: first cut at backporting all the extra bits
Alan Cox [Mon, 21 Jan 2019 22:35:16 +0000 (22:35 +0000)]
v68: first cut at backporting all the extra bits

5 years ago68000: Get tiny68k into a workable state
Alan Cox [Mon, 21 Jan 2019 22:24:21 +0000 (22:24 +0000)]
68000: Get tiny68k into a workable state

v68 will need to be reworked to deal with the changes

5 years agoprocess: Fix SIGSTOP crash and also a debug printf bug
Alan Cox [Mon, 21 Jan 2019 22:23:11 +0000 (22:23 +0000)]
process: Fix SIGSTOP crash and also a debug printf bug

5 years ago68000: remove autogenerated files
Alan Cox [Mon, 21 Jan 2019 22:22:32 +0000 (22:22 +0000)]
68000: remove autogenerated files

5 years ago68000: Switch syscall trap to 12
Alan Cox [Mon, 21 Jan 2019 22:20:37 +0000 (22:20 +0000)]
68000: Switch syscall trap to 12

5 years agopreemption: rabbit also needs the fix
Alan Cox [Mon, 21 Jan 2019 22:19:59 +0000 (22:19 +0000)]
preemption: rabbit also needs the fix

5 years agopreemption: allow pre-emption where the leaving processis marked STOPPED
Alan Cox [Mon, 21 Jan 2019 22:19:23 +0000 (22:19 +0000)]
preemption: allow pre-emption where the leaving processis marked STOPPED

We need this to make SIGSTOP work reliably and properly

5 years agofweep: build for 68000 and adjust constants accordingly.
Alan Cox [Mon, 21 Jan 2019 22:18:29 +0000 (22:18 +0000)]
fweep: build for 68000 and adjust constants accordingly.

Probably wants tuning a bit better (maybe pin low content ?)

5 years agosh: Fix a NULL reference in /bin/sh
Alan Cox [Mon, 21 Jan 2019 22:18:08 +0000 (22:18 +0000)]
sh: Fix a NULL reference in /bin/sh

5 years agotiny68k: adapt C code to proposed new exception logic
Alan Cox [Sun, 20 Jan 2019 23:31:16 +0000 (23:31 +0000)]
tiny68k: adapt C code to proposed new exception logic

5 years ago68000: first cut at rewriting the exception handlers
Alan Cox [Sun, 20 Jan 2019 23:30:17 +0000 (23:30 +0000)]
68000: first cut at rewriting the exception handlers

Move towards a unified sensible model for synchronous exception handling and
try to unify all the trap and stack frame munging we must do.

Untested first cut.

5 years agokernel: make dump_core accessible elsewhere
Alan Cox [Sun, 20 Jan 2019 23:29:53 +0000 (23:29 +0000)]
kernel: make dump_core accessible elsewhere

We will need this for the synchronous trap handlers

5 years agotiny68k: update 68000 state of play
Alan Cox [Sun, 20 Jan 2019 15:28:41 +0000 (15:28 +0000)]
tiny68k: update 68000 state of play

5 years agotiny68k: fix incorrect sr set up at boot
Alan Cox [Sun, 20 Jan 2019 15:21:58 +0000 (15:21 +0000)]
tiny68k: fix incorrect sr set up at boot

5 years agokernel: remove stack size hack for 32bit
Alan Cox [Sun, 20 Jan 2019 15:21:08 +0000 (15:21 +0000)]
kernel: remove stack size hack for 32bit

We leave a sensible minimum value just as a sanity check

5 years ago68000: set sensible stack defaults for binaries
Alan Cox [Sun, 20 Jan 2019 15:19:57 +0000 (15:19 +0000)]
68000: set sensible stack defaults for binaries

Then we can pull the hack out of the kernel.

We are going to change syscall trap in a bit so compatibility is not yet the
slightest concern!

5 years agoutil: build extra tools for 68K, default to 16K stack
Alan Cox [Sun, 20 Jan 2019 15:14:18 +0000 (15:14 +0000)]
util: build extra tools for 68K, default to 16K stack

5 years agoStandalone: update the standard size and chmem
Alan Cox [Sun, 20 Jan 2019 15:14:02 +0000 (15:14 +0000)]
Standalone: update the standard size and chmem

5 years agoutil: chmem and size support for 32bit
Alan Cox [Sun, 20 Jan 2019 15:13:08 +0000 (15:13 +0000)]
util: chmem and size support for 32bit

Also clean up the formatting and other details.

5 years agosyscall_exec32: report ENOMEM properly for outsized binaries
Alan Cox [Sun, 20 Jan 2019 00:49:36 +0000 (00:49 +0000)]
syscall_exec32: report ENOMEM properly for outsized binaries

5 years agolevee: fix build on 68000
Alan Cox [Sun, 20 Jan 2019 00:49:22 +0000 (00:49 +0000)]
levee: fix build on 68000

5 years agoinit: fix alignment
Alan Cox [Sun, 20 Jan 2019 00:48:48 +0000 (00:48 +0000)]
init: fix alignment

5 years agotiny68k: status update
Alan Cox [Sun, 20 Jan 2019 00:48:31 +0000 (00:48 +0000)]
tiny68k: status update

5 years agostartrek: missing fflush
Alan Cox [Sat, 19 Jan 2019 23:38:45 +0000 (23:38 +0000)]
startrek: missing fflush

5 years ago68000: fix longjmp
Alan Cox [Sat, 19 Jan 2019 23:36:30 +0000 (23:36 +0000)]
68000: fix longjmp

This sorts out the weird crashes in shell

5 years agosh: remove old dead ifdeffed stuff
Alan Cox [Sat, 19 Jan 2019 23:36:18 +0000 (23:36 +0000)]
sh: remove old dead ifdeffed stuff

5 years agofsh: fix build on 68000
Alan Cox [Sat, 19 Jan 2019 23:03:08 +0000 (23:03 +0000)]
fsh: fix build on 68000

Forgot to set the define so it build a generic sh

5 years agoutil: fix clean on 68000
Alan Cox [Sat, 19 Jan 2019 23:02:49 +0000 (23:02 +0000)]
util: fix clean on 68000