Alan Cox [Thu, 1 Sep 2016 21:43:00 +0000 (22:43 +0100)]
Makefile: force sysinfoblk.h to be built early
Alan Cox [Thu, 1 Sep 2016 21:16:00 +0000 (22:16 +0100)]
uname: fix the 6809 sysinfo problem
EtchedPixels [Thu, 1 Sep 2016 15:46:32 +0000 (16:46 +0100)]
Merge pull request #418 from
dfffffff/master
refactored ls utility: cleanup, added -R recursive, fixed -r reverse order
dfffffff [Thu, 1 Sep 2016 14:57:52 +0000 (10:57 -0400)]
cleanup, added -R recursive, fixed -r reverse order
Alan Cox [Wed, 31 Aug 2016 21:02:39 +0000 (22:02 +0100)]
init: add terminal setting
We don't support setting parity yet - do we need to ?
Alan Cox [Wed, 31 Aug 2016 19:11:26 +0000 (20:11 +0100)]
Merge branch 'master' of https://github.com/EtchedPixels/FUZIX
Alan Cox [Wed, 31 Aug 2016 19:10:01 +0000 (20:10 +0100)]
tty: level 2 and other improvements
Add what should match the rules for BSD job control. Make sure all
the crap that gets inflicted upon us for this ends up in level2.c
so we don't bloat the smaller platforms.
Do some more preparatory work on the pty support.
Alan Cox [Wed, 31 Aug 2016 19:09:36 +0000 (20:09 +0100)]
process: fix a race in SIGSTOP handling and slightly optimize newproc
EtchedPixels [Wed, 31 Aug 2016 13:45:44 +0000 (14:45 +0100)]
Merge pull request #414 from
dfffffff/master
usage now outputs to stderr
Alan Cox [Wed, 31 Aug 2016 13:40:30 +0000 (14:40 +0100)]
Merge branch 'master' of https://github.com/EtchedPixels/FUZIX
Alan Cox [Wed, 31 Aug 2016 13:39:49 +0000 (14:39 +0100)]
process.c: make SIGSTOP/CONT work properly
I hope anyway 8)
dfffffff [Wed, 31 Aug 2016 12:11:16 +0000 (08:11 -0400)]
usage now outputs to stderr
EtchedPixels [Wed, 31 Aug 2016 10:53:41 +0000 (11:53 +0100)]
Merge pull request #412 from
dfffffff/master
cosmetic
EtchedPixels [Wed, 31 Aug 2016 10:53:17 +0000 (11:53 +0100)]
Merge pull request #413 from vandys/kernel
Minor edit on indentation to make it less misleading
Andy Valencia [Wed, 31 Aug 2016 02:28:55 +0000 (19:28 -0700)]
Minor edit on indentation to make it less misleading
dfffffff [Wed, 31 Aug 2016 02:01:36 +0000 (22:01 -0400)]
cosmetic
Alan Cox [Wed, 31 Aug 2016 00:23:23 +0000 (01:23 +0100)]
stty: struct winsz is needed
Alan Cox [Tue, 30 Aug 2016 23:59:13 +0000 (00:59 +0100)]
buddy: add header
Alan Cox [Tue, 30 Aug 2016 22:10:59 +0000 (23:10 +0100)]
open: fix bug introduced in handling of invalid directory paths during an open
EtchedPixels [Tue, 30 Aug 2016 21:48:00 +0000 (22:48 +0100)]
Merge pull request #408 from beretta42/net_native
net_native: add async event for remote shutdown.
Alan Cox [Tue, 30 Aug 2016 20:56:09 +0000 (21:56 +0100)]
v68: update README
Alan Cox [Tue, 30 Aug 2016 20:25:55 +0000 (21:25 +0100)]
process.c: 32bit clean pointer printing
Alan Cox [Tue, 30 Aug 2016 20:20:24 +0000 (21:20 +0100)]
v68: fix udata naming
Alan Cox [Tue, 30 Aug 2016 20:19:52 +0000 (21:19 +0100)]
68000: set the a5 pointer properly in the sig case
Alan Cox [Tue, 30 Aug 2016 20:08:38 +0000 (21:08 +0100)]
v68: load the uarea shadow when needed
Alan Cox [Tue, 30 Aug 2016 20:08:24 +0000 (21:08 +0100)]
v68: fix tty input
Alan Cox [Tue, 30 Aug 2016 20:06:03 +0000 (21:06 +0100)]
devio: add 32bit pointer support
This means we need a new % code for printk and to stop misusing %x for pointers. Add %p
Alan Cox [Tue, 30 Aug 2016 19:42:04 +0000 (20:42 +0100)]
platform-v68: initial pieces to get it to build/begin booting
Alan Cox [Tue, 30 Aug 2016 19:41:28 +0000 (20:41 +0100)]
tty: fix need_resched ambiguity
Alan Cox [Tue, 30 Aug 2016 19:40:56 +0000 (20:40 +0100)]
v68: first cut at setting up exception vectors
Alan Cox [Tue, 30 Aug 2016 19:40:28 +0000 (20:40 +0100)]
v68: specify number of block devices
Alan Cox [Tue, 30 Aug 2016 19:39:59 +0000 (20:39 +0100)]
68000: imported needed helper routines for gcc, and fix some exports
Alan Cox [Tue, 30 Aug 2016 19:39:37 +0000 (20:39 +0100)]
cpu-68000: add missing endian code
Alan Cox [Tue, 30 Aug 2016 19:39:15 +0000 (20:39 +0100)]
buddy: Further early work - still not usable
Brett Gordon [Tue, 30 Aug 2016 15:33:57 +0000 (11:33 -0400)]
net_native: add async event for remote shutdown.
Alan Cox [Mon, 29 Aug 2016 16:48:16 +0000 (17:48 +0100)]
ps: add ppid support
Alan Cox [Mon, 29 Aug 2016 16:46:10 +0000 (17:46 +0100)]
devsys: check for over end copies for ps
Doesn't matter too much today but it will when we have MMU support
Alan Cox [Mon, 29 Aug 2016 16:13:42 +0000 (17:13 +0100)]
proc.h: add p_event to match kernel changes
Alan Cox [Mon, 29 Aug 2016 15:50:32 +0000 (16:50 +0100)]
init: add tty and geometry setting support
Alan Cox [Mon, 29 Aug 2016 15:49:59 +0000 (16:49 +0100)]
signal: default behaviour for SIGWINCH is ignore
It's not defined by POSIX but this appears to be the convention
Alan
Alan Cox [Mon, 29 Aug 2016 15:49:43 +0000 (16:49 +0100)]
includes: add the extra tty ioctls
Andy Valencia [Mon, 18 Jul 2016 21:55:27 +0000 (14:55 -0700)]
Endian support for size tool
Add support for 6809 variant of executable header; key off
initial byte value to decode as 6809 or Z-80.
Alan Cox [Mon, 29 Aug 2016 11:35:07 +0000 (12:35 +0100)]
Merge branch 'master' of https://github.com/EtchedPixels/FUZIX
Alan Cox [Mon, 29 Aug 2016 11:33:34 +0000 (12:33 +0100)]
z80pack: fix bugs introduced with block param cleanups
Alan Cox [Mon, 29 Aug 2016 11:33:10 +0000 (12:33 +0100)]
inode: fix ref counts from recent ref change
Alan Cox [Mon, 29 Aug 2016 11:32:58 +0000 (12:32 +0100)]
devio: in debug dump R/W counters
Alan Cox [Mon, 29 Aug 2016 10:54:01 +0000 (11:54 +0100)]
pipe: redo reader/writer counting
It's almost as cheap to keep counters as go scanning each time in terms of
memory usage and much much faster.
Alan Cox [Mon, 29 Aug 2016 10:53:31 +0000 (11:53 +0100)]
binmunge: fix missing include
EtchedPixels [Mon, 29 Aug 2016 09:42:49 +0000 (10:42 +0100)]
Merge pull request #404 from
dfffffff/master
added uname command
dfffffff [Mon, 29 Aug 2016 00:29:10 +0000 (20:29 -0400)]
added uname command
Conflicts:
Build/platforms/dfboard.mk
Alan Cox [Sun, 28 Aug 2016 22:51:12 +0000 (23:51 +0100)]
Merge branch 'master' of https://github.com/EtchedPixels/FUZIX
Alan Cox [Sun, 28 Aug 2016 22:50:54 +0000 (23:50 +0100)]
signals: fix other return in chksigs
EtchedPixels [Sun, 28 Aug 2016 22:48:08 +0000 (23:48 +0100)]
Merge pull request #403 from beretta42/dwtimec
dwdate: sets system time from drivewire via userspace
Alan Cox [Sun, 28 Aug 2016 22:45:29 +0000 (23:45 +0100)]
signal: fix handling of pending signals whose default behaviour is ignore
Also noted by David Flamand.
We make chksigs() a generally callable helper so that anyone who needs to
check if a signal is pending can just ask it. This cleans up the (broken)
poking around we were doing as well.
This also cleans up some other freaky corner cases where signals delivered
at the wrong time could be lost.
Alan Cox [Sun, 28 Aug 2016 22:43:51 +0000 (23:43 +0100)]
pipe: first cut at fixing a pile of pipe breakages
There is still lots more to deal with
- introduce a helper to check if a file is open for read/write
- use this to correct the EOF checking
Original bugs noted by David Flamand
EtchedPixels [Sun, 28 Aug 2016 21:53:52 +0000 (22:53 +0100)]
Merge pull request #401 from
dfffffff/master
added missing operation letter
Brett Gordon [Sun, 28 Aug 2016 21:00:43 +0000 (17:00 -0400)]
add man pager file.
Brett Gordon [Sun, 28 Aug 2016 20:48:04 +0000 (16:48 -0400)]
drivewire: init commit "dwdate" - set system time from userspace.
Brett Gordon [Sun, 28 Aug 2016 19:29:47 +0000 (15:29 -0400)]
syscall: stime: return 0 on success.
Brett Gordon [Sun, 28 Aug 2016 19:28:28 +0000 (15:28 -0400)]
stime syscall: ptr params reversed, also finish prototype comment
Alan Cox [Sun, 28 Aug 2016 16:15:51 +0000 (17:15 +0100)]
net_native: space tidying
Brett Gordon [Sun, 28 Aug 2016 16:10:38 +0000 (17:10 +0100)]
net_native: lots of fixes
Merge a load of fixes from Brett Gordon into the net_native code. With these
applied it might even be usable!
There are a couple of changes from Brett's original - we want to report EOF
not EPIPE on reads of a closed socket, and we want to assume that the shutdown
flags will be set by the daemon rather than add extra closed socket checks.
Alan Cox [Sun, 28 Aug 2016 15:59:37 +0000 (16:59 +0100)]
cpu-z80: peephole rules
Alan Cox [Sun, 28 Aug 2016 15:59:23 +0000 (16:59 +0100)]
syscall_net: add a FIXME
Alan Cox [Sun, 28 Aug 2016 15:59:10 +0000 (16:59 +0100)]
syscall_exec32: add some TODO notes
Alan Cox [Sun, 28 Aug 2016 15:58:50 +0000 (16:58 +0100)]
Makefile: add buddy.c
We don't use it yet and it's not finished
dfffffff [Sun, 28 Aug 2016 14:22:04 +0000 (10:22 -0400)]
added missing operation letter
Alan Cox [Fri, 26 Aug 2016 21:01:51 +0000 (22:01 +0100)]
buddy: A segment based MMU buddy allocator
Not yet buildable and certainly not yet debugged.
This is needed for platforms where we must keep our code/data/stack on power
of two boundaries for the MMU hardware in question.
Alan Cox [Fri, 26 Aug 2016 21:01:25 +0000 (22:01 +0100)]
v68: initial bits of 68000 development tree
Nothing to see here yet
Alan Cox [Fri, 26 Aug 2016 21:00:33 +0000 (22:00 +0100)]
68000: Switch to gnu-linux from atari-mint compiler
Alan Cox [Fri, 26 Aug 2016 15:05:22 +0000 (16:05 +0100)]
SmallC: minor adjustments
Alan Cox [Fri, 26 Aug 2016 15:04:36 +0000 (16:04 +0100)]
Merge branch 'master' of https://github.com/EtchedPixels/FUZIX
EtchedPixels [Fri, 26 Aug 2016 14:57:44 +0000 (15:57 +0100)]
Merge pull request #399 from
dfffffff/master
Kernel/lowlevel-6809.s: fixed outd outputting wrong byte, changed code to use less memory
dfffffff [Fri, 26 Aug 2016 14:23:18 +0000 (10:23 -0400)]
Kernel/lowlevel-6809.s: fixed outd outputting wrong byte, changed code to use less memory
EtchedPixels [Fri, 26 Aug 2016 12:57:16 +0000 (13:57 +0100)]
Merge pull request #398 from
dfffffff/master
removed duplicate source file
dfffffff [Sun, 21 Aug 2016 21:06:18 +0000 (17:06 -0400)]
removed duplicate source file
Alan Cox [Fri, 19 Aug 2016 14:07:46 +0000 (15:07 +0100)]
fview: Fix a () to (void) and fix a non-triggerable bug in err printing
EtchedPixels [Fri, 19 Aug 2016 10:29:29 +0000 (11:29 +0100)]
Merge pull request #397 from
dfffffff/master
found this issue when trying to ranlib libm6809.a
dfffffff [Fri, 19 Aug 2016 01:20:24 +0000 (21:20 -0400)]
found this issue when trying to ranlib libm6809.a: 'error: multiple defined label _fp_scan'
EtchedPixels [Thu, 18 Aug 2016 20:02:47 +0000 (21:02 +0100)]
Merge pull request #396 from beretta42/fview2
fview - a simple 1bpp windows .bmp viewer for fuzix
Brett Gordon [Thu, 18 Aug 2016 17:43:56 +0000 (13:43 -0400)]
install test file, remove extra test files.
Brett Gordon [Fri, 8 Jul 2016 14:38:59 +0000 (10:38 -0400)]
more comments, made endian checking run-time
Brett Gordon [Wed, 6 Jul 2016 04:50:34 +0000 (00:50 -0400)]
add copyright, some commenting, lil bugs fixes
Brett Gordon [Wed, 6 Jul 2016 04:48:55 +0000 (00:48 -0400)]
add package file and man page
Brett Gordon [Wed, 6 Jul 2016 03:22:38 +0000 (23:22 -0400)]
now should work on different width screens. ?
Brett Gordon [Tue, 5 Jul 2016 15:15:00 +0000 (11:15 -0400)]
proper (slow) 1bpp mode + boarders + horrible buffer alloc.
Brett Gordon [Tue, 5 Jul 2016 15:11:25 +0000 (11:11 -0400)]
seems ok to add optimization
Brett Gordon [Sun, 3 Jul 2016 23:15:23 +0000 (19:15 -0400)]
fix some more colors/ 16 bpp / dither optional
Brett Gordon [Sun, 3 Jul 2016 19:41:20 +0000 (15:41 -0400)]
playing with palettes
Brett Gordon [Sun, 3 Jul 2016 16:50:19 +0000 (12:50 -0400)]
Init commit, in working state.
Tormod Volden [Sun, 14 Aug 2016 11:47:23 +0000 (13:47 +0200)]
6809: Build levee, sh and kernel with -fno-cprop-registers
Something goes wrong in gcc6809's cprop_hardreg pass if there are
exg,add,exg patterns around.
The kernel seemed fine up to now, but add the flag just to avoid
any undiscovered or future issues.
Remove the gratuitous() workaround from sh/blok.c
Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
EtchedPixels [Sun, 7 Aug 2016 23:52:37 +0000 (00:52 +0100)]
Merge pull request #395 from nealcrook/more-dw
More dw
Neal Andrew Crook [Sun, 7 Aug 2016 21:09:02 +0000 (22:09 +0100)]
Revise todo list and emulator start-up line.
Neal Andrew Crook [Sun, 7 Aug 2016 21:05:46 +0000 (22:05 +0100)]
change register assignment so that checksum is calculated properly.
This bug caused more weird behaviour than I can currently account
for. With the fix, it seems rock-solid.
Neal Andrew Crook [Sun, 7 Aug 2016 21:05:14 +0000 (22:05 +0100)]
whiespace consistency changes. No functional change.
Neal Andrew Crook [Sun, 7 Aug 2016 21:04:09 +0000 (22:04 +0100)]
sync up with version in coco and dragon areas (just a fix to a comment typo)
EtchedPixels [Sun, 7 Aug 2016 11:23:13 +0000 (12:23 +0100)]
Merge pull request #393 from atsidaev/zx_128_betadisk
zx128: Betadisk-128 driver and filesystem building definition
Alexander Tsidaev [Sun, 7 Aug 2016 10:45:16 +0000 (15:45 +0500)]
zx128: filesystem build config
Alexander Tsidaev [Sun, 7 Aug 2016 09:28:59 +0000 (14:28 +0500)]
zx128: Betadisk-128 driver
EtchedPixels [Tue, 2 Aug 2016 23:44:53 +0000 (00:44 +0100)]
Merge pull request #392 from atsidaev/touch_c_fix
touch.c: strerror(...) needs string.h