950610e1eb
Added a kernel worker thread for use when the kernel goes multithreaded.
2012-08-04 18:35:23 +02:00
ee73aa7783
Added a library of functions to simulate atomic operations on memory.
2012-08-04 18:35:23 +02:00
dafe1c499d
Thread secured pipe class.
2012-08-04 18:35:23 +02:00
66d7234ab1
Thread secured logterminal class and implemented VEOF.
2012-08-04 18:35:23 +02:00
bf07674d7a
Thread secured the keyboard driver.
2012-08-04 18:35:23 +02:00
af015491db
sortix/io.cpp is now ready for new system call semantics.
2012-08-04 18:35:23 +02:00
4810ccae39
Added warning in case the laws of logic changes.
2012-08-04 18:35:23 +02:00
2e3843ffce
Fixed incorrect x86-family protection flags.
2012-08-04 18:35:23 +02:00
49ad293d1d
Thread secured the physical page allocator.
2012-08-04 18:35:23 +02:00
459a1b2b3f
Thread secured initfs.
2012-08-04 18:35:23 +02:00
f00c8b3d63
Thread secured the kernel device class.
2012-08-04 18:35:23 +02:00
38349da082
Thread secured the COM driver.
2012-08-04 18:35:22 +02:00
f3f33e22e7
Thread secured ATA driver.
2012-08-04 18:35:22 +02:00
eb4f179330
Added DEBUG_KERNEL option to kernel makefile.
2012-08-04 18:35:22 +02:00
199fec6674
Hack: ReadParamString supports "STOP" because of stdarg bug.
2012-08-04 18:35:22 +02:00
14d709c136
Added support for gcc 4.7.1.
2012-08-04 18:33:47 +02:00
78300931ba
Workarounds for older gcc releases such that endian.h works.
2012-08-02 16:24:51 +02:00
86f8662a4e
The VGA font is now available as /dev/vga.
2012-08-01 13:07:47 +02:00
f3532081aa
Added a filesystem utility class for providing a fixed buffer.
2012-08-01 13:07:31 +02:00
8dc5955f5e
Wrote a driver for the Bochs VBE Extensions (BGA).
2012-07-30 19:03:48 +02:00
cb5a242dfc
Added a text buffer that works with graphical linear frame buffers.
2012-07-30 19:02:05 +02:00
78f0c6c094
The video framework is now exposed as /dev/video.
...
This provides easy user-space access to the framebuffer.
2012-07-30 19:00:24 +02:00
1ce55af846
Created framework for video drivers.
...
This supports dynamic loading and unloading of graphics drivers, mode
switching and detection and flexible kernel access to the framebuffer.
2012-07-30 19:00:24 +02:00
d27a2bdcbf
Better detection of panicing while panicing.
2012-07-30 18:59:13 +02:00
1041823d42
Fixed VGA text buffer rendering non-ascii text incorrectly.
2012-07-30 18:56:47 +02:00
3b0f165c4f
Fixed bad declaration of MapPAT in x86-family/memorymanagement.h.
2012-07-30 18:56:02 +02:00
19b5451f3b
Refactored the PCI code to become a library of utility functions.
2012-07-29 23:45:54 +02:00
5f6ca5e729
Added a kernel endian.h header for easy conversion.
2012-07-29 23:45:54 +02:00
f390418515
Preallocated kernel address space for video memory.
2012-07-24 20:08:51 +02:00
494636b8eb
Added some filesystem utility classes.
2012-07-24 19:56:33 +02:00
143120d160
Added tcgetwinsize(2) for determining terminal resolution.
...
Unfortunately this area is not standardized by POSIX. Linux uses an ioctl
which is not that bad, but I'd like to have a designated function. I'm not
sure if this facility is powerful enough and whether it should be improved.
Also note that I use a struct winsize as on Linux, but I use size_ts instead
for the heck of it. Perhaps I should use another name for the struct.
2012-07-24 18:43:34 +02:00
f5c4b64aff
The console can now be rendered to any text buffer.
...
The console renderer now renders to a text buffer, which can be implemented
on any device, whether it's the VGA text buffer or a bitmap graphics device
with font rendering. This replaces the older code that could only render to
a VGA framebuffer and where the input parsing was tightly coupled with the
device rendering phase.
2012-07-24 14:48:14 +02:00
8c5ab54c9b
Updated refcount.cpp to a newer coding convention.
2012-07-23 00:05:31 +02:00
3907e14cb8
Updated vgaterminal.cpp to newer coding conventions.
2012-07-23 00:05:31 +02:00
d75a7145ef
Updated vga code to newer coding conventions.
2012-07-23 00:05:31 +02:00
b2814db927
Added support for getting a copy of the VGA Font.
...
This will be useful for providing a text-mode like environment (console)
after having switched to graphical mode where the system needs a font.
2012-07-22 15:51:38 +02:00
476ba278da
Improved VGA scroll performance using memcpy(3).
2012-07-17 16:23:00 +02:00
ebc0b064c3
Added a function to get the current address space pointer.
2012-07-06 17:18:07 +02:00
d59c0cb3ed
getpagesize(2) now call Sortix::Page::Size().
2012-07-06 17:18:07 +02:00
22990b77b8
Refactored the internal kernel memory management API.
...
It is now permission-oriented, not just user/kernel oriented.
Added <sys/mman.h> with nice PROT_{READ,WRITE,EXEC,FORK} constants.
2012-07-06 17:18:07 +02:00
ec5fa92761
Programmers can now redirect what the errno macro refers to.
2012-07-06 17:18:06 +02:00
e7baf6a4b0
Updated the kthread compatibility layer with signal support.
2012-07-06 17:17:45 +02:00
c39473157a
Implemented a new initrd format with better inode support.
...
This enables useful features such as directories, CRC32 checksums, and other
useful features. The initrdfs in the kernel is now hooked up against the new
API, although the kernel's current limited FS support is a problem for now.
To work around that, directories are not supported at runtime, although the
internal API understands them wonderfully. This will be fixed when the
kernel gets a real VFS.
2012-07-02 17:40:52 +02:00
2046482e77
Added support for MSRs, PAT and MTRRs.
...
This provides control over the caching of memory, which makes write-combined
IO possible. Graphics drivers can use this to transfer data at a much higher
rate to the video memory.
The implementation is a bit hacky but it'll do for now. It provides enough
support for the experimental VBE driver to work on the real computers I
tested it on, even if the BIOS uses screwed up default MTRRs.
The virtual memory layer now automatically uses the PAT feature if available
but in a backwards compatible manner and otherwise just tries to approximate
PAT features if they are asked for.
2012-07-01 00:55:36 +02:00
b650c9a8f0
Added a O_ACCMODE flag for GNU compatibility.
2012-05-28 22:52:11 +02:00
00869bcdc1
The PIC can now be deprogrammed and reprogrammed.
2012-05-26 14:00:24 +02:00
bd489eb73a
Forward compatbility with the kthread branch.
2012-05-04 13:13:44 +02:00
92c5533820
Improved the implementation of the exec* functions.
2012-04-30 21:10:02 +02:00
93abeda32e
Fixed buffer overflow in devfs.cpp.
2012-04-14 22:03:55 +02:00
3b63a70191
Added Interrupt::SetEnabled(bool).
2012-04-13 17:42:21 +02:00
f59b53ddce
Ported x64 interrupt assembly to the GNU assembler.
2012-04-13 17:34:17 +02:00
8cbf9ff8f0
Added Interrupt::IsEnabled().
2012-04-11 15:46:32 +02:00
36ff6c7f96
seek(2) now correctly rejects a bad whence value.
2012-04-10 13:20:33 +02:00
6367a2352e
Added sforkr(2) that controls the child registers as well.
...
sfork(2) now calls sforkr(2) with the current registers.
This will prove useful in creating threads, where user-space now can fully
control what state the child will start in. This is unlike the Linux clone
system call that accepts a pointer to the child stack; this is more powerful
and somehow simpler. Note that this will create a rather raw thread; no
thread initization has been done by the standard thread API (when it is
implemented), so this feature shouldn't be used by programmers unless they
know what they are doing.
fork(2) now calls sfork(2) directly. Also removed fork(2) and sfork(2) from
the kernel as they are done using sforkr(2) now. So technically they aren't
system calls right now, but that could always change.
2012-04-05 23:00:47 +02:00
6f36ecf0b3
execve(2) now pushes envp to the new stack and sets up registers.
...
This fully implements environmental variables over exec.
2012-04-04 01:49:14 +02:00
05b29ce25a
Renamed rfork(2) to sfork(2) to avoid compatibility issues.
...
sfork is much like rfork except sharing is default for everything.
Eventually, I'll make a rfork(3) wrapper function around sfork(2) to
provide compatibility to BSD programs.
I don't like Linux clone(2): that's some messy function.
2012-04-04 00:29:25 +02:00
95a088fec5
Added a dummy implementation rfork(2) equal to fork(2).
...
Note that in my implementation, you share per default, unless you ask.
2012-04-02 16:30:13 +02:00
b3b1f6af66
Make sure that the right bits.h file is always included.
2012-03-27 16:36:55 +02:00
554b2e44cc
Added a Page::IsAligned inline function.
2012-03-26 16:46:26 +02:00
c62eb09cdc
Added stubs for pread(2) and pwrite(2).
...
These are not implemented yet because the current kernel design is bad.
However, I need the stubs for other code.
2012-03-24 15:23:07 +01:00
1ba4417b19
Removed dead ancient code.
2012-03-22 01:04:27 +01:00
db79994e64
Refactored all the sortix headers into a include directory.
...
Also got rid of trailing white space. That corrupted .git/.
Big ass-commit because of recovered .git directory.
2012-03-22 00:52:29 +01:00
ea11c20689
Added a way for discover refcount of Refcounted classes.
2012-03-21 16:20:41 +01:00
0f099c65ca
The virtual memory fork code can now handle ENOMEM conditions.
...
(forkbombs now don't panic the system!)
2012-03-19 02:39:11 +01:00
c9ab4fa6b4
The sortix vga terminal now can delete tabs and newlines properly.
...
A nice little vga attribute array takes care of remembering what indexes
actually contains a character.
2012-03-18 18:06:13 +01:00
14a9149f17
Added detection for an obscure bug to let users know I know about it.
2012-03-18 17:36:31 +01:00
d564603460
Fixed kernel stack overflow and premature EOF in the unix pipe code.
...
This usually caused the system to lock up when much data was transferred
over pipes, for instance: $ cd /bin ; cat cat | cat
2012-03-18 16:14:26 +01:00
c7c0fc603a
Fixed initial kernel stack being wrong size.
2012-03-17 20:17:12 +01:00
969c0e8048
Cleaned up sortix/kernel.cpp and it now uses the new coding style.
...
All kernel.cpp code related to jssortix was removed as it is not supported
any longer.
2012-03-17 20:11:37 +01:00
4f3e22140c
Fixed x64 memory leaks upon process termination.
2012-03-17 18:14:57 +01:00
0364ce6f55
Added a new COM Port driver.
...
Any detected COM ports available as /dev/comN.
It currently utilizes AGAIN to do polling in user-space. This prevents it
from locking up the whole system and makes it respond to the SIGINT hack.
There is also a more reliable and faster polling-blocking mode, but it locks
up the entire system.
The main interrupt mode is broken, perhaps by a bug in VirtualBox.
2012-03-17 15:48:42 +01:00
1ff0321400
Added a stub for Syscall::Yield() which is used by broken code.
...
The whole system call interface is currently stupid.
2012-03-17 15:14:42 +01:00
798b421d16
Refactored devfs so new devices can easily be added.
2012-03-17 15:11:35 +01:00
5ec4e33196
Removed the last reference to the old kernel HTTP server.
2012-03-16 16:56:16 +01:00
2cdeb575f8
Renamed EACCESS to EACCES.
...
Stupid POSIX.
2012-03-12 01:43:00 +01:00
5b916b5f63
Added F_SETFL and F_GETFL to fcntl(2).
...
This is a bit of a hacky implementation.
2012-03-11 20:29:00 +01:00
868ed43e05
Removed sortix/mxfs.
...
It's unbelievable that this piece of dead code survived 8 months of
development on git. It doesn't even work, despit the comment that it
does.
2012-03-11 18:07:01 +01:00
f01d7951c1
stat(2) now supports directories.
...
A bit hackily, though.
2012-03-11 17:56:20 +01:00
6e536ba8f9
Fixed error codes in sortix/fs/devfs.
2012-03-11 17:55:52 +01:00
245e8bf0e3
Fixed error codes in sortix/fs/initfs.
2012-03-11 17:55:23 +01:00
f45cf8327a
Forgot to enable the /bin /dev hack in sortix/fs/ramfs.cpp.
2012-03-11 17:30:13 +01:00
cadac5ce2a
Moved the /bin and /dev hack into the kernel.
...
This makes ls seem less hacky.
2012-03-11 17:25:32 +01:00
8a9a0c58ea
Added kernelinfo(2), which reads a kernel information string.
...
Currently it lets you query the name of the kernel, its version, and the
build timestamp of the kernelinfo.cpp file.
2012-03-07 18:04:59 +01:00
6bcb3d7384
Added stat(2) and fstat(2) in the kernel.
...
It's a bit hacky, but it works.
lstat(2) currently redirects to stat(2).
2012-03-06 13:10:59 +01:00
4b2c22d480
Added access(2).
2012-03-05 15:46:23 +01:00
d3c4df72b1
Added st_{a,c,m}time to struct stat.
2012-03-04 22:25:47 +01:00
aa72efdb2c
Added <sortix/timeval.h> to the list of public headers.
2012-03-04 22:18:53 +01:00
c0a02248da
Added <sys/time.h>, struct timeval and gettimeofday(3) stub.
2012-03-04 22:05:52 +01:00
5e75f5c428
Added fcntl(3) with FD_GET and FD_SET.
2012-03-04 21:36:40 +01:00
75b14aa821
Refactored the descriptor table and added FD_CLOEXEC and FD_CLOFORK.
...
open(2) now also respects O_CLOEXEC and O_CLOFORK.
2012-03-04 21:06:49 +01:00
777fc04682
Added <time.h>, clock_t, and a stub for clock(3).
2012-03-04 18:38:23 +01:00
ac7e55ffbd
Added st_dev to struct stat.
2012-03-04 16:48:24 +01:00
db7781c54a
Install <sortix/*.h> headers in sysroot.
2012-03-04 16:04:12 +01:00
0ed0082070
Added execv(3) and execve(3).
...
Removed the older libmaxsi system call.
2012-03-02 15:00:11 +01:00
93a9ee334d
Removed a debug ifdef that isn't used.
2012-03-02 14:25:48 +01:00
cfbbf67247
make CALLTRACE=1 to enable printing primitive calltraces upon panic.
...
This won't work with optimizations turned fully on.
Added protection against double panics.
2012-03-02 14:08:25 +01:00
6dd0e586ff
Added protection against bad addrspace vars and bad Page::Put() calls.
...
This will offer protection against the bug fixed in the previous commit.
2012-03-02 13:51:03 +01:00
5e5991ce19
Fixed x86 addrspace destroy code calling Page::Put with garbage.
2012-03-02 13:45:33 +01:00
a7c38cb997
Improved bad currentdir variable panic message.
2012-03-02 00:09:08 +01:00
9c193777a9
Pipes now return 0 (EOF) if no data and no fds can write.
2012-03-01 14:31:12 +01:00
b4f47f0f79
Split descriptor_tables.cpp into a gdt.cpp and idt.cpp.
...
This was about time, since descriptor_tables was a really bad name!
2012-03-01 00:15:28 +01:00
5162a12d03
Further cleanup and refactoring of the interrupt code.
2012-02-29 23:03:40 +01:00
0e48b23429
Refactored the interrupt code to make it cleaner and more flexible.
...
Added support for hooking directly into an interrupt with your own
interrupt handler.
2012-02-29 15:40:30 +01:00
7ebed6dd6a
sortix/io.cpp now uses <sortix/seek.h>.
2012-02-24 22:02:01 +01:00
cbdf54fcdd
Moved SEEK_SET, SEEK_CUR, and SEEK_SET to <sortix/seek.h>.
2012-02-24 17:34:50 +01:00
5de83df6ec
Unified constants from sortix/filesystem.h and <fcntl.h> in <sortix/fcntl.h>.
2012-02-23 13:15:40 +01:00
03273d0076
Added stubs for stat(2), and fstat(2).
2012-02-22 00:30:34 +01:00
edd806fc74
Added a Refcounted class, useful for refcounted objects.
2012-02-21 20:12:52 +01:00
74dab2cbca
Cleaned up the Sortix makefile.
2012-02-13 13:16:43 +01:00
fec176fa15
Refactored the libmaxsi/ directory.
...
C and C++ files are now kept together and so are the mxmpp declarations.
Header files are now stored in include/ and mxmpp'd into preproc/.
All other code now -I ../libmaxsi/preproc.
And other stuff to make this happen, including refactoring Makefile.
2012-02-12 13:20:53 +01:00
4e29f2b907
Fixed spelling error in sortix/*/bits.h.
2012-02-11 21:20:49 +01:00
ead53567a1
Made the compile process a little less verbose.
2012-02-11 19:06:00 +01:00
ed68db03fb
Replaced SORTIX_EXTENSIONS macro with _SORTIX_SOURCE macro.
...
<features.h> declares _SORTIX_SOURCE if no conflicting macros are
declared, such as _GNU_SOURCE.
Fixed g++ automatically declaring _GNU_SOURCE, but Sortix isn't GNU.
Replaced SORTIX_UNIMPLEMENTED macro with __SORTIX_SHOW_UNIMPLEMENTED.
2012-02-11 18:51:55 +01:00
028867ab9d
Removed the unused and deprecated old <libmaxsi/sortix-keyboard.h> API.
...
This has been entirely replaced by stdin and <sys/termmode.h>.
2012-02-10 13:46:26 +01:00
c4264d9c2a
Ported the user-space programs to use stdin for keyboard access.
...
The old keyboard API is hereby deprecated and unused.
2012-02-10 13:46:26 +01:00
fa9c7007b4
Implemented a terminal that reads from keyboard and writes to kernel log.
...
This terminal will provide the keyboard data in a couple modes as /dev/tty.
2012-02-10 13:28:28 +01:00
4ba906a157
Added a class for handling terminal linebuffering.
2012-02-10 13:28:28 +01:00
133fb9871c
Fixed buggy and broken KBKEY_ENCODE and KBKEY_DECODE macros.
2012-02-10 13:28:28 +01:00
5fde8e13ed
Blocking systemcalls now return EBLOCKING instead of EWOULDBLOCK.
...
EWOULDBLOCK is now used when it wanted to block, but didn't.
2012-02-10 13:28:28 +01:00
9bcfdad174
Added protection against running terminated threads.
...
A bool is set when a thread is terminated, which may help detect it.
A cached version of the thread's pid is also kept around.
And lastly, the thread is unsubscribed from events upon destruction.
2012-02-10 13:27:11 +01:00
f6f0d24b5c
Added TERMMODE_NONBLOCK.
...
This allows user-space to query whether data is available.
2012-01-22 18:38:46 +01:00
ecc3114f2a
Refactored the system to use the new Terminal interface.
...
This will allow development of a better terminal providing stdin.
Added new system calls settermmode(2) and gettermmode(2) declared in
<sys/termmode.h>. They allow querying and changing the current mode of
terminals (enabling raw keyboard data, signal handling, line buffering,
UTF-8 encoding stdin, and more). However, all that is unsupported by the
current terminal device driver.
Added KBKEY_ENCODE and KBKEY_DECODE macros to <sys/keycodes.h> which allows
encoding the kbkey format in UTF-32 characters.
2012-01-22 16:48:57 +01:00
ead0e1523f
Refactored the kernel keyboard API, but kept system calls compatible.
...
Caps lock now works as caps lock, not as shift lock.
This new design will allow implementing a working tty, such that stdin is
the only way to access the keyboard, instead of the current hacky way of
using a special system call to read from the keyboard.
Added a new system header file <sys/keycodes.h> defining the constants for
every key on the keyboard. This will be used in future APIs.
The main change is to split the keyboard driver into a class that reads
from the keyboard, while another class handles the translation into
printable characters (if possible). This allows a terminal driver based
on logical key presses and printable characters, instead of a terminal
driver based only on unicode-ish codes.
2012-01-22 15:53:50 +01:00
16dd39e467
Added a nice Log::PrintData function to the kernel.
...
This allows writing binary data to the kernel log.
2012-01-22 02:12:52 +01:00
1f3f85b609
Added a simple utf8 encoder in the kernel.
2012-01-22 00:56:58 +01:00
83aeec2514
Multiple threads can now wait on the same pipe.
2012-01-18 17:19:08 +01:00
ba2cb5103a
Made Device::IsType 'const', because it is.
2012-01-18 01:11:17 +01:00
35fc0f6dfe
Removed the old unused interrupt 177.
2012-01-15 19:00:05 +01:00
93fb66ff90
Refactored Interrupt handlers to provide a user pointer.
2012-01-15 18:32:42 +01:00
c8c34d3cdd
readdirents(2) now properly reports ERANGE.
2012-01-15 00:43:13 +01:00
f476a82498
Fixed idention error in ramfs.
2012-01-14 23:19:31 +01:00
c94f6b64c3
readdirents(2) now returns ERANGE if insufficient storage space
...
was provided, but that the needed size could be copied to d_namelen.
2012-01-14 23:06:18 +01:00
8ef439b328
Fixed NULL-pointer in SysReadDirEnts.
2012-01-14 22:57:17 +01:00
56084556bb
Added stubs for truncate(2) and ftruncate(3).
2012-01-14 16:37:21 +01:00
0519af33ee
Added stubs for rmdir(2) and mkdir(2).
2012-01-14 16:25:28 +01:00
ad200ffa91
Added getpagesize(2) with the new return type size_t.
...
This function is non-standard and size_t is better than int here.
2012-01-08 14:58:57 +01:00
d3ad36b181
Made detected ATA devices available as /dev/ataN block devices.
2012-01-08 14:21:36 +01:00
a3a2226bb5
Implemented a simple ATA PIO Mode driver that can read and write.
...
Read operations are enabled by default, but you must set DISKWRITE=1 in
makeflags before write operations are permitted. This protects against
accidentally corrupting the existing filesystems on the system.
2012-01-08 14:20:39 +01:00
4e0f57f4d7
DevFileWrapper now understands and respects O_TRUNC.
2012-01-08 14:15:51 +01:00
648b324385
Added lseek(2).
2011-12-26 23:12:12 +01:00
36b01eb2d3
Fixed the horrible 'nofoo' bug!
...
When compiled with gcc 4.6.1, 32-bit Sortix would triple fault during
early boot: When the TLB is being flushed, somehow a garbage value had
sneaked into Sortix::Memory::currentdir, and a non-page aligned (and
garbage) page directory is loaded. (Triple fault, here we come!)
However, adding a volatile addr_t foo after the currentdir variable
actually caused the system to boot correctly - the garbage was written
into that variable instead. To debug the problem, I set the foo value
to 0: as long as !foo (hence the name nofoo) everything was alright.
After closer examination I found that the initrd open code wrote to a
pointer supplied by kernel.cpp. The element pointed to was on the
stack. Worse, its address was the same as currentdir (now foo).
Indeed, the stack had gone into the kernel's data segment!
Turns out that this gcc configuration stores variables in the data
segment in the reverse order they are defined in, whereas previous
compilers did the opposite. The hack used to set up the stack during
early boot relied on this (now obviously incorrect) fact.
In effect, the stack was initialized to the end of the stack, not
the start of it: completely ignoring all the nice stack space
allocated in kernel.cpp.
I did not see that one coming.
2011-12-25 03:41:59 +01:00
7bc1fa259e
Made Sortix compatible with gcc 4.6.1.
...
This commit fixes some instances of uninitialized memory.
In addition, the bootstrap tables for x64 are moved around a bit,
in this awful game of placing stuff where it won't collide with grub.
2011-12-25 00:10:56 +01:00
ffe3cc49f2
Now counting memory usage in a more solid manner.
2011-12-23 16:45:07 +01:00
ba9b715c59
Added a global variable at the end of the kernel to detect its size.
...
This lets the kernel use any memory not directly used by it or the
init ramdisk. Although, now we test whether the kernel fits into
the identitymapped area. It can't really grow down there, unless it
wants to collide with user-space. Instead, modules and the like
(when they are invented), should be put in the upper memory. Or in
their own user-space process, yay, microkernel!
2011-12-23 13:09:09 +01:00
0515111314
The initial ramdisk is now mapped onto a special location.
...
This fixes issues where it did not fit into the first few MiB,
or that GRUB loaded it someplace weird.
The kernel heap is now also protected against growing into the
ramdisk and the kernel stack.
2011-12-22 14:13:18 +01:00
a623b1b07e
Fixed registering memstat(2) multiple times.
2011-12-22 12:59:48 +01:00
c8c08e7b90
Moved the make-color-red code into DoWelcome in kernel.cpp.
2011-12-22 12:56:29 +01:00
0f5864602f
Fixed bad filename written in elf.cpp.
2011-12-16 15:35:45 +01:00
7cd28f097c
Program loader now sets the correct program segment type.
2011-12-16 15:33:12 +01:00
acf1eebc98
Added user-space memory allocation.
2011-12-16 13:24:49 +01:00
aae0ed3092
The cursor is disabled when displaying the VGA under JSVM.
2011-12-05 22:13:51 +01:00
2c728fc1f1
VGA now sort-of works under serial-connections.
2011-12-05 21:44:14 +01:00
9e5b9e3767
Hacks to make the JSVM serial driver work better.
2011-12-05 21:14:50 +01:00
cdb873271d
Increased hard-coded maximum JSVM initrd size to 2 MiB 512 KiB.
2011-12-05 19:43:27 +01:00
7e55f325fc
Don't check the initrd on JSVM.
2011-12-05 19:36:15 +01:00
854d9b171a
uptime(2) now reports usecs instead of msecs.
2011-12-04 21:27:21 +01:00
3b8bf9419d
Removed obsolete functions to read from the initrd.
...
Use the initrd filesystem instead.
2011-12-04 15:38:02 +01:00
b0859c6d92
usleep'ing for 0 usecs simply causes a context-switch.
2011-12-02 22:37:17 +01:00
f8129a17b2
Changed the build system for 64-bit quite a bit.
...
The kernel is now compiled 100% as 64-bit code and converted to ELF32.
2011-12-01 23:06:34 +01:00
8e8bb15a1b
Improved corruption panic message in initrd.cpp
2011-12-01 23:03:00 +01:00
b0884584a1
Added a very simple checksum field to the init ramdisk.
...
This uncovers a bug where the bootloader may corrupt the initrd on x64.
This is probably related to our kernel64 hack..
2011-12-01 21:44:11 +01:00
2faafd3f99
Stack is now aligned in 64-bit Sortix threads.
2011-12-01 21:43:35 +01:00
7304c8f528
Fixed uninitialized memory bug in the keyboard driver.
...
This often prevented the letter 'c' from working in x64.
2011-12-01 17:30:14 +01:00
6562da4092
waitpid(2) now returns ECHILD on error.
2011-12-01 14:54:19 +01:00
cf53e4a020
Fixed 64-bit resumed system calls.
2011-12-01 14:29:49 +01:00
c0c20860ed
Lots of improvements to 64-bit Sortix.
...
Fixed 64-bit-ness bug in BSR() and BSF().
Added 64-bit system call stubs in libmaxsi.
Added a Elf64 program loader.
Fixed uninitialized memory bug in the scheduler.
x64/boot.s now takes care of user-space memory permissions.
Fixed bug in x64/syscall.s
That's right. The system now boots in 64-bit mode.
It is horribly unstable, though.
2011-12-01 10:45:44 +01:00
f460c4abec
Added support for 64-bit interrupts.
2011-11-30 23:30:14 +01:00
f0ec4ffacc
Forgot to push %rbp in x64/syscall.s
2011-11-29 11:45:00 +01:00
0b1c2a77c9
x64 now compiles again and runs on real hardware.
...
A nasty bug was fixed that caused triple faults on systems with > 1 GiB RAM.
2011-11-29 01:31:54 +01:00
8c146f14c0
Added uptime(1).
2011-11-28 16:29:53 +01:00
f232a054d3
Added O=-ON and PANIC_SHORT=1 variables to the makefile.
2011-11-28 15:25:00 +01:00
cd936886e6
Writing to /dev/vga hides the cursor until a write to /dev/tty.
2011-11-27 12:03:33 +01:00
4f3a7230b0
Destroying the addressspace after exit prevents leak of at least 16 KiB.
2011-11-27 12:02:34 +01:00
06909c6df9
Fixed panic when executing non-existent file.
2011-11-27 00:20:23 +01:00
bd1b1fe3bc
Added isatty(2), which is used by editor.
2011-11-26 21:00:40 +01:00
3f50a335bb
devfs now correctly manages sizes of files.
...
This fixes a regression caused by the RAM fix a few commits ago.
2011-11-26 21:00:37 +01:00
ce9787a439
Greatly reduced RAM usage in ramfs.
2011-11-26 20:25:25 +01:00
9152443816
Added memstat, a program for displaying memory usage.
2011-11-26 20:14:57 +01:00
b6a0fd0374
Removed deprecated VGA API and moved it to /dev/vga.
2011-11-25 13:38:31 +01:00
724305f3a7
Merge commit '7a233f2938f4e28d61504853e2f6daf4904aab33'
...
Conflicts:
sortix/fs/ramfs.cpp
2011-11-24 15:38:45 +01:00
e72d086a8f
Disallow / in filenames in ramfs.
2011-11-23 17:51:18 +01:00
5bf22b2c4a
ramfs doesn't allow opening files as directories.
2011-11-23 14:14:59 +01:00
2b032b0414
Initial signal support. Please squash improvements into this commit.
2011-11-23 00:19:09 +01:00
46c0cc6a12
-1 is no longer an allowed system call.
...
This treats a nasty bug where forking and exiting before the child
causes the kernel to panic, because the syscall -1 is run, which
just contains a trash value that happened to be 0.
The cause is still unknown. Might be the scheduler.
This won't help me sleep at night.
2011-11-22 22:08:34 +01:00
b15763b2de
Fixed descriptors not being reset by SysExit.
...
Indirectly, this fixes the snake; snake; panic bug.
2011-11-22 19:12:04 +01:00
e234e0a2d4
execve(2) now sets errno=EACCESS on non-buffer files.
2011-11-22 18:21:01 +01:00
9deb183786
The ELF loader now uses errno.
2011-11-22 18:06:40 +01:00
1b56d01f17
Improved error codes in devfs.
2011-11-22 17:56:58 +01:00
9f35df813e
Implemented errno(3), added support in both kernel and utils.
2011-11-22 17:26:47 +01:00
8678ad6444
/ is no longer a valid filename under ramfs.
2011-11-22 15:33:47 +01:00
2c3fb60d52
execve(2) now works relative to current dir.
2011-11-22 15:27:58 +01:00
e8fb8d885b
execve(2) can now load programs from the filesystem.
...
Previously it was restricted to only the ramdisk.
2011-11-22 14:02:33 +01:00
324a9a1a22
Fixed bug in initfs causing panics.
2011-11-22 12:44:23 +01:00
55240fb3f5
Added devfs, with files /tty and /null and mounted it at /dev.
2011-11-21 21:49:13 +01:00
91a97a2667
Mount-points are now handled properly within the kernel.
2011-11-21 19:39:13 +01:00
8b2b52b9f6
Added unlink(2) and rm(1).
2011-11-21 19:01:56 +01:00
d4590cefa1
Added chdir(2), getcwd(2), which mxsh and ls now uses.
2011-11-21 12:19:57 +01:00
eb035d6a4a
Fixed tripple-fault when creating a new frame, after the current was gc'd.
...
We'll need to get rid of that awful VGA sometime soon.
2011-11-21 00:27:10 +01:00
313079483a
Missing files from the previous two commits. Ooops!
2011-11-21 00:19:55 +01:00
0b90ab534f
ramfs now supports O_TRUNC.
2011-11-21 00:12:20 +01:00
Jonas 'Sortie' Termansen