K. Lange
d88712e956
kernel: support syscall/sysret
2023-11-05 14:32:47 +09:00
K. Lange
cf04ba62e8
syscall: Change x86-64 syscall register ABI
2023-11-05 10:40:12 +09:00
K. Lange
613bc73dfd
kernel: properly track time in kernel when returning to userspace
2023-11-02 10:21:00 +09:00
K. Lange
55e873bd6b
x86_64: Fixup error codes for rare/new exceptions
2023-10-25 12:46:36 +09:00
K. Lange
afcf36de70
rpi: advertise correct default speed of miniuart
...
Does not implement setting the baud rate yet, but correctly
advertises that we set it to 921600 by default.
2023-10-09 14:05:36 +09:00
K. Lange
bda8c8d67b
pty: initial support for setting line control
2023-10-08 14:52:34 +09:00
K. Lange
4f36cac62c
kernel: get rid of unused interrupt_registers
...
process_t member was never set, but was referenced in ptrace
core-local member was set in x86-64 but never used
2023-10-06 07:46:11 +09:00
K. Lange
d45a0f2b13
virtio: name worker threads with square brackets
2023-07-22 23:17:39 +09:00
K. Lange
d4315dcb53
aarch64: don't set debug output as primary so we can print early log to framebuffer
2023-07-15 19:25:08 +09:00
K. Lange
7691b1691f
aarch64: don't print every time we do a dtb lookup
2023-07-15 19:24:21 +09:00
K. Lange
e9811a0708
virtio: handle queue overflow
2023-04-30 19:09:49 +09:00
K. Lange
ffb7bc36bb
aarch64: ensure module base address starts aligned
2023-04-18 22:45:19 +09:00
K. Lange
6d2d6edffc
boot: CONSOLE_FLAGS mb2 tag is optional
2023-04-11 17:35:30 +09:00
K. Lange
a4a73a0835
boot: Don't jump back to protected mode in EFI loader
2023-01-05 15:39:44 +09:00
K. Lange
940d82f0f4
terminal: Bitmap font can use uint8_t
...
The baked Deja Vu Sans Mono is 8 pixels wide, so we don't need
to use uint16_t to hold rows of pixel data. Also, splash-log
doesn't render text anymore (it defers to the kernel console),
so we don't need to include the font there any more.
2022-12-13 15:08:56 +09:00
K. Lange
bf2c51426d
kernel: proc->group may have exited
...
Handling of thread parents needs to be better... the initial thread
of a process can disappear, and with pid reuse it could even be
replaced by a different process... this needs to be cleaned up more.
2022-10-29 17:27:05 +09:00
K. Lange
ce625f2137
aarch64: enable userspace access to timer physical counter
2022-08-30 17:12:49 +09:00
K. Lange
1960006ec7
aarch64: dump regs on unknown fatal exceptions
2022-08-30 17:12:18 +09:00
K. Lange
ecbe7f6cbb
aarch64: don't be so intent on running driver bringup on cpu0
2022-08-30 17:11:42 +09:00
K. Lange
c554bf581c
kernel: Support SA_NODEFER
2022-08-06 18:24:06 +09:00
K. Lange
2906476825
kernel: Add sigaction
2022-08-06 18:03:39 +09:00
K. Lange
5ed3dc0640
virtio: Remove unused PCI cap parsing
2022-04-18 10:03:19 -04:00
K. Lange
f44a6bbe9a
aarch64: read rtc time again on qemu
2022-04-17 18:22:53 -04:00
K. Lange
9ecad97b64
virtio: these F keys should work too
2022-04-13 22:20:12 -04:00
K. Lange
71ce811a65
virtio: map command to left super
2022-04-13 18:18:13 -07:00
K. Lange
f144266bc5
aarch64: ac97 driver as module
2022-04-13 16:19:56 -07:00
K. Lange
7fbb7d64d7
aarch64: e1000 driver as module
2022-04-13 16:19:56 -07:00
K. Lange
bea3854b00
aarch64: (wip) support kernel modules
2022-04-13 16:19:56 -07:00
K. Lange
de09abf2dd
x86-64: fixup serial receive loop as this one was kinda bad
2022-03-22 14:13:21 +09:00
K. Lange
5caf43d90a
kernel: x86-64: copy over unmap improvements from aarch64
2022-03-16 08:39:33 +09:00
K. Lange
00075494e6
kernel: aarch64: initial pass at clearing upper page structures on unmap
...
This still has thread issues, but a lot of other things need to be
redesigned to correct them.
Particularly, we have a lot of callers that use `mmu_get_page` without
acquiring any locks on the page directory, so in a multithreaded process
one core may be trying to acquire pages while another is unmapping. In
the normal sbrk setup, this shouldn't happen, especially with the locks
in userspace around malloc, and it shouldn't happen with shm as we avoid
dealing with shm pages in the unmap right now, but from the raw API we
provide it's possible for a crafted program to run into it.
The whole memory management API really needs a redesign, much of what we
expose comes from toaru32, and still leaks details about page size into
other code that shouldn't care.
2022-03-16 08:19:36 +09:00
K. Lange
58a1b6e999
pthread: Remove magic thread exit; pthread_exit() is just syscall_exit
2022-03-12 18:46:28 +09:00
K. Lange
4bcd1b4b89
aarch64: apply e1000 permission changes
2022-03-07 18:08:54 +09:00
K. Lange
2700849ca2
aarch64: clear instruction cache on new pages when forking
2022-03-07 09:20:19 +09:00
K. Lange
4c2ec1f0a0
kernel: add a temporary 'munmap' sysfunc
2022-03-06 15:58:38 +09:00
K. Lange
49fd67bf2d
kernel: relocate some user virtual addresses
2022-03-06 15:58:08 +09:00
K. Lange
f778967edb
aarch64: disable lock debugging
2022-03-06 15:56:53 +09:00
K. Lange
cd04af0a75
aarch64: remove some debug messages
2022-03-04 22:08:05 +09:00
K. Lange
d932559d79
kernel: ensure we're saving/restoring all process state on signals
2022-03-04 22:04:11 +09:00
K. Lange
63fbbec928
kernel: Don't let things clear frames without holding the frame alloc lock
2022-03-04 22:00:37 +09:00
K. Lange
3852646174
kernel: Restart some signals, fix up suspend to work with new signal management
2022-03-04 18:39:38 +09:00
K. Lange
bbebc7c128
kernel: complete redesign of signal handling
2022-03-04 16:12:16 +09:00
K. Lange
b3f6728339
date: add rudimentary support for setting date/time
2022-02-25 21:16:30 +09:00
K. Lange
7d2beb443b
aarch64: expose dtb as a device file
2022-02-23 20:40:53 +09:00
K. Lange
b29edb0db3
rpi400: mini uart driver
2022-02-23 16:33:30 +09:00
K. Lange
94b86558d3
aarch64: cleanup, fix warnings
2022-02-23 10:27:47 +09:00
K. Lange
c96bae5e78
aarch64: temporary device drivers until modules are ready
2022-02-23 10:24:15 +09:00
K. Lange
5a6648a869
rpi400: try to clean up rpi stuff
2022-02-23 09:49:16 +09:00
K. Lange
814f5210ae
aarch64: cleanup some warnings
2022-02-22 17:30:17 +09:00
K. Lange
6c69d235cf
aarch64: implement a better spinlock
2022-02-22 09:53:55 +09:00
