| Rev |
Age |
Author |
Path |
Log message |
Diff |
| 4637 |
5442 d 19 h |
svoboda |
/trunk/kernel/genarch/ |
Implement data transmission to CUDA, use to enable autopolling. Now the driver works both in PearPC and Qemu -M g3beige. |
|
| 4626 |
5446 d 19 h |
svoboda |
/trunk/kernel/genarch/ |
Modify CUDA driver to state machine format, with one interrupt per byte transferred. |
|
| 4610 |
5451 d 21 h |
svoboda |
/trunk/ |
Kernel Mac ADB keyboard driver revived. |
|
| 4311 |
5524 d 4 h |
decky |
/trunk/ |
change the way how input devices are wired together according to ticket #44
(also the proposal http://lists.modry.cz/cgi-bin/private/helenos-devel/2009-March/002507.html) |
|
| 4251 |
5539 d 0 h |
decky |
/trunk/kernel/ |
ia32, amd64: do not reboot via a triple fault (which is nasty), but use the i8042 controller to trigger CPU reset |
|
| 4223 |
5542 d 1 h |
decky |
/trunk/kernel/ |
represent special keystrokes (cursor arrows, page up/down, delete, etc.) as appropriate Unicode characters
do not parse ANSI control sequences in kconsole, but in serial line driver |
|
| 4187 |
5552 d 0 h |
decky |
/trunk/ |
simplify glyph mapping function
draw invalid glyphs with a distinctive color |
|
| 4177 |
5552 d 2 h |
decky |
/trunk/kernel/genarch/ |
use Unicode font glyphs and mapping for framebuffer |
|
| 4148 |
5558 d 5 h |
decky |
/trunk/ |
as kernel little brother drivers are not needed anymore, the device numbers do not have to be correlated between kernel and uspace in any way
introduce new syscall sys_device_assign_devno() for generating system-wide unique device numbers for uspace |
|
| 4119 |
5563 d 7 h |
decky |
/trunk/ |
get back support for framebuffer and keyboard for sparc64 (generic) |
|
| 4114 |
5563 d 20 h |
jermar |
/trunk/ |
Fix one RAW dependency. |
|
| 4094 |
5563 d 22 h |
decky |
/trunk/kernel/genarch/ |
update for the new scheme of wiring device drivers to keyboard/serial modules |
|
| 4092 |
5563 d 22 h |
decky |
/trunk/kernel/genarch/ |
wire keyboard/serial module and device drivers in a more obvious way
(first a device driver is initialized and then it is connected to the generic module) |
|
| 4078 |
5564 d 4 h |
decky |
/trunk/kernel/genarch/ |
add dummy serial line input and output drivers |
|
| 4043 |
5569 d 0 h |
jermar |
/trunk/ |
Serial line module. |
|
| 4042 |
5569 d 0 h |
jermar |
/trunk/ |
Complete emancipation of kernel serial controller drivers (i8042, ns16550 and
z8530). Provide a common keyboard module for PC and Sun keyboards. The serial
line module is still to follow. |
|
| 4022 |
5572 d 19 h |
svoboda |
/trunk/ |
Move multiboot parsing to genarch/*/multiboot and adapt it for use with both ia32 and amd64. Multiboot info parsing now supported on amd64, too. |
|
| 4021 |
5572 d 20 h |
jermar |
/trunk/kernel/ |
There is no need to define ioport{8,16,32}_t types for each architecture separately. |
|
| 3963 |
5582 d 1 h |
jermar |
/trunk/kernel/ |
The kernel i8042 driver should be instance-neutral. |
|
| 3961 |
5582 d 1 h |
jermar |
/trunk/ |
Unbreak sparc64.
Due to the lack of userspace bus drivers for the z8530 and ns16550,
we are currently handling exactly one interrupt from these devices
in userspace. |
|
| 3960 |
5582 d 18 h |
jermar |
/trunk/kernel/genarch/include/kbd/ |
Remove extern declarations. |
|
| 3941 |
5583 d 19 h |
jermar |
/trunk/kernel/ |
Let IRQ claim methods take irq_t * instead of the instance argument. |
|
| 3940 |
5583 d 23 h |
decky |
/trunk/kernel/ |
make hw_area API more generic
this allows mapping of EGA VRAM on ia32/amd64 |
|
| 3938 |
5584 d 0 h |
jermar |
/trunk/kernel/genarch/ |
Embed the ns16550 irq structure into the ns16550 instance structure. |
|
| 3934 |
5584 d 16 h |
jermar |
/trunk/kernel/ |
Improve the ns16550 driver.
- The driver support multiple instances.
- It still remembers the last registered IRQ in a global structure (cannot be
easily fixed now)
- Was converted to use PIO directly |
|
| 3932 |
5584 d 18 h |
jermar |
/trunk/kernel/ |
Kill arch/drivers/ega.h. Provide one-stop shopping
place for architectures with ia32 legacy I/O space. |
|
| 3931 |
5584 d 18 h |
jermar |
/trunk/kernel/genarch/ |
Improve the i8042 kernel driver (phase I).
- Convert the code to be instance-aware
- Still using hardcoded initialization from legacy IO address
- Use elegant i8042 register map for PIO |
|
| 3930 |
5584 d 18 h |
jermar |
/trunk/kernel/ |
Adapt most of the kernel to ioport8_t, ioport16_t and ioport32_t types. |
|
| 3927 |
5584 d 21 h |
jermar |
/trunk/kernel/genarch/ |
Nuke i8042_poll(). |
|
| 3907 |
5588 d 6 h |
decky |
/trunk/kernel/ |
remove forgotten #ifdef arch |
|
| 3906 |
5588 d 17 h |
jermar |
/trunk/kernel/ |
IRQ handlers are using one superfluous argument and an unused elipsis.
On the other hand, IRQ claim functions would need to be passed the instance
argument. |
|
| 3902 |
5589 d 1 h |
jermar |
/trunk/kernel/ |
Introduce a more platform-neutral name for programmed I/O.
The new API looks like pio_read_n() or pio_write_n(), where n is 8, 16 or 32.
The old API (i.e. inb(), inw(), inl(), outb() outw(), outl()) may have made
some people think that the interface is only to be used with the separate I/O
space. That's not the case. This API is to be implemented on all platforms
so that we can finally have really generic kernel device drivers. |
|
| 3876 |
5591 d 20 h |
decky |
/trunk/kernel/genarch/ |
add BGR 888 visual, remove FB_INVERT_ENDIAN |
|
| 3779 |
5631 d 2 h |
jermar |
/trunk/kernel/ |
Only one kernel EGA driver to rule them all. |
|
| 3710 |
5649 d 19 h |
decky |
/trunk/ |
resurrect kernel logo |
|
| 3707 |
5649 d 21 h |
decky |
/trunk/ |
kconsole is optional
kernel & uspace framebuffer rewrite with speedups (some things are slightly broken yet) |
|
| 3704 |
5649 d 21 h |
decky |
/trunk/kernel/genarch/include/kbd/ |
add header |
|
| 3679 |
5659 d 16 h |
rimsky |
/trunk/ |
Added forgotten stuff related to framebuffers with offsets. Got rid of VISUAL_SB1500_PALETTE. |
|
| 3672 |
5660 d 20 h |
jermar |
/trunk/ |
Merge sparc branch to trunk. |
|
| 3655 |
5666 d 20 h |
jermar |
/trunk/kernel/ |
Add additional members to the irq_t structure so that an interrupt-driven driver
does not need to know how to clear the level interrupt. The z8530 was modified
in this way and is much more generic. The ns16550 driver has also been modified,
but awaits testing. The sparc64 interrupt mapping and dispatch code is now using
the new info and calls the clear-interrupt-routine itself. |
|