| Rev |
Age |
Author |
Path |
Log message |
Diff |
| 3985 |
5607 d 18 h |
svoboda |
/trunk/kernel/arch/ia32/src/boot/ |
Rewrite rest of multiboot info parsing to C. |
|
| 3984 |
5607 d 18 h |
svoboda |
/trunk/kernel/ |
The ia32-specific boot code is written completely in assembly for no reason. Start fixing this. For now rewrite the copying of multiboot info to C. |
|
| 3982 |
5608 d 18 h |
jermar |
/trunk/kernel/ |
Do not set sysinfo for uspace drivers directly in the kernel drivers. |
|
| 3973 |
5610 d 2 h |
decky |
/trunk/kernel/ |
kernel memory management revisited (phase 2): map physical memory according to zones
- ia32: register reserved and ACPI zones
- pareas are now used only for mapping of present physical memory (hw_area() is gone)
- firmware zones and physical addresses outside any zones are allowed to be mapped generally
- fix nasty antient bug in zones_insert_zone() |
|
| 3972 |
5610 d 14 h |
decky |
/trunk/kernel/ |
kernel memory management revisited (phase 1): proper support for zone flags
- the zone_t structures are now statically allocated to be easily available
- the locking scheme was simplified
- new flags for non-available zones were introduced
- FRAME_LOW_4_GiB flag is removed, the functionality will be eventually reimplemented using a generic mechanism |
|
| 3964 |
5613 d 0 h |
decky |
/trunk/kernel/ |
switch between the preference of kernel/uspace IRQ hash table |
|
| 3963 |
5614 d 21 h |
jermar |
/trunk/kernel/ |
The kernel i8042 driver should be instance-neutral. |
|
| 3947 |
5615 d 20 h |
jermar |
/trunk/ |
This is the evil commit. In particular, it does:
- introduces more powerful pseudo code for userspace IRQ top-half handlers
- changes the internals of IRQ dispatching
- simplifies the kernel's i8042 driver
- adapts the uspace i8042 driver to make use of the new pseudocode
- breaks all other architectures except ia32
- breaks almost all existing drivers
- breaks switching between uspace and kernel drivers |
|
| 3946 |
5615 d 20 h |
jermar |
/trunk/kernel/arch/ia32/include/ |
Fix ia32's pio_{read,write}_32(). |
|
| 3941 |
5616 d 16 h |
jermar |
/trunk/kernel/ |
Let IRQ claim methods take irq_t * instead of the instance argument. |
|
| 3940 |
5616 d 20 h |
decky |
/trunk/kernel/ |
make hw_area API more generic
this allows mapping of EGA VRAM on ia32/amd64 |
|
| 3936 |
5616 d 21 h |
decky |
/trunk/ |
dash is not a safe character for C preprocessor, use underscore instead |
|
| 3932 |
5617 d 15 h |
jermar |
/trunk/kernel/ |
Kill arch/drivers/ega.h. Provide one-stop shopping
place for architectures with ia32 legacy I/O space. |
|
| 3930 |
5617 d 15 h |
jermar |
/trunk/kernel/ |
Adapt most of the kernel to ioport8_t, ioport16_t and ioport32_t types. |
|
| 3929 |
5617 d 15 h |
jermar |
/trunk/kernel/arch/ |
Introduce ioport8_t, ioport16_t and ioport32_t. These types are to be used with
pio_read_n() and pio_write_n() functions. This breaks everything. |
|
| 3928 |
5617 d 17 h |
jermar |
/trunk/kernel/ |
Kill arch/drivers/i8042.h. |
|
| 3908 |
5620 d 18 h |
decky |
/trunk/ |
overhaul pareas: use one single physical area for the physical address space not belonging to physical memory |
|
| 3906 |
5621 d 14 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. |
|
| 3903 |
5621 d 22 h |
jermar |
/trunk/kernel/arch/ia32/include/ |
Fix ia32 build for Pentium 3 systems. |
|
| 3902 |
5621 d 22 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. |
|
| 3892 |
5622 d 16 h |
jermar |
/trunk/ |
Get rid of CONFIG_DEBUG_AS_WATCHPOINT. |
|
| 3891 |
5622 d 16 h |
jermar |
/trunk/ |
Get rid of CONFIG_SIMICS_FIX. |
|
| 3890 |
5622 d 17 h |
jermar |
/trunk/ |
Get rid of CONFIG_DEBUG_ALLREGS. |
|
| 3889 |
5622 d 22 h |
svoboda |
/trunk/kernel/arch/ia32/include/drivers/ |
Make ia32/amd64 compile again. This is mostly a hack, but a more drastic cleanup is needed to do it the right way. |
|
| 3880 |
5624 d 17 h |
decky |
/trunk/kernel/arch/ |
simplify configuration
introduce arch_construct_function and inb/outb (sometimes empty) on all platforms
various code cleanup |
|
| 3879 |
5624 d 17 h |
decky |
/trunk/kernel/arch/ |
simplify configuration in Makefiles |
|
| 3878 |
5624 d 17 h |
decky |
/trunk/kernel/arch/ |
introduce fncptr_t and ioport_t on all platforms |
|
| 3802 |
5648 d 22 h |
decky |
/trunk/kernel/arch/ |
fallback to VGA registers if VESA routine fails |
|
| 3800 |
5649 d 17 h |
decky |
/trunk/kernel/arch/ |
simplify compile options |
|
| 3790 |
5660 d 1 h |
svoboda |
/trunk/kernel/ |
Make newlines in panic messages consistent. Add periods at end of messages so that it is obvious whether they are printed entirely. |
|
| 3788 |
5660 d 14 h |
svoboda |
/trunk/kernel/ |
Unify newlines in fault_if_from_uspace() format strings. (Remove them and print newline automatically). Also start messages with caps. |
|
| 3779 |
5663 d 23 h |
jermar |
/trunk/kernel/ |
Only one kernel EGA driver to rule them all. |
|
| 3751 |
5672 d 2 h |
svoboda |
/trunk/kernel/arch/ia32/src/drivers/ |
Revive kernel EGA fb driver. Plus fix bug -- was using memcpy() for scrolling instead of memmove(). |
|
| 3709 |
5682 d 18 h |
decky |
/trunk/kernel/ |
invert VGA 3:2:3 palette to reflect changes in revision 3692 |
|
| 3707 |
5682 d 18 h |
decky |
/trunk/ |
kconsole is optional
kernel & uspace framebuffer rewrite with speedups (some things are slightly broken yet) |
|
| 3706 |
5682 d 18 h |
decky |
/trunk/ |
cstyle & small fixes |
|
| 3672 |
5693 d 17 h |
jermar |
/trunk/ |
Merge sparc branch to trunk. |
|
| 3631 |
5704 d 16 h |
jermar |
/trunk/kernel/arch/ia32/src/ |
The IA-32 manual is very secretive about the fact that the SYSENTER instruction
disables interrupts (I found only one evidence of this in the instruction's
pseudo code). The sysenter_handler needs to reenable the interrupts in order to
preserve the preemptive nature of the kernel. |
|
| 3485 |
5754 d 16 h |
jermar |
/trunk/ |
Support for SYSENTER on ia32. |
|
| 3381 |
5789 d 17 h |
jermar |
/trunk/kernel/arch/ia32/include/ |
Prevent 'tmp' in ia32's atomic_lock_arch() from being allocated
to the same register as 'val->count'. |
|