Subversion Repositories HelenOS

Rev

Show changed files | Details | Compare with Previous | Blame | RSS feed

Filtering Options

Rev Age Author Path Log message Diff
4605 6144 d 5 h decky /branches/snapshot/ add task snapshoting branch  
4274 6223 d 21 h decky /trunk/kernel/generic/src/ fix compilation in non-SMP configurations  
4254 6230 d 20 h jermar /trunk/kernel/generic/src/ Address issues with IRQ notifications.  
4248 6231 d 3 h decky /trunk/kernel/generic/src/ipc/ avoid deadlock during cleanup
(the compare function which is called inside hash_table_remove acquires irq->lock)
 
3964 6272 d 5 h decky /trunk/kernel/ switch between the preference of kernel/uspace IRQ hash table  
3947 6275 d 1 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
 
3930 6276 d 20 h jermar /trunk/kernel/ Adapt most of the kernel to ioport8_t, ioport16_t and ioport32_t types.  
3902 6281 d 3 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.
 
3875 6283 d 22 h decky /trunk/kernel/generic/ remove platform dependent constructs from generic code  
3657 6358 d 6 h vana /trunk/ Legacy IRQ support, uspace NS16550 support, some minor changes  
2745 6635 d 6 h decky /trunk/ code cleanup (mostly signed/unsigned)
allow extra compiler warnings
 
2626 6729 d 8 h jermar /trunk/kernel/generic/ Support for 5 payload arguments for IPC irq notifications and the pseudo code.  
2471 6889 d 23 h jermar /trunk/ Greatly improve comments in the IPC layer.
Now I think I finally start to understand our IPC internals :-)
 
2183 6963 d 2 h jermar /trunk/kernel/generic/ Continue to de-oversynchronize the kernel.
- replace as->refcount with an atomic counter; accesses to this
reference counter are not to be done when the as->lock mutex is held;
this gets us rid of mutex_lock_active();

Remove the possibility of a deadlock between TLB shootdown and asidlock.
- get rid of mutex_lock_active() on as->lock
- when locking the asidlock spinlock, always do it conditionally and with
preemption disabled; in the unsuccessful case, enable interrupts and try again
- there should be no deadlock between TLB shootdown and the as->lock mutexes
- PLEASE REVIEW !!!

Add DEADLOCK_PROBE's to places where we have spinlock_trylock() loops.
 
2098 7029 d 0 h decky /trunk/kernel/generic/ avoid conflict with potential keyword  
2071 7040 d 22 h jermar /trunk/ (c) versus (C)  
2064 7054 d 21 h vana /trunk/ SIMICS added as a simulator kernel runs on  
1933 7132 d 21 h jermar /trunk/kernel/generic/ Implement efficient IPC notification cleanup.  
1932 7132 d 22 h jermar /trunk/kernel/ Do not copy notification config out of the IRQ structure.
Add the notify member, that toggles notifications on and off instead.
This is good for preventing other tasks from registering the notification
while the notification config is stored outside the IRQ structure.
It should also help to implement the cleanup code.
 
1923 7133 d 21 h jermar /trunk/ Replace the old IRQ dispatcher and IPC notifier with new implementation.
Note that all architectures except for sparc64 are now broken
and don't even compile.
 

Show All