Unknown W. Brackets
|
8725bfe411
|
Add a priority-indexed ready queue (not used yet.)
Also, not actually only ready threads yet.
|
2013-02-03 13:45:30 -08:00 |
|
raven02
|
24c720892c
|
Add few stubs seen from Dragon Ball Z Tenkaichi Tag Team
|
2013-02-04 03:19:55 +08:00 |
|
Unknown W. Brackets
|
7530a0b6fd
|
When creating threads, etc. restore thread ready.
The old thread was sticking in the RUNNING state.
|
2013-02-02 19:48:23 -08:00 |
|
Unknown W. Brackets
|
dd8002a42c
|
Oops, still need to switch back from interrupts.
There's the real problem: don't leave the thread "running".
|
2013-02-02 19:14:00 -08:00 |
|
Unknown W. Brackets
|
d1c2d5ddc4
|
Loading a savestate crashed with active mipscalls.
|
2013-02-02 18:46:23 -08:00 |
|
Unknown W. Brackets
|
fcdaf3c9c4
|
Return of mipscalls should be handled by actions.
|
2013-02-02 18:09:00 -08:00 |
|
Unknown W. Brackets
|
7055788f0b
|
Simplify argument passing for mipscalls.
|
2013-02-02 18:01:34 -08:00 |
|
Unknown W. Brackets
|
14cfb84633
|
Allow rescheduling in a mipscall, not a callback.
|
2013-02-02 17:57:44 -08:00 |
|
Unknown W. Brackets
|
31eeb3d50b
|
Make sceKernelStartModule() do something.
Doesn't work right, but gets somewhere?
|
2013-02-02 16:22:33 -08:00 |
|
Unknown W. Brackets
|
afccf2fc88
|
Buildfix.
|
2013-01-28 00:48:59 -08:00 |
|
Unknown W. Brackets
|
aeadc80045
|
Recover from an interrupt during a callback.
|
2013-01-27 18:43:38 -08:00 |
|
Unknown W. Brackets
|
dcc9f1ae5f
|
Always check for more callbacks after running one.
|
2013-01-27 18:43:12 -08:00 |
|
Unknown W. Brackets
|
54e6a39e74
|
Optimize callback checking.
Since it's checked every resched, it happens a lot.
This could save as much as 1.6% of time.
|
2013-01-27 17:01:17 -08:00 |
|
Unknown W. Brackets
|
5af5eef908
|
Refactor thread deletion so it can be consistent.
|
2013-01-27 16:55:43 -08:00 |
|
Unknown W. Brackets
|
7f4e322266
|
Cut thread switching time down.
Based on profiling release Windows.
Took these funcs from taking 1.6% to 0.5% time.
|
2013-01-26 11:29:20 -08:00 |
|
Unknown W. Brackets
|
a6094e25ff
|
Get rid of waitTypeStrings and resched reason copy.
|
2013-01-26 11:29:20 -08:00 |
|
Unknown W. Brackets
|
1cb7a88137
|
Don't use a dynamic string for wait debugging.
Happens in release, and shows up on the profiler.
Not huge, but should save ~0.5% of cpu time.
|
2013-01-26 11:29:19 -08:00 |
|
KentuckyCompass
|
9ce1887566
|
fix a bad format string in a DEBUG_LOG
|
2013-01-22 06:39:06 -08:00 |
|
Unknown W. Brackets
|
862c88d61a
|
More warning fixes, switches and printfs, etc.
This time ones not reported by Visual Studio.
|
2013-01-19 18:32:37 -08:00 |
|
Unknown W. Brackets
|
140fd171ef
|
Fix some 64-bit type conversion warnings.
|
2013-01-19 13:48:20 -08:00 |
|
Unknown W. Brackets
|
4a56873541
|
Properly wake semaphores in priority order.
|
2013-01-18 01:38:13 -08:00 |
|
Unknown W. Brackets
|
dcd3045cd5
|
Always use -1 for failed allocs in block allocator.
Before it used 0 and -1, confusing.
|
2013-01-18 01:38:12 -08:00 |
|
Henrik Rydgård
|
e23365dfa9
|
Merge pull request #430 from unknownbrackets/thread-funcs
Remove some thread stack hacks that aren't needed.
|
2013-01-15 23:40:19 -08:00 |
|
Unknown W. Brackets
|
8c07a5c87f
|
Remove some thread stack hacks that aren't needed.
|
2013-01-15 19:46:52 -08:00 |
|
Henrik Rydgard
|
3615b36a82
|
Store fcr0 and fcr31 in context (why was this commented out?)
|
2013-01-10 12:52:30 +01:00 |
|