Age | Commit message (Collapse) | Author | |
---|---|---|---|
2011-12-05 | move some gpu status handling to core | notaz | |
this removes need to update each gpu plugin when this stuff is changed | |||
2011-11-01 | drc/gte: decode MVMVA data deps | notaz | |
2011-11-01 | drc/gte: fix dep list | notaz | |
2011-10-30 | cdriso: support multidisk eboots | notaz | |
2011-10-30 | cdriso: add cbin support | notaz | |
2011-10-30 | psxcounters: use higher precision for vsync timing | notaz | |
don't know why this is needed (having things like BIAS in mind), maybe desyncs with cdrom code or SPU? Tuned this on RE1. | |||
2011-10-30 | psxinterpreter: a bit more accurate Cause reg handling | notaz | |
2011-10-30 | drc: detect sw irqs properly | notaz | |
2011-10-30 | drc: fix various register trashing in mtc0 | notaz | |
2011-10-30 | drc: fix cycle multiplier code | notaz | |
rounding wrong way is enough to cause nasty timing problems | |||
2011-10-30 | drc: fix mtc0 argument | notaz | |
HOST_CCREG is not always cc.. | |||
2011-10-30 | drc: some debug code improvements | notaz | |
2011-10-30 | make root counters use generic event scheduling code | notaz | |
simplifies event handling code a bit | |||
2011-10-30 | drc: always use physical block addresses | notaz | |
there is no reason to use virtual addressing, things can't be remapped on PSX as there is no MMU anyway. Fixes somewhat related invalidation bug along the way. | |||
2011-10-30 | rcnt: remove gate guess for cnt 0 and 1 | notaz | |
at least 1 is wrong according to Ryphecha's info, and breaks at least Wipeout XL. | |||
2011-10-30 | gte: change gte irq workaround | notaz | |
This should be closer to what happens on the real thing. Special thanks to Ryphecha for the idea. | |||
2011-10-30 | drc: detect dead gte reads too | notaz | |
2011-10-30 | drc: add some hack options | notaz | |
intended for caanoo/Wiz, some users want to trade glitches for performance boost. | |||
2011-10-30 | gte: parametrize remaining ops | notaz | |
at least the ones from C source, it looks like there could be more. | |||
2011-10-30 | gte: disable some MAC overflow checks for ARM | notaz | |
also add some missing ones | |||
2011-10-30 | gte: eliminate excessive lims and casts | notaz | |
checking for overflow where it can't occur is a waste of time, get rid of that code. | |||
2011-10-30 | gte: pass a pointer to registers as argument | notaz | |
no more need to load psxRegs in each function and all registers are reachable by ldrh/strh. | |||
2011-10-30 | drc: try to avoid full constant reload | notaz | |
derive from existing ones if possible | |||
2011-10-30 | drc: fix sub assem_debug print | notaz | |
2011-10-30 | drc: avoid invcode checks on close addesses | notaz | |
yes this makes it possible to miss something, but I hope it's not going to happen and will deal with it if it does. | |||
2011-10-30 | gte_arm: some asm experiments | notaz | |
also removes wrong (?) MAC0 saturation. | |||
2011-10-30 | pcnt: measure gte too | notaz | |
2011-10-30 | drc: do MVMVA decoding; parametrize handlers | notaz | |
2011-10-30 | drc: finish GTE reg liveness analysis | notaz | |
2011-10-30 | drc: implement literal dupe merging | notaz | |
doesn't seem to make much difference, but theoretically should help. | |||
2011-10-30 | inline/parametrize rootcounter reads | notaz | |
makes rcnt1 hack impossible though, overclock PSX instead | |||
2011-10-30 | psxcounters: avoid doing excessive updates | notaz | |
this may cause missed interrupts in very rare cases, but performance penalty this makes is surely not worth it. | |||
2011-10-16 | psxmem: set fixed flag on mappings that really need it | notaz | |
also fix unmap | |||
2011-10-10 | psxcounters: change spu update ~2ms | notaz | |
old value ~1.4 was weird anyway | |||
2011-10-10 | drc: check for interrupts after RFE | notaz | |
this missing check wasn't causing too many problems because of events on every hsync, which are gone now. | |||
2011-10-10 | new ability to adjust psx cpu clock | notaz | |
2011-10-10 | get rid of old memhandlers | notaz | |
2011-10-10 | drc: remove some old inline memhandlers | notaz | |
generic code handles it better now. | |||
2011-10-10 | drc: implement memory access speculation | notaz | |
try to guess RAM mirror we are going to hit | |||
2011-10-08 | drc: implement shiftimm constant propagation | notaz | |
it's rare, but it happens. | |||
2011-10-08 | recompilation-caused frameskip workaround | notaz | |
2011-10-08 | spu/cdda: sync cdda to spu playback | notaz | |
this will probably make cdrom/vsync timing desync situation worse (those were never in sync anyway) but fix some cdda related sound stuttering. Those magic values are used trying to preserve other SPU plugin compatibility. | |||
2011-10-08 | try to support more compilers | notaz | |
2011-10-08 | rewrite memhandlers (write) | notaz | |
2011-10-04 | rewrite memhandlers (read) | notaz | |
2011-10-02 | yet more random armv5 tweaks | notaz | |
2011-09-30 | gte_arm: implement RTPS, RTPT | notaz | |
2011-09-27 | psxcounters: avoid update on each hsync | notaz | |
no useful work is done on hsync, no reason to interrupt the recompiler that often, set up events as needed. | |||
2011-09-27 | drc: do gte flag liveness detection | notaz | |
just copy-paste existing code and use _nf handlers from previous patch. | |||
2011-09-27 | gte: split arm code for pollux, generate flagless handlers | notaz | |
flagless are unused but will be used later, at least I hope so. |