aboutsummaryrefslogtreecommitdiff
path: root/libpcsxcore/new_dynarec
AgeCommit message (Collapse)Author
2011-03-22drc: start in more consistent statenotaz
useful when comparing things
2011-03-19drc: try even more to not compile code as 64bitnotaz
..when compiling trash
2011-03-17drc: ujump DS $ra overwrite fix?notaz
2011-03-17minor debug code updatesnotaz
2011-03-15drc: hack for stack-in-RAM-mirror casenotaz
2011-03-15drc: implement ra accesses in ujump DSnotaz
2011-03-10drc: handle some load delay cases with interpreternotaz
place the code together with other PCSX hacks.
2011-03-08pcsxmem: fix ro write handlernotaz
also refactor things a bit (as usual)
2011-03-08drc: update invalid_code for RAM when mirrors are touchednotaz
the memhandlers always check RAM locations, even when something writes to the mirrors.
2011-03-07drc: fix valgrind errorsnotaz
also makes using_tlb constant so that the compiler can remove leftover TLB code.
2011-03-07drc: rm unneeded writebacks in stubs, as suggested by Ari64notaz
also fix leftover moves, missed by last cleanup commit.
2011-03-07frontend: fix valgrind errorsnotaz
also fixes pc build
2011-03-01drc: clear dynarec completely on CD image changenotaz
most likely only BIOS can be reused anyway, no need to stress invalidation code needlessly (even if it does it all well).
2011-03-01gte_neon: hook into recompilernotaz
2011-02-17drc: some PCSX-specific const addr io handlersnotaz
2011-02-17drc: add i/o penalty cyclesnotaz
.. as done by some other PSX emus
2011-02-17drc: don't pass r3 to memhandlersnotaz
also don't recalculate cycles after reads - PCSX doesn't queue any interrupts on reads.
2011-02-17drc: merge register types from Ari64's codenotaz
but don't merge RAM_OFFSET stuff, I don't want to mess with this yet
2011-02-17drc: merge Ari64's patch: 11_reduce_invstub_memory_usagenotaz
2011-02-17drc: merge Ari64's patch: 10_unnecessary_invalidatenotaz
also adjust memhandlers to do inv checks as recompiler no longer does them for unknown writes.
2011-02-17drc: merge Ari64's patch: 08_loop_hoistingnotaz
2011-02-17drc: merge Ari64's patch: 07_clear_cachenotaz
2011-02-17drc: merge Ari64's patch: 05_dont_write_r0notaz
also get rid of unneeded moves in load_assemble
2011-02-17drc: merge Ari64's patch: 02_xor_zeronotaz
2011-02-17drc: get rid of pass 7/provisional_r32 toonotaz
2011-02-17drc: fix p32 32K memset stupidity, don't use p32 insteadnotaz
2011-02-12drc: rm x86 stuff - it's unmaintained herenotaz
Use Ari64's releases if you need x86 stuff.
2011-02-09drc: update memhandler according to pcsxr codenotaz
2011-02-09drc: sync gte with interpreternotaz
2011-02-04drc: handle dummy reads in inline_readstubnotaz
2011-01-28drc: fall back to interpreter on delay slot link reg dependenciesnotaz
perhaps until someone implements this in recompiler used by Tomb Raider 2 at least
2011-01-28cdrom: support cdda playback cursor and autopause/reportnotaz
some differences from PCSX-Reloaded version
2011-01-26drc: invalidate RAM mirrors correctlynotaz
2011-01-26drc: handle $ra accesses in DS (BxxZAL case only)notaz
also restore asserts for rjump and ujump cases.
2011-01-26cdrom: merge most of cdrom.c from latest pcsxr (r62214)notaz
most changes pulled are shalma's work
2011-01-25drc: allow CFC1/2 to r0notaz
2011-01-25drc: make BxxZAL implementation completenotaz
2011-01-24drc: try harder to not compile code as 64bitnotaz
..when compiling trash
2011-01-24drc: fix COP2 decodingnotaz
2011-01-24drc: don't leave until we really need tonotaz
2011-01-24drc: add support for interpreter calls; use them for branches in delay slotsnotaz
2011-01-24fix some alignment issuesnotaz
2011-01-24drc: use correct clear sizenotaz
2011-01-18drc: remove leftover printfnotaz
2011-01-18drc: Ari64's fix for loop reg allocation problemnotaz
2011-01-17Revert "drc: fix a bug with loop reg allocation"notaz
This reverts commit ca7c944853bbb689040c530655e2da231c40db5b in favour of fix by Ari64 himself.
2011-01-16drc: fix a bug with loop reg allocationnotaz
2011-01-16drc: fix some goofage in cop2_{get,put}_dregnotaz
2011-01-16drc: allow reading r31 in delay slotnotaz
as hlide explained it's ok as long as we do DS first.
2011-01-16drc: handle MTC0 in delay slotnotaz
also refactor MTC0 code a bit.