libretro/ppsspp
1
0
Fork 0
mirror of https://github.com/libretro/ppsspp.git synced 2024-12-14 04:28:42 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity
15,234 Commits 19 Branches 54 Tags 247 MiB
eeff110c0f
Commit Graph

559 Commits

Author SHA1 Message Date
Unknown W. Brackets
038394b081 Divide from -1.0 directly in x86 jit vnrcp. 2013-02-19 00:35:15 -08:00
Unknown W. Brackets
a438791e7c Initial (very inefficient) vmmov for x86 jit. 
This makes #464 work (at least LittleBigPlanet), but only in x86 jit.
 2013-02-18 23:21:18 -08:00
Unknown W. Brackets
b8e2177591 Jit vzero/vone, which are easy and common (x86.) 2013-02-18 22:15:47 -08:00
Unknown W. Brackets
a001b8b6f0 Tweak and note vsat0/vsat1 NaN handling. 2013-02-18 22:06:49 -08:00
Unknown W. Brackets
40b2a8dec1 Drop the sign in vsqrt, but not vrsq. 2013-02-18 21:46:33 -08:00
Unknown W. Brackets
2e6f0006fd Oops, correct the bounds check. 2013-02-18 20:43:43 -08:00
Unknown W. Brackets
a3eba1e96e Fix typo, should definitely be VX(). 2013-02-18 20:43:43 -08:00
Unknown W. Brackets
2dfdf3ffeb Implement Comp_VV2Op vfpu ops in the x86 jit. 
Also, some cleanup.  No need for this extra boilerplate, simplify...

This makes the Bink video issue slightly better, in jit only.
 2013-02-18 20:43:28 -08:00
Henrik Rydgard
e32721c72a Merge branch 'master' into armjit-fpu 
Conflicts:
	Core/MIPS/MIPSVFPUUtils.cpp
	Core/MIPS/x86/CompVFPU.cpp
	GPU/GLES/VertexDecoder.cpp
 2013-02-19 00:50:33 +01:00
Unknown W. Brackets
dacbcbdf2b Add a MIPSTables flag for ignoring the prefix. 2013-02-18 01:23:15 -08:00
Unknown W. Brackets
afb7c0b83c Assume prefixes start default until proven wrong. 
Currently this means nothing since the MIPSTables flags are wrong.
It will blow the cache once, after the first vfpu op.
 2013-02-18 01:14:57 -08:00
Unknown W. Brackets
0bfc380575 Try to reuse temp regs for better caching. 2013-02-18 00:32:42 -08:00
Unknown W. Brackets
6855398add Support known prefixes in the vfpu jit. 2013-02-18 00:11:58 -08:00
Unknown W. Brackets
8ea59990ab Make applying prefixes mostly automatic. 
And implement (hopefully) D prefixes.
 2013-02-18 00:11:57 -08:00
Unknown W. Brackets
18c03d0816 Handle temp regs better, no need for direct access. 2013-02-18 00:11:57 -08:00
Unknown W. Brackets
27942606ad Use prefixD directly in jit, just like interp now. 2013-02-17 22:46:34 -08:00
Unknown W. Brackets
08a42a1aaf Preserve orig regs when applying vfpu prefixes. 2013-02-17 22:37:56 -08:00
Unknown W. Brackets
d63548799b Add more temp regs, allow swapping if necessary. 2013-02-17 22:18:46 -08:00
Unknown W. Brackets
7fee4dfd13 Re-enable vdot and vadd/etc. in x86 jit. 2013-02-17 17:53:53 -08:00
Unknown W. Brackets
f532951331 Automatically eat prefixes in x86 jit. 
Simplifies the code and makes it easier to know they're eaten
even for ops not yet jitted.
 2013-02-17 17:53:53 -08:00
Unknown W. Brackets
6191017a2c Fix jit VDot mapping vd incorrectly to a quad. 2013-02-17 17:52:59 -08:00
Unknown W. Brackets
106cbcfc5d Fix possible overlap issue in VDot. 2013-02-16 21:26:32 -08:00
Unknown W. Brackets
0fdc975fde Fix wrong type in x86 jit fpu/vfpu load store. 2013-02-16 20:22:08 -08:00
Unknown W. Brackets
6eae8ed36a Disable VDot and Vec3 in x86 jit, broke things. 2013-02-16 19:57:35 -08:00
Unknown W. Brackets
b27701ac7d Fix VDot returning -0.0 in x86 jit. 2013-02-16 10:37:42 -08:00
Unknown W. Brackets
1c4c5e718b Optimize VDot and VecDo3 to avoid temporaries. 2013-02-16 10:19:05 -08:00
Unknown W. Brackets
0bd382c518 Discard temp regs right away, some helper funcs. 2013-02-16 10:18:13 -08:00
Unknown W. Brackets
0d5da967eb Enable VDot and Vec3 in x86 jit. 2013-02-16 03:27:48 -08:00
Unknown W. Brackets
35537b3c97 Add TEMP0 fpu regs to x86 like in armjit. 
But... will probably need more and the ability to swap into memory
if we want to deal with prefixes.
 2013-02-16 03:27:03 -08:00
Henrik Rydgard
909b768f47 Don't need separate variables for writemask. Some optimizations. 2013-02-16 09:28:55 +01:00
Henrik Rydgard
b8abb77eee More armjit-fpu work - dot product working for example. Add some non working DISABLEd stuff too. 2013-02-16 09:27:48 +01:00
Unknown W. Brackets
be8ddf12aa Don't go out of bounds applying vfpu swizzle. 2013-02-15 23:43:40 -08:00
Henrik Rydgard
0ee7578d68 Merge branch 'master' into armjit-fpu 2013-02-15 23:09:59 +01:00
Henrik Rydgard
d22e258943 Don't need separate variables for writemask. Some optimizations. 2013-02-15 22:56:38 +01:00
Henrik Rydgard
44e4ba8772 Merge branch 'master' into armjit-fpu 2013-02-15 21:42:44 +01:00
Unknown W. Brackets
e42e7bf22e Don't flush all regs in mfvc, just prefixes. 2013-02-15 09:50:07 -08:00
Unknown W. Brackets
f95e66eb98 Forget cached prefixes when calling generic. 
It may eat them, or maybe always does?
 2013-02-15 08:35:34 -08:00
Unknown W. Brackets
2b441f1638 Initial implementation of jit vadd/vsub/vdiv/vmul. 2013-02-15 08:35:34 -08:00
Unknown W. Brackets
b9506c9568 Minor cleanup for vdot in x86 jit. 2013-02-15 08:35:34 -08:00
Unknown W. Brackets
ccad259ae5 Keep track of VFPU prefixes and flush them in jit. 2013-02-15 08:35:33 -08:00
Unknown W. Brackets
f6f2927526 Add curlies around DISABLE/CONDITIONAL_DISABLE. 2013-02-15 08:35:33 -08:00
Unknown W. Brackets
4eca76e0cc Check for s/t/d prefix reg changes in jit. 2013-02-14 00:27:09 -08:00
Unknown W. Brackets
3b58cc27bd Oops, vfpu was missing CONDITIONAL_DISABLEs. 2013-02-14 00:27:09 -08:00
Unknown W. Brackets
abe390e6f3 Add some checks for fpu/vfpu writing to $0. 2013-02-14 00:27:09 -08:00
Unknown W. Brackets
4789a8e5eb Oops, can't have CONDITIONAL_DISABLE here, no op. 2013-02-14 00:27:08 -08:00
Henrik Rydgard
30318a4a4d Merge branch 'master' into armjit-fpu 
Conflicts:
	Core/MIPS/x86/CompFPU.cpp
 2013-02-13 20:47:41 +01:00
Unknown W. Brackets
f1386dfca1 Add a quick optimization to the x86 fpu comps. 2013-02-13 02:21:26 -08:00
Unknown W. Brackets
19cc652a37 Correct NaN handling in fpu comparisons. 2013-02-13 01:54:07 -08:00
Unknown W. Brackets
3cab6986c5 Jit the FPU comparisons on x86. 
Probably not too fast.  Also, NaN handling seems wrong?
 2013-02-13 00:55:10 -08:00
Henrik Rydgard
2c01b36585 Some FPU optimization 2013-02-12 00:58:31 +01:00
Henrik Rydgard
4eb89e6aec Merge branch 'master' into armjit-fpu 2013-02-11 19:22:14 +01:00
Henrik Rydgard
3ce4a8a719 Allow switching 2xSSAA on and off ingame. Add Show FPS menu option. 2013-02-11 19:02:38 +01:00
Unknown W. Brackets
7c428bfeba Fix immediate div CMP. 2013-02-10 10:02:55 -08:00
Unknown W. Brackets
e0ebfd2211 Jit div/divu in x86. 2013-02-10 09:36:41 -08:00
Unknown W. Brackets
9bb78ce2ec Jit madd/msub in x86. 2013-02-10 08:45:35 -08:00
Henrik Rydgard
3a11b030d6 Merge branch 'master' into armjit-fpu 
Conflicts:
	Core/MIPS/ARM/ArmCompFPU.cpp
	Core/MIPS/x86/CompFPU.cpp
 2013-02-10 15:57:16 +01:00
Henrik Rydgard
f75d14d3b5 ARM FPU jit work 2013-02-10 15:53:56 +01:00
Henrik Rydgard
78923f5538 Jit a little more (vfpu single load/store, transfer instructions) 2013-02-10 12:14:55 +01:00
Unknown W. Brackets
eb84c2f00a Validate jumps in jit slowmem mode. 
This makes it easier to see what is going on in the emulator debugger.
 2013-02-09 23:11:26 -08:00
Unknown W. Brackets
71c85ccf33 In jit slowmem, verify actual address. 
Oops, it could crash if it was near the boundary.
Well, it still could if it were very near, but that's rare.
 2013-02-09 23:08:57 -08:00
Henrik Rydgard
377c94b125 JIT x86: cvt.s.w 2013-02-06 20:29:49 +01:00
Lewis Robbins
442e64cd84 compiler warning and const top-level const 2013-02-05 17:54:29 +00:00
lioncash
025a1351b4 Get rid of unused iterators. 
Also fix the formatting in 3 sprintf calls.
 2013-02-04 08:49:58 -05:00
Unknown W. Brackets
6bee870ac9 Fix CompShiftVar for x86 jit. 
In case rd == rs, need to load ECX first.  I can't find anything
else wrong with it for it to be disabled.
 2013-02-02 14:02:07 -08:00
Unknown W. Brackets
f777c872e6 Jit unaligned reads/writes. 
This mostly just improves perf on debug, not really on the map for release.
 2013-02-02 13:12:34 -08:00
Unknown W. Brackets
bab7947be6 Read delay slots as instructions not mem. 
Just in case - could be a jump target, maybe?  Never seen it, though.
 2013-02-02 11:46:35 -08:00
Unknown W. Brackets
44b5adeaac Properly jit the break instruction. 
Otherwise, it just keeps on going past it.
We never want to hit this anyway, but it's good to know if we do.
 2013-02-01 00:49:14 -08:00
Henrik Rydgard
c97f63a9d9 Minor armjit opt 2013-01-30 20:01:42 +01:00
Henrik Rydgard
90b11bba37 Implement mult, multu, mflo/hi, mtlo/hi in x86 JIT 2013-01-29 00:48:42 +01:00
Unknown W. Brackets
6d7a8d9b1a Apply the memview mask to jit immediates too. 2013-01-26 23:54:43 -08:00
Unknown W. Brackets
a7b5433ba7 Make sure fastmem isn't confused by rs changing. 2013-01-26 23:18:50 -08:00
Unknown W. Brackets
a89d61463e Make the VFPU jit use far jumps for memory access. 2013-01-26 23:08:19 -08:00
Unknown W. Brackets
0e8e9697c5 Add lv.q/sv.q support to the x86 jit. 2013-01-26 10:09:18 -08:00
Unknown W. Brackets
b77ce99d01 Oops, no slow read for immediates usually. 2013-01-26 09:27:52 -08:00
Unknown W. Brackets
9cd5836b85 Rename WriteFinish() to Finish() is safe mem. 
It's nothing to do with writing.
 2013-01-26 09:09:47 -08:00
Unknown W. Brackets
3e419f513a Refactor jit safe memory reads without dup code. 
But, maybe too automagical...
 2013-01-26 08:42:34 -08:00
Unknown W. Brackets
b7ef3e7bef Make sure to log / check bad immediate mem access. 
Although, theoretically, this should never happen.
Also, definitely time to refactor.
 2013-01-25 23:06:43 -08:00
Unknown W. Brackets
3418383917 Immediately break on bad mem access in jit slowmem. 2013-01-25 22:52:51 -08:00
Unknown W. Brackets
db5fa233a8 Make sure we don't mark a reg dirty on noop. 2013-01-25 22:34:01 -08:00
Henrik Rydgard
2738417040 VFPU JIT: start setting up infrastructure. very incomplete. vdot works if undisabled, but isn't complete. 2013-01-26 01:34:19 +01:00
Henrik Rydgard
68991511ee Split out the FPU reg cache into its own file too. 2013-01-26 01:34:19 +01:00
Henrik Rydgard
ad5e2b58c6 Separate the two regcaches before doing major surgery to FPURegCache. 2013-01-26 01:34:18 +01:00
Henrik Rydgard
dd149a50a3 Must flush FPR regcache before thrashing the fp regs 2013-01-25 19:55:30 +01:00
Henrik Rydgard
aabc0aa9ef Quick implementation of LV.Q and SV.Q in x86/x64 JIT 2013-01-25 19:50:30 +01:00
Henrik Rydgård
0f080aeaaa Merge pull request #492 from unknownbrackets/jit-minor 
ALU jit optimizations
 2013-01-25 01:01:34 -08:00
Unknown W. Brackets
a7c6f46829 Optimize and/or 0 to just a mov in x86 jit. 2013-01-25 00:25:40 -08:00
Unknown W. Brackets
ab9bea068c Jit reg+reg compile time, and avoid flushing EDX. 2013-01-25 00:16:55 -08:00
Unknown W. Brackets
ce5f393fb8 Hit immediates in the ALU better and more simply. 2013-01-25 00:16:55 -08:00
Unknown W. Brackets
d1909a1581 Add a quick disable define for nice delay slots. 2013-01-24 19:11:03 -08:00
Unknown W. Brackets
75cbe18afc Simplify nice delay slot detect, and yes for noop. 
NOOP seems very common so this should already benefit speed a bit.
 2013-01-24 08:29:32 -08:00
Unknown W. Brackets
2eba209f64 Move around the jit nice delay slot logic. 
Nice delay slots don't not save flags, they run before the CMP.
 2013-01-24 07:31:51 -08:00
Unknown W. Brackets
3444fc8981 Avoid some memory writes on jr. 
Should improve tight mips function loops a bit.
 2013-01-24 01:23:50 -08:00
Unknown W. Brackets
c1757ee166 Check downcount in jit after a syscall. 2013-01-23 22:25:35 -08:00
Unknown W. Brackets
0e33923844 Belt and suspenders check for branch ops. 2013-01-22 08:11:37 -08:00
Unknown W. Brackets
c324983340 Make the jit support bltzal and friends. 
Fixes problems with jit in games.  Android changes completely untested.
 2013-01-22 08:04:01 -08:00
Unknown W. Brackets
a9d0390426 Adjust downcount before syscalls, not after. 
This makes jit slightly slower for syscalls, but it's minor and makes
sure jit and interpreter timing are determistically the same.
 2013-01-21 22:57:53 -08:00
Unknown W. Brackets
566b7a0910 A branch was missing inDelaySlot, refactor it. 2013-01-21 22:45:07 -08:00
Unknown W. Brackets
c897e6446a Don't over decr downcount when hitting a jit bp. 2013-01-21 19:41:12 -08:00
Unknown W. Brackets
8438371941 Read memory in the jit dispatcher same as lw. 
Just for consistency.  One less op, maybe faster, probably same.
 2013-01-21 17:51:14 -08:00
Unknown W. Brackets
1485b0865c Improve the speed of branch debugging a bit. 2013-01-20 19:48:55 -08:00
Unknown W. Brackets
dd69694302 Add some optional logging to debug jit branching. 2013-01-20 19:48:55 -08:00
Unknown W. Brackets
776eb8ab2e Simplify CompileDelaySlot(). 2013-01-20 19:48:54 -08:00
Unknown W. Brackets
df06bb5624 Add some checks to make sure ZERO is never set. 2013-01-20 19:48:53 -08:00
Unknown W. Brackets
a43078ab68 Same optimization for FPU load / store. 2013-01-20 13:16:41 -08:00
Unknown W. Brackets
f5963df0dc Optimize write to a single x64 op too. 2013-01-20 13:06:19 -08:00
Unknown W. Brackets
e8dc99328a Avoid using EAX as a temporary where possible. 
All the regs should be indirect addressing compatible.  So if it's
in a reg, let's use that instead of EAX.
 2013-01-20 12:57:14 -08:00
Unknown W. Brackets
eaa24ee047 Use EDX as a temporary for sb, and jit it. 2013-01-20 12:25:08 -08:00
Unknown W. Brackets
30f6a4ba87 Fix stupid stupid typo breaking slowmem jit. 2013-01-20 09:39:13 -08:00
Unknown W. Brackets
da22eb8adf Make swc1 and lwc1 fast even without fastmem. 2013-01-20 02:07:00 -08:00
Henrik Rydgård
2cb830510c Merge pull request #444 from unknownbrackets/jit-minor 
Don't muck with currentMIPS->r directly in the slowmem jit
 2013-01-20 01:58:48 -08:00
Unknown W. Brackets
53600161ba Don't write anything to a bad static pointer. 
Not sure what came over me...
 2013-01-20 00:19:18 -08:00
Unknown W. Brackets
385417effe Log jit misses at runtime instead of compile time. 
Really, it could be very different after all... this shouldn't be
all that slow, I guess.
 2013-01-19 20:11:17 -08:00
Unknown W. Brackets
75a3872923 Log missed jit ops for poor man profiling. 2013-01-19 19:58:25 -08:00
Unknown W. Brackets
e78223d2c0 Since flipping the op is easy, also do lb/lh. 2013-01-19 16:25:57 -08:00
Unknown W. Brackets
5e8f1917ee Fix 64-bit memory dereferencing. 2013-01-19 16:25:57 -08:00
Unknown W. Brackets
90e6f0b7df Optimize static memory read/writes in jit as well. 
Like the arm jit does.
 2013-01-19 16:25:56 -08:00
Unknown W. Brackets
c64966c16e Oops, lost the CONDITIONAL_DISABLE. 2013-01-19 16:25:56 -08:00
Unknown W. Brackets
37fb64ac83 Fast path scratchpad too, shouldn't be expensive. 2013-01-19 11:11:45 -08:00
Unknown W. Brackets
72e547420d Refactor jit slowmem, add lbu to jit since easy. 2013-01-19 11:11:45 -08:00
Unknown W. Brackets
5305017fc3 Properly save registers before the slowmem call. 2013-01-19 11:11:44 -08:00
Unknown W. Brackets
f1295f6262 Don't muck with currentMIPS->r in the slowmem jit. 2013-01-19 11:11:44 -08:00
Henrik Rydgard
229d4e9f32 Buildfix 2013-01-19 13:10:52 +01:00
Henrik Rydgård
c20cef2399 Merge pull request #440 from unknownbrackets/jit-minor 
Micro optimizations to x86 jit mem when fastmem is off
 2013-01-19 02:04:34 -08:00
Unknown W. Brackets
2ad77aa9c8 Gotta flush before the call, too. 2013-01-19 01:53:11 -08:00
Unknown W. Brackets
09422d5adb Avoid a func if possible when fastmem is off. 2013-01-19 01:02:47 -08:00
Unknown W. Brackets
bc75b68c36 Also jit sh and lhu, which are pretty common ops. 2013-01-18 23:10:51 -08:00
Unknown W. Brackets
d5ae85201c Optimize sw/lw even under safe memory. 
They're very common instructions, so shaving cycles helps.
 2013-01-18 23:10:50 -08:00
Unknown W. Brackets
11c5cdfdb0 Refactor out all the CheckJitBreakpoint()s. 
Ah, much cleaner.
 2013-01-18 21:33:23 -08:00
Unknown W. Brackets
5080285e54 Add breakpoints for delay slots. 
Wanted to do this in CompileAt(), darn not nice delay slots.
 2013-01-18 21:12:58 -08:00
Unknown W. Brackets
40ae3dfe45 Correctly break at branch points in x86 jit. 2013-01-18 21:12:53 -08:00
Unknown W. Brackets
beac991a9e Clear jit cache when changing breakpoints. 
For now, only when paused.  I don't think clearing the cache while
running is an awesome idea.
 2013-01-18 20:12:53 -08:00
Unknown W. Brackets
a9293c8923 Add breakpoints to x86 jit for easier debugging. 
They should be really fast so leaving them on in release.
 2013-01-18 20:10:37 -08:00
Henrik Rydgard
674911ddba Move downcount into MIPSState for efficiency, enable block linking. 
On ARM JIT we can now reach it through the cpu context reg.
 2013-01-12 00:44:18 +01:00
Henrik Rydgard
ea3055322c Oops 2013-01-09 00:45:54 +01:00
Henrik Rydgard
f5c94775b9 Cleanup 2013-01-09 00:12:02 +01:00
Henrik Rydgard
76481a300c Icache must be invalidated. Jit now starts to run, but there's no cube in cube.elf! 2013-01-08 23:52:11 +01:00
Henrik Rydgard
8915677241 More progress but it weirds out... 2013-01-08 21:24:42 +01:00
Henrik Rydgard
e3a4ed510c Fix bug in x86 jit :) 2013-01-08 19:30:28 +01:00
Henrik Rydgard
8c06edc47b It's getting close to the first totally unoptimized jit run. 2013-01-08 17:03:17 +01:00
Henrik Rydgard
b3fd1ff34c Lots of various work on the ARM jit. It executes a couple of blocks now. 2013-01-08 13:49:52 +01:00
Henrik Rydgard
5a7f4acc06 More armjit work 2013-01-08 00:26:42 +01:00
Henrik Rydgard
a2ff416534 Rename files. Rewrite ArmRegCache from scratch. 2013-01-07 22:33:09 +01:00
Henrik Rydgard
93470e0c1f Remove some old PowerPC references... 2013-01-05 19:52:11 +01:00
Arthur Blot
be91ad2f9c Fixed $zr being non-zero after loading instruction 2013-01-03 16:14:45 +01:00
Henrik Rydgard
6b64554e64 Fix savestates under JIT 2013-01-03 12:46:10 +01:00
Henrik Rydgård
f9133c1a56 Reset JIT before saving/loading state 2012-12-29 00:10:43 +01:00
Henrik Rydgård
ed68dea0d5 JIT: Ignore branches in delay slots. Not sure if this is 100% correct. 2012-12-26 08:37:53 +01:00
Henrik Rydgård
d8da92befa Fix timing issue wiith JIT on non-Windows platforms 2012-12-25 09:02:02 +01:00
Henrik Rydgard
f5efd6f2b1 JIT: add option to enable/disable jit of memory instructions. for crash safety. 2012-12-21 16:50:13 +01:00
Sacha
3c903dda24 Improve portability to future platforms. Make use of new USING_GLES2 define throughout PPSSPP. Also use ARM definition where suitable. Remove some redundancy with includes. 2012-11-26 13:25:14 +10:00
Henrik Rydgard
45d42f09a1 Remove x86disasm, experiment with build order 2012-11-20 00:48:13 +01:00
Henrik Rydgard
f326c36220 Some cleanup, re-enable some apparently disabled jit ops 2012-11-18 23:14:22 +01:00
Henrik Rydgard
7720dc3f60 Various warning, logging, jit fixes 2012-11-17 19:56:28 +01:00
Henrik Rydgard
d485b76e11 Jit fixes, test update 2012-11-12 14:35:10 +01:00
Henrik Rydgard
9bc7385502 Power test working 2012-11-11 19:32:27 +01:00
Sacha
41f5abab31 PPSSPP ported to Blackberry10 
Now builds on Playbook and Dev Alpha
Make emulator more compatible with other OS (case sensitivity, defines, includes)
Uses Android's code paths and backend
 2012-11-05 23:09:49 +10:00
Henrik Rydgard
64cc573703 Switch to "GPL 2.0 or later" for various reasons. I wrote most of the code I imported from Dolphin (which is GPL2-but-not-later), so it should be OK. 2012-11-04 23:24:00 +01:00
MrOrdinaire
1ab9943643 fix missing #include's and missing CMake linking directives 2012-11-03 10:33:24 +08:00
Henrik Rydgard
4f7ad15758 Add snapshot of the whole source code. 2012-11-01 16:19:01 +01:00