Commit Graph

14035 Commits

Author SHA1 Message Date
Rohit Nirmal 4c14ebdf32 Remove pointless initializations. 2014-08-27 20:36:49 -05:00
Rohit Nirmal 9e340b1c81 Initialize variables to silence garbage return warning from scan-build. 2014-08-27 20:35:20 -05:00
Rachel Bryk a3cfc98f26 Allow system time to move forward during netplay. 2014-08-27 16:45:15 -04:00
degasus 8b0ad5daec glx: fix shutdown hang 2014-08-27 18:16:56 +02:00
Pierre Bourdon 7d05ebbc9b Merge pull request #888 from FioraAeterna/fmulinterp
Fix another absent-minded typo in the fmul interpreter patch
2014-08-27 10:57:09 +02:00
Fiora 7e07acbf3f Fix another absent-minded typo in the fmul interpreter patch 2014-08-26 23:00:11 -07:00
shuffle2 061f2058c2 Merge pull request #887 from FioraAeterna/fmulinterp
Bugfixes for fmul rounding
2014-08-26 21:50:43 -07:00
Fiora 1a0a33518b Bugfixes for fmul rounding
Fix the places I forgot to add Force25Bit, and fix an incredibly silly typo bug
2014-08-26 21:37:45 -07:00
Rachel Bryk 31353573cb Change default netplay buffer to 5. 2014-08-26 21:50:30 -04:00
comex 75b9d6da48 Merge pull request #867 from comex/gpu-determinism-pr1
Some small preparatory changes to VideoCommon for GPU thread determinism mode.
2014-08-26 20:17:27 -04:00
Fiora 7dbc623dc0 JIT: Initial FPRF support
Doesn't support all the FPSCR flags, just the FPRF ones.
Add PPCAnalyzer support to remove unnecessary FPRF calculations.

POV-ray benchmark with enableFPRF forced on for an extreme comparison:
Before: 1500s
After, fmul/fmadd only: 728s
After, all float: 753s

In real games that use FPRF, like F-Zero GX, FPRF previously cost a few percent
of total runtime.

Since FPRF is so much faster now, if enableFPRF is set, just do it for every
float instruction, not just fmul/fmadd like before. I don't know if this will
fix any games, but there's little good reason not to.
2014-08-26 10:57:03 -07:00
comex e31d6feaa2 Unify three types of non-FIFO requests to the GPU thread around Common::Event and Common::Flag.
The only possible functionality change is that s_efbAccessRequested and
s_swapRequested are no longer reset at init and shutdown of the OGL
backend (only; this is the only interaction any files other than
MainBase.cpp have with them).  I am fairly certain this was entirely
vestigial.

Possible performance implications: efbAccessReady now uses an Event
rather than spinning, which might be slightly slower, but considering
the slow loop the flags are being checked in from the GPU thread, I
doubt it's noticeable.

Also, this uses sequentially consistent rather than release/acquire
memory order, which might be slightly slower, especially on ARM...
something to improve in Event/Flag, really.
2014-08-26 12:43:39 -04:00
comex de7294ecc1 Add Flag support to ChunkFile.h 2014-08-26 12:43:39 -04:00
comex 45a4236283 A tiny restructuring to allow inlining of FifoCommandRunnable. Probably useless. 2014-08-26 12:43:39 -04:00
comex 14125cf951 Refactor SetCpStatus into two functions for from-GPU and from-CPU mode rather than a boolean parameter.
This shouldn't affect functionality.  I'm not sure if the breakpoint
distinction is actually necessary (my commit messages from the old
dc-netplay last year claim that breakpoints are broken anyway, but I
don't remember why), but I don't actually need to change this part of
the code (yet), so I'll stick with the trimmings change for now.
2014-08-26 12:43:39 -04:00
Dolphin Bot f52888d3ec Merge pull request #884 from FioraAeterna/ppcfpopt
PPCFP: add comment
2014-08-26 18:28:26 +02:00
Fiora 288babf414 PPCFP: add comment 2014-08-26 09:08:22 -07:00
Fiora 90324f3809 JIT64: add mulhwx implementation 2014-08-26 01:09:04 -07:00
Rachel Bryk 5d20455e7d Disable all logs by default. 2014-08-26 03:59:17 -04:00
Pierre Bourdon 81048f38f8 Merge pull request #875 from lioncash/statics
DiscIO: Make the unordered set in IsSoundFile static
2014-08-26 09:35:46 +02:00
Lioncash 5082afa670 DiscIO: Move VolumeDirectory off of raw pointers 2014-08-26 00:23:16 -04:00
Fiora aaca1b01e5 JIT64: clean up and unify float load/store code
While we're at it, support a bunch of float load/store variants that weren't
implemented in the JIT. Might not have a big speed impact on typical games but
they're used at least a bit in povray and luabench.

694 -> 644 seconds on povray.
2014-08-25 19:51:40 -07:00
Lioncash f18fec81fe DiscIO: Make the unordered set in IsSoundFile static
Doesn't need to be instantiated every time the function is called.
2014-08-25 19:56:09 -04:00
Pierre Bourdon 25bd734578 Merge pull request #838 from Tilka/unused
CommandProcessor: mark some functions as static
2014-08-25 21:24:15 +02:00
Tillmann Karras 07c7e6f35e CommandProcessor: mark some functions as static 2014-08-25 21:09:42 +02:00
Lioncash 44ee2f20b9 Merge pull request #874 from FioraAeterna/fixidiocy
JIT: fix incredibly silly mistake in fmul rounding patch
2014-08-25 13:22:33 -04:00
Lioncash 8a77fe0539 Merge pull request #865 from lioncash/debugger-stuff
DolphinWX: Use wxGraphicsContext in the Code View for the debugger.
2014-08-25 13:21:32 -04:00
Fiora f04e362721 JIT: fix incredibly silly mistake in fmul rounding patch 2014-08-25 10:10:28 -07:00
Pierre Bourdon c9f3fd356b Merge pull request #872 from dolphin-emu/revert-828-pp-shader-catch-broken-config
Revert "Catch broken configurations inside of the Post Processing shader...
2014-08-25 14:34:04 +02:00
Pierre Bourdon bf93920c05 Revert "Catch broken configurations inside of the Post Processing shaders." 2014-08-25 14:33:41 +02:00
Dolphin Bot 2f2f992bc7 Merge pull request #828 from Sonicadvance1/pp-shader-catch-broken-config
Catch broken configurations inside of the Post Processing shaders.
2014-08-25 09:17:30 +02:00
comex c87a0021f4 Merge pull request #866 from comex/warnings2
Remove unused variable m_zero.
2014-08-24 17:02:59 -04:00
comex 6574682ff5 Remove unused variable m_zero. 2014-08-24 16:22:19 -04:00
comex d128795594 Merge pull request #862 from comex/registersinuse
Reduce my idiocy in register saving code.
2014-08-24 16:16:32 -04:00
comex a7752f49be Merge pull request #861 from comex/warnings
Fix warnings for OS X
2014-08-24 16:15:58 -04:00
comex b458d1a367 Merge pull request #859 from comex/doubleclick
Improve wording of a particularly atrocious message.
2014-08-24 16:14:03 -04:00
comex 80da767576 Improve wording of a particularly atrocious message.
(Now without gettextize.)
2014-08-24 16:00:58 -04:00
comex cf01f47b52 Fix bloody printf specifiers.
In particular, even in code that only runs on x86-64, you can't use
PRIx64 for size_t because, on OS X, one is unsigned long and the other
is unsigned long long and clang whines about the difference.  I guess
you could make a size_t specifier macro, but those are horribly ugly, so
I just used casting.

Anyone want to make a nice (and slow) template-based printf?

Now without bare 'unsigned'.
2014-08-24 15:56:41 -04:00
Lioncash f239ea3853 DolphinWX: Parenthesize some expressions in CodeView.cpp 2014-08-24 15:40:19 -04:00
Lioncash c3e41809d9 DolphinWX: Move the CodeView debugger view over to wxGraphicsContext
This is a more advanced drawing 'backend' over the previous one and allows us to control things like transparency and anti-aliasing, etc.
2014-08-24 15:40:05 -04:00
Pierre Bourdon 9ff7125786 Merge pull request #810 from lioncash/controller-interface
InputCommon: Don't base default radius of analog sticks off of their name
2014-08-24 19:58:25 +02:00
Pierre Bourdon ebf1b98106 Merge pull request #834 from FioraAeterna/fixfmulrounding
JIT64: Fix fmul rounding issues
2014-08-24 19:49:56 +02:00
Fiora 4d7b1275c9 Interpreter: apply the same odd rounding to single multiplies as the JIT 2014-08-24 10:28:52 -07:00
Fiora 4f18f6078f JIT64: Fix fmul rounding issues
Thanks to magumagu's softfp experiments, we know a lot more about the Wii's
strange floating point unit than we used to. In particular, when doing a
single-precision floating point multiply (fmulsx), it rounds the right hand
side's mantissa so as to lose the low 28 bits (of the 53-bit mantissa).

Emulating this behavior in Dolphin fixes a bunch of issues with games that
require extremely precise emulation of floating point hardware, especially
game replays. Fortunately, we can do this with rather little CPU cost; just ~5
extra instructions per multiply, instead of the vast load of a pure-software
float implementation.

This doesn't make floating-point behavior at all perfect. I still suspect
fmadd rounding might not be quite right, since the Wii uses fused instructions
and Dolphin doesn't, and NaN/infinity/exception handling is probably off in
various ways... but it's definitely way better than before.

This appears to fix replays in Mario Kart Wii, Mario Kart Double Dash, and
Super Smash Brothers Brawl. I wouldn't be surprised if it fixes a bunch of
other stuff too.

The changes to instructions other than fmulsx may not be strictly necessary,
but I included them for completeness, since it feels wrong to fix some
instructions but not others, since some games we didn't test might rely on
them.
2014-08-24 10:28:52 -07:00
Pierre Bourdon aaff5a0afb Merge pull request #856 from FioraAeterna/ppcfpopt
JIT: faster PPC_FP code
2014-08-24 19:25:56 +02:00
Dolphin Bot f31ebd23bb Merge pull request #864 from FioraAeterna/avx2bmi
Add AVX2/BMI1/BMI2 detection support
2014-08-24 18:55:01 +02:00
Fiora ce6d09ca5d Add AVX2/BMI1/BMI2 detection support
Also clean up the formatting in a bit of the CPU detection code.
2014-08-24 09:14:54 -07:00
Pierre Bourdon b749a74d33 Merge pull request #863 from lioncash/remhost
UnitTests: Remove now non-existent UpdateStatusBar from StubHost.
2014-08-24 14:41:13 +02:00
Lioncash f200a03d97 UnitTests: Remove now non-existent UpdateStatusBar from StubHost. 2014-08-24 08:34:25 -04:00
comex d19ec35363 Reduce my idiocy in register saving code.
(1) Rename ABI_ALL_CALLEE_SAVED to ABI_ALL_CALLER_SAVED, because that's
what it was actually defined as (and used as).  Derp.

(2) RegistersInUse is always used for the purpose of saving registers
before calling a C++ function in the middle of a JIT block (without
flushing).  There is no need to save callee-saved registers in this
case.  Change the name to CallerSavedRegistersInUse and mask with
ABI_ALL_CALLER_SAVED.

Nothing obvious broke when starting up a Melee game.  (I added a test
for anything actually being masked out; it happens, but in this
particular case seemed to occur at most a few dozen times per second, so
the actual performance benefit is probably negligible.)
2014-08-23 15:46:10 -04:00