Commit Graph

11015 Commits

Author SHA1 Message Date
Matthew Parlane
d261dfaf46 Upgrade libusb to 1.0.16 2013-09-30 18:36:54 +13:00
Ryan Houdek
3c53f2e5e0 [Android] Fix Fastmem on Android 4.2 2013-09-29 20:53:32 -05:00
comex
1a008b9e62 Fix use of ABI_GetAlignedFrameSize. 2013-09-29 16:36:26 -04:00
comex
ccbf2ac21a Match ABI_AlignStack with ABI_RestoreStack properly.
The relevant function is entirely unused, so it shouldn't have any
effect.
2013-09-29 14:59:13 -04:00
Rachel Bryk
b0200219dd Add literally a million blank inputs in netplay when a wiimote changes reporting mode, just to make nsmbw sync. 2013-09-28 23:39:29 -04:00
comex
a2fd52ef4f Move existing Melee AR codes to be for revision 0 and add JMC's for revision 2.
He claims they don't work on that revision either, but from a quick
Google they (most of them?) seem to be intended for it, so keep them in
case they're useful for anyone.
2013-09-28 23:38:40 -04:00
comex
853392b790 Use a separate section for enabled Gecko codes, like AR.
This properly fixes default gecko codes.

It makes perfect sense to have two separate cheat windows and two
separate code paths for the different code formats, right?
2013-09-28 23:38:40 -04:00
comex
f57ff0a569 Support a gcm revision-specific game ini for cheats + partially fix gecko codes in default ini.
The local ini is not revision-specific because it would require renaming
everything.  Meh.
2013-09-28 23:38:25 -04:00
comex
1ed06f1dc4 Reset wxTAB_TRAVERSAL.
Fixes issue 3903.
2013-09-28 23:00:56 -04:00
kostamarino@hotmail.com
cb715f4fc9 Gameini database update. Enforce LLE recompiler in Wind Waker to prevent a situation where LLE interpreter might be enforced without wanting it. If the user selected in the general sound settings LLE interpreter, closed the window to apply it and then reopened it to select HLE, running Wind Waker afterwards would enforce LLE interpreter on boot. 2013-09-28 16:47:42 +03:00
Rachel Bryk
9c53a21c18 Allow setting DSP settings via game ini. 2013-09-28 08:31:28 -04:00
Rachel Bryk
785171abb4 Change iTLBHack to a bool.
It is only used as a bool.

Fixes issue 6668.
2013-09-28 08:07:23 -04:00
Lioncash
1c4c00d171 [Android] Return true upon displaying the exit confirmations dialog in EmulationActivity.java. Now multiple games can be launched like normal. Previously it was returning false. 2013-09-27 18:17:29 -04:00
Lioncash
69d2ecff52 Simplify SwitchPage() within GameListActivity.java. We no longer need the first switch statement, since case 0 will be hit when SwitchPage(0) is called from the folder browser. This means we can also get rid of recreateFragment(), since the only time this needs to be called is within the onCreate() function. When SwitchPage(0) is called, SwitchPage will perform basically the same actions as recreateFragment() would; thus the reason it can be removed. 2013-09-27 17:16:59 -04:00
kostamarino@hotmail.com
ad73efda57 Gameini database update.
Fixes issue 6662.
Fixes most of the games in issue 6666.
2013-09-27 18:39:13 +03:00
Rachel Bryk
cbd366236a Allow loading save states via drag and drop. 2013-09-27 08:38:12 -04:00
skidau
9a2c7df8dc Added a small disc access delay to fix the missing music in Super Monkey Ball 2. 2013-09-27 20:35:27 +10:00
Lioncash
689aca2788 [Android] Remove commented out code within EmulationActivity.java. Since the back button handling has a specific purpose, this is no longer needed. 2013-09-26 23:40:39 -04:00
Lioncash
f6a8733660 [Android] Tiny inline documentation cleanup. 2013-09-26 23:32:41 -04:00
Lioncash
8aba748735 [Android] Fix a situation within the FolderBrowser where the application would crash. listFiles() returns null when either the File object it's called on isn't a directory or if an I/O error happens (in their infinite wisdom, they actually thought NOT throwing an exception was a cool way to handle this. How about that?). In the case of trying to access system directories as a normal user, an I/O error will occur due to permission access rights. This fixes that. 2013-09-26 23:06:10 -04:00
comex
bea76ac129 No need to std::move a return value. Thanks Billiard. 2013-09-26 21:15:35 -04:00
Lioncash
4542b9fcbb [Core] Fix a memory leak in NetPlayServer.cpp in function OnData(). 2013-09-26 15:42:22 -04:00
Ryan Houdek
e37cb1fc76 Implement CR1 for the intepreter. To be honest I have no idea why this was never done previously, all it is is copying four bits from the FPSCR register to CR1. This fixes issue 2390. 2013-09-26 18:09:25 +00:00
Ryan Houdek
4efc3e6c8f Quick build fix. 2013-09-26 07:50:24 +00:00
Ryan Houdek
feaf65f2ae [Android] The dynamic UBO access isn't actually fixed, contrary to what rev cd646d8e236 said. I presumed it fixed with v4x drivers, but I didn't have the LG G2 with me to test 100% at the time. This won't afflict any Adreno device with v4x drivers since UBOs are disabled for them since they are _broken_. 2013-09-26 07:46:56 +00:00
comex
ecca0045a9 Fix Imm8 check.
(I blame whoever made it take a u8 despite logically being a s8.)
2013-09-25 14:29:17 -04:00
Ryan Houdek
691f76b826 [ARM] Implement CR1 setting for the few floating point instructions that I have setting the flags. For the rest, drop to interpreter if it sets CR1. At that point it'll spam a panic alert. I don't quite understand why Interpreter and JIT64/IL don't do this yet, it's a simple 4 bit copy. 2013-09-25 18:17:05 +00:00
comex
54843ad1e8 Need to reload from XMM0 in this case. 2013-09-25 14:16:20 -04:00
Rachel Bryk
74ee85aaac Fix linux, attempt 2. 2013-09-25 06:22:11 -04:00
Rachel Bryk
f77b3ccf8a Fix linux, probably. 2013-09-25 06:13:44 -04:00
Rachel Bryk
93f7622e16 Fix changing wiimotes when starting netplay.
AccessWiiMote() crashed for some users. Not sure why...
2013-09-25 05:44:16 -04:00
comex
ebe4448749 Save only the registers that need to be saved rather than going through ProtectFunction. 2013-09-25 03:15:53 -04:00
comex
2a339c926e Fastmem writes for x86-64. 2013-09-25 03:15:53 -04:00
comex
18abc33306 2x banner images! 2013-09-25 03:06:27 -04:00
comex
0787019cbe Check in 2x versions of Clean*. Thanks, MaJoR1! 2013-09-25 02:06:53 -04:00
comex
693cee36cb Hack up wx to support @2x images in the toolbar. Oh please let's switch to Qt... 2013-09-25 02:03:32 -04:00
Ryan Houdek
624c92f97e [ARM] fresx/fnmaddsx/fselx/frsqrtex/fnmaddx implementations. 2013-09-25 03:00:57 +00:00
comex
a38821fc38 Comment out wx change to set WS_POPUP.
Fixes issue 6651, apparently.  We don't use wxGLCanvas, so the line
isn't necessary in the first place.
2013-09-24 22:48:31 -04:00
Ryan Houdek
ae75f92b2a [ARM] psq_lx/psq_lux/psq_stx/psq_stux implementations. Four more instructions that JIT64 doesn't have. 2013-09-25 02:22:52 +00:00
Ryan Houdek
cd646d89e2 [Android] The issue with the dynamic UBO access on Adreno platforms was fixed with v41 of the video drivers. v41 and above of the video drivers fix the spiky polygon problems that are noticed ingames. 2013-09-25 01:50:02 +00:00
Ryan Houdek
357a7707a6 [ARM] ps_cmpu0/ps_cmpu1/ps_cmpo0/ps_cmpo1 implementations. 2013-09-24 21:13:33 +00:00
Ryan Houdek
74bc855f20 [ARM] ps_res implementation. 2013-09-24 21:00:50 +00:00
Ryan Houdek
3b1b0d3fb5 [ARM] ps_div implementation. 2013-09-24 20:46:57 +00:00
Ryan Houdek
94a731b49d [ARM] fctiwx implementation. 2013-09-24 20:37:10 +00:00
Ryan Houdek
482170c3ea [ARM] Implement subfic with optimizations stolen from JIT64. 2013-09-24 19:01:03 +00:00
Ryan Houdek
405aa30cb8 [ARM] Fix fastmem... 2013-09-24 18:03:06 +00:00
Ryan Houdek
8e2e5a4e70 [ARM] Have both fastmem and non-fastmem paths for floating point loadstores because fastmem is completely broken garbage on Android at this point in time. 2013-09-24 17:40:12 +00:00
Ryan Houdek
5866859ff0 Screw you comex, this doesn't even make any damn sense. 2013-09-24 17:25:13 +00:00
Ryan Houdek
eb6ed3e42a [ARM] Change all floating point loadstores to fastmem implementations except lfs since all floating point accesses tend to be to RAM space. lfs tends to get used to write quickly to the gatherpipe and other places, look at the JIT64 implementation to see how to make it quicker. 2013-09-24 05:41:58 +00:00
comex
29dc253fde Improve context structure handling on non-Windows.
Instead of copying data into and out of a fake CONTEXT structure with
only a few entries, use the platform specific structure directly with a
typedef and macros.  This is needed because fastmem writes need to be
able to access any register from BackPatch.  It adds a fair number of
repetitive defines, but it's better than the alternative.
2013-09-24 01:38:27 -04:00