Commit Graph

5980 Commits

Author SHA1 Message Date
Gregory Hainaut 45930ecbf3 gsdx: value was wrongly overwritten
Need a lots of tests to ensure at least no regression on cache behavior

Issue #332
2015-03-17 09:39:50 +01:00
Gregory Hainaut 3b5367c5b7 Merge pull request #452 from turtleli/fix-linux-build-avx2-with-tsx-disabled
GSdx: Linux: Fix build failure on AVX2 processor with disabled TSX, enable SSE4.2, SSE4.1, SSSE3 builds
2015-03-01 17:53:21 +01:00
Jonathan Li 1222bcbf6e GSdx: Comment out TransactionScope code
The TransactionScope class is compiled but never used. Comment it out
for now.
2015-03-01 16:40:48 +00:00
Gregory Hainaut c1604c4b38 gsdx🧵 only keep std::mutex/std::condition_variable
Keep pthread for thread until std::thread is really working
2015-03-01 17:29:40 +01:00
3kinox 7f1e907616 + Correct faulty behaviour of std thread by using native posix handle.
+ This is not yet perfect. Really, this standard seems like a load of crap to me in fact...
At least it works now. Should test again when gcc 5 & new c++ libs gets out.. Until then, it will do.
2015-03-01 17:29:39 +01:00
Gregory Hainaut b68270ded1 Merge pull request #448 from PCSX2/lilypad-linux-port
Lilypad linux port
Basic support of Keyboard and Joystick

No gui to configure anything. Not well tested yet.
2015-03-01 16:52:23 +01:00
Gregory Hainaut 05eff696d3 Merge branch 'Leucos8-master' 2015-03-01 16:49:37 +01:00
Leucos 4587a1e64e IT_it translation update 2015-03-01 16:44:59 +01:00
Jonathan Li 13c0e6f5ef GSdx: Explain TransactionScope changes 2015-02-28 14:21:44 +00:00
3kinox 4211d51f07 64 bits: solve various of type problems 2015-02-27 22:07:13 +01:00
Gregory Hainaut 283f5236a5 Merge branch 'xsacha-icore-cleanup'
From PR #266

I drop sVU modification because sVU will be dropped (ifdef out) on the future
2015-02-27 21:24:58 +01:00
Sacha 4bb41bc267 pcsx2: More cleanup of iCore.
Greg: Drop Svu changes, Svu will be dropped in the future anyway
2015-02-27 21:20:11 +01:00
Sacha b2738fdd5b Remove 3dnow (unused, obsolete).
Greg: merge the removal of EMMS stuff
2015-02-27 21:12:34 +01:00
Gregory Hainaut f254d44030 CDVDlinuz: fix gtk header include
It doesn't fix GTK3 build but it allows to see the real error now.
2015-02-27 10:11:38 +01:00
Gregory Hainaut fb100e05f2 cmake: improve previous commit
Avoid to set -m32 two times

Fix issue #463
2015-02-27 10:02:38 +01:00
Gregory Hainaut b5612ec622 cmake: always set -m32 for 32 bits build
It fixes cross-compilition issue when users/env set CC/CXX variables
2015-02-22 22:17:38 +01:00
Avi Halachmi (:avih) 6b48784709 gzip iso: reduce seek time on fragmented files by OS prefetch on win32
ask the OS to asynchronously fetch the next chunk from disk before we actually
use it. We then cancel this request and ignore the data before the next extract
but it's very likely that when extract tries to read from disk the data will
already be cached by the OS.

This frequently allows to overcome HDD seek time (mostly due to fragmentation)
of the compressed file.
2015-02-22 16:53:51 +02:00
Avi Halachmi (:avih) 5e88afe978 patches: print the actual patch lines only in verbose mode 2015-02-22 16:53:49 +02:00
Jonathan Li 329953ebc3 GSdx: Linux: Enable SSE4.2, SSE4.1 and SSE3 builds 2015-02-22 11:18:23 +00:00
turtleli 9b572d2d47 GSdx: Linux: Fix AVX2 CPU build with TSX disabled
On Linux, CPUs with AVX2 instruction sets that have TSX disabled (by
microcode update or otherwise) fail to build GSdx. The __RTM__ macro is
undefined, with leads to the TSX RTM instruction set (_xbegin, _xend,
_xabort, etc.) being unavailable.

Modify the preprocessor check so that the RTM instructions are only used
if available.
2015-02-22 11:18:23 +00:00
Avi Halachmi (:avih) 56b2c009cd patches: unify patch application verbosity
PCSX2 applies patches from several sources: GameIndex.dbf, cheats folder,
widescreen patches folder and widescreen patches zip (cheats_ws.zip).

all these patche sources are typically applied whenever any kind of
configuration changes (e.g. aspect ratio etc), depending on config.

typically we want to display all this info at the console/log only on boot,
unless we're in Devel mode or if the user enables the verbose console.

this patch unifies all the verbosity related prints to use PatchesCon, where
this object is controlled according to the conditions (devel mode, boot time,
console mode).

it replaces passing around a "verbose" and includes some more prints which did
not adhere to this policy earlier (e.g. the patches lines were only printed
if verbose console was enabled - now they also print on boot).

warnings and errors are unmodified and always print to the console.
2015-02-21 16:19:24 +02:00
Gregory Hainaut 330d14941f gsdx-linux: support dump mode on linux
It could be useful to analyze GS dump. Warning it consumes a lot of
disk space.
2015-02-21 13:51:06 +01:00
Gregory Hainaut 8d76586dec lilypad: fix windows build
I think lilypad used libc instead of libc++
2015-02-21 11:08:45 +01:00
Gregory Hainaut 642371996a lilypad: add joy support based on linux evdev
I wanted to play with linux interface :)

It is far from perfect but joysticks are detected. The biggest issue is the correct generation of the config file!
2015-02-21 11:08:45 +01:00
Gregory Hainaut 3a2e8f38df lilypad: removed old linux/LilyPad.cpp file 2015-02-20 23:05:21 +01:00
Gregory Hainaut 8350dc2c68 lilypad: remove debug printf 2015-02-20 23:05:21 +01:00
Gregory Hainaut 13bb1ae028 lilypad: Keep original KeyboardQueue (more ifdef)
It would be nice if windows support basic std::mutex

The reverse fifo queue is kept in Linux/KeyboardQueue.cpp
2015-02-20 23:05:21 +01:00
Gregory Hainaut 349bb810f0 lilypad: merge LilyPad.cpp (ifdef, ifdef)
Avoid duplication for linux
2015-02-20 23:05:21 +01:00
Gregory Hainaut 7985f5114c cmake: enable lilypad
I manage to have x/start button working with 0/1 key. Here the reference configuration.

[General\ Settings]
Force\ Cursor\ Hide=0
Mouse\ Unfocus=0
Background=0
Multiple\ Bindings=0
DirectInput\ Game\ Devices=0
XInput=0
DualShock\ 3=0
Multitap\ 1=0
Multitap\ 2=0
Escape\ Fullscreen\ Hack=0
Disable\ Screen\ Saver=0
Logging=0
Save\ State\ in\ Title=0
GH2=0
Turbo\ Key\ Hack=0
Vista\ Volume=0
Close\ Hacks=0
Keyboard\ Mode=16
Mouse\ Mode=0
Volume=0
[Pad\ 0\ 0]
Mode=1
Auto\ Analog=0
[Pad\ 0\ 1]
Mode=1
Auto\ Analog=0
[Pad\ 0\ 2]
Mode=1
Auto\ Analog=0
[Pad\ 0\ 3]
Mode=1
Auto\ Analog=0
[Pad\ 1\ 0]
Mode=1
Auto\ Analog=0
[Pad\ 1\ 1]
Mode=1
Auto\ Analog=0
[Pad\ 1\ 2]
Mode=1
Auto\ Analog=0
[Pad\ 1\ 3]
Mode=1
Auto\ Analog=0
[Device\ 0]
Display\ Name=displayName
Instance\ ID=instanceID
Product\ ID=deviceID
API=16
Type=1
Binding 0=0x00040030, 0, 31, 65536, 0, 0, 0
Binding 1=0x00040031, 0, 19, 65536, 0, 0, 0
Binding 2=0x00040002, 0, 50, 65536, 0, 0, 0
Binding 3=0x00040003, 0, 51, 65536, 0, 0, 0
Binding 4=0x00040004, 0, 52, 65536, 0, 0, 0
Binding 5=0x00040005, 0, 53, 65536, 0, 0, 0
Binding 6=0x00040006, 0, 54, 65536, 0, 0, 0
Binding 7=0x00040007, 0, 55, 65536, 0, 0, 0
Binding 8=0x00040008, 0, 56, 65536, 0, 0, 0
Binding 9=0x00040009, 0, 57, 65536, 0, 0, 0
Binding 10=0x0004000C, 0, 58, 65536, 0, 0, 0
Binding 11=0x0004000D, 0, 59, 65536, 0, 0, 0
Binding 12=0x0004000E, 0, 60, 65536, 0, 0, 0
Binding 13=0x0004000F, 0, 61, 65536, 0, 0, 0
Binding 14=0x00200010, 0, 62, 65536, 0, 0, 1
Binding 15=0x00200011, 0, 63, 65536, 0, 0, 1
Binding 16=0x01020012, 0, 33, 65536, 0, 0, 13172
Binding 17=0x02020012, 0, 35, 65536, 0, 0, 13172
Binding 18=0x02020014, 0, 39, 65536, 0, 0, 13172
Binding 19=0x02020015, 0, 38, 65536, 0, 0, 13172
2015-02-20 23:05:21 +01:00
Gregory Hainaut 42b9932bd0 lilypad: add code to bind key 2015-02-20 23:05:21 +01:00
Gregory Hainaut 3096e94616 lilypad : plug a basic keyboard device
Add a reverse fifo to transfer event from core to plugin

Long story short, on linux there is only one event queue in core
2015-02-20 23:05:21 +01:00
Gregory Hainaut aec43f0bee lilypad: add x11 display pointer
If we move all window management hack into the core it would be useless (for input)
2015-02-20 23:05:21 +01:00
Gregory Hainaut 087f252909 lilypad: add plumbering config
Note: it miss interaction with any dialog
2015-02-20 23:05:21 +01:00
Gregory Hainaut 4f013a2960 lilypad: import common utility to manage config
Based work on spu2x equivalent file. Only convert the file to a CfgHelper object
2015-02-20 23:05:21 +01:00
Gregory Hainaut 156f66ef62 lilypad: make it compile on linux
* Ifdef windows code
* Add a windows crap to linux crap translation layer
2015-02-20 23:05:21 +01:00
Gregory Hainaut 3661f51bcb lilypad: minor c++ fixes
Must be checked on windows
2015-02-20 23:05:21 +01:00
Gregory Hainaut 147353a177 lilypad: add cmake file 2015-02-20 23:05:21 +01:00
Gregory Hainaut 76ff56a4e3 lilypad: add a linux fork
Yes it is bad!
2015-02-20 23:05:21 +01:00
Gregory Hainaut 2471306fc4 cmake: actually onepad requires X11 2015-02-20 23:05:21 +01:00
Gregory Hainaut 1655055f61 onepad: update old comment 2015-02-20 23:05:21 +01:00
Gregory Hainaut 2c481113ad update sv_SV mo file 2015-02-20 19:54:58 +01:00
Gregory Hainaut e0f56282de cmake: no preferred-stack-option in asan
It conflicts with the global definition

I don't remember why this option was set on GSdx. Potentially it could be dropped (or fixed correctly)

Anyway, it will help to enable Address Sanitizer on Linux Build
2015-02-20 19:44:51 +01:00
Gregory Hainaut 866f052e34 Merge pull request #442 from pgert/master
sv_SE update
2015-02-20 19:39:55 +01:00
ramapcsx2 5c6b5374af Less verbose SPR stall control logging, avoids fps dives in ST:Voyager 2015-02-15 18:09:14 +01:00
Avi Halachmi (:avih) f2657ae450 gzip iso: 2nd attempt to fix non-english file names (windows only) 2015-02-10 22:49:29 +02:00
pgert 9a60050e93 Update pcsx2_Main.po 2015-02-06 16:33:34 +01:00
pgert aa7e745b43 Update pcsx2_Iconized.po 2015-02-06 16:32:06 +01:00
Avi Halachmi (:avih) f3bb434b27 CDVD: more descriptive browse dialog title 2015-02-05 04:15:49 +02:00
ramapcsx2 f88bea5caf PCSX2 will now default to using the preset system for new installs. That
looks far more friendly to new users when they go to configure the
emulator for the first time."
2015-02-01 19:49:10 +01:00
Avi Halachmi (:avih) c14a4f274f GameIndex.dbf: sort entries 2015-02-01 20:18:59 +02:00