| Age | Commit message (Collapse) | Author |
|---|
|
|
|
After seeing the regressions from multitap I took a quick look at the
code, and there's quite a bit of cleanup needed.
This fixes the initial really obvious stuff, a useless copy and a
duplicate in_keystate that can cause compilation issues, along with
bringing styling closer to the pcsx standard so it's a bit more readable.
The multitap changeset also seems to have been done with an editor that
mangled line endings, I'll address that in a followup commit.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
```
frontend/libretro.o: In function `retro_deinit':
libretro.c:(.text+0x3769): undefined reference to `deinit_vita_mmap'
collect2: error: ld returned 1 exit status
make: *** [pcsx_rearmed_libretro.so] Error 1
```
Introduced #68
|
|
|
|
|
|
|
|
untested...
|
|
|
|
|
|
higher C std
|
|
Split update_controller_port_device() into
update_controller_port_variable() updating the controllers from the
variables, update_controller_port_device() updating the controllers from
retro_set_controller_port_device() and controller_port_variable()
sharing getting the controller variable between the two.
Avoid setting to PSE_PAD_TYPE_NONE controllers whose variable is set to
'default' when updating them.
|
|
Check that 'environ_cb' is valid before using it in update_multitap()
and update_controller_port_device() as they may be called from
retro_set_controller_port_device() before 'environ_cb' is set.
|
|
Change the way the key of the controler type variables is passed to one
which doesn't crash.
|
|
Add the 'default' value to controller type variables and set it as the
default value and implements the retro_set_controller_port_device()
function. When the variable's value is 'default', the value set via the
function is used, otherwise the value of the variable is used.
This allows the frontends to use retro_set_controller_port_device() or
the variables, pleasing both the users and frontends prefering the
standard API and the ones prefering the custom API.
|
|
Add the 'auto' multitape vairable value.
This allows to automatically detect multitaps based on the number of
plugged controllers.
|
|
Allow to set no controller to a port.
This will be used in the next commit to allow auto detecting whether
multitaps are plugged.
|
|
Refactor the way the controller type variables are read as we were doing
8 times almost exactly the same thing.
This makes the code more readable and maintainable.
|
|
pass pad index in pad datas structure.
|
|
|
|
|
|
|
|
fix analog value
|
|
|
|
add game/core option to enabled or disabled multitap.
Test with micro machine V3 and 5 devices, 2 multitap enabled, all
device react good i can’t test the 6 7 8 select player but i can select
8 gamer. The 2 multitap works fine.
i detect a bug, when we change the game pad type to analog, the library
crash, see it next time
|
|
|
|
|
|
|
|
github issue #68
|
|
|
|
|
|
the frontend instead.
- dynamically allocate the recompiler cache, this will allow using the
recompiler even when the .bss section is relocated far from the .text
section (for example when using the hombrew loader).
|
|
reported by KP142EH12 and perhaps somebody else I forgot
also take the opportunity to update libpicofe
|
|
|
|
|
|
|
|
- use linearAlloc for vout_buf.
- use svcFlushProcessDataCache to flush only the required range,
invalidating icache doesn't seem to be required.
|
|
|
|
needed svc access.
|
|
|
|
add missing files
|
|
|
|
|
|
|
