Homebrew Official Retroarch WiiU (wip.)

[Radius4]

Honestly that doesn't change anything, it's a spare time thing for me.

The issue is known (all build failures are reported) and should be easy to fix:
http://p.0bl.net/53843

command.c: In function 'command_network_init':
command.c:382:28: error: implicit declaration of function 'socket_init' [-Werror=implicit-function-declaration]
int fd = socket_init((void**)&res, port,
^~~~~~~~~~~
command.c:383:16: error: 'SOCKET_TYPE_DATAGRAM' undeclared (first use in this function)
NULL, SOCKET_TYPE_DATAGRAM);
^~~~~~~~~~~~~~~~~~~~

If it's not fixed I'll look at the commits to find if something was recently changed, I think this is the command API (for remote control and things like that) and maybe it was enabled on wiiu builds accidentally.

 

[the_randomizer]

Well the issue is known and should be easy to fix:
http://p.0bl.net/53843

command.c: In function 'command_network_init':
command.c:382:28: error: implicit declaration of function 'socket_init' [-Werror=implicit-function-declaration]
int fd = socket_init((void**)&res, port,
^~~~~~~~~~~
command.c:383:16: error: 'SOCKET_TYPE_DATAGRAM' undeclared (first use in this function)
NULL, SOCKET_TYPE_DATAGRAM);
^~~~~~~~~~~~~~~~~~~~

Yeeeeeeah, I don't know anything about compiling, I'm afraid. I'll wait till whenever it gets fixed. Sorry. We only suggested you hack your Wii U since the no. of people who have a Wii U is in short supply. Nothing will convince it seems.

 

[Radius4]

It seems if HAVE_NETWORK_CMD is always defined so this ifdef does not prevent this code from being built for wiiu
https://github.com/libretro/RetroArch/blob/master/command.c#L378

Not sure why the recent changes to command.c cause this to happen though...

It was always like that

https://github.com/libretro/RetroArch/blob/master/Makefile.common#L1432
HAVE_NETWORK_CMD is pretty redundant...

 

[the_randomizer]

It seems if HAVE_NETWORK_CMD is always defined so this ifdef does not prevent this code from being built for wiiu
https://github.com/libretro/RetroArch/blob/master/command.c#L378

Not sure why the recent changes to command.c cause this to happen though...

What happens now, might I ask?

 

[Radius4]

You wait

 

[rioray]

Honestly that doesn't change anything, it's a spare time thing for me.

The issue is known (all build failures are reported) and should be easy to fix:
http://p.0bl.net/53843

command.c: In function 'command_network_init':
command.c:382:28: error: implicit declaration of function 'socket_init' [-Werror=implicit-function-declaration]
int fd = socket_init((void**)&res, port,
^~~~~~~~~~~
command.c:383:16: error: 'SOCKET_TYPE_DATAGRAM' undeclared (first use in this function)
NULL, SOCKET_TYPE_DATAGRAM);
^~~~~~~~~~~~~~~~~~~~

If it's not fixed I'll look at the commits to find if something was recently changed, I think this is the command API (for remote control and things like that) and maybe it was enabled on wiiu builds accidentally.

The culpit is this commit
https://github.com/libretro/RetroArch/commit/bb9efe9bf1020a8812566c307db734138427c6e9
removing HAVE_COMMAND on line 194 cause this.
easy fix is to and 2 &&defined(HAVE_COMMAND) line https://github.com/r-type/RetroArch/commit/547557d44319eb87d148750feeabe7554a494bfe

 

[the_randomizer]

You wait

*Sigh* I'm so effing sorry

 

[Radius4]

lol, @rioray already fixed it, I opened a PR on his behalf.

 

[the_randomizer]

lol, @rioray already fixed it, I opened a PR on his behalf.

That was fast...dayum...

 

[ShadowOne333]

lol, @rioray already fixed it, I opened a PR on his behalf.

Praise the lords and saviours of RA Wii U, @Radius4 and @rioray

 

[Radius4]

I did nothing lol
My server broke, I'm just checking stuff via the github web interface.

 

[the_randomizer]

I did nothing lol
My server broke, I'm just checking stuff via the github web interface.

Even so, we need more support on the Wii U side of things, that's gonna be a helluva task... >< I know some people but...I don't have the balls to ask them.

 

[rioray]

I did nothing lol
My server broke, I'm just checking stuff via the github web interface.

BTW thanks @Radius4 for the keymapper (and all your work in RA side) ,as it is a very cool feature that serve all the computer core (and more) !
For my part to be honest , I did nothing on the RA side , just some contribution in CORE side.

 

[the_randomizer]

Wish there were more people willing to help the Wii U side like you guys I don't want this port to die.

 

[cucholix]

I suggested a quick fix for random crashes, I'm not sure if it would even work, but would be worth a shot: force closing content before swaping cores.

 

[the_randomizer]

I suggested a quick fix for random crashes, I'm not sure if it would even work, but would be worth a shot: force closing content before swaping cores.

Will that be in the new nightly build?

 

[cucholix]

Will that be in the new nightly build?

No lol, it was just an idea from my experience swaping cores , it seems that closing content before swaping to another core make it more stable, totally anecdotal evidence, I lack the skills to code complex stuff, the most I've achieved is edit some obvious things. Sorry :c

 

[Radius4]

I think forcing the user to close content might be good too, I found some other issues on some other platforms connected to this.
I don't think everyone will agree though...

 

[the_randomizer]

I think forcing the user to close content might be good too, I found some other issues on some other platforms connected to this.
I don't think everyone will agree though...

Would would it be a bad thing to force close content if it means higher stability? If not everyone agrees, that's their problem. Either that, or it can be an option to turn off or on.

 

[ShadowOne333]

I think forcing the user to close content might be good too, I found some other issues on some other platforms connected to this.
I don't think everyone will agree though...

Forcing close content before core swap would be the best way to go, I think for other ports as well.