V8 Forwarders crashing back to system menu

[GoogleCodeExporter]

Hi Dimok,

Just tested both V8 forwarders on my buddies Wii.  Both crash back to the 
system menu.  This was the case with V7 forwarders as well.  I have reinstalled 
V5 for now on his machine which works.  Any ideas?

Original issue reported on code.google.com by gsoud...@cox.net on 22 Sep 2010 at 1:50

[GoogleCodeExporter]

Does he have the IOS58?

Original comment by dimok...@hotmail.de on 22 Sep 2010 at 5:12

[GoogleCodeExporter]

Yes he has the latest IOS58 v6176.

Original comment by gsoud...@cox.net on 23 Sep 2010 at 12:20

[GoogleCodeExporter]

I am 80% sure it's something with his system. I can't go and fix something that 
i can't reproduce sorry.

Original comment by dimok...@hotmail.de on 23 Sep 2010 at 8:31

• Changed state: WontFix

[GoogleCodeExporter]

OK Dimok.  I think it might be his hard drive not being detected by the 
forwarder.  I had a look at the forwarder code main.c.  If I had to guess his 
usb drive is not being detected.  In this case the code will exit via 
SYS_ResetSystem(SYS_RETURNTOMENU, 0, 0).  Which I think returns to the system 
menu.  This is exactly what I am seeing.

I am getting the same result with Tantrics Wiimc forwarder V1.4 on his system.  
It simply returns to the system menu.  I have him running V1.1 of the Wiimc 
forwarder for now.  Tell you what, I will try to run the apps off an SD card on 
his system and see what happens.  I will test both Wiixplorer and Wiimc 
forwarders with an SD card.

I am going on vacation for a week so I will not test this until I return.  
Thanks and I will keep you posted...

Original comment by gsoud...@cox.net on 23 Sep 2010 at 10:19

[GoogleCodeExporter]

Hi Dimok,

I finally got around to testing this on my buddies wii.  Interesting results to 
say the least.  As I stated the V8 forwarder just returns to the system menu 
without launching the app.  However if I run off an SD card with the USB 
disconnected the app launches and runs fine from the forwarder.  If I run the 
forwarder with both the SD card inserted and the USB drive connected, the app 
launches and runs fine with both the SD card and USB drive mounted.  The 
insteresting part is that with both SD and USB attached, the app is run off the 
USB drive not the SD card.  Usually forwarders look for the SD card first and 
then run the app from there.

So I am correct in my assumption that the forwarder is not detecting the USB 
drive and therefor bails out bails out back to the system menu.  As I stated I 
was having the same problem with Tantrics V1.4 forwarder.  However his V1.5 
does not return to the system menu like V1.4.  It runs the app but then errors 
with can not find a valid SD or USB device.  However with both SD and USB 
connected, V1.5 launches the app and runs fine.  Just like Wiixplorer the app 
shows both SD and USB devices.  However wiimc will run the app off the SD not 
the USB as Wiixplorer does.

I have no doubt the fact that my buddies drive is a green and can go to sleep 
is contributing to this problem.  Tantric tells me that HBC 1.8 makes things 
worse because it actually puts the drive to sleep when launching an app.  This 
make no sense to me why HBC would do this but I assume he is correct.  
Wiixplorer runs fine from HBC on his system it is just running from the 
forwarder.  With all this said is there anything you can do to wake the drive 
up when running the app from the forwarder?  Thanks for your support...

Original comment by gsoud...@cox.net on 26 Oct 2010 at 10:04

[GoogleCodeExporter]

Hi Dimok,

Just wanted to let you that I have tried my buddies USB drive on my Wii.  The 
results are the same the forwarder just returns to the system menu.  So it is 
not related to his system.  Just thought I would pass it on...

Original comment by gsoud...@cox.net on 2 Nov 2010 at 4:52

[GoogleCodeExporter]

This is usb device specific issue and not an issue of WiiXplorer but the libogc 
usbstorage handle.
I will compile the next forwarder version with new libogc and than this issue 
might already be fixed even without me doing anything. If you want you can 
compile the current forwarder with new libogc from SVN and see if this issue is 
gone.

I am sure that it is usb device or system related because the forwarder works 
fine for me and many others.

Original comment by dimok...@hotmail.de on 4 Nov 2010 at 4:30

[GoogleCodeExporter]

Dimok this is a usb device specific issue.  I do not believe that the new 
version of libogc will correct this issue.  I believe it is a timing issue.  I 
had the same problem with Tantrics wiimc forwarder V4.  However he corrected 
this issue with V5.  He added a 10 second loop to allow the device time to be 
detected.  Please look at the code changes he made that corrected this 
problem...

http://code.google.com/p/wiimc/source/diff?spec=svn561&r=561&format=side&path=/t
runk/libs/forwarder/source/fileop.c

Original comment by gsoud...@cox.net on 4 Nov 2010 at 8:54

[GoogleCodeExporter]

If it's only a timing issue you should have pointed it out earlier. I'll add 
the wait loop for the next forwarder release. It's a quick change actually.

Original comment by dimok...@hotmail.de on 4 Nov 2010 at 8:58

[GoogleCodeExporter]

Thanks Dimok,

Please understand that it is my friends drive and I have been running back and 
forth.  I finally just took this drive home with me and I still have it.  I 
believe I stated that Tantric had corrected this problem in the second 
paragraph of comment 5.  However I had not examined his code to see how he 
fixed the problem.  Sorry I guess I should of done that.

Even after Tantric fixed the forwarder the app still could not find this drive. 
 The forwarder would find the drive but the app would not.  In fact wiimc could 
not even find his drive when running HBC 1.0.8.  So he added in a 5 second wait 
loop in the app code in r664/r665.  This change allows the app to detect this 
drive when launched from the forwarder.  I still have to remove the 
no_ios_reload line in meta.xml for this drive.  If not the app will not find 
the drive when launched from HBC even with the wait loop in r664/665.  Tantric 
has found a bug in HBC and the implimentation of no_ios_reload that when 
corrected should allow me to add back in the no_ios_reload.

I am not sure that you need the loop to be 10 seconds for the forwarder.  I 
think 5 should do the trick.  However once the drive is found the loop exits so 
it really does not matter.

I will keep my friend running your V5 forwarder for now and await the next 
release from you.  Thanks for your patience and all you do...

Original comment by gsoud...@cox.net on 4 Nov 2010 at 9:25

[GoogleCodeExporter]

Hi Dimok,

My assumption was correct.  It was a timing issue.  I added in wait/retry loop 
for 5 seconds and the forwarder now finds his drive.  I have attached a diff 
file if you want to use it.  Thanks for all your help...

Original comment by gsoud...@cox.net on 7 Dec 2010 at 2:56

[GoogleCodeExporter]

The thing is that i don't want to add a 5 secs delay in there. People starting 
WiiXplorer with Forwarder from SD and don't have USB inserted will always have 
5 secs delay which sucks.

Original comment by dimok...@hotmail.de on 11 Dec 2010 at 10:01

[GoogleCodeExporter]

Hi Dimok,

Sorry for the slow response, I have been very busy.  Sorry that the diff file 
was not a proper patch.  I have deleted it from my last post.  Yes I also 
notice that SD without USB users would have to wait 5 seconds and this does 
suck.  However I have solved that.  These patches were tested against R207 and 
work for both SD and USB with no delays.  I tried R208 but had problems with 
SD.  SD access will lock up my Wii with the DVD slot flashing.  This happens 
both from HBC and the forwarder.  You might want to check this.

Please check the two attachments which are proper patches and let me know what 
you think...

Original comment by gsoud...@cox.net on 13 Dec 2010 at 11:29

Attachments:

• devicemounter.patch
• main.patch

[GoogleCodeExporter]

Alright. Good idea to search first the SD. Why didn't I think of that 
myself?...lol.

Anyway. I will add something very similar but...well let's say with my words 
;). Anyway thanks for doing the tests and providing the solution.

Original comment by dimok...@hotmail.de on 14 Dec 2010 at 7:57

[GoogleCodeExporter]

Sounds good to me.  I will be interested in seeing how you code this.  Let me 
know when you have made the changes so I can have a look...

Original comment by gsoud...@cox.net on 14 Dec 2010 at 3:44