<no_ios_reload/> in meta.xml causing hard drive not detected

[dagger2b]

Sorry about that must be getting tired. I will remove that old link. Can you do the same in your reply…

[aka107]

@fattest wrote:

seems none of the apps like my usb hard drive. i copied the apps folder from my sd card to my usb drive and none of my forwarders work, and hbc does not mount my usb drive. priiloader says can not mount fat volume. any ideas?

Of course, that change is in libogc not in wiimc, I’ve improved libogc with fattest tests and now that improve is in libogc but all applications need to be recompiled using last libogc including hbc.

mrgreg
About your problem using usb to load wiimc I guess it’s a time issue, probably ios58 need 1 or 2 secs to init the ehci module, . Try this new beta but I think is not going to work:
http://www.mediafire.com/?96ajme2j7sm19bf

[Anonymous]

@rodries wrote:

Of course, that change is in libogc not in wiimc, I’ve improved libogc with fattest tests and now that improve is in libogc but all applications need to be recompiled using last libogc including hbc.

thanks for verifying, this is what i speculated in my following post. i will wait till things are recompiled before moving apps from my sd card to my usb drive.

[dagger2b]

Thanks Rodries,

I will try and make it over to his place this evening and test out the new beta. I think you are correct in your assumption about a timing issue. This might explain why removing allows his drive to be detected from HBC because that must take time to reload the ios. However when running from the forwarder do you think that this also is the problem? When the app is run from the forwarder, is meta.xml even used? In this case will an iso reload occur? Thanks…

[dagger2b]

Hi Rodries,

OK I grabbed his hard drive a brought it home with me. Two reasons for this. First I do not have to keep running back and forth to test this. Secondly it is a good test to see if his wii is causing the problem. I tested his drive on my Wii with the latest that you sent me on mediafire. I did not bother with SD card tests. The results are the same as before…

The following tests were run with in meta.xml…
________________________________________________________

1. Running from USB

a. Launching app from HBC – Fails with “Could not find a valid SD or USB device”.
b. Launching app from CF – Fails with “Could not find a valid SD or USB device”.

The following tests were run without in meta.xml…
________________________________________________________

1. Running from USB

a. Launching app from HBC – Successfully runs and mounts USB.
b. Launching app from CF – Fails with “Could not find a valid SD or USB device”.

So once again his drive just does not like the because removing it allows success from HBC. However again it fails when run from the forwarder. Again I will ask if meta.xml is referenced when running from a channel forwarder? In this case does does an ios reload occur? I have to run out for a while but will be back this evening. I will check for a reply from you then…

[rodries]

HBC by default does an IOS reload before launching an application. To suppress the IOS reload (and thereby keep DVD support), we can add the no_ios_reload flag to meta.xml.

The channel forwarder does not do an IOS reload. When you launch the channel, IOS58 is loaded once, and is not reloaded before the application is.

So to sum it up, when you add no_ios_reload it makes HBC act just like the channel forwarder.

[dagger2b]

Tantric,

Thanks for your explanation. Makes perfect sense now. When running from HBC without no_ios_reload, HBC does an automatic ios reload allowing the drive to be detected. The forwarder ignores meta.xml and never does an ios reload which causes the drive to not be detected. The forwarder is able to find the USB drive and launch the app without doing an ios reload so why can’t the app find the USB drive without an ios reload? Also why does running from HBC or the forwarder using the SD card allow the USB drive to be detected with or without no_ios_reload? Is there anyway I can force the forwarder to do an ios reload perhaps through code using ios_reload(). Thanks for your support…

[dagger2b]

Tantric and Rodries,

I just noticed that Tantric made a change to fileopp.cpp in r665 that will try and mount the USB device for up to 5 seconds. Is there any chance you could release a beta7 so I can test my friends drive again? It would be greatly appreciated…

[aka107]

@mrgreg wrote:

Tantric and Rodries,

I just noticed that Tantric made a change to fileopp.cpp in r665 that will try and mount the USB device for up to 5 seconds. Is there any chance you could release a beta7 so I can test my friends drive again? It would be greatly appreciated…

The beta:
http://www.mediafire.com/?ka5ly1w7xf40dh5

[dagger2b]

Rodries thanks for uploading the new beta for me. Well I have good news and bad. Here are my results…

The following tests were run with in meta.xml…
________________________________________________________

1. Running from USB

a. Launching app from HBC – Fails with black screen never displays “Could not find a valid SD or USB device”. 😮
b. Launching app from CF – Success the USB drive is found and mounted. 😀

The following tests were run without in meta.xml…
________________________________________________________

1. Running from USB

a. Launching app from HBC – Successfully runs and mounts USB. 😀
b. Launching app from CF – Success the USB drive is found and mounted. 😀

The wait loop has succeeded in allowing the device to be found and mounted when running from the forwarder. This is great news now he can run the forwarder. Thanks guys.

I am confused with the results of 1a running from HBC with no_ios_reload/> in meta.xml. This should now work because as Tantric stated “when you add no_ios_reload it makes HBC act just like the channel forwarder.” Since the forwarder now works so should running from HBC. Any idea why?

I allowed the wii to sit on the black screen for about 15 minutes when running test 1a. The app is not locked up because when I unplug the usb cable from the wii, I instantly get the failure message “Could not find a valid SD or USB device”. It is as if the error message was there all along but not brought to the foreground.

Now that the forwarder is working I can live with removing no_ios_reload from meta.xml. However can we fix the error message and bring it to the foreground? Then possibly post one more beta with this fixed? Thanks again guys for all you do…

[aka107]

@mrgreg wrote:

I allowed the wii to sit on the black screen for about 15 minutes when running test 1a. The app is not locked up because when I unplug the usb cable from the wii, I instantly get the failure message “Could not find a valid SD or USB device”. It is as if the error message was there all along but not brought to the foreground.

This problem is because of the new hard reset, when hardreset try to remount the device and the device is locked and the call to get the device info (IOS_Ioctl(fd, USBV5_IOCTL_GETDEVPARAMS, io_buffer, 0x20, buffer, desc_out_size);) is not in async mode, so IOS_Ioctl wait for ever.

Try this version: http://www.mediafire.com/?dqgw6uf9c79fom6

[dagger2b]

I just tried this new version with my buddies drive. The test is run from HBC with no_ios_reload included in meta.xml. The result is the same as before. It just goes to black screen and does not display the error message. Pulling the usb cable or powering down the drive brings the message up immediately.

[rodries]

How long did you wait?

[dagger2b]

Just tried it again and waited 5 minutes. Same result just black screen until I power down the drive or pull the cable.

[Author]

Posts