wololo.net/talk

Posted: **Wed Jan 16, 2013 12:56 am**

Are you reaching the menu at that point, or not? because the long list of "sceIoDread_Vita start" tends to say you are in the menu already (which is a good sign).

I am seeing 2 errors here, one in the "update_thread" (I believe that one is part of the menu), but also a breakpoint happening above that.

Posted: **Wed Jan 16, 2013 1:15 am**

wololo wrote:Are you reaching the menu at that point, or not? because the long list of "sceIoDread_Vita start" tends to say you are in the menu already (which is a good sign).

I am seeing 2 errors here, one in the "update_thread" (I believe that one is part of the menu), but also a breakpoint happening above that.

Nope,no menu, only blackscreen, and the breakpoint, it isnt my own so i dont know where it comes from

with pymenu
http://pastebin.com/yCucd08d
edit:wagic similar result^

Posted: **Wed Jan 16, 2013 1:32 am**

it seems "Test sceIoDread" failed.
This is usually the case when you are missing critical functions. Does the log complain about missing NIDS at the top of the log file? could you copy/paste those? The addresses are a reference to the NIDS defined in sdk_hbl.S, so it would be useful if you copied that too.

Posted: **Wed Jan 16, 2013 1:40 am**

wololo wrote:it seems "Test sceIoDread" failed.
This is usually the case when you are missing critical functions. Does the log complain about missing NIDS at the top of the log file? could you copy/paste those? The addresses are a reference to the NIDS defined in sdk_hbl.S, so it would be useful if you copied that too.

Right, I see,
the "HBL FUNCTION MISSING" part

Code: Select all

HBL Function missing at 0x00010028, this could lead to trouble if syscall estimates do not work
HBL Function missing at 0x00010030, this could lead to trouble if syscall estimates do not work
HBL Function missing at 0x00010038, this could lead to trouble if syscall estimates do not work
HBL Function missing at 0x00010048, this could lead to trouble if syscall estimates do not work
HBL Function missing at 0x00010058, this could lead to trouble if syscall estimates do not work
HBL Function missing at 0x000100A0, this could lead to trouble if syscall estimates do not work
HBL Function missing at 0x000100C8, this could lead to trouble if syscall estimates do not work
HBL Function missing at 0x000100E0, this could lead to trouble if syscall estimates do not work
HBL Function missing at 0x000100E8, this could lead to trouble if syscall estimates do not work
HBL Function missing at 0x00010128, this could lead to trouble if syscall estimates do not work
HBL Function missing at 0x00010130, this could lead to trouble if syscall estimates do not work
HBL Function missing at 0x00010180, this could lead to trouble if syscall estimates do not work
HBL Function missing at 0x00010188, this could lead to trouble if syscall estimates do not work
HBL Function missing at 0x00010190, this could lead to trouble if syscall estimates do not work
HBL Function missing at 0x00010198, this could lead to trouble if syscall estimates do not work
HBL Function missing at 0x000101A0, this could lead to trouble if syscall estimates do not work
HBL Function missing at 0x000101A8, this could lead to trouble if syscall estimates do not work
HBL Function missing at 0x000101B0, this could lead to trouble if syscall estimates do not work
HBL Function missing at 0x000101B8, this could lead to trouble if syscall estimates do not work
HBL Function missing at 0x000101C0, this could lead to trouble if syscall estimates do not work

here's my sdk_hbl:
http://pastebin.com/pFFBKTQ0

Yep it does corresponds with sceIoDread among others,the game sadly does not import them
here's sdk.s:
http://pastebin.com/fBYyyeqe

I think those can be gotten via p5 stubs save dialog ? But this game gets stuck in the loop in state 2

Posted: **Wed Jan 16, 2013 1:48 am**

Kankertje wrote: Yep it does corresponds with sceIoDread among others,the game sadly does not import them
here's sdk.s:
http://pastebin.com/fBYyyeqe

I think those can be gotten via p5 stubs save dialog ? But this game gets stuck in the loop in state 2

Basically without sceIoDread, HBL will not work correctly, so you *need* to get p5 stubs code to work.
About save dialog being stuck, I have a discussion on the subject started here: viewtopic.php?f=4&t=11097

I ran into a simlliar issue a while ago, and never found the time to investigate deeply. Lots of things can still be attempted. Try to unload the save dialog before loading it, try to kill some rendering component that is blocking it (can you maybe PM wth about this, I think he ran into similar issues once, where the current GE list was not processed yet, which was blocking a bunch of things)

Posted: **Wed Jan 16, 2013 1:52 am**

wololo wrote:
Kankertje wrote: Yep it does corresponds with sceIoDread among others,the game sadly does not import them
here's sdk.s:
http://pastebin.com/fBYyyeqe

I think those can be gotten via p5 stubs save dialog ? But this game gets stuck in the loop in state 2
Basically without sceIoDread, HBL will not work correctly, so you *need* to get p5 stubs code to work.
About save dialog being stuck, I have a discussion on the subject started here: viewtopic.php?f=4&t=11097

I ran into a simlliar issue a while ago, and never found the time to investigate deeply. Lots of things can still be attempted. Try to unload the save dialog before loading it, try to kill some rendering component that is blocking it (can you maybe PM wth about this, I think he ran into similar issues once, where the current GE list was not processed yet, which was blocking a bunch of things)

Yeah you can notice my post in that topic, it was the same game

, but never got that one working

I guess ill ask around and try what you said, thanks for advice

edit: also not sure if it's important, but the loop runs fine first time, then it relocates stub adresses, and then the loop starts again and get stuck at "2"

Posted: **Wed Jan 16, 2013 3:17 am**

oops, just realized now you were the same person. Sad, I thought we had a "trend"...
Yes, the fact that it works once, then not a second time, is an interesting thing. I don't know how much it helps us, but it tends to say we are changing the state of the system somehow..
Although, I am not sure anymore why 2 loops are needed... haven't looked at the code in a while

Posted: **Wed Jan 16, 2013 11:43 am**

wololo wrote:oops, just realized now you were the same person. Sad, I thought we had a "trend"...
Yes, the fact that it works once, then not a second time, is an interesting thing. I don't know how much it helps us, but it tends to say we are changing the state of the system somehow..
Although, I am not sure anymore why 2 loops are needed... haven't looked at the code in a while

Looking at hbl code in loader.c
it goes like this:
p5_get_stubs:

p5_copy_stubs_savedata
Copy stubs for the savedata dialog with autoload mode
this works fine.

p5_copy_stubs_savedata_dialog
Copy stubs for the savedata dialog with save mode
this doesn't work

So i got this idea of commenting out the second loop, and now i got into "Wagic:Loading Resources"

, but not further
Gonna try some menu's to see what it does
edit: menu from monster hunter vhbl ends up with the update_thread exception mentioned before, and lots of
sceIoDopen_Vita start
sceIoDopen_Vita Done
sceIoDread_Vita start
sceIoDread_Vita Done
then blackscreen

tried wmenu 0.1 from hbl rev 91, loaded up fine and now showing folders/files too!
previously when i had wmenu 0.2 running it never showed and files/folders
so i assume the first loop got the Dread and Dopen function

when trying to run homebrew, results in bus error

Code: Select all

File is [ms0:/PSP/GAME/gpSP/EBOOT.PBP]
-->ALLOCATING MEMORY from partition 2, size 0x00000027... Found free block 1
Got block from kernel with UID 0x04816979
sceIoOpen override: menu.cache become ms0:/PSP/SAVEDATA/censored/menu.cache
_hook_sceKernelExitGame called
Threads cleanup
cleaning audio threads
Exception - Bus error (instr)
Thread ID - 0x04818E61
Th Name   - user_main
EPC       - 0x0812D848
Cause     - 0x10000018
BadVAddr  - 0x00000000
Status    - 0x00088613
zr:0x00000000 at:0x00000001 v0:0x00000000 v1:0x00000001
a0:0x00000000 a1:0xDEADBEEF a2:0xDEADBEEF a3:0xDEADBEEF
t0:0xDEADBEEF t1:0xDEADBEEF t2:0xDEADBEEF t3:0xDEADBEEF
t4:0x882F8700 t5:0x00088613 t6:0x00004000 t7:0x0812D848
s0:0x00000001 s1:0x00010200 s2:0x00000000 s3:0x04818E61
s4:0x00000000 s5:0x08910000 s6:0x09FFD4BC s7:0x08910000
t8:0xDEADBEEF t9:0xDEADBEEF k0:0x09FFDF00 k1:0x00000000
gp:0x08914810 sp:0x09FFD360 fp:0x08910000 ra:0x09F75770

This might be related, I think I got same error when trying to exit the menu, or when trying to exit the hbl test homebrew

But atleast some progress, I guess

edit2: @ the stuck loop part, i noticed the parameter p5_open_savedata(PSP_UTILITY_SAVEDATA_SAVE) <-
this game only has autosave/autoload(might be not related at all, just a wild guess,im a noob

), and via google i found out that there also exists PSP_UTILITY_SAVEDATA_AUTOSAVE
but if i put in that, it crashes right after at

and with regular p5_open_savedata(PSP_UTILITY_SAVEDATA_SAVE), is the p5_close_savedata part important ? wouldnt it work without it ? because it gets the stubs, so only the closing is really a problem

Posted: **Thu Jan 17, 2013 11:33 am**

At this point I would say: comment out the "closing" part, and see if it takes you closer to running the menu.
But I think if you don't close the save dialog, that you might not have control of the screen...

Give it a try, at worst it might help you understand better what's blocking you

Posted: **Thu Jan 17, 2013 12:12 pm**

wololo wrote:At this point I would say: comment out the "closing" part, and see if it takes you closer to running the menu.
But I think if you don't close the save dialog, that you might not have control of the screen...

Give it a try, at worst it might help you understand better what's blocking you

Tried that, didnt really help, I found few more usermode exploits and a particular one had pretty straight-forward hbl porting so now im going to focus on that
When i compile with "distrib" i get few exceptions but i get into wmenu, if i do regular debug compile, it gets same exceptions but never starts the menu
edit:correction, now it seems to work with nonids compile too

Also no homebrew works, gonna post log in few minutes

edit2: heres the full log, loaded up wmenu and tried to run test homebrew
http://pastebin.com/hp2PLNxE

wololo.net/talk

hbl porting help

Re: hbl porting help

Re: hbl porting help

Re: hbl porting help

Re: hbl porting help

Re: hbl porting help

Re: hbl porting help

Re: hbl porting help

Re: hbl porting help

Re: hbl porting help

Re: hbl porting help