mirror of
https://github.com/hrydgard/ppsspp.git
synced 2026-04-24 21:56:10 +03:00
[GH-ISSUE #718] missing save screen issue (Suzumiya) #207
Labels
No labels
Atrac3+
Audio
CPU emulation
D3D11
D3D9 (removed)
Depth / Z
Feature Request
Font Atlas
GE emulation
Guardband / Range Culling
HLE/Kernel
I/O
Input/Controller
MP3
Multithreading
Needs hardware testing
Networking/adhoc/infrastructure
No Feedback / Outdated?
OpenGL
PGF / sceFont
PSMF / MPEG
Platform-specific (Android)
Platform-specific (Windows)
Platform-specific (iOS)
PowerVR GPU
SDL2
Saving issue
User Interface
Vulkan
arm64jit
armjit
armv6
x86jit
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
starred/ppsspp#207
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Originally created by @sum2012 on GitHub (Feb 17, 2013).
Original GitHub issue: https://github.com/hrydgard/ppsspp/issues/718
Game:Suzumiya Haruhi No Tsuisou -ULJS00371
Version:ppsspp-v0.6-504-g9338cf9-windows-x86
(Edit 513 also crash)
JIT crash log: https://gist.github.com/sum2012/4969720
interpreter log:https://gist.github.com/sum2012/4969729
interpreter mode ony can play to attach image.

@unknownbrackets commented on GitHub (Feb 17, 2013):
Have you enabled "Fast memory (dynarec, unstable)"? If yes, it's expected that these lines will crash:
On PCs, slowmem shouldn't be that much slower than fastmem, and it should not crash in this case.
-[Unknown]
@sum2012 commented on GitHub (Feb 17, 2013):
It is default option.
Turn off fix crash.
Now close the issue ?
(Also need make pull request in disable Fast memory in default ?)
@hrydgard commented on GitHub (Feb 17, 2013):
Well, there's still some bug somewhere - the game should not try to write/read invalid addresses, so something has gone wrong earlier.
@unknownbrackets commented on GitHub (Feb 17, 2013):
Most probably with the font or psmf stuff. But it's not a jit bug is what I meant.
-[Unknown]
@sum2012 commented on GitHub (Feb 17, 2013):
@unknownbrackets
I have changed to write/read invalid addresses issue
@sum2012 commented on GitHub (Mar 2, 2013):
v0.6-843 can play futher but random crash

https://gist.github.com/sum2012/0bded2c58b994eac096f
@sum2012 commented on GitHub (Mar 4, 2013):
Sometime another error:black screen after
https://f.cloud.github.com/assets/2177532/213084/94b3e83a-8355-11e2-81ba-95534c7b2994.jpg
log:
https://gist.github.com/sum2012/5c8bcfbc69cdd3fd1144
@unknownbrackets commented on GitHub (Jun 9, 2013):
Is this game working any better now?
-[Unknown]
@sum2012 commented on GitHub (Jun 9, 2013):
ppsspp-v0.7.6-1245
same black screen after save screen.
https://gist.github.com/sum2012/2c8626869962d070a895
Will make VS 2010 debug log when I have time
Some interest
42:16:212 Test Thread I[HLE]: HLE\sceIo.cpp:541 stderr:
C++ runtime abort: internal error: static object marked for destruction more than once
42:16:212 Test Thread E[HLE]: HLE\sceKernelModule.cpp:1256 UNIMPL sceKernelStopUnloadSelfModuleWithStatus(00000001, 00000000, 00000000, 00000000, 00000000): game has likely crashed
42:16:212 Test Thread E[HLE]: HLE\sceKernelMemory.cpp:644 UNIMPL sceKernelPrintf(libc:%s: something wrong 0x%08X, 08ac49e0, 00000000, 00000000)
42:16:212 Test Thread E[CPU]: Mips\MIPSInt.cpp:195 BREAK!
@unknownbrackets commented on GitHub (Jun 9, 2013):
That is very interesting. Looks possibly sceFont related...
-[Unknown]
@sum2012 commented on GitHub (Jun 10, 2013):
@unknownbrackets
Can you correctly update sceKernelPrintf from this
http://forums.ppsspp.org/showthread.php?tid=3125
to me to test.Thanks.
@unknownbrackets commented on GitHub (Jun 10, 2013):
That version of sceKernelPrintf() is unfortunately not correct. Just need to whitelist 'X' really...
But it's basically saying it got a 0 when not expecting it.
-[Unknown]
@sum2012 commented on GitHub (Jun 11, 2013):
JPCSP also have this probrem in new version.
It is about sceUtilityOsk change.Will keep you update when it fixed
@sum2012 commented on GitHub (Jun 12, 2013):
@unknownbrackets JPCSP fix this probrem in
http://code.google.com/p/jpcsp/source/detail?r=3230
Do PPSSPP need do the same ?
@unknownbrackets commented on GitHub (Jun 13, 2013):
I'm not sure it's the same problem then, because that mostly seems to make JPCSP use logic more similar to PPSSPP's current logic (although I'm not sure.)
It does seem to ignore negative values, though, you can try this:
Core/Dialog/PSPOskDialog.cpp:
-[Unknown]
@sum2012 commented on GitHub (Jun 13, 2013):
No help
@unknownbrackets commented on GitHub (Sep 2, 2013):
How does this look now? Still the same?
-[Unknown]
@sum2012 commented on GitHub (Sep 2, 2013):
same.Can I use JPCSP trace log (with some modules) to solve ?
@unknownbrackets commented on GitHub (Sep 2, 2013):
Hmm. It might help, but I'm not sure which functions to hook. Maybe sceSasSetVoicePCM... but not really sure.
-[Unknown]
@sum2012 commented on GitHub (Sep 10, 2013):
@unknownbrackets I should give you see newest debug log first.
https://docs.google.com/file/d/0B3OaSdeV0L8kbVhtRWNZNm1jY3M/edit?usp=sharing
still sceSasSetVoicePCM ?
@unknownbrackets commented on GitHub (Sep 11, 2013):
Well, the game says this:
That's bad. It's within the OSK and PPGe stuff, so likely to be related. It's probably something to do with sceGe, but that will be a lot to log...
It could even just be because the OSK proceeds through the statuses too quickly.
Does it happen if you comment out these lines in Core/Util/PPGeDraw.cpp?
-[Unknown]
@sum2012 commented on GitHub (Sep 11, 2013):
No help
https://docs.google.com/file/d/0B3OaSdeV0L8kLVVDMXMzMVVYVmc/edit?usp=sharing
@sum2012 commented on GitHub (Sep 11, 2013):
Maybe we should review which commit break OSK.
Fast test
good
ppsspp-v0.7.5-276-gb200005
bad
ppsspp-v0.7.5-318-g570ddb2
@unknownbrackets commented on GitHub (Sep 11, 2013):
Before those commits we were overwriting memory and corrupting the game's stack. Maybe there's still an issue, but it was definitely more incorrect before those changes.
-[Unknown]
@sum2012 commented on GitHub (Sep 11, 2013):
git bisect result

edit: this commit ?
github.com/hrydgard/ppsspp@27cf72afb6PPSSPP v0.7.5-280-g7baf27b First OSK break
https://docs.google.com/file/d/0B3OaSdeV0L8kaC1JX2p5V0hQZHc/edit?usp=sharing
PPSSPP v0.7.5-277-gf8d726c Last OSK good
https://docs.google.com/file/d/0B3OaSdeV0L8kZ01GbWVlRFlNQ2M/edit?usp=sharing
@unknownbrackets commented on GitHub (Sep 11, 2013):
Well, the field length is correct, but I found one thing. Does #3732 fix it?
-[Unknown]
@sum2012 commented on GitHub (Sep 11, 2013):
Well done,finally fix it
:)
@sum2012 commented on GitHub (Sep 11, 2013):
Post a good picture to prove

@hrydgard commented on GitHub (Sep 11, 2013):
OK, I close it then :) (I merged the fix)
@sum2012 commented on GitHub (Oct 23, 2013):
Oh no ,the game still invalid addresses
last time I only lucky no probrem.
Firstly I give back PPSSPP v0.9.1-762-gbd8a4ec log (OSK fixed version)
After OSK, invalid addresses
case 1:https://gist.github.com/sum2012/67b36e6bb11a93b289b3
case 2:https://gist.github.com/sum2012/51299f817b3be3edba98
Good log:https://gist.github.com/sum2012/7bcd52366290f0a6e381
should re-open the issue
@sum2012 commented on GitHub (Oct 23, 2013):
PPSSPP v0.9.1-2105-gb0d7f75 log:
bad1:https://gist.github.com/sum2012/fb84e85e0361ec701bec
bad2:https://gist.github.com/sum2012/3cd7779b4dad802b68e4
good:https://gist.github.com/sum2012/f2ff6f4f7d6823c279a5
debug (stop at first error):https://docs.google.com/file/d/0B3OaSdeV0L8kYU91YUMxOV9vLTA/edit?usp=sharing
@unknownbrackets commented on GitHub (Oct 23, 2013):
So it's random? Is it affected by multithreading or buffered rendering settings?
-[Unknown]
@sum2012 commented on GitHub (Oct 23, 2013):
Yes it is random,don't affected by multithreading , buffered rendering or JIT settings.
@sum2012 commented on GitHub (Oct 27, 2013):
Check with real PSP.It is missing this screen when no save.

Click Circle then psp create data.
If PPSSPP use Real psp save ,ppsspp would not have probrem (tested 4 times)
JPCSP trace log (before OSK)
https://gist.github.com/sum2012/d0641205e8a5fefad647
Real PSP save (before OSK)
https://drive.google.com/file/d/0B3OaSdeV0L8kV2ZJQlhKd3lsWUE/edit?usp=sharing
Edit:PPSSPP do not create save before OSK
@sum2012 commented on GitHub (Oct 27, 2013):
PPSSPP log (no save, before OSK)
Info log
https://gist.github.com/sum2012/c72d9900a34bcbfee046
Debug log (no save ,before OSK)
https://drive.google.com/file/d/0B3OaSdeV0L8kdFRlQnc2WXBUeFE/edit?usp=sharing
tag @shenweip whether he has idea to fix
@sum2012 commented on GitHub (Oct 27, 2013):
Add sceUtilitySavedataGetStatus in JPCSP trace log
https://gist.github.com/sum2012/30cf7a85403b8406959f
@unknownbrackets commented on GitHub (Jan 30, 2014):
Just in case: does it work any better with softgpu?
-[Unknown]
@sum2012 commented on GitHub (Jan 30, 2014):
No help
@sum2012 commented on GitHub (Feb 8, 2014):
Well done

PPSSPP v0.9.6-802 The save screen apeear
Log:https://gist.github.com/sum2012/4763b79579b8d07c5ed8
The game should no problem.
@unknownbrackets commented on GitHub (Feb 8, 2014):
Awesome. That makes it seem likely that #5360 will be improved as well.
-[Unknown]