Jump to content
559 posts in this topic

Recommended Posts

Posted (edited)
11 minutes ago, jalavoui said:

idk what ure doing but this only happens if on acel mode? is frame working solo?

 

no fb+gfx

 

WindowServer-2026-05-14-200422.ips WindowServer-2026-05-14-200412.ips WindowServer-2026-05-14-200402.ips WindowServer-2026-05-14-200350.ips WindowServer-2026-05-14-200345.ips

fb.log.zip x.log.zip

Edited by Stezza88

the issue is likely validateDetailedTiming failing for ALL modes leaving CoreDisplay with zero valid modes

That validateDisplayMode → return 0xe00002c2 is the trigger — it causes getPixelInformation for framebuffer 0 failed

Posted (edited)

this is well know see from wserver log. it misses tgl acel files.

mayce cause u're pushing wserver?

 

image.thumb.png.e81aeafc33bf1f28654870cca1ff0ee5.png

 

for setuptiming this var exists at function end and is used in vblank

image.png.f8258e96b338276d572c9a44000196ae.png

here it is

image.png.db5218c3ace6acce72576f6775471b63.png

 

i just saw xlog

u're making wserver crazy

just stop messing

 

and also this acess issue

(Sandbox) Sandbox: WindowServer(417) deny(1) file-map-executable /Library/Extensions/AppleIntelTGLGraphicsMTLDriver.bundle/Contents/MacOS/AppleIntelTGLGraphicsMTLDriver

 

why i like logs? they show the bug

here's your display bug:

image.png.874bf1fd5d76d2cbb68395d10f370525.png

idk how u did this but its dam bugged

also here

image.png.9124a977856007a2ff4aac0bf2c174f9.png

 

it's identified - try fix the cause

Edited by jalavoui

One remaining item — SSDT-GPIO is still missing. If the touchpad still doesn't work after this reboot, that means VoodooGPIO can't find the interrupt pin automatically. In that case you'd need to dump your DSDT and find the GPIO pin number for the I2C touchpad device

Posted (edited)
35 minutes ago, jalavoui said:

u need to remove class structs from ngreen. e.g AppleIntelPlane* as wrong offsets. keep non class only like crtparms, etc

 

this is not pipeindex !

 

image.thumb.png.2fe8402db20f29f05dd262fbfd64ab7e.png

my database says that you're both wrong, just in different ways

Edited by Visual Ehrmanntraut
Posted (edited)

this could be a confirmation

Kernel-2026-05-14-221934.panic

 

IOFramebuffer::open called something via a function pointer / virtual dispatch that resolved to 0x2ed040c0 — a non-mapped userspace address. This is a corrupted vtable or callback pointer, not a logic error.

Key question: does 0x2ed040c0 look like an Intel MMIO register value that could have been written into a vtable slot by one of NootedGreen's

Edited by Stezza88
Posted (edited)

It's like slot machine AI ... i need rehab

 

And i have a plan to accomplish rehab : study the goods, study the bads .. and try to improve it

 

I am trainee of develop angular UI since last month and this is the most beautiful and challenging exercise that i could dream : make the biggest UI works, an entire OS.

 

Thanks to everybody.

Edited by Stezza88
  • Like 1
52 minutes ago, Stezza88 said:

It's like slot machine AI ... i need rehab

 

because it's not real AI, it's the autocomplete on your phone but with TBs of scrapped data and pirated works as its dataset...

10 hours ago, Visual Ehrmanntraut said:

because it's not real AI, it's the autocomplete on your phone but with TBs of scrapped data and pirated works as its dataset...

 

I'm not of the same advise. It makes me horny when starts to use nm, otool etc...

Posted (edited)

Ok, i need to resrore a working version in local ...

 

FB-only is loadin (black screen or fragmented), i wait for logs..

 

Still some problems but I've got logs...

fb.log.zip x.log.zip Lilu_1.7.2_23.6.txt

Edited by Stezza88
On 5/14/2026 at 12:49 PM, jalavoui said:

did u try goto bios and add + mem ?

if the driver trys write pass this adress and can't u will never fix it

image.png.114682887ba9fcc2be4ec0655531a589.png

 

this needs tobe solved, check what happens when u dont have this bug in log

 

image.thumb.png.a7035a9c09a3cde41bdcc7a893560c5d.png

 

I never went to bios.....

Key finding: Sandbox: WindowServer deny(1) file-map-executable /Library/Extensions/AppleIntelTGLGraphicsMTLDriver.bundle — Sandbox blocks WS from mapping the Metal driver as executable. That's a secondary issue (separate from the kernel-side getBlit3DContext).

Posted (edited)

       0x1239b5000 -        0x123e34fff com.apple.driver.AppleIntelTGLGraphicsMTLDriver (16.0.32) <2b849e57-ac6a-3216-a67b-38e080603585> /Library/GPUBundles/AppleIntelTGLGraphicsMTLDriver.bundle/Contents/MacOS/AppleIntelTGLGraphicsMTLDriver
    "CFBundleIdentifier" : "com.apple.driver.AppleIntelTGLGraphicsMTLDriver",
    "path" : "\/Library\/GPUBundles\/AppleIntelTGLGraphicsMTLDriver.bundle\/Contents\/MacOS\/AppleIntelTGLGraphicsMTLDriver",
    "name" : "AppleIntelTGLGraphicsMTLDriver",
 

WindowServer-2026-05-16-152646.ips MTLCompilerService-2026-05-16-152623.ips

fb.log.zip x.log.zip Lilu_1.7.2_23.6.txt

Edited by Stezza88
Posted (edited)

use nblue patches them u get sandbox permission error

u have old plist patch logic that i replace by editing info.plist

but u need nblue kexts not those ure using

 

when u're ready move to icl. i ieft a kp to fix

keep tgl as we might need to inject code from tgl into icl

 

what i notice is code of dmc firmware in tgl as little adress changes regarding icl

hwInitializeCState()+ related functions dont match linux

that is the cause of your display issues

this is "fine" for frameb mode on some displays but not for acel mode

 

i left this as coment

image.png.b0c2bc67c2698de3eb94515614347195.png

Edited by jalavoui
×
×
  • Create New...