Mobility Radeon - 10.6.6/7 guide for QE/CI - testing 10.7

[kizwan]

I don't know why but I can't post on mucha's post , so I'll post here what is happening me.

The new user lounge where mucha post the guide is a moderated forum. You can post there & your post only appear after moderator approved it. Your post is now visible at Post #34.

 

If you don't mind, I have a suggestion for you. Try get LVDS & VGA working with Peregrine first. I believe you can get both working. When you have LVDS & VGA, then you can concentrate on the HDMI using other framebuffer like Shrike. I know your focus is HDMI but at least get the LVDS & VGA working first.

[wmarsh]

The only things left to do is put the right Encoder & Transmitter values in the kext file. Are you doing from scratch again?

Tried it, did not work

 

One thing that is different about our cards is the order they appear in radeondump:

 

bash-3.2# ./radeondump < ./VBIOS/9552.0301.00E0.vga.rom 
ATOM BIOS Rom: 
SubsystemVendorID: 0x1028 SubsystemID: 0x02aa
IOBaseAddress: 0xde00
Filename: BR31245C.001
BIOS Bootup Message: 
Dell_Roberts_M92S_GDDR3 M92 GDDR3 64bit 450e/600m						   

Connector at index 0 type: VGA (1)
Connector's i2cid: 91
Connector at index 1 type: LVDS (7)
Connector's i2cid: 90

Bcc9 listed the ports in the order in his tool.

So I am considering trying switching the order in the patch.

10 00 00 00 10 00 00 00 00 01 00 00 00 10 00 02
02 00 00 00 40 00 00 00 09 01 00 00 12 00 01 01

I am also wondering if in fact I should try other Framebuffers. I have never gotten output of any kind with Peregrine and only got external with Shrike. Its easy enough to recompile chameleon trunk to inject different FB, in fact crazybirdy has posted his loaders for different FB in mucha's thread - he has his HD 4330 working.

 

Any thoughts?

 

[EDIT] Attached find ATIFramebuffer.kext with all birds patched with above patch and modified chameleon trunk bootloaders. (Kabyl does not work for me).

[kizwan]

Tried it, did not work

 

One thing that is different about our cards is the order they appear in radeondump:

 

 

Bcc9 listed the ports in the order in his tool.

So I am considering trying switching the order in the patch.

Don't compare with my card. I have completely different card. You're worrying for something that is not important. Please observed mine:-

ATOM BIOS Rom: 
SubsystemVendorID: 0x1028 SubsystemID: 0x02bd
IOBaseAddress: 0x0000
Filename: BR034783.006
BIOS Bootup Message: 
BR034783-006 M92 DDR3 64bit 500e/800m									   

Connector at index 0 type: VGA (1)
Connector's i2cid: 95
Connector at index 1 type: HDMI-A (11)
Connector's i2cid: 91
Connector at index 2 type: LVDS (7)
Connector's i2cid: 90

senseid = (i2cid & 0xf) + 1
VGA:  (95 And 0xf) + 1 = 6
HDMI: (91 And 0xf) + 1 = 2
LVDS: (90 And 0xf) + 1 = 1

All of my connections are working after patching the Shrike connections table. As you can see I didn't follow the order but still working for me:-

Patched connection table:-
--> 00037298	 02  00  00  00  40  00  00  00  09  01  00  00  12  00  00  01 	(LVDS)
--> 000372A8	 10  00  00  00  10  00  00  00  00  01  00  00  00  10  01  06 	(VGA)
--> 000372B8	 00  08  00  00  00  02  00  00  00  01  00  00  20  01  02  02 	(HDMI)

I am also wondering if in fact I should try other Framebuffers. I have never gotten output of any kind with Peregrine and only got external with Shrike. Its easy enough to recompile chameleon trunk to inject different FB, in fact crazybirdy has posted his loaders for different FB in mucha's thread - he has his HD 4330 working.

Based on my test, any "birds" framebuffer will work but yeah, try other framebuffer if you want.

Any thoughts?

Not for the moment. I'll probably get an idea after I see your IORegistry dump.

 

EDIT: Everything already in order for you. Just need to figure out the correct Encoder & Transmitter values (for correct link for your card). That's all.

 

I have another suggestion for you. Try "brute-force" method. These are LVDS connection I found in the 10.6.6 ATIFramebuffer file:-

02 00 00 00 40 00 00 00 29 00 00 00 00 01 01 03
02 00 00 00 40 00 00 00 09 00 00 00 00 01 01 03
02 00 00 00 40 00 00 00 09 01 00 00 22 01 00 07
02 00 00 00 40 00 00 00 09 01 00 00 02 01 00 05
02 00 00 00 40 00 00 00 09 01 00 00 00 00 00 05

Disconnect the external monitor. Try one by one of the above LVDS connection. Just change the last bytes (senseid) with your LVDS senseid value. Hopefully one of them will make your LVDS working.

[wmarsh]

Not for the moment. I'll probably get an idea after I see your IORegistry dump.

ioreg is a moving target, depending how I boot.

Attached find ioreg when booting Flicker with above patch, no EDID injection, and Video Bios in extra in the format GraphicsEnabler looks for (1002_9552_02aa1028.rom)

 

I have no video with this setup, but I have it set to start Terminal on login so I can type blind and generate data.

EDIT: Everything already in order for you. Just need to figure out the correct Encoder & Transmitter values (for correct link for your card). That's all.

My dmesg is posted above. I think they are correct.

http://www.insanelymac.com/forum/index.php...t&p=1656253

I have another suggestion for you. Try "brute-force" method. These are LVDS connection I found in the 10.6.6 ATIFramebuffer file:-

02 00 00 00 40 00 00 00 29 00 00 00 00 01 01 03
02 00 00 00 40 00 00 00 09 00 00 00 00 01 01 03
02 00 00 00 40 00 00 00 09 01 00 00 22 01 00 07
02 00 00 00 40 00 00 00 09 01 00 00 02 01 00 05
02 00 00 00 40 00 00 00 09 01 00 00 00 00 00 05

Disconnect the external monitor. Try one by one of the above LVDS connection. Just change the last bytes (senseid) with your LVDS senseid value. Hopefully one of them will make your LVDS working.

Already did that when bcc9 first posted his guide, before dong and mucha gave us more input on LVDS.

 

I am going to try more tommorrow. Probably finish going through the birds, then maybe try again with EDID injected one of the 2 ways. I appreciate ideas, just frustrated and tired.

[kizwan]

ioreg is a moving target, depending how I boot.

IORegistry contain important information & useful when troubleshooting. The IORegistry dump you gave doesn't contain the part I want to look. Why don't you setup remote connection? Connect to your OS X remotely via VNC & use IORegistryExplorer to dump the IORegistry. You said you were able to get external monitor working. Can I get IORegistry dump when the external monitor working?

[tommac_osx86]

I also have a ATI Mobility Radeon HD 4330 (512 MB) in my Lenovo IdeaPad U450P - M23L6GE. At the moment I'm still running on windows 7. I tried to install SL on it, but only got black display with white scratches on internal display, so I stopped two month ago. I followed this thread for weeks until now.

 

Now it is time for me to check, if I can switch to Snow Leo with fully supported graphics (LVDS and HDMI).

 

Which steps do I have to prepare for a successful switch?

Can I dump any information on windows 7, so that anyone of you can check which and in what order I do have to patch any working ATI-Kext?

 

I really read nearly all comments, but there is so much heavy stuff in there, I really do not know from which point I could start of testing.

 

I will write a full guide for my card/Lenovo, if you anybody could provide a little help.

[kizwan]

I really read nearly all comments, but there is so much heavy stuff in there, I really do not know from which point I could start of testing.

Please read mucha's guide. You'll find what you need there. Sorry, no "lightweight" stuff if you really want to get it working. I think better start with bcc9's guide first, & after that read mucha's guide.

[wmarsh]

Crazybirdy reports in mucha's thread his Mobility Radeon HD 4330 works with:

 

02 00 00 00 40 00 00 00 09 01 00 00 12 00 00 07 LVDS
10 00 00 00 10 00 00 00 00 01 00 00 00 10 01 01 VGA

 

The difference between his hardware and mine is his LVDS is on port 0 and mine is on port 1

 

So I would conclude our method works well for any ATI Mobility with LVDS on port 0

 

Port 1 is a work in progress; I have not given up.

 

I have been trying:

10 00 00 00 10 00 00 00 00 01 00 00 00 10 00 02
02 00 00 00 40 00 00 00 09 01 00 00 12 00 01 01

 

However, this gives KP in Shrike (see above test proposed -- I have not finished all FB yet)

 

And I noticed that no Apple FB have Control Flag 0x0040 on port 1.

But there are LVDS on port 1

 

So as a shot in the dark I am considering:

10 00 00 00 10 00 00 00 00 01 00 00 00 10 00 02
02 00 00 00 00 01 00 00 09 01 00 00 12 00 01 01

Copying the port 1 from Kakapo, Raven or Shrike

(and trying these FB rather than Peregrine)

 

Any thoughts?

[c0ke]

Do you have another computer? The best way to work/troubleshoot the problem is to connect remotely from another computer. You can dump the IORegistry, check whether necessary kexts loaded or not, etc with this method. I'm guessing the ATIFramebuffer.kext is not loaded. Please execute this command to refresh the kext cache:-

touch /System/Library/Extensions

Then reboot. Make sure you boot with GraphicsEnabler=Yes. Make sure Peregrine framebuffer is use by checking in IORegistry. If you can give me the IORegistry dump, I can look into it further.

 

Hi, I've tried again with no luck. And once the grey apple and spinning wheel disappears and a completely blank/grey screen appears, it seems the system hangs and I must force it to shut down.

 

I think my dump is slightly different than the one you get, if you want I could send it to you by mail, since I cannot write a pm to you..

[mucha]

I have been trying:

10 00 00 00 10 00 00 00 00 01 00 00 00 10 00 02
02 00 00 00 40 00 00 00 09 01 00 00 12 00 01 01

 

However, this gives KP in Shrike (see above test proposed -- I have not finished all FB yet)

 

And I noticed that no Apple FB have Control Flag 0x0040 on port 1.

But there are LVDS on port 1

 

 

 

Any thoughts?

 

I don't know,based on my experience i'm just guessing your problem is one of two or maybe both:

 

1- the bootloader:

i was using Anval bootloader before a then Kabyl's one and during my test every time i got ATiFramebuffer.kext & ATI4600Controler.kext both installed my boot process just stuck a SLOT---1 message with no disk activity ( no VNC) and when only one of the two was installed i was getting the same but with backgroud sound (VNC working)but no acceleration due to missing kext

by luck only i tried a chameleon rev 747 based bootloader and i got both ATIFramebuffer & ATI4600Controller loaded & i could reach the desktop

 

2-the ControlFlag:

as you have guessed & as i said on my guide there is two possible value for ATY,ControlFlag for LVDS: 0x40 & 0x100.

[wmarsh]

Hi, I've tried again with no luck. And once the grey apple and spinning wheel disappears and a completely blank/grey screen appears, it seems the system hangs and I must force it to shut down.

 

I think my dump is slightly different than the one you get, if you want I could send it to you by mail, since I cannot write a pm to you..

c0ke: you and I have same hardware.

The problem is our LVDS is port 1; kizwan's is port 0

 

I may have a fix, see my last post in this thread, but I have not tested it yet.

I invite you to join me in testing.

 

I don't know,based on my experience i'm just guessing your problem is one of two or maybe both:

 

1- the bootloader:

i was using Anval bootloader before a then Kabyl's one and during my test every time i got ATiFramebuffer.kext & ATI4600Controler.kext both installed my boot process just stuck a SLOT---1 message with no disk activity ( no VNC) and when only one of the two was installed i was getting the same but with backgroud sound (VNC working)but no acceleration due to missing kext

by luck only i tried a chameleon rev 747 based bootloader and i got both ATIFramebuffer & ATI4600Controller loaded & i could reach the desktop

 

2-the ControlFlag:

as you have guessed & as i said on my guide there is two possible value for ATY,ControlFlag for LVDS: 0x40 & 0x100.

The bootloader is a real issue. Kabyls does not work for me either. I recompiled chameleon trunk from google code for all the birds and am now using. Don't know version number but as I svn the code it should be current. Binaries are posted above.

 

I think the control flag is real issue and am excited to try this when I get home tonight.

[tommac_osx86]

Unfortunately I cannot reply to this thread within the new users section

 

So, if any admin reads this, this can be moved, if you want.

 

@crazybirdy:

I change the framebuffer to Peregrine for 4330.

 

Finally, my Mobility Radeon HD 4330 (1002_9552) is working with QE & CI on Internal LVDS screen now.

Could you please provide some more info on how you got your graphics fully supported?

 

Or do I just take your one of your boot-files (i386-4330-Peregrine?) and place it within the chameleon bootloader?

Do i also have add or patch any ATIFramebuffer.kext or delete something else?

At the end add "GraphicsEnabler=Yes" to "apple.boot.com....", correct?

 

On which laptop hardware is your system running. Mine is Lenovo Ideapad U450p with a dedicated ATI Mobility Radeon HD 4330 512 MB Ram. My first try with a retail Snow Leo on it resulted a black display with little horizontal white scatches while installing via 8GB USB thumb drive. How do I get a "normal" output, so I can install my Lenovo laptop?

 

I could try to dump the videobios with everest within Windows 7, if this helps.

 

Regards, tommac

[kizwan]

I think my dump is slightly different than the one you get, if you want I could send it to you by mail, since I cannot write a pm to you..

Please compare yours & the one I extracted from Inspiron 1545 BIOS:-

ATOM BIOS Rom: 
SubsystemVendorID: 0x1028 SubsystemID: 0x02aa
IOBaseAddress: 0x0000
Filename: BR32831.001 
BIOS Bootup Message: 
BR32831-001 M92 DDR2 64bit 450e/500m										

Connector at index 0 type: VGA (1)
Connector's i2cid: 91
Connector at index 1 type: LVDS (7)
Connector's i2cid: 90

Which part are different? The IOBaseAddress?

 

c0ke: you and I have same hardware.

The problem is our LVDS is port 1; kizwan's is port 0

This is dump from my graphic card's ROM. My LVDS at index 2:-

ATOM BIOS Rom: 
SubsystemVendorID: 0x1028 SubsystemID: 0x02bd
IOBaseAddress: 0x0000
Filename: BR034783.006
BIOS Bootup Message: 
BR034783-006 M92 DDR3 64bit 500e/800m									   

Connector at index 0 type: VGA (1)
Connector's i2cid: 95
Connector at index 1 type: HDMI-A (11)
Connector's i2cid: 91
Connector at index 2 type: LVDS (7)
Connector's i2cid: 90

[Ccc]

......

 

Our thanks to dong (for his RadeonHD.kext we have tried to obsolete and his tool)

to bcc9 for inventing the method.

to mucha for the last piece to the puzzle

 

My dell inspiron 1464 (mobility hd4330, 512M 0x95521002) finally supported with QE/CI.

I followed exactly mucha's instructions.

one thing to mention, my LVDS port is port 2, if it helps.

LVDS and VGA works since I do not have hdmi monitor to test, so I changed only those two sections.

I tried Quali, flicker, kakapo, shrike and peregrine.

I did not get well until I notice that there's NullCPUPowerManagement.kext playing around causing blackscreen or timed out. After remove the kext the graphics worked, both internal and VGA!

 

Many thanks people!

[kizwan]

My dell inspiron 1464 (mobility hd4330, 512M 0x95521002) finally supported with QE/CI.

I followed exactly mucha's instructions.

one thing to mention, my LVDS port is port 2, if it helps.

LVDS and VGA works since I do not have hdmi monitor to test, so I changed only those two sections.

I tried Quali, flicker, kakapo, shrike and peregrine.

I did not get well until I notice that there's NullCPUPowerManagement.kext playing around causing blackscreen or timed out. After remove the kext the graphics worked, both internal and VGA!

 

Many thanks people!

 

kizwan, your dump is the same as mine, port 0-vga-i2cid95, port 1=hdmi=i2cid91, port2-lvds-i2cid90

Congratulation! I believe your HDMI should work too. HDMI works nicely on my Dell.

[wmarsh]

@crazybirdy:

Could you please provide some more info on how you got your graphics fully supported?

crazybird has the details for his patch posted in mucha's thread, read down further

He used method similiar to kizwan's but with different encoder/decoder to support his hardware

[wmarsh]

I don't know,based on my experience i'm just guessing your problem is one of two or maybe both:

 

1- the bootloader:

i was using Anval bootloader before a then Kabyl's one and during my test every time i got ATiFramebuffer.kext & ATI4600Controler.kext both installed my boot process just stuck a SLOT---1 message with no disk activity ( no VNC) and when only one of the two was installed i was getting the same but with backgroud sound (VNC working)but no acceleration due to missing kext

by luck only i tried a chameleon rev 747 based bootloader and i got both ATIFramebuffer & ATI4600Controller loaded & i could reach the desktop

 

2-the ControlFlag:

as you have guessed & as i said on my guide there is two possible value for ATY,ControlFlag for LVDS: 0x40 & 0x100.

Thank you again for your great guide and advice.

 

I agree the bootloader is a big issue and I also have not been able to get ATY injected with Kabyl. I recompiled chameleon trunk current downloaded from google code and only modifying ATI.C for all the birds and have been using this. Boot files included in test post above.

 

I think the ControlFlag is the real issue and testing will prove if my theory is right.

ControlFlag for LVDS on index 0 is 0x40

ControlFlag for LVDS on index 1 (?and higher??) is 0x100

 

Please compare yours & the one I extracted from Inspiron 1545 BIOS:-

ATOM BIOS Rom: 
SubsystemVendorID: 0x1028 SubsystemID: 0x02aa
IOBaseAddress: 0x0000
Filename: BR32831.001 
BIOS Bootup Message: 
BR32831-001 M92 DDR2 64bit 450e/500m										

Connector at index 0 type: VGA (1)
Connector's i2cid: 91
Connector at index 1 type: LVDS (7)
Connector's i2cid: 90

Which part are different? The IOBaseAddress?

 

 

This is dump from my graphic card's ROM. My LVDS at index 2:-

ATOM BIOS Rom: 
SubsystemVendorID: 0x1028 SubsystemID: 0x02bd
IOBaseAddress: 0x0000
Filename: BR034783.006
BIOS Bootup Message: 
BR034783-006 M92 DDR3 64bit 500e/800m									   

Connector at index 0 type: VGA (1)
Connector's i2cid: 95
Connector at index 1 type: HDMI-A (11)
Connector's i2cid: 91
Connector at index 2 type: LVDS (7)
Connector's i2cid: 90

Sorry my mistake.

 

Still, c0ke and I have basically the same hardware, so we are likely to have same problem.

 

Mucha thinks we should try the other ControlFlag for LVDS (0x100)

 

Please compare yours & the one I extracted from Inspiron 1545 BIOS:-

ATOM BIOS Rom: 
SubsystemVendorID: 0x1028 SubsystemID: 0x02aa
IOBaseAddress: 0x0000
Filename: BR32831.001 
BIOS Bootup Message: 
BR32831-001 M92 DDR2 64bit 450e/500m										

Connector at index 0 type: VGA (1)
Connector's i2cid: 91
Connector at index 1 type: LVDS (7)
Connector's i2cid: 90

Which part are different? The IOBaseAddress?

Here is mine. Yes the base address is different

bash-3.2# ./radeondump < ./VBIOS/9552.0301.00E0.vga.rom
ATOM BIOS Rom:
SubsystemVendorID: 0x1028 SubsystemID: 0x02aa
IOBaseAddress: 0xde00
Filename: BR31245C.001
BIOS Bootup Message:
Dell_Roberts_M92S_GDDR3 M92 GDDR3 64bit 450e/600m						  

Connector at index 0 type: VGA (1)
Connector's i2cid: 91
Connector at index 1 type: LVDS (7)
Connector's i2cid: 90

This is dump from my graphic card's ROM. My LVDS at index 2:-

Sorry for my misunderstanding, I thought yours was at index 0

 

However, this answers my earlier question that the order of the lines does not reflect the order in bcc9's tool.

[kizwan]

These are my conectors in my Vbios:

$ ./radeondump < 9553.0301.00E0.vga.rom 

ATOM BIOS Rom: 
SubsystemVendorID: 0x1025 SubsystemID: 0x0205
IOBaseAddress: 0x2000
Filename: BR33359D.bin
BIOS Bootup Message: 
Acer_JV50_MV_M92M2_XT_DDR3 M92 DDR3 64bit 680e/800m						 

Connector at index 0 type: LVDS (7)
Connector's i2cid: 96
Connector at index 1 type: VGA (1)
Connector's i2cid: 90
Connector at index 2 type: HDMI-A (11)
Connector's i2cid: 91

 

And the radeon's log:

RHDAtomOutputAllocFree
Mapping DIG1 encoder to KLDSKP_UNIPHYE
Crtc[0]: found native mode from Monitor[LVDS Panel]: 
Modeline "1366x768"  72  1366 1414 1446 1486  768 771 775 806
Connector "PANEL" uses Monitor "LVDS Panel":
RHDValidateScaledToMode
rhdModeValidateCrtc
DxModeValid: ATOM CRTC 1
rhdAtomOutputModeValid
Listing modesetting layout:
ATOM CRTC 1: tied to Atom PLL 1 and LUT A:
Outputs: AtomOutputtainCountUniphyE (PANEL)
ATOM CRTC 2: unused
Unused Outputs: AtomOutputDACA, AtomOutputUniphyA

 

I tried patching Shrike, Vervet and Peregrine:

 

Shrike = Both displays were detected, however there wasn't any signal in both display (black screens) (checked via VNC)

02 00 00 00 40 00 00 00 09 01 00 00 12 00 00 07 LVDS
10 00 00 00 10 00 00 00 00 01 00 00 00 10 01 01 VGA
00 08 00 00 00 02 00 00 00 01 00 00 20 01 02 02 HDMI

Based on the dmesg:-

CRTC0 > DIG1 > UNIPHYE > LVDS: Encoder=0x00 Transmitter=0x12

CRTC1 > DAC > DACA > VGA: Encoder=0x10 Transmitter=0x00

CRTC1 > DIG2 > UNIPHYA > HDMI: Encoder=0x01 Transmitter=0x10

 

So the connections table should be:-

02 00 00 00 40 00 00 00 09 01 00 00 12 00 00 07 (LVDS)

10 00 00 00 10 00 00 00 00 01 00 00 00 10 01 01 (VGA)

00 08 00 00 00 02 00 00 00 01 00 00 10 01 02 02 (HDMI)

[wmarsh]

I just received the following from a new user (pebuls4090) who apparently cannot post in this thread.

 

Hi I've sended a pm to Coke and now to you.

 

QUOTE QUOTE Hi! Firs of all I'm sorry for my bad english.I'm a newbie in the forum and I can't post in the topic about ati4330.My dell 1545 is the same like yours.I've read all that you and the other ones posted.Let me know you how I was able to get qe/ci with the last instructions - I've removed the Lion's kexts and put back the snow leo 10.6.6 kexts (Ati4600Controller.kext with dev id added into info.plist, AtiSupport.kext and patched for our card AtiFramebuffer.kext-I've patched this one like it was described in kizwan's instructions).GraphicsEnabler=Yes.Chameleon is ver.2.0 rc 5 -package made by D-r Hurt.I've run kext utility rebooted.The screen was grey.Then I've rebooted with -v -x and repaired permission with disk utility.Reboot and the screen was black.Shut down and connect the external monitor- black on external and grey screen on internal.While the external was connected I've rebooted with -x and voala- my laptop's screen was with full resolution,but external was black.Second reboot without any flags (except GraphicsEnabler=Yes) and absolutely everything works on laptop's monitor -dvd,dashboard's effects,preview,transparent toolbar with green apple,but my external monitor is black and I don't need it actually.This was my short guide and I hope it will be usefull for all of you.I may upload any files if you need it.

 

Hi, thank you for your support!!!

I don't know why you cannot write on the topic, if you want I can report your experience so it could be usefull for other users!

I'll try to follow your steps as soon as I'll have time to spend!

 

 

I can't write on the topic, because I'm a new in the forum. Ok If you want you can report my explanatation.Here is my Ati4330 files (these are at this moment in System/Library/Extensions and Snow leo files.I've installed 10.6.0 retail,then updated to 10.6.6 and installed chameleon from ati4330 folder. I've installed kexts from Snow leo folder,rebooted and then installed ati kexts. I wish you success!

 

--------------------

Dell Inspiron 1545 Core 2 Duo 2,2Ghz; 4 GB SDRAM;

ATI Mobility Radeon HD4330 512MB

AppleAirPortBrcm4311.kext WIFI; VoodooPS2.kext; VoodooHDA.kext audio;

Dualboot: Win7 & Snow Leopard 10.6.6(all working except sleep)

 

He reports using

Patched connection table:-
CODE--> 00010F60 02 00 00 00 40 00 00 00 09 01 00 00 02 01 00 01 (LVDS)
--> 00010F70 10 00 00 00 10 00 00 00 00 01 00 00 00 10 00 02 (VGA)

 

This did not work for me.

I suspect multiple hardware versions of Inspiron 1545

I am trying to get his radeondump to help understanding.

[hjs89]

Based on the dmesg:-

CRTC0 > DIG1 > UNIPHYE > LVDS: Encoder=0x00 Transmitter=0x12

CRTC1 > DAC > DACA > VGA: Encoder=0x10 Transmitter=0x00

CRTC1 > DIG2 > UNIPHYA > HDMI: Encoder=0x01 Transmitter=0x10

 

So the connections table should be:-

02 00 00 00 40 00 00 00 09 01 00 00 12 00 00 07 (LVDS)

10 00 00 00 10 00 00 00 00 01 00 00 00 10 01 01 (VGA)

00 08 00 00 00 02 00 00 00 01 00 00 10 01 02 02 (HDMI)

 

Thank you! I've done some test without any luck.

 

I've tested:

 

- Peregrine = LDVS + HDMI = Ldvs working, hdmi display recognized but without signal.

- Motmot = LDVS + VGA = Ldvs working, vga display recognized but without signal.

- Shrike = LDVS + VGA + HDMI = Ldvs and hdmi recognized but without signal. Vga not tested.

- Vervet = LDVS + VGA + HDMI + BLANK = Ldvs and hdmi recognized without signal. VGA working but with EDID switched with Lvds (It's like happened me some time ago with HDMI). Maybe if I start with hdmi It'll work with the same issue.

[KUGLee]

I can confirm that pebuls4090's method works. I used kizwan's values for patching.

Dell Inspiron 1525 ATI Radeon HD 4330 (512MB)

[kizwan]

Thank you! I've done some test without any luck.

 

I've tested:

 

- Peregrine = LDVS + HDMI = Ldvs working, hdmi display recognized but without signal.

- Motmot = LDVS + VGA = Ldvs working, vga display recognized but without signal.

- Shrike = LDVS + VGA + HDMI = Ldvs and hdmi recognized but without signal. Vga not tested.

- Vervet = LDVS + VGA + HDMI + BLANK = Ldvs and hdmi recognized without signal. VGA working but with EDID switched with Lvds (It's like happened me some time ago with HDMI). Maybe if I start with hdmi It'll work with the same issue.

Try change ControlFlags for LVDS from 0x40 to 0x100:-

02 00 00 00 00 01 00 00 09 01 00 00 12 00 00 07 (LVDS)

 

BTW, I'm not using Kabyl's boot file anymore since I saw a System Profiler's screenshots at mucha's thread. I compiled Chameleon 2 from official trunk with these changes in ati.c file:-

Original: const char *ati_efidisplay_0[] = { "@0,ATY,EFIDisplay", "TMDSB" };

Mod: const char *ati_efidisplay_0[] = { "@0,ATY,EFIDisplay", "LVDS" };

 

and

 

Original: { 0x10029553, "Peregrine"} ,

Mod: { 0x10029553, "Shrike"} ,

 

For the changes from TMDSB to LVDS, I have no idea whether it effected or not on LVDS functionality. I'm too lazy to test. I changed it because I saw in mucha's screenshot of the IORegistry (ATY,EFIDisplay).

[wmarsh]

I can report significant progress on the difficult version of Inspiron 1545

 

I am now fully connected using patch:

10 00 00 00 10 00 00 00 00 01 00 00 00 10 00 02
02 00 00 00 00 01 00 00 09 01 00 00 12 00 01 01

using any of the birds except Kakapo and Shrike as evidenced by this ioreg

	| |   +-o AGP@1  <class IOPCIDevice, id 0x1000001c6, registered, matched, a$
| |   | +-o IOPCI2PCIBridge  <class IOPCI2PCIBridge, id 0x100000210, regist$
| |   |   +-o VID@0  <class IOPCIDevice, id 0x1000001c7, registered, matche$
| |   |	 +-o ATI4600Controller@1  <class ATI4600Controller, id 0x1000002$
| |   |	 +-o ATY,Peregrine@0  <class IOFBStub, id 0x1000002c1, registere$
| |   |	 | +-o ATY_Peregrine  <class ATIFramebuffer, id 0x1000002d3, reg$
| |   |	 |   +-o AMDNDRVService  <class AtiAppServices, id 0x1000002d6, $
| |   |	 |   +-o AppleMCCSControlModule  <class AppleMCCSControlModule, $
| |   |	 |   | +-o AppleMCCSParameterHandler  <class AppleMCCSParameterH$
| |   |	 |   +-o AppleUpstreamUserClientDriver  <class AppleUpstreamUser$
| |   |	 |   +-o IOFramebufferI2CInterface  <class IOFramebufferI2CInter$
| |   |	 |   +-o IOFramebufferUserClient  <class IOFramebufferUserClient$
| |   |	 +-o ATY,Peregrine@1  <class IOFBStub, id 0x1000002c2, registere$
| |   |	 | +-o ATY_Peregrine  <class ATIFramebuffer, id 0x1000002dc, reg$
| |   |	 |   +-o AppleMCCSControlModule  <class AppleMCCSControlModule, $
| |   |	 |   | +-o AppleMCCSParameterHandler  <class AppleMCCSParameterH$
| |   |	 |   +-o AppleUpstreamUserClientDriver  <class AppleUpstreamUser$
| |   |	 |   +-o AMDNDRVService  <class AtiAppServices, id 0x1000002e2, $
| |   |	 |   +-o IOFramebufferI2CInterface  <class IOFramebufferI2CInter$
| |   |	 |   +-o IOFramebufferUserClient  <class IOFramebufferUserClient$
| |   |	 +-o ATISupport  <class ATISupport, id 0x1000002be, registered, $
| |   |	 +-o ATIRadeonX2000  <class ATIRadeonX2000, id 0x1000002bf, regi$
| |   |	   +-o IOService  <class IOService, id 0x100000303, registered, $
| |   |	   | +-o IOHWSensor  <class IOHWSensor, id 0x10000030d, register$
| |   |	   +-o ATIR6002DContext  <class ATIR6002DContext, id 0x10000032e$
| |   |	   +-o ATIR6002DContext  <class ATIR6002DContext, id 0x100000330$

However, I still have black screen on LVDS. System is running and I have access to terminal.

 

Interestingly, if I add EDID to com.apple.Boot.plist, I no longer connect!

 

Next I intend to try this EDID fix

http://www.insanelymac.com/forum/index.php?showtopic=208410

[Ccc]

Congratulation! I believe your HDMI should work too. HDMI works nicely on my Dell.

 

 

Did you noticed that the fan keeps spinning all the time?

the cpu usage is not high, so does the graphics i think.

do you have any solution for that?

[crazybirdy]

crazybirdy has the details for his patch posted in mucha's thread, read down further

He used method similiar to kizwan's but with different encoder/decoder to support his hardware

 

Great, I can post here now.

 

Base on mucha's guide, I wrote a guide in chinese as http://bbs.pcbeta.com/thread-848362-1-1.html.

Now we got port 0, port 1, and port 2 work with QE & CI for mobility radeon HD 4330.

 

Here is the details of port 1 FYI. The hardware is Dell 1440. Hope it's helpful to you.

 

ATOM BIOS Rom:

SubsystemVendorID: 0x1028 SubsystemID: 0x02cf

IOBaseAddress: 0xde00

Filename: BR32939.003

BIOS Bootup Message:

Dell_Alba_M92LP_DDR2 M92 DDR2 64bit 450e/500m

 

Connector at index 0 type: VGA (1)

Connector's i2cid: 91

Connector at index 1 type: LVDS (7)

Connector's i2cid: 90

 

Port 0　　VGA : SenseLine = (91 & 0xf)+1 = 2

Port 1　　LVDS : SenseLine = (90 & 0xf)+1 = 1

 

Attaching Output AtomOutput UniphyE to Connector PANEL

ObjectTable - size: 134, BIOS - size: 65536 TableOffset: 40 object_header_end: 174

DisplayPathObjectTable: entries: 2 version: 1

DisplaPathTable[0]: size: 10 DeviceTag: 0x1 ConnObjId: 0x3105 NAME: VGA GPUObjId: 0x1100

GraphicsObj[0] ID: 0x2115 Type: 0x2 ObjID: 0x15 ENUM: 0x1 NAME: INTERNAL_KLDSCP_DAC1

DeviceIndex: 0x0

DisplaPathTable[1]: size: 10 DeviceTag: 0x2 ConnObjId: 0x310e NAME: LVDS GPUObjId: 0x1100

GraphicsObj[0] ID: 0x2121 Type: 0x2 ObjID: 0x21 ENUM: 0x1 NAME: UNIPHY2

DeviceIndex: 0x1

Output: AtomOutput DACA[0x 1] - adding devices:

Output: AtomOutput UniphyE[0x b] - adding devices:

Sensing DACA on Output AtomOutput DACA

DAC_LoadDetection Successful

BIOSScratch_0: 0x40000

rhdAtomBIOSScratchDACSenseResults: RHD_SENSED_NONE

I2C device "RHD I2C line 0:ddc2" registered at address 0xA0.

rhdAtomLvdsTimings: LVDS Modeline: 1366x768 73000 1366 (1366) 1414 1446 (1506) 1506 768 (768) 771 777 (803) 803

Mapping DIG1 encoder to KLDSKP_UNIPHYE

Crtc[0]: found native mode from Monitor[LVDS Panel]:

Modeline "1366x768" 73 1366 1414 1446 1506 768 771 777 803

Connector "PANEL" uses Monitor "LVDS Panel":

DxModeValid: ATOM CRTC 1

Listing modesetting layout:

ATOM CRTC 1: tied to Atom PLL 1 and LUT A:

Outputs: AtomOutput UniphyE (PANEL)

ATOM CRTC 2: unused

Unused Outputs: AtomOutput DACA

 

　　CRTC0>DIG1>UNIPHYE>LVDS　 Encoder = 0x00　　Transmitter = 0x12

　　CRTC1>DAC > DACA >VGA　　Encoder = 0x10　　Transmitter = 0x00

 

And the patch is as below, use Peregrine and the Kabyl's boot on my post in mucha's thread post#20.

02 00 00 00 40 00 00 00 09 01 00 00 12 00 01 01 LVDS

10 00 00 00 10 00 00 00 00 01 00 00 00 10 00 02 VGA