Jump to content

R9 380, can't boot directly, have to do it via Intel 4600, Ghost monitor.

pcpaul

By pcpaul,
in ATi

 Share
146 posts in this topic

Recommended Posts

pcpaul
pcpaul
Posted
Posted

Ghost Monitor is solved, main problem is not being able to fully boot with AMD set as a primary gpu in BIOS.

 

if I try to boot directly ( in bios pci set as primary gpu) via 380, after verbose screen monitor loses signal. If I set primary graphics as intel 4600, then I have to stare at the black screen until OS overtakes it, because 380 is not primary hence it only works once fully booted to OS.

Reason for this is because El Capitan detects "TWO" monitors, while I only have ONE!

So if I boot with pci as primary, it shows verbose but when OS has to overtake it, it loses signal because OS boots to non-existent monitor, it thinks that I have a 30.5-inch monster between my legs, but I don't, it's only a meager 27-inch

 

here's a screenshot, 30.5-inch monitor on the right is not real, it's El Capitan's drunken delusional mirage... mirage to which it tries to boot.... and apparently boots "successfully", hence I lose a signal on the REAL monitor.... 

 

Screen_Shot_2015_07_22_at_2_08_51_PM.png

 

reconnecting monitor doesn't help... re-detecting displays doesn't help either.

Link to comment
Share on other sites
pcpaul
pcpaul
Posted
  • Author
Posted

What framebuffer are you using? I remember I had issues with a ghost monitor when I accidentally edited my framebuffer with two connectors that had the same sense-id. What model is your GPU (full model, as in manufacturer, etc.) and what ports does it have?

Now i'm using Labrador frame buffer and ghost monitor disappeared, but I can't boot while AMD is set as primary, because after verbose screen monitor loses signal, if I boot with intel as primary, then once OS is loaded I can disconnect HDMI cable from intel and connect it to AMD and it works, so something doesn't allow AMD to pass through "a barrier" right in between the verbose screen transitioning to OS screen.

 

Model is MSI GV-R938G1 GAMING-4GD/F4

Connectors are DP, HDMI, DVI-D and DVI-I

 

I've already worked out connectors for patching and tried various patched buffers by myself, but I'm stuck with AMD set as primary losing signal right after verbose, in yosemite it works without issues even HDMI audio works with simples Clover injection.

 

01 displayport      11 02 01 01

02 HDMI-type-A   21 03 02 03
03 DVI-D              10 00 03 05
04 DVI-I                12 04 04 06 
05 DVI-I                00 10 05 06
 
Plus after new El Capitan update PB2, I can't even use internet because only fakesmc is loaded and no other kexts are accepted, tried moving them to S/L/E and L/E, fixed permissions, rebuild cache, nothing helps.... any idea here?
Link to comment
Share on other sites
pcpaul
pcpaul
Posted
  • Author
Posted

See Post #2, same sense-id.

Oh, now I see, that's way Labrador made that ghost monitor disappear, since I only edited it for HDMI and it doesn't have the same sense-ids.

Ok, that's a step forward, now the only issue is monitor losing the signal once verbose screen is over if I boot with AMD as primary.

 

Here's my original Labrador buffer:

 

Personality: Labrador unmodified

ConnectorInfo count in decimal: 6
Disk offset in decimal 1850800
0000000    02000000 00010000 00010103 21030204
0000010    02000000 00010000 00010203 11020103
0000020    00040000 04030000 00010307 10000301
0000030    00040000 00010000 00010403 20010505
0000040    00040000 04030000 00010507 12040402
0000050    00040000 00010000 00010603 22050505
0000060 

 

This one is patched for HDMI:

 

Personality: Labrador
ConnectorInfo count in decimal: 6
Disk offset in decimal 1850800
0000000    00080000 00020000 00010107 21030203
0000010    02000000 00010000 00010203 11020103
0000020    00040000 04030000 00010307 10000301
0000030    00040000 00010000 00010403 20010505
0000040    00040000 04030000 00010507 12040402
0000050    00040000 00010000 00010603 22050505

0000060  

 

 

Here are all Buffers for 9000 controller, Exmoor and Baladi are not being accepted,  

 

Kext /System/Library/Extensions/AMD9000Controller.kext/Contents/MacOS/AMD9000Controller

Personality: Exmoor
ConnectorInfo count in decimal: 6
Disk offset in decimal 1850448
0000000    02  00  00  00  00  01  00  00  00  01  01  03  22  05  02  04
0000010    02  00  00  00  00  01  00  00  00  01  02  03  12  04  01  03
0000020    00  04  00  00  04  03  00  00  00  01  03  07  11  02  03  01
0000030    00  04  00  00  00  01  00  00  00  01  04  03  21  03  05  05
0000040    00  04  00  00  04  03  00  00  00  01  05  07  10  00  04  02
0000050    00  04  00  00  00  01  00  00  00  01  06  03  20  01  05  05
0000060
Personality: OPM
ConnectorInfo count in decimal: 6
Disk offset in decimal 1850544
0000000    00  04  00  00  04  03  00  00  00  00  00  01  10  00  01  01
0000010    00  04  00  00  04  03  00  00  00  00  00  01  20  01  02  02
0000020    00  04  00  00  04  03  00  00  00  00  00  01  11  02  03  03
0000030    00  04  00  00  04  03  00  00  00  00  00  01  21  03  04  04
0000040    00  04  00  00  04  03  00  00  00  00  00  01  12  04  05  05
0000050    00  04  00  00  04  03  00  00  00  00  00  01  22  05  06  06
0000060
Personality: GreyhoundS
ConnectorInfo count in decimal: 1
Disk offset in decimal 1850640
0000000    00  08  00  00  04  02  00  00  00  01  01  00  12  04  03  01
0000010
Personality: Labrador
ConnectorInfo count in decimal: 6
Disk offset in decimal 1850800
0000000    02  00  00  00  00  01  00  00  00  01  01  03  21  03  02  04
0000010    02  00  00  00  00  01  00  00  00  01  02  03  11  02  01  03
0000020    00  04  00  00  04  03  00  00  00  01  03  07  10  00  03  01
0000030    00  04  00  00  00  01  00  00  00  01  04  03  20  01  05  05
0000040    00  04  00  00  04  03  00  00  00  01  05  07  12  04  04  02
0000050    00  04  00  00  00  01  00  00  00  01  06  03  22  05  05  05
0000060
Personality: Basset
ConnectorInfo count in decimal: 4
Disk offset in decimal 1850896
0000000    02  00  00  00  00  01  00  00  00  01  01  03  21  03  02  02
0000010    02  00  00  00  00  01  00  00  00  01  02  03  11  02  01  01
0000020    00  04  00  00  04  03  00  00  00  01  03  07  10  00  03  03
0000030    00  04  00  00  04  03  00  00  00  01  04  07  20  01  04  04
0000040
Personality: Greyhound
ConnectorInfo count in decimal: 6
Disk offset in decimal 1851088
0000000    00  04  00  00  04  03  00  00  00  01  01  07  10  00  01  03
0000010    00  04  00  00  04  03  00  00  00  01  02  07  20  01  02  04
0000020    00  04  00  00  04  03  00  00  00  01  03  07  11  02  03  01
0000030    00  04  00  00  04  03  00  00  00  01  04  07  21  03  04  02
0000040    00  04  00  00  04  03  00  00  00  01  05  07  12  04  05  05
0000050    00  04  00  00  04  03  00  00  00  01  06  07  22  05  06  06
0000060
Personality: Baladi
ConnectorInfo count in decimal: 6
Disk offset in decimal 1851216
0000000    00  04  00  00  04  03  00  00  00  01  03  00  12  04  03  03
0000010    00  04  00  00  04  03  00  00  00  01  01  00  11  02  01  01
0000020    00  04  00  00  04  03  00  00  00  01  02  00  21  03  02  02
0000030    00  04  00  00  04  03  00  00  00  01  04  00  22  05  04  04
0000040    00  04  00  00  04  03  00  00  00  01  05  00  10  00  05  05
0000050    00  04  00  00  04  03  00  00  00  01  06  00  20  01  06  06
0000060
Link to comment
Share on other sites
vusun123
vusun123
Posted
Posted

Again, same sense id:

0000000    00080000 00020000 00010107 21030203
0000010    02000000 00010000 00010203 11020103
If your HDMI port has sense id = 03, then patch the second line, not the first, or you can patch every ports on your GPU to avoid this issue
  • Like 1
Link to comment
Share on other sites
pcpaul
pcpaul
Posted
  • Author
Posted

 

Again, same sense id:

0000000    00080000 00020000 00010107 21030203
0000010    02000000 00010000 00010203 11020103
If your HDMI port has sense id = 03, then patch the second line, not the first, or you can patch every ports on your GPU to avoid this issue

 

Oh god, my eyes no longer work, actually mind.... ok gonna patch this one this way and reboot to see if makes a difference, thanks pointing out.

 

Original:

0000000    02000000 00010000 00010103 21030204

0000010    02000000 00010000 00010203 11020103
0000020    00040000 04030000 00010307 10000301
0000030    00040000 00010000 00010403 20010505
0000040    00040000 04030000 00010507 12040402
0000050    00040000 00010000 00010603 22050505
0000060 

 

This one is patched for HDMI and DVI-I

 

Personality: Labrador
ConnectorInfo count in decimal: 6
Disk offset in decimal 1850800
0000000    02000000 00010000 00010103 21030204  
0000010    00080000 00020000 00010107 21030203  HDMI
0000020    00040000 04030000 00010307 10000301
0000030    00020000 04020000 00010203 12040406 DVI_I
0000040    00040000 04030000 00010507 12040402
0000050    00040000 00010000 00010603 22050505

0000060

 

changed my mind, it seems I can't escape same sense-ids on this one, because there are two 05 ids too.... Edit" noticed that it doesn't, I really need to take a break, lol 

 

Ok, Gonna take OPM then

 Original

Personality: OPM

ConnectorInfo count in decimal: 6
Disk offset in decimal 1850544
0000000    00040000  04030000  00000001  10000101  
0000010    00040000  04030000  00000001  20010202  
0000020    00040000  04030000  00000001  11020303 
0000030    00040000  04030000  00000001  21030404  
0000040    00040000  04030000  00000001  12040505  
0000050    00040000  04030000  00000001  22050606  
0000060
 
Modified

 Original

Personality: OPM

ConnectorInfo count in decimal: 6
Disk offset in decimal 1850544
0000000    00040000  04030000  00010001  11020101  DP
0000010    00080000  00020000  00010001  21030203  HDMI
0000020    04000000  14020000  00010001  10000305  DVI-D
0000030    00020000  04020000  00010001  12040406  DVI-I
0000040    00040000  04030000  00000001  20010502  rearranged from second place
0000050    00040000  04030000  00000001  22050604
0000060
 
 

Oh, BTW regarding that DVI-I connector with the same sense-ids in previous post,  on my card it is actually a one physical DVI-I connector: DP, HDMI, DVI-D, DVI-I. 4 in total.

Link to comment
Share on other sites
toleda
toleda
Posted
Posted

Current process is unlikely to succeed.  Suggestion, evaluate one connector at a time. Start with DP.  If you don't have a DP display, get a mDP2HDMI passive adapter. Test for the framebuffer that boots to the Desktop (native AMD9000Controller.kext). Reply with the working framebuffer(s) and the radeon_bios_decode output from your VBIOS.

Link to comment
Share on other sites
pcpaul
pcpaul
Posted
  • Author
Posted

Current process is unlikely to succeed.  Suggestion, evaluate one connector at a time. Start with DP.  If you don't have a DP display, get a mDP2HDMI passive adapter. Test for the framebuffer that boots to the Desktop (native AMD9000Controller.kext). Reply with the working framebuffer(s) and the radeon_bios_decode output from your VBIOS.

Unfortunately I don't have a DP. 

Tested all with HDMI and none can boot to Desktop. I can only boot to desktop via Intel gpu. The only way to boot to desktop via AMD is by deleting either AMD9000Controller.kext or AMDRadeonx4000.kext, of course that way a card is useless. 

 

Radeon_bios_decode output:

 

ATOM BIOS Rom: 

SubsystemVendorID: 0x1458 SubsystemID: 0x22bb

IOBaseAddress: 0xe000

Filename: R938G14D.F4 

BIOS Bootup Message: 

GV-R938G1 GAMING-4GD/F4                                                     

 

PCI ID: 1002:6939

Connector at index 0

Type [@offset 42704]: DisplayPort (10)

Encoder [@offset 42708]: INTERNAL_UNIPHY1 (0x20)

i2cid [@offset 42806]: 0x90, OSX senseid: 0x1

Connector at index 1

Type [@offset 42714]: HDMI-A (11)

Encoder [@offset 42718]: INTERNAL_UNIPHY1 (0x20)

i2cid [@offset 42833]: 0x92, OSX senseid: 0x3

Connector at index 2

Type [@offset 42724]: DVI-D (3)

Encoder [@offset 42728]: INTERNAL_UNIPHY (0x1e)

i2cid [@offset 42860]: 0x94, OSX senseid: 0x5

Connector at index 3

Type [@offset 42734]: DVI-I (2)

Encoder [@offset 42738]: INTERNAL_UNIPHY2 (0x21)

i2cid [@offset 42897]: 0x95, OSX senseid: 0x6

Connector at index 4

Type [@offset 42744]: DVI-I (2)

Encoder [@offset 42748]: INTERNAL_KLDSCP_DAC1 (0x15)

i2cid [@offset 42897]: 0x95, OSX senseid: 0x6

Link to comment
Share on other sites
pcpaul
pcpaul
Posted
  • Author
Posted

I'm thinking Greyhound or Baladi would be suitable for your system. I'd like you to post output of your rom with redsock bios decoder

Turns out, that I've up to this time been running El Capitan with boot argument "rootless=1", So I'm about to retest every single buffer again with correct boot argument.

 

 

Here's the redsock output:

GV-R938G1 GAMING-4GD/F4                                                     

 
Subsystem Vendor ID: 1458
       Subsystem ID: 22bb
Object Header Structure Size: 305
Connector Object Table Offset: 48
Router Object Table Offset: 0
Encoder Object Table Offset: e2
Display Path Table Offset: 12
Connector Object Id [19] which is [DISPLAY_PORT]
encoder obj id [0x20] which is [iNTERNAL_UNIPHY1 (osx txmit 0x11 [duallink 0x1] enc 0x2)] linkb: false
Connector Object Id [12] which is [HDMI_TYPE_A]
encoder obj id [0x20] which is [iNTERNAL_UNIPHY1 (osx txmit 0x21 [duallink 0x1] enc 0x3)] linkb: true
Connector Object Id [4] which is [DVI_D]
encoder obj id [0x1e] which is [iNTERNAL_UNIPHY (osx txmit 0x10 [duallink 0x0] enc 0x0)] linkb: false
Connector Object Id [2] which is [DVI_I]
encoder obj id [0x21] which is [iNTERNAL_UNIPHY2 (osx txmit 0x12 [duallink 0x2] enc 0x4)] linkb: false
Connector Object Id [2] which is [DVI_I]
encoder obj id [0x15] which is [iNTERNAL_KLDSCP_DAC1 (osx txmit 0x00 enc 0x10?)] linkb: false
Link to comment
Share on other sites
TheRacerMaster
TheRacerMaster
Posted
Posted

Try the 4-port framebuffer (Basset) with this patched version:

00 04 00 00 04 03 00 00 00 01 00 00 11 02 01 01 // DP
00 08 00 00 00 02 00 00 00 01 00 00 21 03 02 03 // HDMI
00 02 00 00 04 02 00 00 00 01 00 00 10 00 03 05 // DVI-D (assuming single link)
04 00 00 00 14 02 00 00 00 01 00 00 12 04 04 06 // DVI-I (assuming dual link)

Gigabyte specs say the DVI-I point is dual-link and didn't say anything for the DVI-D port, so I assumed it was single-link.

Link to comment
Share on other sites
vusun123
vusun123
Posted
Posted

DVI-D and DVI-I from 2013 till now is dual-link only, single-link is history. There's an image which helps you identify whether is your DVI port dual-link or single-link, I forgot where to find it

This is what I came up with OP's rom:

00040000 04030000 0001 0000 11 02 01 01               //DP

00080000 04020000 0001 0000 21 03 02 03               //HDMI

04000000 14020000 0001 0000 10 00 03 05               //DVI-D

04000000 14020000 0001 0000 12 04 04 06               //DVI-I

04000000 14020000 0001 0000 00 10 05 06               //DVI-I (DAC)

Link to comment
Share on other sites
pcpaul
pcpaul
Posted
  • Author
Posted

Thanks guys, but none of that works. Must be something with 4000kext or AMDFramebuffer, because on Yosemite I don't need any injection, literally, I can boot without even injecting "ATI Inject" and all ports work, all.  Even tested with Displayport.


Start with DP.  If you don't have a DP display, get a mDP2HDMI passive adapter. Test for the framebuffer that boots to the Desktop (native AMD9000Controller.kext). Reply with the working framebuffer(s).

 

Got DP adapter, though not mini because my card doesn't have mini, it's normal sized, So booted via DP and none of the frame buffers work, even tried patched, none work, after verbose monitor loses connection.

 

Baladi should be perfect for DP on my card, it literally matches it, number by number. No luck, must be screwed up kexts by apple, so far I'm only aware of one guy having this card on a mac, though I don't know specifics how he boots, I bet his doesn't work either, normally that's, most likely he boots having one monitor hooked up to AMD another to Intel..... because I can boot via intel, then reconnect monitor to AMD and that way AMD works perfectly without any injection, but it refuses to boot on it's own, signal is lost after verbose. 

 

Personality: Baladi

ConnectorInfo count in decimal: 6
Disk offset in decimal 1851216
0000000    00  04  00  00  04  03  00  00  00  01  03  00  12  04  03  03
0000010    00  04  00  00  04  03  00  00  00  01  01  00  11  02  01  01
0000020    00  04  00  00  04  03  00  00  00  01  02  00  21  03  02  02
0000030    00  04  00  00  04  03  00  00  00  01  04  00  22  05  04  04
0000040    00  04  00  00  04  03  00  00  00  01  05  00  10  00  05  05
0000050    00  04  00  00  04  03  00  00  00  01  06  00  20  01  06  06
0000060
 
My card's DP
01 displayport 11 02 01 01
Link to comment
Share on other sites
vusun123
vusun123
Posted
Posted

 

Thanks guys, but none of that works. Must be something with 4000kext or AMDFramebuffer, because on Yosemite I don't need any injection, literally, I can boot without even injecting "ATI Inject" and all ports work, all.  Even tested with Displayport.

 

Got DP adapter, though not mini because my card doesn't have mini, it's normal sized, So booted via DP and none of the frame buffers work, even tried patched, none work, after verbose monitor loses connection.

 

Baladi should be perfect for DP on my card, it literally matches it, number by number. No luck, must be screwed up kexts by apple, so far I'm only aware of one guy having this card on a mac, though I don't know specifics how he boots, I bet his doesn't work either, normally that's, most likely he boots having one monitor hooked up to AMD another to Intel..... because I can boot via intel, then reconnect monitor to AMD and that way AMD works perfectly without any injection, but it refuses to boot on it's own, signal is lost after verbose. 

 

Personality: Baladi

ConnectorInfo count in decimal: 6
Disk offset in decimal 1851216
0000000    00  04  00  00  04  03  00  00  00  01  03  00  12  04  03  03
0000010    00  04  00  00  04  03  00  00  00  01  01  00  11  02  01  01
0000020    00  04  00  00  04  03  00  00  00  01  02  00  21  03  02  02
0000030    00  04  00  00  04  03  00  00  00  01  04  00  22  05  04  04
0000040    00  04  00  00  04  03  00  00  00  01  05  00  10  00  05  05
0000050    00  04  00  00  04  03  00  00  00  01  06  00  20  01  06  06
0000060
 
My card's DP
01 displayport 11 02 01 01

 

Did you try using FB Rotate ? That app is pretty useful in these situation. Also what smbios are you using ?

Link to comment
Share on other sites
pcpaul
pcpaul
Posted
  • Author
Posted

Did you try using FB Rotate ? That app is pretty useful in these situation. Also what smbios are you using ?

Thanks for a suggestion, I've just tried it and it doesn't work, apparently after verbose screen OS doesn't boot, because I've my account set up to auto-login, so I've tried to manually invoke sleep via a shortcut (command+option+f12) and it didn't work and then tried the obvious, that's to change volume, so while changing speakers didn't produce any sound.

 

So that means, with AMD set as primary, OS after verbose won't load at all, this is where it ends, right after trying to switch from verbose to GUI.

 

My smbios is iMac 14.2, also tried macpro6.1.

 

I'm attaching my two Clover logs, both using the same settings, booting to Yosemite and El Capitan with AMD set as primary. Yosemite boots, El Capitan monitor loses signal after verbose ( as of now it seems that's where the boot ends completely, after verbose screen that is).

 

So a summary so far is... AMD set as primary prevents El Capitan to boot after verbose screen ends, though that's a guess, only Intel goes through the full boot/load process and that way AMD works, if I reconnect a cable from Intel to it. 

Clover logs.zip

Link to comment
Share on other sites
TheRacerMaster
TheRacerMaster
Posted
Posted

DVI-D and DVI-I from 2013 till now is dual-link only, single-link is history. There's an image which helps you identify whether is your DVI port dual-link or single-link, I forgot where to find it

This is what I came up with OP's rom:

00040000 04030000 0001 0000 11 02 01 01               //DP

00080000 04020000 0001 0000 21 03 02 03               //HDMI

04000000 14020000 0001 0000 10 00 03 05               //DVI-D

04000000 14020000 0001 0000 12 04 04 06               //DVI-I

04000000 14020000 0001 0000 00 10 05 06               //DVI-I (DAC)

Yeah, oops, assumed single-link still existed since one of the DVI ports on my 7970 was single-link. Turns out R9-290 and above always have dual-link DVI.

Link to comment
Share on other sites
vusun123
vusun123
Posted
Posted

Thanks for a suggestion, I've just tried it and it doesn't work, apparently after verbose screen OS doesn't boot, because I've my account set up to auto-login, so I've tried to manually invoke sleep via a shortcut (command+option+f12) and it didn't work and then tried the obvious, that's to change volume, so while changing speakers didn't produce any sound.

 

So that means, with AMD set as primary, OS after verbose won't load at all, this is where it ends, right after trying to switch from verbose to GUI.

 

My smbios is iMac 14.2, also tried macpro6.1.

 

I'm attaching my two Clover logs, both using the same settings, booting to Yosemite and El Capitan with AMD set as primary. Yosemite boots, El Capitan monitor loses signal after verbose ( as of now it seems that's where the boot ends completely, after verbose screen that is).

 

So a summary so far is... AMD set as primary prevents El Capitan to boot after verbose screen ends, though that's a guess, only Intel goes through the full boot/load process and that way AMD works, if I reconnect a cable from Intel to it. 

Try Mac Pro 3,1 as it won't load AGPM which will help debugging a bit easier. Mac Pro 6,1 requires a patch on AGDP to prevent its Dual AMD Firepro Policy to fire up. So by your assumption, Yosemite is ok but El Capitan's 9000 kexts are messed up right ?

Link to comment
Share on other sites
pcpaul
pcpaul
Posted
  • Author
Posted

Try Mac Pro 3,1 as it won't load AGPM which will help debugging a bit easier. Mac Pro 6,1 requires a patch on AGDP to prevent its Dual AMD Firepro Policy to fire up. So by your assumption, Yosemite is ok but El Capitan's 9000 kexts are messed up right ?

Tried today 3.1, 15.1 (with both board IDs (with appropriate serials for each) matching in Device policy info plist, tried 6.1 with AGDP edit (config1=none). Nothing worked.

 

Yes, El Capitan's kexts must be messed up, but not sure which, it could be 9000 or AMDRadeonX4000 and AMDFramebuffer, because Yosemite uses AMDFramebuffer and all ports work, even HDMI audio. I don't know, if I boot with intel as primary and once OS is loaded, reconnect cable to AMD, AMD works. So maybe those kexts have nothing to do with it, but something forbids for OS to boot after verbose with AMD as a primary.

Tried also rolling back kexts from yosemite, doesn't work, It boots to OS fully, but graphics have no acceleration and redrawing is at the the snail's pace, not usable at all.

 

Here's clover's log with 3.1, if it helps with anything at all.

debug.log.zip

Link to comment
Share on other sites
pcpaul
pcpaul
Posted
  • Author
Posted

Tried to edit DisplayEDIT for my monitor, no luck, edit works, but AMD the same story, no signal after verbose.

 

Also notice that DisplayEDIT doesn't match between Windows and Yosemite/El capitan

 

Here is El Capitan's/Yosemite:

AP///////wBBDNLArU8AACQYAQOAPCJ4Ki+lpVRQnicQUFS9SwDRwJUAlQ+zAIHAgYCBQAEBAjqAGHE4LUBYLEUAVlAhAAAeAAAA/wBVSzExNDM2MDIwMzk3AAAA/ABQSEwgMjczVjUKICAgAAAA/QA4TB5TEQAKICAgICAgAYECAyLxTwECAwUGBxAREhMUFRYfBCMJFweDAQAAZQMMABAAAjqAGHE4LUBYLEUAVlAhAAAejArQiiDgLRAQPpYAVlAhAAAYAR0AclHQHiBuKFUAVlAhAAAejArQkCBAMSAMQFUAVlAhAAAYAAAAAAAAAAAAAAAAAAAAAAAAAAAAGQ==

Here's Windows:

AP///////wBBDNLArU8AACQYAQOAPCJ4Ki+lpVRQnicQUFS9SwDRwJUAlQ+zAIHAgYCBQAEBAjqAGHE4LUBYLEUAVlAhAAAeAAAA/wBVSzExNDM2MDIwMzk3AAAA/ABQSEwgMjczVjUKICAgAAAA/QA4TB5TEQAKICAgICAgAYE=

Why's Windows shorter? Haven't tried with windows edit to boot, but probably there's no point, I guess.

 

Also tried to boot with my old monitor, the same thing, loses signal at the end of verbose...

 

Attaching IOREGlog

ioregSaved.txt

Link to comment
Share on other sites
vusun123
vusun123
Posted
Posted

EDID is a monitor's data. If you are hooking multiple monitors to your card then of course OS X will inject every EDIDs possible from your monitors. In Windows you have to pick up each monitors to get their EDID, and it's 128 bytes in HEX

Link to comment
Share on other sites
pcpaul
pcpaul
Posted
  • Author
Posted

So as of now I call {censored} on anyone saying that they run 300 series fine on El Capitan, unless I get details, most likely in all "fine" cases it is booted with intel set as a primary or maybe both being actively used during boot, didn't try to use both, don't care, it just plainly stupid to connect one monitor to R9 380 and other to intel, either it works on one or it doesn't..... anyways all I need is jut one working monitor on R9 380.

 

No full support for R9 380 on El Capitan unless proven otherwise....

 

Done with this fruitless waste of time, it just sucks.... Yosemite works OOB, f*** the Capitan... 

 

Either apple messed it up on purpose for Beta and will bring it back later or it's done for good, in that case Hello Windows next year or so.

Link to comment
Share on other sites
vusun123
vusun123
Posted
Posted

So as of now I call {censored} on anyone saying that they run 300 series fine on El Capitan, unless I get details, most likely in all "fine" cases it is booted with intel set as a primary or maybe both being actively used during boot, didn't try to use both, don't care, it just plainly stupid to connect one monitor to R9 380 and other to intel, either it works on one or it doesn't..... anyways all I need is jut one working monitor on R9 380.

 

No full support for R9 380 on El Capitan unless proven otherwise....

 

Done with this fruitless waste of time, it just sucks.... Yosemite works OOB, f*** the Capitan... 

 

Either apple messed it up on purpose for Beta and will bring it back later or it's done for good, in that case Hello Windows next year or so.

Why would you try El Capitan on new hardware anyway ? If you are looking for stability then Yosemite should be enough

Link to comment
Share on other sites
pcpaul
pcpaul
Posted
  • Author
Posted

Why not, since in yosemite it's supported oob.

 

Noticed that info.plist data for Tonga differs in El Capitan and Yosemite in AMDRadeonX4000.kext, sensor properties are missing altogether in El Capitan, triedto match it up, but didn't help.

 

Here's Yosemite's

<string>AMDTongaGraphicsAccelerator</string>
<key>IODVDBundleName</key>
<string>AMDRadeonVADriver</string>
<key>IOKitDebug</key>
<integer>0</integer>
<key>IOMatchCategory</key>
<string>IOAccelerator</string>
<key>IOPCIMatch</key>
<string>0x69201002 0x69211002 0x69301002 0x69381002 0x69391002</string>
<key>IOProbeScore</key>
<integer>200</integer>
<key>IOProviderClass</key>
<string>IOPCIDevice</string>
<key>IOSourceVersion</key>
<string>0.0.0.0.0</string>
<key>IOVARendererID</key>
<integer>16908288</integer>
<key>cail_properties</key>
<dict>
<key>CAIL_DisableAcpPowerGating</key>
<integer>0</integer>
<key>CAIL_DisableSAMUPowerGating</key>
<integer>0</integer>
<key>CAIL_DisableSysClockGating</key>
<integer>1</integer>
<key>CAIL_DisableUVDPowerGating</key>
<integer>1</integer>
<key>CAIL_DisableVCEPowerGating</key>
<integer>0</integer>
</dict>
<key>sensor-properties</key>
<array>
<dict>
<key>device_type</key>
<data>Z3B1LXNlbnNvcg==</data>
<key>location</key>
<string>GPU</string>
<key>name</key>
<string>gpu-sensor</string>
<key>polling-period</key>
<data>AAAAAQAAAAA=</data>
<key>reg</key>
<data>AAAAAg==</data>
<key>sample-period</key>
<data>AAAAAACYmAA=</data>
<key>sensor-id</key>
<data>AAAABg==</data>
<key>version</key>
<data>AAAAAg==</data>
<key>zone</key>
<data>AAAAAg==</data>
</dict>
</array>
</dict>

here's El Capitan's

<key>AMDTongaGraphicsAccelerator</key>
<dict>
<key>ATIEnableWideBlitSupport</key>
<true/>
<key>ATIUseTearingWideBlit</key>
<false/>
<key>CFBundleIdentifier</key>
<string>com.apple.AMDRadeonX4000</string>
<key>GpuDebugPolicy</key>
<integer>0</integer>
<key>IOClass</key>
<string>AMDTongaGraphicsAccelerator</string>
<key>IODVDBundleName</key>
<string>AMDRadeonVADriver</string>
<key>IOKitDebug</key>
<integer>0</integer>
<key>IOMatchCategory</key>
<string>IOAccelerator</string>
<key>IOPCIMatch</key>
<string>0x69201002 0x69211002 0x69301002 0x69381002 0x69391002</string>
<key>IOProbeScore</key>
<integer>200</integer>
<key>IOProviderClass</key>
<string>IOPCIDevice</string>
<key>IOSourceVersion</key>
<string>0.0.0.0.0</string>
<key>IOVARendererID</key>
<integer>16908288</integer>
<key>cail_properties</key>
<dict>
<key>CAIL_DisableAcpPowerGating</key>
<integer>0</integer>
<key>CAIL_DisableJpegEngine</key>
<integer>1</integer>
<key>CAIL_DisableSAMUPowerGating</key>
<integer>0</integer>
<key>CAIL_DisableSysClockGating</key>
<integer>1</integer>
<key>CAIL_DisableUVDPowerGating</key>
<integer>0</integer>
<key>CAIL_DisableVCEPowerGating</key>
<integer>0</integer>
</dict>
</dict
>
 
Link to comment
Share on other sites
 Share
Go to topic listing
×
×
  • Create New...