Jump to content

OpenCore General Discussion

dgsga

By dgsga,
in OpenCore

 Share
8,809 posts in this topic

Recommended Posts

jsl2000
jsl2000
Posted
Posted (edited)
3 hours ago, vandroiy2012 said:

 

Try agdpmod=pikera bootarg. 

In my Ryzen 1700X hackintosh it works for AMD 7950 GPU, but -wegoff or removal of WhateverGreen.kext did not work !

Edited by jsl2000
Link to comment
Share on other sites
bittantone
bittantone
Posted
Posted

With OC 0.5.1, I'm having a strange behavior.

 

When I set ScanPolicy = 0, OC finds everything (Catalina on NVMe, Windows on another SSD, the Catalina USB installer and even Clover/OpenCore on external USBs). Instead, if I set ScanPolicy to the default value 983299, it does not find any entry. Why is this happening?

 

Moreover, I'd like to start macOS by default. I tried to set BootApple = YES and to remove ShowPicker entry, but it does not work (no schema error at startup). Can someone give me any hint?

 

PS: I have an H97M-plus which does not have native NVRAM. Thank you

config.plist

Link to comment
Share on other sites
iGPU
iGPU
Posted
Posted (edited)
On 10/18/2019 at 3:34 PM, bittantone said:

With OC 0.5.1, I'm having a strange behavior.

 

When I set ScanPolicy = 0, OC finds everything (Catalina on NVMe, Windows on another SSD, the Catalina USB installer and even Clover/OpenCore on external USBs). Instead, if I set ScanPolicy to the default value 983299, it does not find any entry. Why is this happening?

 

Moreover, I'd like to start macOS by default. I tried to set BootApple = YES and to remove ShowPicker entry, but it does not work (no schema error at startup). Can someone give me any hint?

 

Default boot drive is what you set in the MacOS: System Preferences/Startup Disk.

 

The ScanPolicy is a sum of the various drives you want scanned. See the docs for the values. I like using 3080963.

 

Edited by iGPU
Link to comment
Share on other sites
bittantone
bittantone
Posted
Posted (edited)
7 hours ago, Andrey1970 said:

 

What? Where you take this nonsense?

 

Sorry, I thought that BootApple was a property that I could set to force OC to load macOS by default.
Unfortunately I don't have the NVRAM, so actually each time I boot up I need to choose macOS by pressing 2 within the timeout. This is really annoying, is there a way to set macOS as the default boot option?

 

7 hours ago, iGPU said:

 

Default boot drive is what you set in the MacOS: System Preferences/Startup Disk.

 

The ScanPolicy is a sum of the various drives you want scanned. See the docs for the values. I like using 3080963.

 

Default boot drive seems to work only when native NVRAM is supported, isn't it? Theoretically, the default scan policy 0xF0103 should imply that APFS partition located on NVME are scanned...so I can't see why in my case it is not finding macOS Catalina

Edited by bittantone
  • Like 1
Link to comment
Share on other sites
bittantone
bittantone
Posted
Posted

I've tried to play around with the ScanPolicy and things are becoming more and more obscure here.

 

I set ScanPolicy = 1793, which should correspond to

  • Allow any kind of devices (SATA, NVME, USB, ecc)
  • Allow only APFS, HFS, ESP

The strange thing here is that OC finds Windows Boot Manager (this was expected, since I enabled ESPs), but macOS is not found. The driver ApfsDriverLoader.efi is of course loaded. I attach the log. Thanks

opencore-2019-10-19-072722.txt

Link to comment
Share on other sites
bittantone
bittantone
Posted
Posted
13 minutes ago, Download-Fritz said:

@Mihoko Okayami it's just empty because you don't use the DEBUG version

@bittantone Nothing is obscure, you entered the hex instead of the decimal value

 

Sorry, but in the config.plist I have the following key: 

<key>ScanPolicy</key>
<integer>1793</integer>

Unless OC interprets 1793 as an hex, it is actually a decimal value.

 

I'm conducting more experiments and it seems that this issue is related to the fact that macOS is installed on an NVME. I tried to install macOS on a SATA, and it is correctly detected with the same ScanPolicy. Any idea?

Link to comment
Share on other sites
mhaeuser
mhaeuser
Posted
Posted

@bittantone Indeed, I could have sworn I saw a decimal log call, but it's hex, so your setup is correct. What fails is retrieving the necessary information of the disk as indicated by:

Quote

08:067 00:040 OCPI: ReadDisk1 Device Error
08:106 00:039 OCPI: Failed to retrieve disk info

I'll push a workaround idea to a test branch soon. Are you familiar with compiling OC yourself? I do not have a verified build environment (I'm on Windows and we had interesting quirks with builds from there before). Otherwise you'll have to wait for someone to wake up to compile it.

  • Like 2
Link to comment
Share on other sites
bittantone
bittantone
Posted
Posted
12 minutes ago, Download-Fritz said:

@bittantone Indeed, I could have sworn I saw a decimal log call, but it's hex, so your setup is correct. What fails is retrieving the necessary information of the disk as indicated by:

I'll push a workaround idea to a test branch soon. Are you familiar with compiling OC yourself? I do not have a verified build environment (I'm on Windows and we had interesting quirks with builds from there before). Otherwise you'll have to wait for someone to wake up to compile it.

 

Yep, I actually build myself OC at each release. Please let me know if I can be useful in any way or if you need more informations 

 

Thanks!

Link to comment
Share on other sites
danif
danif
Posted
Posted
34 minutes ago, Download-Fritz said:

@bittantone Please compile master OpenCorePkg with this branch of OcSupportPkg and post a log whether it works or not: https://github.com/acidanthera/OcSupportPkg/tree/disk_test

 

Hi, my problem with scanpolicy is this:

 

I cannot use any DEVICE_LOCK bit because my nvme path is kind of different from what oc expects I guess --> PciRoot(0x0)/Pci(0x2,0x3)/Pci(0x0,0x0)/Unit(0x1)/HD(6,GPT,...)

 

Does this workaround target this problem too?

Thanks

Link to comment
Share on other sites
danif
danif
Posted
Posted
[mention=115038]danif[/mention] No, it does not. Frankly, I haven't ever seen a DP like that before. Because of how "generic" it is, I do not believe it will be considered in the future, sorry
Ok thanks, no worries.

Well just in case, if there were an easy way to hard code the string i need to search, just let me know how the change would look like. That way I could compile a personal oc version with that feature.

Enviado desde mi Mi A2 Lite mediante Tapatalk

[mention=115038]danif[/mention] No, it does not. Frankly, I haven't ever seen a DP like that before. Because of how "generic" it is, I do not believe it will be considered in the future, sorry
Ok thanks, no worries.

Well just in case, if there were an easy way to hard code the string i need to search, just let me know how the change would look like. That way I could compile a personal oc version with that feature.

Enviado desde mi Mi A2 Lite mediante Tapatalk

Link to comment
Share on other sites
mhaeuser
mhaeuser
Posted
Posted
4 minutes ago, danif said:

Well just in case, if there were an easy way to hard code the string i need to search, just let me know how the change would look like. That way I could compile a personal oc version with that feature.

The check is here: https://github.com/acidanthera/OcSupportPkg/blob/5e0f400c69115b2d21f402f3ba694bc595e23ba9/Library/OcBootManagementLib/PolicyManagement.c#L78

The most trivial change would be to add a case for MSG_DEVICE_LOGICAL_UNIT_DP, but unsure of how much (security-related) use that is, really. If you're concerned with security, you should probably integrate edk2's NvmExpressDxe into your FW (this is out of scope for this thread though, so please do not ask questions about that here)

  • Like 1
Link to comment
Share on other sites
bittantone
bittantone
Posted
Posted (edited)
3 hours ago, Download-Fritz said:

@bittantone Please compile master OpenCorePkg with this branch of OcSupportPkg and post a log whether it works or not: https://github.com/acidanthera/OcSupportPkg/tree/disk_test

 

Looks like it's NOT working. Now OC is stuck.

 

I'm also experiencing some problems with CMOS. After a certain number of boots with OC, I can't access the BIOS settings until I pull and reinsert the CMOS battery and all the SATA drives.

opencore-2019-10-19-131722.txt

Edited by bittantone
Link to comment
Share on other sites
dogansan
dogansan
Posted
Posted
On 10/13/2019 at 9:33 PM, dogansan said:

thanks for help. I deleted FakeSMC and ACS6x.kext from L/E and made off SSDT in config.plist. Now, I see some progress but still no full booting and login. I am using a USB as boot device , it seems my signature is not seen at the bottom, my systems spec are as follows:

Desktop PC : Intel Core i7 3370K  | ASUS Maximus Formula V |ASUS R7 370 | DDR3 (2x8GB) RGB | 82579V Gigabit LAN  | Broadcom BCM43228 802.11a/b/g/n+ Bluetooth | macOS Catalina 10.15.1 Beta 19B68f, Linux , Mojave  and Windows Enterprise 10 Build 1903 (Bootloader : Clover r5080 + OpenCore 0.5.1 not yet)

20191013_211249.jpg

I solved the hanging at apfs boot module start during booting by activating SSDT-EC-USBX.aml abd adding nv_disable=1 to boot args. It now boots ,now I have to activate Radeon R9-270

Link to comment
Share on other sites
mhaeuser
mhaeuser
Posted
Posted

@bittantone sorry, I had two copy+paste mistakes: https://github.com/acidanthera/OcSupportPkg/commit/0c20f3f5cf860d51961d8b0c1bbcaf36eee6e571

but considering the first branch is not even reached, I get the feeling this will not work either. Please report back whether it works, but a different kind of workaround is going to take a bit more time to implement

  • Like 1
Link to comment
Share on other sites
 Share
Go to topic listing
×
×
  • Create New...