Jump to content

Clover General discussion

ErmaC

By ErmaC,
in Clover

 Share
30960 posts in this topic

Recommended Posts

apianti
apianti
Posted
Posted

Nah, you are using the wrong URL. That URL gives you exactly what it's supposed to, the changes made at the HEAD commit. change your URL to either https://sourceforge.net/p/cloverefiboot/code/ or https://sourceforge.net/p/cloverefiboot/code/HEAD/tree/.

  • Like 1
PMheart
PMheart
Posted
Posted

Nah, you are using the wrong URL. That URL gives you exactly what it's supposed to, the changes made at the HEAD commit. change your URL to either https://sourceforge.net/p/cloverefiboot/code/ or https://sourceforge.net/p/cloverefiboot/code/HEAD/tree/.

 

Great! It works. Thx apianti.

liujianwei
liujianwei
Posted
Posted

When I use Clover 3333 install OS X 10.12.3 (16D32),it stuck at progress bar.

When I use Clover 4305 install OS X 10.12.3 (16D32),it works.

what is the difference between Clover 3333 and Clover 4305 make this happen?

Thank you.

smolderas
smolderas
Posted
Posted

When I use Clover 3333 install OS X 10.12.3 (16D32),it stuck at progress bar.

When I use Clover 4305 install OS X 10.12.3 (16D32),it works.

what is the difference between Clover 3333 and Clover 4305 make this happen?

Thank you.

A whole 972 commits of fixes...

  • Like 2
ITzTravelInTime
ITzTravelInTime
Posted
Posted

please help us, we are trying to get clover to work on ryzen cpus for some kernel testing using the kernel that worked for other amd sse 4.1 and sse4.2 capable cpus, we are also working with some kernel developers, but we first need to get the boot loader to work, so i ask clover developers to help up us. the erro we are getting now is displayed on that screen, it seems to be a memory allocation error or something related to cpu clock (the same screen may also appear with unstable overlcloks, also consider that the ryzen clock system is very different from the old am3 cpus)

 

if you have problem with the image here is the url:

https://dl.dropboxusercontent.com/s/76t448hafn27shj/post-598519-0-79895300-1489399096.jpg

 

post-598519-0-79895300-1489399096.jpg

  • Like 2
smolderas
smolderas
Posted
Posted

I am asking Slice for explain how his C source code,logic gate,binary change my life,I do not need arabic numerals that misleading Slice C source code.

Well I don't know any numbering system in another language, especially in mandarin.

Anyway, if you happen to spend 2 seconds for searching, you would have find the expanation you are looking for

  • Like 1
Slice
Slice
Posted
Posted

Hi, I've corrected the mistakes in kernel_patcher.c.  :) The old discussion can be found at here.

 

Please check the attachment.

 

attachicon.gifnew_kernelPm_kernel_patcher.c.zip

I may accept this if someone else confirmed it working.

please help us, we are trying to get clover to work on ryzen cpus for some kernel testing using the kernel that worked for other amd sse 4.1 and sse4.2 capable cpus, we are also working with some kernel developers, but we first need to get the boot loader to work, so i ask clover developers to help up us. the erro we are getting now is displayed on that screen, it seems to be a memory allocation error or something related to cpu clock (the same screen may also appear with unstable overlcloks, also consider that the ryzen clock system is very different from the old am3 cpus)

 

if you have problem with the image here is the url:

https://dl.dropboxusercontent.com/s/76t448hafn27shj/post-598519-0-79895300-1489399096.jpg

 

 

It's a pity I have no test machine to known what is the problem. Some user/coder/hacker with such cpu should inform me what to change in Clover.

Hi. I think there's a typo in kext_inject.c:

UINT8   KBESieDebugSearchEXT[]   = { 0xE8, 0x47, 0x00, 0x00, 0x00, 0xE9, 0x09, 0x00, 0x00, 0x00, 0x48, 0x8B, 0x7D, 0xE8, 0xE8, 0xD9 };
UINT8   KBESieDebugReplaceEXT[]  = { 0xE8, 0x47, 0x00, 0x00, 0x00, 0x90, 0x90, 0x00, 0x00, 0x00, 0x48, 0x8B, 0x7D, 0xE8, 0xE8, 0xD9 };

Note that now we don't NOP out the jmpq fully. (E9 09 00 00 00 -> 90 90 00 00 00 , should be 5x NOP actually.)

 

And fixed here... (Also with a minor variable renaming, don't care about that. ;))

 

attachicon.gifnew_kext_inject.c.zip

Thanks, accepted.

@Slice

i made new v3 pkg file. completed all

Thanks, accepted!

Is there anyway to get the log from the part of Clover that does the kext patching? This is the stuff that occurs after the clover boot log and before the kernel boot log.

 

Or is there a way to redirect the console output to the serial port instead of the screen? If the console output could be redirected, then couldn't it be redirected to a memory buffer, maybe the same buffer that holds the boot log or a new buffer for this log? It may be necessary to allocate extra space for this log before writing log information there if it is not possible to make a growing buffer.

Yes, possible. Post#50 Non-Standard Legacy Boot Files

The command to build is

./ebuild.sh -D DEBUG_ON_SERIAL_PORT

I think it works also for UEFI but not sure.

  • Like 2
Zenith432
Zenith432
Posted
Posted

referring to new_kernelPm_kernel_patcher.c.zip

I may accept this if someone else confirmed it working.

It does not work on my system.

Darwin 16.4

i5-7600

Z270 chipset

MSR 0xE2 locked in bios so need KernelPM patch.

Current KernelPM patch by RehabMan works.

With new patch I get KP.

Attached is image of KP. Sorry for botched focus.

Had to restore working Clover from Windows.

post-446538-0-42656900-1489779940_thumb.jpg

  • Like 2
liujianwei
liujianwei
Posted
Posted

Well I don't know any numbering system in another language, especially in mandarin.

Anyway, if you happen to spend 2 seconds for searching, you would have find the expanation you are looking for

Clover_v2.3k_r3333.zip modify in 2016
AppleKeyStore:operation failed
USBMSC Identifier (non-unique)
busy timeout[0], (60s),kextd wait (0):'AppleACPICPU'
post-236960-0-72041500-1490150747_thumb.jpeg
Clover_v2.4k_r4035.zip modify in 2017
this link works
liujianwei
liujianwei
Posted
Posted

please help us, we are trying to get clover to work on ryzen cpus for some kernel testing using the kernel that worked for other amd sse 4.1 and sse4.2 capable cpus, we are also working with some kernel developers, but we first need to get the boot loader to work, so i ask clover developers to help up us. the erro we are getting now is displayed on that screen, it seems to be a memory allocation error or something related to cpu clock (the same screen may also appear with unstable overlcloks, also consider that the ryzen clock system is very different from the old am3 cpus)

 

if you have problem with the image here is the url:

https://dl.dropboxusercontent.com/s/76t448hafn27shj/post-598519-0-79895300-1489399096.jpg

 

post-598519-0-79895300-1489399096.jpg

RAX RBX 's R acronym for CPU register. 

INT 0EH acronym for floppy Interrupt,means usb clover Interrupt.

the clover give me the same problem once I used c2d.

  • Like 1
Sherlocks
Sherlocks
Posted
Posted

Hi, with the latest version of clover spotlight stop to work, someone else have this problem?

This issue was fixed in r4041 commit.

Wait new build or do build clover from clover script.

 

나의 LG-F410S 의 Tapatalk에서 보냄

Philip Petev
Philip Petev
Posted
Posted

remove .metadata_never_index (hidden) from the root of your hd if present.

 

Yeah, that will do, but won't be enough. In most cases, removing that file doesn't automatically enable the indexing, so the following command is required as well:

sudo mdutil -i on /
  • Like 1
Slice
Slice
Posted
Posted

If you give me some instruments to debug I'll be glad to help you

You have to be a programmer to do this.

Briefly:

Clover's boot file consists of three packed parts. Look ebuild.sh

    echo Compressing DUETEFIMainFv.FV ...
    "$BASETOOLS_DIR"/LzmaCompress -e -o "${BUILD_DIR}/FV/DUETEFIMAINFV${TARGETARCH}.z" "${BUILD_DIR}/FV/DUETEFIMAINFV${TARGETARCH}.Fv"

    echo Compressing DxeCore.efi ...
    "$BASETOOLS_DIR"/LzmaCompress -e -o "${BUILD_DIR}/FV/DxeMain${TARGETARCH}.z" "$BUILD_DIR_ARCH/DxeCore.efi"

    echo Compressing DxeIpl.efi ...
    "$BASETOOLS_DIR"/LzmaCompress -e -o "${BUILD_DIR}/FV/DxeIpl${TARGETARCH}.z" "$BUILD_DIR_ARCH/DxeIpl.efi"

    echo "Generate Loader Image ..."
        "$BASETOOLS_DIR"/GenFw --rebase 0x10000 -o "$BUILD_DIR_ARCH/EfiLoader.efi" "$BUILD_DIR_ARCH/EfiLoader.efi"
        "$BASETOOLS_DIR"/EfiLdrImage -o "${BUILD_DIR}"/FV/Efildr32 \
        "${BUILD_DIR}"/${TARGETARCH}/EfiLoader.efi                \
        "${BUILD_DIR}"/FV/DxeIpl${TARGETARCH}.z                   \
        "${BUILD_DIR}"/FV/DxeMain${TARGETARCH}.z                  \
        "${BUILD_DIR}"/FV/DUETEFIMAINFV${TARGETARCH}.z
        cat $BOOTSECTOR_BIN_DIR/start32H.com2 $BOOTSECTOR_BIN_DIR/efi32.com3 \
        "${BUILD_DIR}"/FV/Efildr32 > "${BUILD_DIR}"/FV/boot

When boot file (it is actually DUET) is started then it unpacks all three parts in their own memories.

Clover/CloverEFI/OsxEfiLdr/EfiLoader.c

When you get the red panic you can see the panic address. This address corresponds to one of these parts with its own memory location.

When you find this address is in the unpacked module then you can disassemble this module and find a place in sources where it happens.

After that I will be ready to continue the dialog.

  • Like 2
liujianwei
liujianwei
Posted
Posted

 

4305 is ontheway

svn diff -r 3333:4035 https://svn.code.sf.net/p/cloverefiboot/code/ > huUuUuUge.diff

 

vimdiff /Users/cuthead/FSInject-64.efi /Users/cuthead/Downloads/FSInject-32.efi

clover different is BOOTX64.efi because clover is boot.

  • Like 1
 Share
Go to topic listing
×
×
  • Create New...