Jump to content

Chameleon 2.4svn Official PKG Installer


ErmaC
4,261 posts in this topic

Recommended Posts

@ErmaC

Yesterday: I told you that i was having an instant reboot on rev2640, but took it back later on after realizing that it could just been a bad download since it was all Ok when i re-download the trunk again.

 

Today: I will take it back again lol.... The instant reboot issue actually came from FileNVRAM.dylib https://public.xzenue.com/downloads/?dir=FileNVRAM/trunk from 79 up to 81 which i updated after installing rev2640. The good version was 77

 

Now everything is clear :)

Edited by luxinhaus
Link to comment
Share on other sites

I want to translate , can someone link me the files :)

for both of the projects (clover and chameleon) we use Pootle --> http://pootle.zetam.org/pootle/projects/chameleon/

 

ErmaC

@ErmaC

Yesterday: I told you that i was having an instant reboot on rev2640, but took it back later on after realizing that it could just been a bad download since it was all Ok when i re-download the trunk again.

 

Today: I will take it back again lol.... The instant reboot issue actually came from FileNVRAM.dylib https://public.xzenue.com/downloads/?dir=FileNVRAM/trunk from 79 up to 81 which i updated after installing rev2640. The good version was 77

 

Now everything is clear :)

Can you describe the problem...?

 

With version up to 2636+ and latest FileNVRAM(81)... bad

With version previous of 2636 and latest FileNVRAM(81) all good

With version 2640+ and old FileNVRAM (77) ok

 

Is correct?

Link to comment
Share on other sites

@ErmaC

 

Il be back... heading home :)

 

Update:

Testing FileNVRAM.dylib from https://public.xzenue.com/downloads/?dir=FileNVRAM/trunk , using chameleon 2381 up to 2640:

 

SnowLeopard =  :thumbsup_anim: All Works!

 

Higher OS Build:

81:thumbsdown_anim: Instant restart after "Decompressing Kernel Using lzss"

80:thumbsdown_anim: Instant restart after "Decompressing Kernel Using lzss"

79:thumbsdown_anim: Instant restart after "Decompressing Kernel Using lzss"

77:thumbsup_anim: OK!

Edited by luxinhaus
  • Like 2
Link to comment
Share on other sites

SnowLeopard =  :thumbsup_anim: All Works!

 

Higher OS Build:

81:thumbsdown_anim: Instant restart after "Decompressing Kernel Using lzss"

80:thumbsdown_anim: Instant restart after "Decompressing Kernel Using lzss"

79:thumbsdown_anim: Instant restart after "Decompressing Kernel Using lzss"

77:thumbsup_anim: OK!

 

 

Thx for the info and test.

Useful info BTW.

 

I recently talk with meklort and soon him is going to enchange the chameleon loading module system.

 

So with out put hand on the FileNVRAM module we can't do to much from Chameleon side (I think).


 

Other question... still the glitches in the chameleon TEXT mode?

 

Cordially

ErmaC

  • Like 2
Link to comment
Share on other sites

No glitches on my side, one user reported it still exist though

Pre-built r2625 still has these white blocks. Not every entry has it though. Only some of them.

I can confirm other issues are gone. :)

So you think volumename lenght matters hence the glitch? Im guessing here lol.
Link to comment
Share on other sites

What does klibc.dylib & uClibcxx.dylib modules do? are they obligatory?

Almost nothing..... until you are writing a new module that use C++. This because you can do that if you want w/o modify the "boot" file..... adding new funtionalities to existing bootloader.

Link to comment
Share on other sites

@ErmaC

Hello again... found another issue on my old ICH7 board

 

revision i tried: 2633 up to 2643 :thumbsdown_anim:

post-1406307-0-96540600-1427266312_thumb.jpg

 

Last know working for me was 2529 :thumbsup_anim:

Can you list all revision you would want me to test so we can pinpoint the bad commit

 

Thanks

Edited by luxinhaus
Link to comment
Share on other sites

@ErmaC

Hello again... found another issue on my old ICH7 board

 

revision i tried: 2633 up to 2643  :thumbsdown_anim:

attachicon.gifich7error.jpg

 

Last know working for me was 2529  :thumbsup_anim:

What revision would you want me to test so we can pinpoint the bad commit

 

Thanks

 

Hi luxinhaus.

Sorry but I'm really busy... Pls this are the changes for the trunk starting from 2529 until 2558

tonight or tomorrow I try to see where the issue id located.

 

 

2.558 Improve GMA detection for HD4x00 HSW cards.

 

2.557 Added new IDs for nVidia cards.

 

2.556 Refactor function patch_nvidia_rom (Credits Clover Team)

 

2.555 Update HDA data for internal HDAEnabler injector

 

2.554 Typo.

 

2.548 Move define stuff from cpu.h to platform.h

 

2.547 Update cpu define (Bungo)

 

2.546 Typo

 

2.545 Temp. comment out .IABootFiles hidden folder on boot partition. (credit Bungo)

 

2.544 Added Logging start time (credit Bungo)

 

2.543 Improve detection for HD4200, HD4400 and HD4600 (gma.c)

 

2.541 Added getRTCdatetime() Int 1Ah function 02h - RTC service (Credits to Bungo)

 

2.540 Typo, indenting and whitespace

 

2.539 Typo, indenting and whitespace

 

2.538 Typo (Credits to Bungo)

 

2.535 Use caching with ExFat filesystem by Zenith432

 

2.533 Update Chameleon.xcodeproj

 

2.532 Add sectorsize utility by Micky1979 see: http://www.insanelymac.com/forum/topic/302764-sectorsize-v11-a-tool-to-detect-the-physical-and-logical-sector-size-of-hard-disk-advanced-format-supported/

 

2.529 Update languages for pkg Installer

 

 

ErmaC

Link to comment
Share on other sites

@ErmaC

Hello again... found another issue on my old ICH7 board

 

revision i tried: 2633 up to 2643 :thumbsdown_anim:

attachicon.gifich7error.jpg

 

Last know working for me was 2529 :thumbsup_anim:

Can you list all revision you would want me to test so we can pinpoint the bad commit

 

Thanks

Hi ! on your picture  "size 0x0 file hfs.c" <-- you installed osx system on hfs ? 

Link to comment
Share on other sites

@ErmaC

Please give me time on this as i do not have the computer beside me. I will post my findings maybe tomorrow :)

 

Thanks

No problem.

I have marked the "safe" IMHO commit (ex typo only)

Anyway I have compiled some of it ("AS IS" I mean directly from the SVN)...

 

2.558 Improve GMA detection for HD4x00 HSW cards.

2.557 Added new IDs for nVidia cards.

2.556 Refactor function patch_nvidia_rom (Credits Clover Team)

2.555 Update HDA data for internal HDAEnabler injector

Typo.

2.548 Move define stuff from cpu.h to platform.h

2.547 Update cpu define (Bungo)

Typo

2.545 Temp. comment out .IABootFiles hidden folder on boot partition. (credit Bungo)

2.544 Added Logging start time (credit Bungo)

2.543 Improve detection for HD4200, HD4400 and HD4600 (gma.c)

2.541 Added getRTCdatetime() Int 1Ah function 02h - RTC service (Credits to Bungo)

2.540 Typo, indenting and whitespace

Typo, indenting and whitespace

Typo (Credits to Bungo)

Use caching with ExFat filesystem by Zenith432

2.533 Update Chameleon.xcodeproj

2.532 Add sectorsize utility by Micky1979...

Update languages for pkg Installer

 

ErmaC

Link to comment
Share on other sites

@Ermac

I tried all the versions you mentioned all the way up to 2588 and its all good  :thumbsup_anim:

I currently have no xcode in this computer so i cannot test other revisions, but im definitely sure that 2617  :thumbsdown_anim:  has the error.

Is it possible for you to send me those that might have the bad commit from version 2590 to 2616 please

Il do it once i get home, but cant able to test it until Monday.

 

Thanks

Edited by luxinhaus
Link to comment
Share on other sites

i have 3 problems

 

My mobo is EP45-UD3R rev 1.1 with ICH10R.

AHCI is set in BIOS

 

 

  1. Installing Chameleon using package installer gives a weird issue that BIOS seems stuck at recognising disk. Difference of MBR between manual installation and package installer is value at 00001ce. Package installer sets “80” and manual installation sets “00”
  2. “Memory allocation error at line 924 of hfs.c”  issue with r2625. As luxinhaus noted, r2588 has no problem.
  3. “Boot0: GPT”, “Boot0: Error” issue when i install r2588 manually into HDD. i have read somewhere its because of sector size. Couldnt remember exactly. Anyway i thought r2588 solved this. My HDD’s sector size is 4096. USB disk has no problem since it is 512. Edit: [Solved using TM's trick "dd boot1h /dev/diskXs2"]

 

 

Help me please to solve those issues.

thanks.

Link to comment
Share on other sites

i have 3 problems

 

My mobo is EP45-UD3R rev 1.1 with ICH10R.

AHCI is set in BIOS

 

 

  1. Installing Chameleon using package installer gives a weird issue that BIOS seems stuck at recognising disk. Difference of MBR between manual installation and package installer is value at 00001ce. Package installer sets “80” and manual installation sets “00”
  2. “Memory allocation error at line 924 of hfs.c”  issue with r2625. As luxinhaus noted, r2588 has no problem.
  3. “Boot0: GPT”, “Boot0: Error” issue when i install r2588 manually into HDD. i have read somewhere its because of sector size. Couldnt remember exactly. Anyway i thought r2588 solved this. My HDD’s sector size is 4096. USB disk has no problem since it is 512. Edit: [solved using TM's trick "dd boot1h /dev/diskXs2"]

 

 

Help me please to solve those issues.

thanks.

Inside your Chameleon backup folder there will be the Extra folder containing the log made by r2625 installer, you can kindly post it?

 

1. and .3:

Note: dd boot1h /dev/diskXs2" is not a TM's trick (do not believe what they say):

Andvanced Format Hard drive use 4096kb boot sector (the disk require to be umounted) and Chameleon Installer no longer use dd. Probably your target disk was your start-up Volume that can't be umounted to solve the problem....simply

...problem can be circumvented by installing the bootloader from another partition..

Link to comment
Share on other sites

@Ermac

I tried all the versions you mentioned all the way up to 2588 and its all good  :thumbsup_anim:

I currently have no xcode in this computer so i cannot test other revisions, but im definitely sure that 2617  :thumbsdown_anim:  has the error.

 

===

timerValue          2

intermediate         0x0000000024C011BC

saveTime              0x000000001C4EDC0D

 

Memory allocation error! Addr: 0xdeadbeef, Size: 0x0, File: hfc.c, Line: 924

This is a non recoverable error! System HALTED!!!

===

 

“Memory allocation error at line 924 of hfs.c”  issue with r2625. As luxinhaus noted, r2588 has no problem.

i386-2610.zip Typo (remove struct as already defined by typedef)

2.609 - Correct Kabini FB name was "Abini"

- Changed cards IDs order now is low to hight

- Add new cards ID (kNull FB)

2.608 Add IDs for GeForce GTX 960 and Quadro k1200 (nvidia.c)

i386-2607.zip Indent

2.606 DEBUG option for last booted partition

2.605 Temp disabled feature introduced in commit 2562 (Added ability to auto-select last booted partition as the boot volume), cause AllocateKernelMemory error.

i386-2604.zip Fix instant restart after kexts are being force loaded (Introduced from commit 2561)

2.603 Sanitize 16-bit assembly (Merge from Zenith432 branch)

i386-2602.zip Fix bug in malloc_init, use bios for enableA20() (Merge from Zenith432 branch)

2.601 Update ATI/AMD gfx FB (Credits to pjalm)

2.600 Update Intel gfx cards IDs.

2.599 Add visible output on bad kernel compression.

i386-2598.zip Slightly improve decive selection name - add os version. (credits to meklort)

i386-2593.zip Typo.

i386-2592.zip Merge Interrupt Management changes from Zenith432 branch

2.591 Fix KernelPanic for x32 machine located in fake_efi.c (Big thanks for testing to luxinhaus)

2.590 Typo.

 

ErmaC

Link to comment
Share on other sites

Inside your Chameleon backup folder there will be the Extra folder containing the log made by r2625 installer, you can kindly post it?

 

1. and .3:

Note: dd boot1h /dev/diskXs2" is not a TM's trick (do not believe what they say):

Andvanced Format Hard drive use 4096kb boot sector (the disk require to be umounted) and Chameleon Installer no longer use dd. Probably your target disk was your start-up Volume that can't be umounted to solve the problem....simply

...problem can be circumvented by installing the bootloader from another partition..

 

thank you for your reply.

 

i'm sorry for not giving credit where credit is due. i just found that trick on TM's site.

Thanks to people who found out that trick.

 

ok.

As memory allocation error is related to r2602 commit and AF Hard drive problem is solved, i have one issue left here that is related to package installer. its same for r2625 and r2588. Didnt try other revs, but i expect same result.

 

let me explain the problem

 

- i've run official package installer to install chameleon bootloader into usb flash disk.

- After that, rebooted pc to boot into newly installed usb boot loader. But it stuck at POST screen. Seems BIOS couldnt recognise new USB booter disk or something. It stuck there forever. if i remove the new USB booter disk, it went ok and boots. So i guess something is happened during package installer.

- if i install chameleon manually using tools (fdisk440, dd, cp) , it went ok and boots

- So i retrieved MBR of new USB disk after package installer installation and manual installation. then compared them

- Only difference is like below

maveas-Mac-Pro:Desktop mavea$ diff hex.MBR.Package hex.MBR.Manual 
29c29
< 00001c0 ff ff ee fe ff ff 01 00 00 00 47 6c 78 00 80 00
---
> 00001c0 ff ff ee fe ff ff 01 00 00 00 47 6c 78 00 00 00
 

You can see value "80" at 00001ce in MBR for Package Installer installation and "00" for manual installation.

 

 is this related to setting active paritition? i notified package installer sets active partition. other than that its mostly same as maual installation.

 

thanks

  • Like 1
Link to comment
Share on other sites

×
×
  • Create New...