Jump to content
vit9696

AptioMemoryFix

495 posts in this topic

Recommended Posts

The shell stopped working years ago, somewhere before builds 2000 of clover.

Ok, gonna check all that.

Edit : ok the scan option brought it back but it still crashes. I get a fix cursor on the upper left corner. I think it's again a memory issue because it does launch fine from clover in legacy mode.

 

Works fine for me. Just used it. Use it like probably once a week. Maybe you need to test one of the other shells, there are several variants because of bugs in various firmwares, they don't all work. Also there is an EFI shell and a UEFI shell, only one type will work and no idea which without trying (there's variants of each of these is what I meant, lol). I would have to think though that because you can get it to work straight from firmware that the issue is actually a filesystem (or other clover loaded) driver and it's crashing trying to enumerate block io/volumes. Try removing individual drivers to see if one is causing the issue.

 

EDIT: Also as I said the AptioMemoryFix from this topic (also can be included in package from external) has a slide calculator that works kinda like KASLR but removing places the kernel can't be placed, if you don't provide a slide=N boot argument.

Share this post


Link to post
Share on other sites
Advertisement

Ok, I launched shell from my BIOS and tried to launch again the shellX64U, it crashed with the same static cursor.

Rebooted and tried with shellX64 (no U) and I could run "a shell in my shell" so I removed the U one from Tools\ and voilà, the shell launched fine from clover. I could run memmap and see a quite different memory occupation.
In the same move I tried again the Aptiomemoryfix from this thread without slide parameter and it booted fine.
So 2 very old problems solved at once, thank you :)

Share this post


Link to post
Share on other sites

 

Ok, I launched shell from my BIOS and tried to launch again the shellX64U, it crashed with the same static cursor.

Rebooted and tried with shellX64 (no U) and I could run "a shell in my shell" so I removed the U one from Tools\ and voilà, the shell launched fine from clover. I could run memmap and see a quite different memory occupation.
In the same move I tried again the Aptiomemoryfix from this thread without slide parameter and it booted fine.
So 2 very old problems solved at once, thank you :)

 

Haha, awesome, didn't think something was right. Glad it's working. However, please check your hibernation/sleep, it may be broken now that you have working NVRAM, lol.

 

EDIT: Oh, the tradeoffs!  :wallbash:

 

EDIT2: Appears your problem was actually you can't use the UEFI shell, only EFI shell. Depends on the firmware like I said, unsure why it wouldn't work on UEFI firmware but whatever. Or also maybe you have them named backwards, unsure, there could be a million reasons.

Share this post


Link to post
Share on other sites

Actually I had already noticed this shell version issue once but had not made the relation with this clover issue, I don't think they are named backward, I never messed with that.

About hibernation and sleep I can't test it, it NEVER worked in UEFI (was ok in legacy, depending on the OSX version). The mobo deeply crashes. On wake up it's stuck in a power loop : goes on for a few seconds then off, then on again… indefinitely. Forcing shutdown doesn't help, I must cut power with the back PSU button.

Share this post


Link to post
Share on other sites

Actually I had already noticed this shell version issue once but had not made the relation with this clover issue, I don't think they are named backward, I never messed with that.

About hibernation and sleep I can't test it, it NEVER worked in UEFI (was ok in legacy, depending on the OSX version). The mobo deeply crashes. On wake up it's stuck in a power loop : goes on for a few seconds then off, then on again… indefinitely. Forcing shutdown doesn't help, I must cut power with the back PSU button.

 

Well that appears that is because the other AptioFixes were breaking your runtime. You should be able to get it working now, the only issue is that because of not moving the runtime areas at all the kernel may crash because the memory map needs altered to move the stored runtime areas to their new positions after resuming from hibernation (since they most likely change every boot). That may be a difficult problem to solve though....

Share this post


Link to post
Share on other sites

 

EDIT2: Appears your problem was actually you can't use the UEFI shell, only EFI shell. Depends on the firmware like I said, unsure why it wouldn't work on UEFI firmware but whatever. Or also maybe you have them named backwards, unsure, there could be a million reasons.

 

I have that same problem with my old Biostar TZ68K+, my solution was also to remove Shell64U.efi. The shell doesn't work with my Gigabyte Z68AP-D3 either (tested after the UEFI update, which was total {censored} so went back to legacy).

Share this post


Link to post
Share on other sites

I have that same problem with my old Biostar TZ68K+, my solution was also to remove Shell64U.efi. The shell doesn't work with my Gigabyte Z68AP-D3 either (tested after the UEFI update, which was total {censored} so went back to legacy).

 

I used to have a Z68 and they were amongst the first to get UEFI, starting with that HybridEFI {censored} from gigabyte. Only the EFI shell worked. However I now have a Z77 and both work... lol. So who the knows what is going on, just depends on the firmware I guess. And that Z68 UEFI from gigabyte is the biggest joke ever, that thing made my computer reboot loop every restart, I tried to go back to the previous version and it failed to flash one of the chips so it was stuck in a permanent boot loop. I talked to Gigabyte and they said that I voided my warranty for flashing the UEFI firmware, even though it never told me that anywhere. They did not replace the board, and I will never buy one of their products again. Switched to ASRock, have two, built five others, none have any issues really. I think I had a problem in windows for like three days after an update where my computer would randomly just hibernate after sleeping for a while, but I think that was windows cause its fine again now.

Share this post


Link to post
Share on other sites

Oh I remember those bug ridden and unfinished "UEFI" firmwares from Gigabyte. I also had problems with their mobos and even though they couldn't fix them they still tested the very out-of-warranty motherboards and even sent me another similar model for free.

Still, even though I've had very good experience with their support I won't buy another product from them, in the last 6 years I've had 3 partially broken and dead Gigabyte mobos, and that's quite enough for me.

 

Sorry for the off topic!

Share this post


Link to post
Share on other sites

So broken? No hibernate mode 25 too?

 

Nop, same.

But I guess there are tons of things that can interfere with this. My ACPI tables are probably pretty messed up considering that I don't feel brave enough to manually patch them. My GFX are AMD RX 580 + NV 9500GT. I think that's not the case anymore, but I remember AMD got trouble with sleep mode for years.

Share this post


Link to post
Share on other sites

So sorry for repeating this over and over and over again, but:  Is my intel hd4600(with connector)+1050ti final cut crashing issue related to aptiofix and maybe fixable from here? Or did Apple just never test a hd4600 OpenCL processor next to nvidia/amd, since there is no Apple model providing that?

 

Or said in another way: Final Cut will crash, if you use your intel hd graphics + nvidia both as output devices.

Share this post


Link to post
Share on other sites

So sorry for repeating this over and over and over again, but:  Is my intel hd4600(with connector)+1050ti final cut crashing issue related to aptiofix and maybe fixable from here? Or did Apple just never test a hd4600 OpenCL processor next to nvidia/amd, since there is no Apple model providing that?

 

Or said in another way: Final Cut will crash, if you use your intel hd graphics + nvidia both as output devices.

 

No, it's a good question but it's also hard to answer and requires lots of research and tests, and having that hardware helps to develop. I believe that the problem may be related to a larger overall problem of intel graphics having minor issues and a few that have major.

Share this post


Link to post
Share on other sites

So sorry for repeating this over and over and over again, but:  Is my intel hd4600(with connector)+1050ti final cut crashing issue related to aptiofix and maybe fixable from here? Or did Apple just never test a hd4600 OpenCL processor next to nvidia/amd, since there is no Apple model providing that?

 

Or said in another way: Final Cut will crash, if you use your intel hd graphics + nvidia both as output devices.

It is possible that the crashes of the final cut are related to the decoding of the video. You should try this:

https://github.com/vit9696/Shiki

Share this post


Link to post
Share on other sites

ok. stinga11 i already set to intel decoding. Also it actually is encoding. for h264. That is not the problem. Also the dual opencl device configuration works fine in most other opencl apps, like luxmark etc.

Share this post


Link to post
Share on other sites

Whenever I activate my Thunderbolt card in BIOS, I am not able to boot. I get

 

AptioMemoryFix(R12): Starting \System\Library\CoreServices\boot.efi

AMF: Not using custom memory pool - Not Found

does printf work??

 

I also tried some older versions

 

I tried OsxAptioFix2Drv-free2000.efi and it gives me this error.

Freeing low memory (up to 0x20000000)...

Result=0

OsxAptiofixDrv: Starting overrides for \System\Library\CoreServices\boot.efi

Using reloc block: no, hibernate wake:no

does printf work??

 


The other solution was to try OsxAptioFixDrv-64.efi along with Test2-2016.efi

 

This also failed with

 

OsxAptioFixDrv: Starting overrides for \System\Library\CoreServices\boot.efi

Using reloc block: yes, hibernate wake: no

OsxAptioFixDrv: AllocateRelocBlock(): can not allocate relocation block (0x19D65

pages below 0x100000000): Not Found

 
 
It works fine under windows but I can't even boot into OSX unless I disable the Thunderbolt board. I am running an X99 Asus Rampage V Extreme with the Asus Thunderbolt 3 board

Share this post


Link to post
Share on other sites

I just moved slots for the Thunderbolt board. It still gives me the same boot error

 

AptioMemoryFix(R12): Starting \System\Library\CoreServices\boot.efi

AMF: Not using custom memory pool - Not Found

does printf work??

 

When I disable it in BIOS, it boots fine. This is driving me crazy (well... crazier  :) )

Share this post


Link to post
Share on other sites

I found a version of OsxAptioMemoryFix-64 that someone said worked with his Thunderbolt board. He did not have an X99 system. When I tried it with mine, I did not get the error I always get about printf. I got the plus signs across the screen like it was going to boot but then the screen went black and it rebooted. It did the same thing even when I disabled Thunderbolt on BIOS so something is not compatible with my system, but just in case it can help someone else, here it is.

OsxAptioFixDrv-64.efi

Share this post


Link to post
Share on other sites

Hi vpac I to got same problem but only if I boot with igpu HD 530 ....Boots with no probs with egpu Nvidia have you tried booting with egpu good luck

Share this post


Link to post
Share on other sites

Hi vpac I to got same problem but only if I boot with igpu HD 530 ....Boots with no probs with egpu Nvidia have you tried booting with egpu good luck

Niknod:

 

What is egpu?

 

I don't have an NVidia board. I have an RX480

Share this post


Link to post
Share on other sites

@vpac,

 

The problem is that your runtime is being relocated, so something is not functioning correctly by the time the kernel is started, you need to use either AptioMemoryFix or AptioFix3 drivers. Also since you have an X99, you may not have very much luck getting a working system if your memory is too fragmented to place the kernel. Check your memory map from the EFI shell from the clover GUI and see if you have any regions below 0x100000000, with like more than 0x10000 pages...

Share this post


Link to post
Share on other sites

Published the source code of my changes here:

https://github.com/vit9696/AptioFix

Enjoy =)

 

I have a question: Why do you require a disabled VT-d?  

 

 

VT-d disabled in BIOS if present

 

Sorry for the noob question, but that option always worked fine here... Also do VMs not profit from this option? 

 

EDIT: Oh I don't have a DMAR SSDT at all :) So maybe it was never actually working LOL

No, it's a good question but it's also hard to answer and requires lots of research and tests, and having that hardware helps to develop. I believe that the problem may be related to a larger overall problem of intel graphics having minor issues and a few that have major.

 

Uhm, I now finally disabled VT-d in my bios (never had a DMAR table though) :blush: , and now Final Cut does not crash with HD4600-connector-enabled ACPI anymore, it "only" shows corrupted graphics on the previews. So that's kind of interesting. It seems like Apple uses the Intel gfx for the little previews, and dedicated GPU for the raw calculations.

 

Any idea now how to debug this any further? How could I test if some memory area was filled accidently.

 

Using v12 of AptioMemoryFix (compiling produces useless symbolic links only here)

Share this post


Link to post
Share on other sites

I have a question: Why do you require a disabled VT-d?  

 

Sorry for the noob question, but that option always worked fine here... Also do VMs not profit from this option? 

 

EDIT: Oh I don't have a DMAR SSDT at all :) So maybe it was never actually working LOL

 

Uhm, I now finally disabled VT-d in my bios (never had a DMAR table though) :blush: , and now Final Cut does not crash with HD4600-connector-enabled ACPI anymore, it "only" shows corrupted graphics on the previews. So that's kind of interesting. It seems like Apple uses the Intel gfx for the little previews, and dedicated GPU for the raw calculations.

 

Any idea now how to debug this any further? How could I test if some memory area was filled accidently.

 

Using v12 of AptioMemoryFix (compiling produces useless symbolic links only here)

 

Strange, I have never had to disable VT-d or drop my DMAR, but it causes others many issues. So strange. I have to image that the corruption in graphics is related to the same sort of corruptions the HD3000/4000 have, maybe all the intel integrateds do. I think that the graphics reserves a bunch of memory that is discarded as not needed by the kernel because the apple firmware does not set them up in the same way and that memory needs preserved properly.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

  • Recently Browsing   0 members

    No registered users viewing this page.

×