This was totally it. I feel dumb because the speed should have stuck out to me. Thanks!It may be necessary to switch to 1 Gbps manually, then switch back to Auto.
View attachment 9156
This was totally it. I feel dumb because the speed should have stuck out to me. Thanks!It may be necessary to switch to 1 Gbps manually, then switch back to Auto.
View attachment 9156
Little update:I responded to something similar a couple of posts above:
Ryzen 7000 Testing
About what? Can you send a link? AMD Addresses Controversy: RDNA 3 Shader Pre-Fetching Works Fine https://www.tomshardware.com/news/amd-addresses-controversy-rdna-3-shader-pre-fetching-works-fineforum.amd-osx.com
Let's see if this helps.
I’ll provide a new version that writes to both dmesg and system log files. Really hoping someone can find my name in one of those logs.Little update:
I have tried runa NVRAM reset (for another reason) and magically, the speed in both direction is back!
@CaseySJ Thanks for the support, hopefully tonight I will perform an update from Big sur to Ventura and test the KEXT (always if Amazon manages to deliver my 6900Xt...)
Here's a new version that writes to both log files...I’ll provide a new version that writes to both dmesg and system log files. Really hoping someone can find my name in one of those logs.
hi I can't see any refer to CaseySJHere's a new version that writes to both log files...
Noooooooo!hi I can't see any refer to CaseySJ
To get logs from RestrictEvents you need:Noooooooo!
Okay, back to the drawing board.
According to comments in the RestrictEvents source code:To get logs from RestrictEvents you need:
- Debug or Release version of DebugEnhancer
- Debug version of RestrictEvents
And a boot argument: -revdbg
Then you can get the logs with this command: (sudo dmesg | grep -i RestrictEvents) > ~/Desktop/RestrictEventsLog.txt
Currently Pallas will not provide updates to any machine with gibraltar ProductType but non-ap (e.g. J137AP) and non-VMM HWModelStr.
To workaround this issue we hook sysctls used by softwareupdated and com.apple.Mobile to report VMM-x86_64 in HWModelStr.
The VMM model is chosen if the hypervisor sysctl returns true.
rerouteHvVmm
. This function is called if the following condition is met: if ((getKernelVersion() >= KernelVersion::Monterey ||
(getKernelVersion() == KernelVersion::BigSur && getKernelMinorVersion() >= 4)) &&
(revsbvmmIsSet || revassetIsSet))
rerouteHvVmm(patcher);
revassetIsSet
.sudo dmesg | grep Casey > ~/Documents/dmesg.txt
I'll test this after my new motherboard arrives, I somehow broke it when upgrading the GPU...@Lorys89 / All,
This version of RestrictEvents disables the call to rerouteHvVmm (see post directly above for details). If you use MacPro7,1 and Ventura then feel free to try this version.
It is still a work in progress so we should not expect it to work. However, it will generate log messages if we add Boot Argument -revdbg. To grab the log, we can type this after booting into macOS:
Bash:sudo dmesg | grep Casey > ~/Documents/dmesg.txt
It may also be necessary to add DebugEnhancer kext from Acidanthera GitHub.
Ouch! I recently broke the PCI slot retention tab on an older motherboard when attempting to remove the GPU, but fortunately that’s not a functional problem. I was very annoyed at myself nevertheless.I'll test this after my new motherboard arrives, I somehow broke it when upgrading the GPU...
I took advantage of this situation to upgrade to a PCIe 4.0 motherboard (Gigabyte B550M Gaming X V2) from my Gigabyte B450M DS3H so I can run my new GPU at full speed. The same old motherboard had around the same price, so it just wasn't worth it. The new one also has BIOS Flashback so I have something to try in case something like this happens againOuch! I recently broke the PCI slot retention tab on an older motherboard when attempting to remove the GPU, but fortunately that’s not a functional problem. I was very annoyed at myself nevertheless.
Are you replacing like for like, or switching to a different make/model?
[ 0.732879]: RestrictEvents rev: @ Casey - revsbvmm is 0
[ 0.732882]: RestrictEvents rev: @ (DBG) Casey - revsbvmm is 0
[ 0.732884]: RestrictEvents rev: @ Casey - Setting revsbvmmIsSet and revassetIsSet to False because AMD CPU
[ 0.732887]: RestrictEvents rev: @ (DBG) Casey - Setting revsbvmmIsSet and revassetIsSet to False because AMD CPU
As discussed previously in this thread, whenRestrictEventskext is enabled on an AMD platform, we lose the ability to performincrementalmacOS updates. Instead, the full 12GB installer is downloaded, which wastes network bandwidth, time and disk space.
But if RestrictEvents is disabled, then incremental update works?This issue also happens on a friend's Intel desktop when using the MacPro7,1 SMBIOS, so it's not AMD exclusive.
Thanks! This shows that the rerouteHvVmm method was not called, as we had expected. But that in itself did not help.@CaseySJ
Code:[ 0.732879]: RestrictEvents rev: @ Casey - revsbvmm is 0 [ 0.732882]: RestrictEvents rev: @ (DBG) Casey - revsbvmm is 0 [ 0.732884]: RestrictEvents rev: @ Casey - Setting revsbvmmIsSet and revassetIsSet to False because AMD CPU [ 0.732887]: RestrictEvents rev: @ (DBG) Casey - Setting revsbvmmIsSet and revassetIsSet to False because AMD CPU
With latest kext incremental upgrade doesn't work (from 13.01 to 13.1)