Kernel 6.9-rc1 vmnet compile fails #239
Comments
|
Maybe the following will help (untested): |
|
I'm aware if this build failure but unfortunatley I haven't found time to look into it yet. I'll have to check what exactly does the code need to take |
Applied this, and vmnet compiled successfully with 6.9-rc1, and brief test with Win10 Guest (F40/Plasma/X11 host) showed everything appears to be working as it should.. Thanks.. |
|
@rgadsdon Maybe I'm too much of a n00b, but where did you apply this? |
Have you been able to figure out where to apply this change? Thanks. |
|
If I understand correctly, The /vmnet-only/vmnetInt.h file must be edited and some lines commented out but the problem is that in vmware-host-modules-w17.5.1/vmnet-only/vmnetInt.h there aren't such lines, which makes it impossible to follow the tasks. So in my case I end up with a bunch of problems: |
I applied the patch mentioned in #239 (comment) but sadly with no luck :( vmware-host-modules-w17.5.1 with kernel 6.8.7-300 (FC40). Am seeing a few warnings and then failure with this error: |
|
Salut, I had the same issue and fixed it by changing the line in error (vmware-host-modules-w17.5.1/vmmon-only/common/vmx86.c:52) from |
Yes - that did the trick - thank you! |
I made it, vmmon is fully loaded and working as expected, however, vmnet is loaded but It's not working very well: This is the output of vmware.service: When I try to load vmnet module I get this: It keeps trying to load the module kernel but no expires. It's freaky, because the module seems loaded: It's ClearLinux 6.8.8-1430.native kernel with VMware workstation 17.5.1 |
|
Yeah, while the compile works, and my VM seems to run ok, there's a pile of errors in syslog. The module seems to break the host system as it refuses to shutdown cleanly after modules are loaded |
|
dmesg output |
|
same on 6.9rc7, asm/timex.h to linux/timex.h seems haven't done the trick I see now the line 52 comes already modified |
|
any news? this is my output on this morning compilation:
|
|
I'm sorry, I was quite busy last few weeks. I hope to get to fixing this within next few days. |
Thanks, looking forward to a fix. |
That worked as far as I can tell, nothing complaining in dmesg, and vmware seems to be working. |
|
@rgadsdon perfect thank you a lot
|
|
it works @LightJack05 :: thanks tried on Kernel 6.9 :: I'm gonna compile 6.9.1 |
|
nan0desu's patch works for me (VMware Workstation Pro 17.5.2 and kernel 6.9.1-1 on openSUSE Tumbleweed) |
|
Please, I need help on this |
There's no point in using vmware-modconfig, it will not work. |
Hi @Cris70, please how did you apply the patch. I am having trouble getting mine to work and apply the patch. |
If you want to use vmware-modconfig, you have to use the second install method under point 0 (This is also what I do) to replace the original source files with the patched ones. |
Oh my, I feel like screaming @Cris70, it worked, thanks a lot. Do have and enjoy the rest of your day. God bless you. |
|
Hello everyone, is anyone having this issue. I am unable to launch virtual network editor on my vmware-workstation 17.5.2. When i click on Edit > vVirtual Network Editor, nothing happens. |
Same here, but you should open another report rather then hijacking this one. |
|
@K3lvinRob3rt same here also, no logs in console |
I confirm it is working. but I am experiencing high boot time and high shutdown time. See for example: I think it may be related to the patch. |
Worked perfect for 6.9.1-arch1-2 |
|
Hi, try running Virtual Network Editor and you will see it is not working ... in my opinion the patch needed to compile modules under kernel 6.9 breaks something in the vmware workstation so Virtual Network Editor no longer works. The only way to recover it is downgrade to the kernel 6.8.9 |
This worked flawlessly on OpenSUSE Tumbleweed with:
Using the branch tmp/workstation-17.5.0-k6.8 |
|
Thank you for all the "works for me" reports but the problem is that we are talking about synchronization primitives here. And as these guard the access to device list, collisions are expected to be quite rare under normal circumstances. Therefore even replacing thise lock/unlock calls with no-op (i.e. "do nothing) would almost certainly pass these simple tests that it builds and running a VM does not show any issue. |
|
Here's the process for anyone lurking here who has to use VMWare :-) Kinda sad in this day and age, VMWare can't keep up with Linux kernel development on a paid product. |
|
I used |
|
I am on Arch, |
This works for me Fedora 40, 6.9.4-200.fc40.x86_64 + VMware Workstation Pro 15.5.7 build-17171714 |
Me too on Pop!_OS 22.04 LTS - 6.9.3-76060903-generic - VMware Workstation 17.5.2 build-23775571 |
|
After the recent takeover of VMware, I'm hoping this issue is not proving too much of a challenge to fix 'completely', now? |
Probably the opposite. Broadcom doesn't take over companies in order to improve anything. They're raising prices on their enterprise stuff 2-10x or more, they made VMWare Fusion free which means it's probably dead, and who knows about Workstation, a cheap prosumer product that's likely low or 0 priority for them. |
This code change worked for me with Kernel version 6.9.5 |
Thank you! This is very useful. |
|
Has anyone managed to get vmware-network (sudo vmware-networks --start) working on their system? |
and others
vmmon compiles OK, but vmnet:
The text was updated successfully, but these errors were encountered: