By: Etienne Lorrain (etienne_lorrain.delete@this.yahoo.fr), January 14, 2021 3:02 am
Room: Moderated Discussions
Jörn Engel (joern.delete@this.purestorage.com) on January 13, 2021 8:49 am wrote:
> Maynard Handley (name99.delete@this.name99.org) on January 9, 2021 3:12 pm wrote:
> >
> > I ask this seriously: what does achieving raw metal linux achieve over VM linux?
>
> Maynard has demonstrated multiple times that he doesn't want to learn,
> but to argue. So this answer is explicitly for everyone else. ;)
>
> It is almost always about control.
> - I regularly see seconds of latency for high-priority realtime threads in VMs. I used to
> see similar problems on bare metal - until I fixed the bugs. Fixing hypervisor bugs in a
> Linux hypervisor is already too painful to consider, with OSX it is plain impossible.
> - There is no NMI timer interrupt, but you can simulate one by using performance counters. But inside
> a VM those counters don't work, so your interrupt only fires if you force the VM to burn CPU.
> - Every hypervisor intercepts some keys. I've had to crawl through menus with a mouse to piece
> together sysrq-something or ctrl-alt-del. Bare metal simply doesn't annoy you in such ways.
> - If you want to use the entire system, particularly all of RAM, the hypervisor will
> get in the way. You can pin two threads to two different CPUs inside the VM, but the
> hypervisor is still allowed to schedule those threads on the same physical CPU. Swap
> will happen in ways you cannot control. The entire VM may get killed without warning.
And another problem with Virtual Managers (but this one not only on Macs):
You might want to check your whole DRAM for failing bits, for instance using memtest under BIOS, and having a virtual (i386 real-mode) machine created by the EFI BIOS (with maximum 4 Gbytes of DRAM) will lead to a "memory error" or crash when attempting to check the DRAM at address 0x1_0000_0000...
I suppose someone will say that Macs have a perfect DRAM and never develop any randomly failing bit.
> Maynard Handley (name99.delete@this.name99.org) on January 9, 2021 3:12 pm wrote:
> >
> > I ask this seriously: what does achieving raw metal linux achieve over VM linux?
>
> Maynard has demonstrated multiple times that he doesn't want to learn,
> but to argue. So this answer is explicitly for everyone else. ;)
>
> It is almost always about control.
> - I regularly see seconds of latency for high-priority realtime threads in VMs. I used to
> see similar problems on bare metal - until I fixed the bugs. Fixing hypervisor bugs in a
> Linux hypervisor is already too painful to consider, with OSX it is plain impossible.
> - There is no NMI timer interrupt, but you can simulate one by using performance counters. But inside
> a VM those counters don't work, so your interrupt only fires if you force the VM to burn CPU.
> - Every hypervisor intercepts some keys. I've had to crawl through menus with a mouse to piece
> together sysrq-something or ctrl-alt-del. Bare metal simply doesn't annoy you in such ways.
> - If you want to use the entire system, particularly all of RAM, the hypervisor will
> get in the way. You can pin two threads to two different CPUs inside the VM, but the
> hypervisor is still allowed to schedule those threads on the same physical CPU. Swap
> will happen in ways you cannot control. The entire VM may get killed without warning.
And another problem with Virtual Managers (but this one not only on Macs):
You might want to check your whole DRAM for failing bits, for instance using memtest under BIOS, and having a virtual (i386 real-mode) machine created by the EFI BIOS (with maximum 4 Gbytes of DRAM) will lead to a "memory error" or crash when attempting to check the DRAM at address 0x1_0000_0000...
I suppose someone will say that Macs have a perfect DRAM and never develop any randomly failing bit.
Topic | Posted By | Date |
---|---|---|
Question to Torvalds | Paul | 2020/11/14 04:08 AM |
Question to Torvalds | Linus Torvalds | 2020/11/14 02:12 PM |
Question to Torvalds | never_released | 2020/11/14 05:12 PM |
Question to Torvalds | Doug S | 2020/11/15 09:55 AM |
Question to Torvalds | never_released | 2020/11/15 12:31 PM |
Question to Torvalds | Doug S | 2020/11/16 10:46 AM |
Question to Torvalds | Maxwell | 2020/11/16 11:49 AM |
Question to Torvalds | never_released | 2020/11/16 04:25 PM |
Question to Torvalds | lyra64 | 2020/11/23 11:23 AM |
Question to Torvalds | me | 2020/11/22 12:11 PM |
Question to Torvalds | James | 2020/11/25 06:59 AM |
Question to Torvalds | bakk | 2021/01/09 03:35 PM |
Question to Torvalds | Maynard Handley | 2021/01/09 04:12 PM |
He asked disingenuously (NT) | JS | 2021/01/09 08:33 PM |
He asked disingenuously | Maynard Handley | 2021/01/10 10:51 AM |
He asked disingenuously | JS | 2021/01/10 03:50 PM |
He asked disingenuously | Maynard Handley | 2021/01/10 06:02 PM |
Question to Torvalds | anon | 2021/01/10 07:01 PM |
Question to Torvalds | Maynard Handley | 2021/01/10 07:59 PM |
Question to Torvalds | anon | 2021/01/11 09:56 AM |
Question to Torvalds | Jukka Larja | 2021/01/12 05:50 AM |
Question to Torvalds | anon2 | 2021/01/10 07:21 PM |
Question to Torvalds | Maynard Handley | 2021/01/10 08:15 PM |
Question to Torvalds | Maynard Handley | 2021/01/10 08:22 PM |
Question to Torvalds | anon2 | 2021/01/10 08:47 PM |
Question to Torvalds | Maynard Handley | 2021/01/10 09:28 PM |
Question to Torvalds | anon2 | 2021/01/10 10:36 PM |
Question to Torvalds | Jukka Larja | 2021/01/11 06:21 AM |
Question to Torvalds | Maynard Handley | 2021/01/11 10:33 AM |
Question to Torvalds | anon2 | 2021/01/11 10:40 PM |
Question to Torvalds | Jukka Larja | 2021/01/12 06:05 AM |
Question to Torvalds | Maynard Handley | 2021/01/12 09:42 AM |
Question to Torvalds | Jukka Larja | 2021/01/12 11:15 AM |
Question to Torvalds | Maynard Handley | 2021/01/12 12:07 PM |
Question to Torvalds | Jukka Larja | 2021/01/13 06:24 AM |
Question to Torvalds | Michael S | 2021/01/13 08:45 AM |
Question to Torvalds | Ungo | 2021/01/13 07:34 PM |
Question to Torvalds | Jörn Engel | 2021/01/13 09:49 AM |
Question to Torvalds | Etienne Lorrain | 2021/01/14 03:02 AM |
Question to Torvalds | dmcq | 2021/01/14 08:26 AM |
Question to Torvalds | Jörn Engel | 2021/01/14 11:42 AM |
Question to Torvalds | dmcq | 2021/01/14 12:13 PM |
Question to Torvalds | Jukka Larja | 2021/01/15 06:57 AM |
Question to Torvalds | dmcq | 2021/01/15 08:27 AM |
Question to Torvalds | Anne O. Nymous | 2021/01/15 12:19 PM |
Question to Torvalds | dmcq | 2021/01/15 02:58 PM |
Question to Torvalds | Jukka Larja | 2021/01/15 09:04 PM |
Question to Torvalds | dmcq | 2021/01/16 02:50 AM |
Question to Torvalds | Jukka Larja | 2021/01/16 09:37 PM |
Question to Torvalds | dmcq | 2021/01/17 06:39 AM |
Question to Torvalds | Adrian | 2021/01/17 08:46 AM |
Question to Torvalds | dmcq | 2021/01/17 09:36 AM |
Question to Torvalds | Jukka Larja | 2021/01/17 09:35 AM |
Question to Torvalds | dmcq | 2021/01/17 10:01 AM |
Question to Torvalds | Jukka Larja | 2021/01/17 10:52 AM |
Question to Torvalds | Doug S | 2021/01/14 10:37 AM |