6 ms·

Given the focus on security, don't expect to see SMP anytime soon if ever. It's a huge source of potential security bugs and attacks on the kernel.

ceratopisan 11y ago
I've never seen or expected to see "limit yourself to one processor for better security" as actual advice. Maybe this is sarcasm that I'm misreading?
- armitron 11y ago
  Not at all, SMP opens a huge can of worms both for userland and kernel security. Suddenly you have all sorts of weird race conditions to deal with and in some kernels (cough XNU) logic bugs that stem directly from how kernel subsystems interact with threads. I don't blame OpenBSD for not going down that route.
  - cm3 11y ago
    Which is why DragonflyBSD was forked during the disagreement on FreeBSD's SMPng. Matt implemented more of a message passing system instead and Sepherosa Ziehau made all kinds of improvements to share as little as possible between cores and thus enable higher scalability in particular for networking.
- _delirium 11y ago
  OpenBSD supports multiple cores fine as far as userland goes. If you have 8 cores and start 20 userland processes, they'll be scheduled on all 8 cores. But yes, the kernel itself is not as fine-grain multithreaded as some others. Whether that matters depends on whether waiting on the kernel turns out to be a bottleneck for your workload.
  - chriscappuccio 11y ago
    The kernel is undergoing rapid work to distribute load among cores for major subsystems right now...
    - mioelnir 11y ago
      But in practice, partial fine-grained locking is as useless as partial security. Which is why nobody cares.
  - cm3 11y ago
    It might be ok for compute-only tasks that spread calculations over threads or processes in userland, but anything involving input/outpout (e.g. networking) will block.
cm3 11y ago
As with all things in computer science, there haven been successful implementations of anything that we see as unsolvable in a Unix-derivative. It's sad that so little of even non-research results has been incorporated into mainstream operating systems. I'm sure nickpsecurity can expand on this in detail, but let's just say Burroughs, VMS, Eros, L4 are all things that come to minds as ancient production quality software which solved issues fundamentally. For a newer strain have a look at Barrelfish.

ceratopisan 11y ago

I've never seen or expected to see "limit yourself to one processor for better security" as actual advice. Maybe this is sarcasm that I'm misreading?

armitron 11y ago
Not at all, SMP opens a huge can of worms both for userland and kernel security. Suddenly you have all sorts of weird race conditions to deal with and in some kernels (cough XNU) logic bugs that stem directly from how kernel subsystems interact with threads. I don't blame OpenBSD for not going down that route.
- cm3 11y ago
  Which is why DragonflyBSD was forked during the disagreement on FreeBSD's SMPng. Matt implemented more of a message passing system instead and Sepherosa Ziehau made all kinds of improvements to share as little as possible between cores and thus enable higher scalability in particular for networking.
_delirium 11y ago
OpenBSD supports multiple cores fine as far as userland goes. If you have 8 cores and start 20 userland processes, they'll be scheduled on all 8 cores. But yes, the kernel itself is not as fine-grain multithreaded as some others. Whether that matters depends on whether waiting on the kernel turns out to be a bottleneck for your workload.
- chriscappuccio 11y ago
  The kernel is undergoing rapid work to distribute load among cores for major subsystems right now...
  - mioelnir 11y ago
    But in practice, partial fine-grained locking is as useless as partial security. Which is why nobody cares.
- cm3 11y ago
  It might be ok for compute-only tasks that spread calculations over threads or processes in userland, but anything involving input/outpout (e.g. networking) will block.

cm3 11y ago

As with all things in computer science, there haven been successful implementations of anything that we see as unsolvable in a Unix-derivative. It's sad that so little of even non-research results has been incorporated into mainstream operating systems. I'm sure nickpsecurity can expand on this in detail, but let's just say Burroughs, VMS, Eros, L4 are all things that come to minds as ancient production quality software which solved issues fundamentally. For a newer strain have a look at Barrelfish.