qyliss changed the topic of #spectrum to: A compartmentalized operating system | https://spectrum-os.org/ | Logs: https://logs.spectrum-os.org/spectrum/
klltkr has joined #spectrum
klltkr has quit [Remote host closed the connection]
Madars has quit [Ping timeout: 256 seconds]
Madars has joined #spectrum
tilpner_ has joined #spectrum
tilpner has quit [Ping timeout: 260 seconds]
tilpner_ is now known as tilpner
<cole-h> Sounds interesting. Are the implications just security ones, or are there other uses for Spectrum?
<qyliss> I don't think there's any real new functionality described there, except for security?
<cole-h> Just making sure my post-work brain was in the right place :P
<qyliss> but the security benefits sound great, and really useful to us and feasible to implement
<qyliss> SECRETMEM_UNCACHED is something you'd want to use more on an application level, but it's also quite exciting
<qyliss> stop your password being side-channeled by cache by just never storing it in cache!
<cole-h> :P
cole-h has quit [Ping timeout: 264 seconds]
<IdleBot_6d92ac96> A bit annoying that now you are back to needing to make decision what to do if memory allocation fails! You could have zero locked quota, for example…
<qyliss> you never didn't need to make a decision about memory allocation failing
<qyliss> you could just get away with it a bit more often
<IdleBot_6d92ac96> If a Linux system does not have carefully chosen OOM adjustments, one can panic! on failed malloc (I know I am mixing my metaphors here) with few realistic drawbacks…
pie_ has quit [Quit: pie_]
pie_ has joined #spectrum
mossad has joined #spectrum
cole-h has joined #spectrum
adisbladis has quit [Quit: ZNC 1.8.2 - https://znc.in]
adisbladis has joined #spectrum
<V> the kernel won't panic?
<V> also I strongly recommend turning off overcommit and the OOM killer
<V> that way failed allocations are at least somewhat predictable
<IdleBot_6d92ac96> Kernel can always panic someone else!
bqv has quit [Quit: WeeChat 2.9]
bqv has joined #spectrum