Stuff like this is why hppa and itanic are explicitly excluded from @musllibc's future arch targets. https://twitter.com/kittenpies3/status/879024041809633283 …
Generality has high cost, and when it's for the sake of one or two wacky obsolete outliers, it's definitely not worth it.
-
-
I remember back when I was in HS, people hyped Itanic, and my thoughts at the time were more "yeah that is gonna sink". x86-64 came along...
-
Designing new hardware features to provide needed safety/security properties/mitigations is a complete non-starter.
-
If for no other reason than the state of the patent system. Any attempt to do so is just a grab for monopolization.
-
Viable hardening or privilege models need to be equally implementable everywhere, not limited to one vendor's machines.
-
at the time, the cost of everyone jumping to a new arch seemed too steep for what one could get from it (DOS wasn't even fully dead yet...)
-
Yes, that's a huge problem in itself. I was probably wrong to jump just on wacky hardening features at hw level...
-
...when Itanic also overlapped somewhat into the domain of "wacky new computation model" stuff.
-
yeah, would need fancy/new compiler magic to make it work well; and for the older software which worked (Win32 only), it takes a perf hit...
- 1 more reply
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.