I think v1 is going to pretty universally require application-level fixes.
-
-
-
An MSR to disable branch predictor should suffice. Need lawyers forcing Intel to do it (and fix the bug in future cpus).
-
So you want to end all speculation past a cmp?
-
Until there's hardware to isolate the entire speculative cache hierarchy to be non-observable, YES.
-
No, except for SGX, there is no architectural protection domain here. A process is allowed to observe its own cache misses.
End of conversation
New conversation -
-
-
Ah. They finally finished reading my write up ? https://docs.google.com/document/d/14k1CQANtgh6XTVK9qeWoF7Ouc1IbOXhkQ0hbLvQlJGk/view?usp=sharing …
-
Google Docs is miserably slow, ironically due to Spectre mitigation (javascript.options.baselinejit=false). PDF?
-
Do not expect too much.
End of conversation
New conversation -
-
-
Those fixes have to wait for Blood Lake and Swan Lake.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
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.