Re: [PATCH v5 1/2] x86/speculation: apply IBPB more strictly to avoid cross-process data leak

[Jiri Kosina]

On Sun, 21 Oct 2018, Pavel Machek wrote:

> Imagine JIT running evil code (flash, javascript). JIT will prevent evil 
> code from doing ptrace() (or maybe there is syscall filter in effect or 
> something like that), but if evil code can poison branch buffers and do 
> timings, security problem stays.

JITs sort of remove the traditional unix security domain boundary between 
mutually (un)trusted code (processess and threads), that's a more general 
problem, yes.

> Do we need prctl(I_DONT_RUN_EVIL_CODE)?

That's basically the level of fine-graining Tim's followup patchset 
(that's currently being discussed) is eventually going to achieve.

Thanks,

-- 
Jiri Kosina
SUSE Labs