Re: [PATCH V2 9/9] x86/speculation: Remove all ANNOTATE_NOSPEC_ALTERNATIVE directives

[Peter Zijlstra]

On Tue, Apr 07, 2020 at 03:52:11PM +0200, Peter Zijlstra wrote:
> On Tue, Apr 07, 2020 at 09:31:42AM +0200, Alexandre Chartre wrote:
> 
> > -	ANNOTATE_NOSPEC_ALTERNATIVE
> >  	ALTERNATIVE_2 __stringify(ANNOTATE_RETPOLINE_SAFE; jmp *\reg),	\
> >  		__stringify(RETPOLINE_JMP \reg), X86_FEATURE_RETPOLINE,	\
> >  		__stringify(lfence; ANNOTATE_RETPOLINE_SAFE; jmp *\reg), X86_FEATURE_RETPOLINE_AMD
> 
> Possibly we can write this like:

.macro OOL_RETPOLINE_JMP reg:req
SYM_FUNC_START(__x86_retpoline_jmp_\reg)
	CFI_STARTPROC
	RETPOLINE_JMP \reg
	CFI_ENDPROC
SYM_FUNC_END(__x86_retpoline_jmp_\reg)
.endm

> 	ALTERNATIVE("", "lfence", X86_FEATURE_RETPOLINE_AMD);
> 	ALTERNATIVE("jmp *\reg", "jmp __x86_retpoline_jmp_\reg", X86_FEATURE_RETPOLINE);
> 
> With an out-of-line copy of the retpoline, just like the THUNKs the
> compiler uses, except of course, it can't be those, because we actually
> want to use the alternative to implement those.
> 
> By moving the retpoline magic out-of-line we ensure it has a unique
> address and the ORC stuff should work.
> 
> I'm just not sure what to do about the RETPOLINE_CALL variant.

Duh, something like so:

	ALTERNATIVE("", "lfence", X86_FEATURE_RETPOLINE_AMD);
	ALTERNATIVE("call *\reg", "call __x86_retpoline_jmp_\reg", X86_FEATURE_RETPOLINE);