Re: [Qemu-devel] [RFC PATCH 2/3] tcg: Add support for fence generation in x86 backend

[Richard Henderson]

On 05/25/2016 12:25 PM, Alex Bennée wrote:
> That would solves the problem of converting the various backends
> piecemeal - although obviously we should move to all backends having
> "native" support ASAP. However by introducing expensive substitute
> functions we will slow down the translations as each front end is
> expanded to translate the target barrier ops.

Obviously.  We could in fact do that all up front if desired.  It doesn't take 
long to look up the barrier instructions for each isa.


> Should we make the emitting of the function call/TCGop conditional on
> MTTCG being enabled? If we are running in round-robin mode there is no
> need to issue any fence operations.

Probably.  But to keep the translators clean we should probably hide that 
within tcg_gen_fence().


r~