On Fri, 2020-02-14 at 08:59 -0800, Sean Christopherson wrote:
> On Fri, Feb 14, 2020 at 10:56:08AM -0500, Qian Cai wrote:
> > arch/x86/kvm/emulate.c: In function 'x86_emulate_insn':
> > arch/x86/kvm/emulate.c:5686:22: error: cast between incompatible
> > function types from 'int (*)(struct x86_emulate_ctxt *)' to 'void
> > (*)(struct fastop *)' [-Werror=cast-function-type]
> > rc = fastop(ctxt, (fastop_t)ctxt->execute);
> >
> > Fixes: 3009afc6e39e ("KVM: x86: Use a typedef for fastop functions")
> > Signed-off-by: Qian Cai <cai@xxxxxx>
> > ---
> > arch/x86/kvm/emulate.c | 8 +++++---
> > 1 file changed, 5 insertions(+), 3 deletions(-)
> >
> > diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c
> > index ddbc61984227..17ae820cf59d 100644
> > --- a/arch/x86/kvm/emulate.c
> > +++ b/arch/x86/kvm/emulate.c
> > @@ -5682,10 +5682,12 @@ int x86_emulate_insn(struct x86_emulate_ctxt *ctxt)
> > ctxt->eflags &= ~X86_EFLAGS_RF;
> >
> > if (ctxt->execute) {
> > - if (ctxt->d & Fastop)
> > - rc = fastop(ctxt, (fastop_t)ctxt->execute);
>
> Alternatively, can we do -Wno-cast-function-type? That's a silly warning
> IMO.
I am doing W=1 on linux-next where some of the warnings might be silly but the
recent commit changes all warnings to errors forces me having to silence those
somehow.
>
> If not, will either of these work?
>
> rc = fastop(ctxt, (void *)ctxt->execute);
>
> or
> rc = fastop(ctxt, (fastop_t)(void *)ctxt->execute);
I have no strong preference. I originally thought just to go back the previous
code style where might be more acceptable, but it is up to maintainers.
>
> > - else
> > + if (ctxt->d & Fastop) {
> > + fastop_t fop = (void *)ctxt->execute;
> > + rc = fastop(ctxt, fop);
> > + } else {
> > rc = ctxt->execute(ctxt);
> > + }
> > if (rc != X86EMUL_CONTINUE)
> > goto done;
> > goto writeback;
> > --
> > 1.8.3.1
> >
