Re: [PATCH 7/8] tools/objtool: Check for use of the ENQCMD instruction in the kernel

[Fenghua Yu]

Hi, Josh,

On Thu, Sep 23, 2021 at 05:55:40PM -0700, Josh Poimboeuf wrote:
> On Thu, Sep 23, 2021 at 03:26:14PM +0000, Fenghua Yu wrote:
> > > > +		} else if (op2 == 0x38 && op3 == 0xf8) {
> > > > +			if (insn.prefixes.nbytes == 1 &&
> > > > +			    insn.prefixes.bytes[0] == 0xf2) {
> > > > +				/* ENQCMD cannot be used in the kernel. */
> > > > +				WARN("ENQCMD instruction at %s:%lx", sec->name,
> > > > +				     offset);
> > > > +
> > > > +				return -1;
> > > > +			}
> > > 
> > > The only concern here is if we want it to be fatal or not. But otherwise
> > > this seems to be all that's required.
> > 
> > objtool doesn't fail kernel build on this fatal warning.
> > 
> > Returning -1 here stops checking the rest of the file and won't report any
> > further warnings unless this ENQCMD warning is fixed. Not returning -1
> > continues checking the rest of the file and may report more warnings.
> > Seems that's the only difference b/w them.
> > 
> > Should I keep this "return -1" or not? Please advice.
> 
> I'd say remove the "return -1" since it's not a fatal-type analysis
> error and there's nothing to prevent objtool from analyzing the rest of
> the file.

Sure. It does make sense to remove "return -1". I will remove it.

Thanks.

-Fenghua