On Sun, Dec 16, 2018 at 2:19 PM David Miller wrote:
>
> From: Paolo Abeni
> Date: Sun, 16 Dec 2018 19:55:40 +0100
>
> > Again, I messed it! I'm really sorry to waste everybody's time.
> > I was unable to give proper coverage with different configs. I tested
> > vs.:
> >
> > CONFIG_IPV6=ymn
> > CO
From: Paolo Abeni
Date: Sun, 16 Dec 2018 19:55:40 +0100
> Again, I messed it! I'm really sorry to waste everybody's time.
> I was unable to give proper coverage with different configs. I tested
> vs.:
>
> CONFIG_IPV6=ymn
> CONFIG_INET=yn
>
> but
>
> # CONFIG_RETPOLINE is not set
>
> fooled me
On Sat, 2018-12-15 at 13:23 -0800, David Miller wrote:
> From: Paolo Abeni
> Date: Fri, 14 Dec 2018 11:51:56 +0100
>
> > The spectre v2 counter-measures, aka retpolines, are a source of measurable
> > overhead[1]. We can partially address that when the function pointer refers
> > to
> > a builti
From: Paolo Abeni
Date: Fri, 14 Dec 2018 11:51:56 +0100
> The spectre v2 counter-measures, aka retpolines, are a source of measurable
> overhead[1]. We can partially address that when the function pointer refers to
> a builtin symbol resorting to a list of tests vs well-known builtin function
>
The spectre v2 counter-measures, aka retpolines, are a source of measurable
overhead[1]. We can partially address that when the function pointer refers to
a builtin symbol resorting to a list of tests vs well-known builtin function and
direct calls.
Experimental results show that replacing a singl
