Re: [Xen-devel] [PATCH v19 12/14] x86/VPMU: Merge vpmu_rdmsr and vpmu_wrmsr

["Jan Beulich" <JBeulich@xxxxxxxx>]

>>> On 17.03.15 at 15:54, <boris.ostrovsky@xxxxxxxxxx> wrote:
> ---
> Changes in v19:
> * const-ified arch_vpmu_ops in vpmu_do_wrmsr
> * non-changes:
>    - kept 'current' as a non-initializer to avoid unnecessary initialization
>      in the (common) non-VPMU case

Afaict there's nothing at all keeping the compiler to still schedule part
or all of the involved insns ahead of that first goto. And if it was an
initializer, there would equally be nothing keeping the compiler from
scheduling part or all of those insns after the if(). So I don't see why
not making it the initializer is better (other than increasing line count
and patch size). If you really wanted to do something to help the
compiler, flagging the if() condition with likely() would perhaps be the
thing to do.

Jan


_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel