Re: [Xen-devel] [PATCH v2 1/6] x86/vmx: Fix handing of MSR_DEBUGCTL on VMExit

["Jan Beulich" <JBeulich@xxxxxxxx>]

>>> On 30.05.18 at 19:34, <andrew.cooper3@xxxxxxxxxx> wrote:
> @@ -117,6 +115,9 @@ integer_param("debug_stack_lines", debug_stack_lines);
>  static bool opt_ler;
>  boolean_param("ler", opt_ler);
>  
> +/* LastExceptionFromIP on this hardware.  Zero if LER is not in use. */
> +uint32_t __read_mostly ler_msr;

Hmm, this is still uint32_t rather than unsigned int, while you did change
calc_ler_msr()'s return type.

> +void percpu_traps_init(void)
> +{
> +    subarch_percpu_traps_init();
> +
> +    if ( !opt_ler )
> +        return;
> +
> +    if ( !ler_msr && (ler_msr = calc_ler_msr()) )
> +        setup_force_cpu_cap(X86_FEATURE_XEN_LBR);

I assume it was on purpose that you've adjusted the commit message
rather than the code here regarding the possibility of pointless multiple
invocation?

With preferably the type inconsistency addressed
Reviewed-by: Jan Beulich <jbeulich@xxxxxxxx>

Jan



_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxxx
https://lists.xenproject.org/mailman/listinfo/xen-devel