On 10/04/2018 07:05 AM, Sebastian Andrzej Siewior wrote:
> Most users of __raw_xsave_addr() use a feature number, shift it to a
> mask and then __raw_xsave_addr() shifts it back to the feature number.
>
> Make __raw_xsave_addr() use the feature number as argument.
This generally looks like a nice cleanup. Thanks for taking a look at it!
> diff --git a/arch/x86/kernel/fpu/xstate.c b/arch/x86/kernel/fpu/xstate.c
> index 87a57b7642d36..38d0b5ea40425 100644
> --- a/arch/x86/kernel/fpu/xstate.c
> +++ b/arch/x86/kernel/fpu/xstate.c
> @@ -811,10 +811,8 @@ void fpu__resume_cpu(void)
> *
> * Note: does not work for compacted buffers.
> */
> -void *__raw_xsave_addr(struct xregs_state *xsave, int xstate_feature_mask)
> +void *__raw_xsave_addr(struct xregs_state *xsave, int feature_nr)
> {
Could we call this 'xfeature_nr' consistently?
> - int feature_nr = fls64(xstate_feature_mask) - 1;
> -
> if (!xfeature_enabled(feature_nr)) {
> WARN_ON_FPU(1);
> return NULL;
> @@ -842,6 +840,7 @@ void *__raw_xsave_addr(struct xregs_state *xsave, int xstate_feature_mask)
> */
> void *get_xsave_addr(struct xregs_state *xsave, int xstate_feature)
> {
> + int feature_nr;
> /*
> * Do we even *have* xsave state?
> */
> @@ -869,7 +868,8 @@ void *get_xsave_addr(struct xregs_state *xsave, int xstate_feature)
> if (!(xsave->header.xfeatures & xstate_feature))
> return NULL;
>
> - return __raw_xsave_addr(xsave, xstate_feature);
> + feature_nr = fls64(xstate_feature) - 1;
> + return __raw_xsave_addr(xsave, feature_nr);
> }
Should we also be using a feature number for get_xsave_addr()? In other
words, could you take a look and see how widely we should be doing this
kind of conversion and not just limit it to __raw_xsave_addr()?
