From: Wei Liu Subject: Re: [DOCSDAY PATCH] xen: rework the comments for struct xen_domctl_vnuma. Date: Thu, 26 Feb 2015 12:15:17 +0000 Message-ID: <20150226121517.GC1824@zion.uk.xensource.com> References: <20150225153316.9677.27792.stgit@Solace.station> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: Content-Disposition: inline In-Reply-To: <20150225153316.9677.27792.stgit@Solace.station> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org To: Dario Faggioli Cc: Elena Ufimtseva , Keir Fraser , Andrew Cooper , Xen-devel , Jan Beulich , Wei Liu List-Id: xen-devel@lists.xenproject.org On Wed, Feb 25, 2015 at 04:33:17PM +0100, Dario Faggioli wrote: > In fact: vnode_to_pnode is an array, not a mask; there was a > typo in the one about vmemrange; there was no indication > of the data directions. > > Signed-off-by: Dario Faggioli > Cc: Wei Liu > Cc: Jan Beulich > Cc: Andrew Cooper > Cc: Keir Fraser > Cc: Elena Ufimtseva > --- > xen/include/public/domctl.h | 32 +++++++++++++++++++++----------- > 1 file changed, 21 insertions(+), 11 deletions(-) > > diff --git a/xen/include/public/domctl.h b/xen/include/public/domctl.h > index b3413a2..ad76853 100644 > --- a/xen/include/public/domctl.h > +++ b/xen/include/public/domctl.h > @@ -958,27 +958,37 @@ typedef struct xen_domctl_vcpu_msrs xen_domctl_vcpu_msrs_t; > DEFINE_XEN_GUEST_HANDLE(xen_domctl_vcpu_msrs_t); > #endif > > -/* > - * Use in XEN_DOMCTL_setvnumainfo to set > - * vNUMA domain topology. > - */ > +/* XEN_DOMCTL_setvnumainfo: specifies a virtual NUMA topology for the guest */ > struct xen_domctl_vnuma { > + /* IN: number of vNUMA nodes to setup. Shall be greater than 0 */ > uint32_t nr_vnodes; > + /* IN: number of memory ranges to setup */ > uint32_t nr_vmemranges; > + /* > + * IN: number of vCPUs of the domain (used as size of the vcpu_to_vnode > + * array declared below). Shall be equal to the domain's max_vcpus. > + */ > uint32_t nr_vcpus; > - uint32_t pad; > + uint32_t pad; /* must be zero */ > + > + /* > + * IN: array for specifying the distances of the vNUMA nodes > + * between each others. Shall have nr_vnodes*nr_vnodes elements. > + */ > XEN_GUEST_HANDLE_64(uint) vdistance; > + /* > + * IN: array for specifying to what vNUMA node each vCPU > + * belongs. Shall have nr_vcpus elements. > + */ > XEN_GUEST_HANDLE_64(uint) vcpu_to_vnode; > - > /* > - * vnodes to physical NUMA nodes mask. > - * This kept on per-domain basis for > - * interested consumers, such as numa aware ballooning. > + * IN: array for specifying on what physical NUMA node > + * each vNUMA node is placed. Shall have nr_vnodes elements. > */ At the very least this hunk should be backported to 4.5? > XEN_GUEST_HANDLE_64(uint) vnode_to_pnode; > - > /* > - * memory rages for each vNUMA node > + * IN: array for specifying the memory ranges. > + * Shall have nr_vmemranges elements. Inconsistent filling column setup? This line wraps at around 55 characters while others appear to be around 80. I think it might be better to wrap around 72. Wei. > */ > XEN_GUEST_HANDLE_64(xen_vmemrange_t) vmemrange; > };