From mboxrd@z Thu Jan 1 00:00:00 1970 From: Keir Fraser Subject: Re: [PATCH v3 03/13] xen: introduce cpumask_from_bitmap Date: Thu, 25 Apr 2013 18:20:27 +0100 Message-ID: References: <1366903403.20256.565.camel@zakaz.uk.xensource.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <1366903403.20256.565.camel@zakaz.uk.xensource.com> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org To: Ian Campbell Cc: "Keir (Xen.org)" , Stefano Stabellini , "Tim (Xen.org)" , xen-devel , Julien Grall , Jan Beulich List-Id: xen-devel@lists.xenproject.org On 25/04/2013 16:23, "Ian Campbell" wrote: > On Thu, 2013-04-25 at 16:13 +0100, Keir Fraser wrote: >> On 25/04/2013 15:51, "Ian Campbell" wrote: >> >>> On Thu, 2013-04-25 at 15:44 +0100, Ian Campbell wrote: >>>> On Thu, 2013-04-25 at 15:42 +0100, Ian Campbell wrote: >>>>> And it really wants to act on sets of real physical cpus (hence >>>>> cpumask_t)? >>>> >>>> Actually, vcpus. Perhaps cpumask_t is the wrong datatype then? >>> >>> x86 defines for_each_set_bit in terms of find_first_bit and >>> find_next_bit (both of which we have on ARM). Seems like we should just >>> carry that over? >> >> Yup, seems a shame to duplicate it into asm-arm, but I can live with it. I >> can imagine some arch wanting a specialised version of it sometime. > > We could move it to generic code, perhaps with a containing ifndef for > this next arch... Happy to see that, and have everyone #include xen/bitops.h rather than asm/bitops.h while we're at it. -- Keir > Ian. >