From mboxrd@z Thu Jan 1 00:00:00 1970 From: Dario Faggioli Subject: Re: [PATCH v5 08/24] libxl: introduce libxl__vnuma_config_check Date: Tue, 17 Feb 2015 16:44:56 +0000 Message-ID: <1424191494.4235.92.camel@citrix.com> References: <1423770294-9779-1-git-send-email-wei.liu2@citrix.com> <1423770294-9779-9-git-send-email-wei.liu2@citrix.com> <21726.1811.311219.353545@mariner.uk.xensource.com> <20150213151251.GE13644@zion.uk.xensource.com> Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="===============6335025405817176559==" Return-path: In-Reply-To: <20150213151251.GE13644@zion.uk.xensource.com> Content-Language: en-US List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org To: Wei Liu Cc: "JBeulich@suse.com" , Andrew Cooper , "xen-devel@lists.xen.org" , "ufimtseva@gmail.com" , Ian Jackson , Ian Campbell List-Id: xen-devel@lists.xenproject.org --===============6335025405817176559== Content-Language: en-US Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="=-kFkTzUPmrMwmMR7cAZRl" --=-kFkTzUPmrMwmMR7cAZRl Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable On Fri, 2015-02-13 at 15:12 +0000, Wei Liu wrote: > On Fri, Feb 13, 2015 at 02:15:47PM +0000, Ian Jackson wrote: > > Wei Liu writes ("[PATCH v5 08/24] libxl: introduce libxl__vnuma_config_= check"): > > > This function is used to check whether vNUMA configuration (be it > > > auto-generated or supplied by user) is valid. > >=20 > > This looks plausible, but I think you should explain what the impact > > of this patch is. Presumably the intent is to replace various later > > failures with ERROR_FAIL with something more useful and more > > specific ? > >=20 >=20 > Yes, providing more useful error message is on aspect. Another aspect is > just to do sanity check -- passing an invalid layout to guest doesn't > make much sense. >=20 I agree with Wei. There are a lot of possible variants and combinations of all these parameters, and the earlier we assess the entire set makes sense the better. > > Are there any cases which this new check forbids but which are > > currently accepted by libxl ? If so then we have to think about > > compatibility. > >=20 >=20 > First thing is there is no previous supported vNUMA interface in > toolstack so there won't be a situation where previous good config > doesn't pass this check. >=20 > Second thing is if user supplies a config without vNUMA configuration > this function will not get called, so it won't have any effect. >=20 Indeed. > > Also I would like to see an ack from the authors of the vnuma support, > > as I'm not familiar enough with vnuma to fully understand the > > semantics of the new checks. > >=20 >=20 > Elena and Dario, what do you think? >=20 I made some comments on the code, but, those aside, the checks Wei performs are the correct ones for me. Regards, Dario --=-kFkTzUPmrMwmMR7cAZRl Content-Type: application/pgp-signature; name="signature.asc" Content-Description: This is a digitally signed message part Content-Transfer-Encoding: 7bit -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iEYEABECAAYFAlTjcAYACgkQk4XaBE3IOsSxigCgrmwsI4UtpinemitLt/zy4+7Z lWMAnRtmxuNlYIHOWXgl31owZ9jnOOL3 =Ly0O -----END PGP SIGNATURE----- --=-kFkTzUPmrMwmMR7cAZRl-- --===============6335025405817176559== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel --===============6335025405817176559==--