From: Steffen Klassert Subject: Re: [PATCH 1/2] crypto: Implement a generic crypto statistics Date: Wed, 31 Jan 2018 09:33:37 +0100 Message-ID: <20180131083337.rvshu72x5xhh22ef@gauss3.secunet.de> References: <1515700617-3513-1-git-send-email-clabbe@baylibre.com> <3297958.ndliyBY1jk@tauon.chronox.de> <20180112090730.GA15630@Red> <16302559.hxlYq3rOsN@tauon.chronox.de> <20180118085813.GA1988@Red> <20180126154328.GA4002@gondor.apana.org.au> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Cc: LABBE Corentin , , , , , To: Herbert Xu Return-path: Content-Disposition: inline In-Reply-To: <20180126154328.GA4002@gondor.apana.org.au> Sender: linux-kernel-owner@vger.kernel.org List-Id: linux-crypto.vger.kernel.org On Fri, Jan 26, 2018 at 11:43:28PM +0800, Herbert Xu wrote: > On Thu, Jan 18, 2018 at 09:58:13AM +0100, LABBE Corentin wrote: > > > > I have two way of adding a new netlink request > > - keep the current patch and simply add a new CRYPTO_MSG_GETSTAT which use the same function than CRYPTO_MSG_GETALG > > => minimal changes, in fact CRYPTO_MSG_GETSTAT and CRYPTO_MSG_GETALG would be the same, but it is easy for userspace to test presence of stat. > > - Create a new CRYPTO_MSG_GETSTAT which imply lot of code and add a new crypto_user_stat.c > > => this imply also to change makefile (rename crypto_user.c to crypto_user_base.c) since crypto_user.ko is made of two files. > > > > Which one do you prefer ? > > I think you should check with the crconf author, Steffen Klassert > since that will be affected by this change. I still haven't had time to look into the details, but I'm fine with everything that does not break existing userspace. Everything else would be a 'no go'.