From: PrasannaKumar Muralidharan Subject: Re: virtio:rng: Virtio RNG devices need to be re-registered after suspend/resume Date: Fri, 3 Nov 2017 21:03:12 +0530 Message-ID: References: <1509703041-6281-1-git-send-email-Jim.Quigley@oracle.com> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" To: Matt Mackall , Herbert Xu , "open list:HARDWARE RANDOM NUMBER GENERATOR CORE" , Jim Quigley Return-path: Received: from mail-io0-f174.google.com ([209.85.223.174]:44190 "EHLO mail-io0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753635AbdKCPdN (ORCPT ); Fri, 3 Nov 2017 11:33:13 -0400 Received: by mail-io0-f174.google.com with SMTP id m16so7132646iod.1 for ; Fri, 03 Nov 2017 08:33:13 -0700 (PDT) In-Reply-To: Sender: linux-crypto-owner@vger.kernel.org List-ID: Hi Jim, Have second thoughts on this. On 3 November 2017 at 20:55, PrasannaKumar Muralidharan wrote: >> >> It would be cleaner to just get rid of probe_common() altogether in that >> case, and do whatever >> needs to be done in virtrng_probe()/virtrng_restore() respectively, but > > That's correct. > >> I didn't want to change code >> affecting the normal probe path as well as suspend/resume. Is it OK to >> leave it that way to avoid >> the more extensive changes ? > > Personally I would prefer to do the cleaner way. It is up to the > virtio and hwrng maintainer. You are trying to restore the status quo of the driver that was before the commit 5c06273401. It is really important and valid. The driver's suspend/resume code is not in best shape but that is not what you are trying to solve. I am fine with this change. Regards, PrasannaKumar