From mboxrd@z Thu Jan 1 00:00:00 1970 From: Stephan Mueller Subject: Re: [PATCH v3 0/3] crypto: drbg - revamp locking Date: Sun, 19 Apr 2015 17:35:23 +0200 Message-ID: <2916811.EN7EiDNNhd@myon.chronox.de> References: <1702481.h2hMdNkdaF@myon.chronox.de> <20150419054931.GC11255@gondor.apana.org.au> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7Bit Cc: linux-crypto@vger.kernel.org To: Herbert Xu Return-path: Received: from mail.eperm.de ([89.247.134.16]:34211 "EHLO mail.eperm.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751648AbbDSPf3 (ORCPT ); Sun, 19 Apr 2015 11:35:29 -0400 In-Reply-To: <20150419054931.GC11255@gondor.apana.org.au> Sender: linux-crypto-owner@vger.kernel.org List-ID: Am Sonntag, 19. April 2015, 13:49:31 schrieb Herbert Xu: Hi Herbert, > On Sat, Apr 18, 2015 at 07:35:11PM +0200, Stephan Mueller wrote: > > Hi, > > > > the current implementation of the DRBG generates a shadow copy of its > > DRBG state for each incoming request. The idea is that only a short > > term lock is needed to spawn the shadow copy. The drawback is that > > if multiple parallel requests come in, the generated DRBG shadow > > states only differ by a high-resolution timer that was mixed in during > > the shadow state generation. > > > > This patch now removes this shadow state and introduces a mutex > > to serialize all requests to one DRBG instance. > > > > The patch was fully CAVS tested and demonstrates that the DRBG still > > complies with the standard. > > All applied. Thanks -- can you please push your tree? There is one more mutex_init missing: in drbg_healthcheck. This missing mutex_init will cause a segfault when booting the kernel with fips=1 -- Ciao Stephan