From: Herbert Xu Subject: Re: [PATCH 1/2] crypto: stm32 - Fix uninitialized data usage Date: Sat, 7 Oct 2017 12:19:58 +0800 Message-ID: <20171007041958.GF29803@gondor.apana.org.au> References: <20170912093553.2580986-1-arnd@arndb.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Lionel Debieve , "David S. Miller" , Maxime Coquelin , Alexandre Torgue , linux-crypto@vger.kernel.org, linux-arm-kernel@lists.infradead.org, linux-kernel@vger.kernel.org To: Arnd Bergmann Return-path: Received: from orcrist.hmeau.com ([104.223.48.154]:57408 "EHLO deadmen.hmeau.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750765AbdJGEUb (ORCPT ); Sat, 7 Oct 2017 00:20:31 -0400 Content-Disposition: inline In-Reply-To: <20170912093553.2580986-1-arnd@arndb.de> Sender: linux-crypto-owner@vger.kernel.org List-ID: On Tue, Sep 12, 2017 at 11:35:38AM +0200, Arnd Bergmann wrote: > The error handling in stm32_hash_irq_thread passes > uninitialized data into stm32_hash_finish_req, as gcc > points out: > > drivers/crypto/stm32/stm32-hash.c: In function 'stm32_hash_irq_thread': > drivers/crypto/stm32/stm32-hash.c:1088:2: error: 'err' may be used uninitialized in this function [-Werror=maybe-uninitialized] > > I could not tell what data should be passed there instead, > so this changes the code to always pass zero, making it > well-defined, though possibly still wrong. Please check. > > Signed-off-by: Arnd Bergmann This is already fixed in cryptodev. -- Email: Herbert Xu Home Page: http://gondor.apana.org.au/~herbert/ PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt