From: Herbert Xu Subject: Re: Fixing gave up waiting for init of module libcrc32c. Date: Sat, 20 Mar 2010 11:45:31 +0800 Message-ID: <20100320034531.GA31798@gondor.apana.org.au> References: <20100319234044.GA5127@jenkins> <20100320010115.GA30561@gondor.apana.org.au> <20100320010849.GA30654@gondor.apana.org.au> <20100319.194612.226782074.davem@davemloft.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: brandon@ifup.org, linux-crypto@vger.kernel.org, rusty@rustcorp.com.au To: David Miller Return-path: Received: from rhun.apana.org.au ([64.62.148.172]:60376 "EHLO arnor.apana.org.au" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1751386Ab0CTDpj (ORCPT ); Fri, 19 Mar 2010 23:45:39 -0400 Content-Disposition: inline In-Reply-To: <20100319.194612.226782074.davem@davemloft.net> Sender: linux-crypto-owner@vger.kernel.org List-ID: On Fri, Mar 19, 2010 at 07:46:12PM -0700, David Miller wrote: > > Even if userland "does" this, nothing stops another thread > from rmmod'ing the dependency in between the modloads. > > We have to handle this case properly in the kernel, no matter > what. Right, if that had happened I think the reasonable response would be to fail the subsequent load for lack of dependency rather than looping around forever. However, I don't think that's what Brandon is seeing here, unless there is something performing the rmmod? Cheers, -- Visit Openswan at http://www.openswan.org/ Email: Herbert Xu ~{PmV>HI~} Home Page: http://gondor.apana.org.au/~herbert/ PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt