From: Alexander Shishkin Subject: Re: [RFC][PATCH] crypto: disable tests when they are not enabled Date: Mon, 24 May 2010 15:20:50 +0300 Message-ID: <20100524122050.GB27432@shisha.kicks-ass.net> References: <1274693724-23831-1-git-send-email-virtuoso@slind.org> <20100524112726.GB29848@gondor.apana.org.au> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: linux-kernel@vger.kernel.org, linux-crypto@vger.kernel.org To: Herbert Xu Return-path: Content-Disposition: inline In-Reply-To: <20100524112726.GB29848@gondor.apana.org.au> Sender: linux-kernel-owner@vger.kernel.org List-Id: linux-crypto.vger.kernel.org On Mon, May 24, 2010 at 09:27:26 +1000, Herbert Xu wrote: > On Mon, May 24, 2010 at 12:35:24PM +0300, Alexander Shishkin wrote: > > Don't compile in and don't schedule cryptomanager's tests when > > CONFIG_CRYPTO_TESTS is not set. This saves quite some time during > > kernel initialization. > > You're disabling the self-testing infrastructure. This is not > the same thing as the tcrypt module. It is my intention to disable these tests since in my running them at every boot doesn't add any value. I thought about adding a separate Kconfig option for that, too. I want to make sure it's a sensible thing to do. > Is this for fast booting? Yes, it saves ~180 miliseconds for me at prng_mod_init(), for example. Regards, -- Alex