From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755270AbaEOPNF (ORCPT ); Thu, 15 May 2014 11:13:05 -0400 Received: from casper.infradead.org ([85.118.1.10]:59357 "EHLO casper.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755165AbaEOPNC (ORCPT ); Thu, 15 May 2014 11:13:02 -0400 Date: Thu, 15 May 2014 17:12:56 +0200 From: Peter Zijlstra To: Johannes Weiner Cc: Tejun Heo , Ingo Molnar , linux-kernel@vger.kernel.org, "Rafael J. Wysocki" , Juri Lelli , Bruce Allan Subject: Re: [REGRESSION] funny sched_domain build failure during resume Message-ID: <20140515151256.GR30445@twins.programming.kicks-ass.net> References: <20140509160455.GA4486@htj.dyndns.org> <20140514140034.GM30445@twins.programming.kicks-ass.net> <20140514170238.GB15690@htj.dyndns.org> <20140514171051.GX30445@twins.programming.kicks-ass.net> <20140514223648.GH15690@htj.dyndns.org> <20140515144113.GA2886@cmpxchg.org> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="z7K7m04Qp/nn6yf9" Content-Disposition: inline In-Reply-To: <20140515144113.GA2886@cmpxchg.org> User-Agent: Mutt/1.5.21 (2012-12-30) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org --z7K7m04Qp/nn6yf9 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline On Thu, May 15, 2014 at 10:41:14AM -0400, Johannes Weiner wrote: > Is it necessary that resume paths allocate at all? Freeing at suspend > what you have to reallocate at resume is asking for trouble. It's not > just higher order allocations, either, even order-0 allocations are > less reliable without GFP_IOFS. So I think this should be avoided as > much as possible. Well, in my case its because suspend does an effective hot-unplug of all CPUs in the system except CPU0. And if the cpu topology changes I need to allocate new data structures and free the old ones. The reverse is true on resume, it hot-plugs all CPUs again, same story. --z7K7m04Qp/nn6yf9 Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.12 (GNU/Linux) iQIcBAEBAgAGBQJTdNl4AAoJEHZH4aRLwOS6FWkP/1R6lWdKqo+mYmk2OoKuZnHo BfTR1VTKRS5cBcEgMHPDz45G8ctXuiy98CoxoZCjpZaqpUe5i0+jCa5OSQZnvCze qN8ap2glV4AappFmgfn0xL3Jf1jNNj0GxHWDNHlCOmu9JH8YO7ax30Ks1H3W1tP/ fAMGi/r0N7TpeASKIIY/F/h8bgWQ89fiHJG4IBnd5DhVebqOAIB8nx6pjx7E/kKt s8k2JKAXlbpHj1mjnWbskEC6rx7O/eAOW6XfbIEAiWuWLzx1l9OvRfv4wIw+BnYp fRXn7+laQcLmJgArBOWzGgEdIvi1Jz2Ggk/1E8j+O005IiMMy5uy4H0h1cyxuMEU L7SegF4hj+vGvFixOJIogX36K0dFapPepdpWeE0Ukvx/MxBpPslEWO6S48DLjOBV jcUn5eGM+2XwgbQZRnUM8swAlmykEkMQPba4Fn5jVTd084ySTxCzDH6eu6riL+4/ 9OX7Cl4i77BzfQR02nX7GibaMilGHY7FceEAc2YEc6IHLUNSo/aQT/hwtJToAlj9 n68XlfSDPtYgWH3lvBrQkpYsp/4kMtf9IdX81WaLeoPGgfa2UdusfqrAid3hqwXU Qfaszud0NQnGFabksStNhZyGrKxGu3TB+seKhuTladsQrqAvIglm+Es03ZDSd1MO HlbGRUmbRgYYUAC/gLj1 =c2hW -----END PGP SIGNATURE----- --z7K7m04Qp/nn6yf9--