From: Andi Kleen Subject: Re: [PATCH 5 of 7] x86/paravirt: add register-saving thunks to reduce caller register pressure Date: Fri, 06 Feb 2009 08:28:26 +0100 Message-ID: <87fxisvxb9.fsf@basil.nowhere.org> References: <7d4a052c499dbfe5e816.1233182105@abulafia.goop.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: In-Reply-To: <7d4a052c499dbfe5e816.1233182105@abulafia.goop.org> (Jeremy Fitzhardinge's message of "Wed, 28 Jan 2009 14:35:05 -0800") Sender: linux-kernel-owner@vger.kernel.org To: Jeremy Fitzhardinge Cc: Ingo Molnar , Zachary Amsden , Rusty Russell , Ian Campbell , the arch/x86 maintainers , linux-kernel@vger.kernel.org, Ravikiran Thirumalai , Xen-devel List-Id: xen-devel@lists.xenproject.org Jeremy Fitzhardinge writes: > One of the problems with inserting a pile of C calls where previously > there were none is that the register pressure is greatly increased. > The C calling convention says that the caller must expect a certain > set of registers may be trashed by the callee, and that the callee can > use those registers without restriction. This includes the function > argument registers, and several others. > > This patch seeks to alleviate this pressure by introducing wrapper > thunks that will do the register saving/restoring, so that the > callsite doesn't need to worry about it, but the callee function can > be conventional compiler-generated code. I had something like this (but not as generic) for the old pre unification semaphores on x86-64 too. I think it's a generally useful facility for lots of things (in fact I've been asking gcc developers for providing this in a general way for some time). So I would suggest to move the necessary macros to do that outside paravirt.h into a separate include so that it can be used independently and not call it PV_*. -Andi -- ak@linux.intel.com -- Speaking for myself only.