From mboxrd@z Thu Jan 1 00:00:00 1970 From: Zachary Amsden Subject: Re: [patch 13/21] Xen-paravirt: Add nosegneg capability to the vsyscall page notes Date: Thu, 15 Feb 2007 22:06:37 -0800 Message-ID: <45D549ED.4000206@vmware.com> References: <20070216022449.739760547@goop.org> <20070216022531.495288020@goop.org> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20070216022531.495288020@goop.org> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xensource.com Errors-To: xen-devel-bounces@lists.xensource.com To: Jeremy Fitzhardinge Cc: xen-devel@lists.xensource.com, Ian Pratt , virtualization@lists.osdl.org, linux-kernel@vger.kernel.org, Chris Wright , Andi Kleen , Andrew Morton , Christian Limpach List-Id: virtualization@lists.linuxfoundation.org Jeremy Fitzhardinge wrote: > Add the "nosegneg" fake capabilty to the vsyscall page notes. This is > used by the runtime linker to select a glibc version which then > disables negative-offset accesses to the thread-local segment via > %gs. These accesses require emulation in Xen (because segments are > truncated to protect the hypervisor address space) and avoiding them > provides a measurable performance boost. > > Signed-off-by: Ian Pratt > Signed-off-by: Christian Limpach > Signed-off-by: Chris Wright > Acked-by: Zachary Amsden We would like to see this by dynamic, but that is much more difficult to achieve, and seeing your recent linker issues, I don't think this should gate merging this code. The performance loss for us I believe to be negligible, and the fix is quite a bit more complicated than something achievable in the .21 timeframe. Zach