From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jeremy Fitzhardinge Subject: Re: [patch 1/2] Relocate VDSO ELF headers to match mapped location with COMPAT_VDSO Date: Wed, 04 Apr 2007 23:46:29 -0700 Message-ID: <46149B45.2070502@goop.org> References: <20070405063152.9227D180064@magilla.sf.frob.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20070405063152.9227D180064@magilla.sf.frob.com> Sender: linux-kernel-owner@vger.kernel.org To: Roland McGrath Cc: Andi Kleen , Andrew Morton , virtualization@lists.osdl.org, lkml , Zachary Amsden , Jan Beulich , "Eric W. Biederman" , Ingo Molnar List-Id: virtualization@lists.linuxfoundation.org Roland McGrath wrote: > The patch looks nice and clean. However, it does not relocate the symbol > table(s) values. I thought that was done in an earlier version of this I > saw, but I might be misremembering. Though not fatal, this is a regression > from the previous CONFIG_COMPAT_VDSO behavior. It will show up in things > like __kernel_* name display in backtraces. Hm, OK. It does, but I wasn't sure if it would matter. It should be fairly simple to fix up. > If with your other patch > CONFIG_COMPAT_VDSO will become other than a rarely-used compatibility > option, then this should be fixed. Note that with your second patch this > will also break the symbol values in the randomly-located vma vdso; > non-ancient glibc doesn't care if the vdso isn't mapped where its phdrs > say, but everything does still care that the symbol tables in an ELF file > use addresses matching the phdrs in the same file. > I did the second patch because I could, and to see if it would provoke some comment. But effectively removing a kernel config option seems like a good idea to me. J