From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Mosberger Date: Mon, 12 Jul 2004 20:08:54 +0000 Subject: Re: serious performance regression due to NX patch Message-Id: <16626.61398.906723.300371@napali.hpl.hp.com> List-Id: References: <200407100528.i6A5SF8h020094@napali.hpl.hp.com> <20040711123803.GD21264@devserv.devel.redhat.com> <20040712182431.GB28281@infradead.org> <16626.57892.533203.683465@napali.hpl.hp.com> In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: Ingo Molnar Cc: davidm@hpl.hp.com, Christoph Hellwig , Jakub Jelinek , suresh.b.siddha@intel.com, jun.nakajima@intel.com, Andrew Morton , Linus Torvalds , linux-ia64@vger.kernel.org, linux-kernel@vger.kernel.org, Ulrich Drepper >>>>> On Mon, 12 Jul 2004 15:54:56 -0400 (EDT), Ingo Molnar said: Ingo> On Mon, 12 Jul 2004, David Mosberger wrote: Ingo> is it an issue? Each new port will have PT_GNU_STACK, unless they base Ingo> themselves on old compilers. >> PT_GNU_STACK is pure bloat on new architectures (and ia64). Ingo> EF_IA_64_LINUX_EXECUTABLE_STACK is using elf_ex->e_flags. I Ingo> did it the same way for x86 originally, but the tools people Ingo> specifically rejected it as a hack. We dont control the ELF Ingo> specification, but a new gcc section like PT_GNU_STACK is fair Ingo> game. So it might be 'bloat' but it's clean and doesnt try to Ingo> hijack. I know. All I'm saying is that when the stack permissions match the permission which the platform uses by default, the PT_GNU_STACK header should be omitted. --david