From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Woodhouse Date: Tue, 12 Oct 2004 21:50:24 +0000 Subject: Re: [PATCH] Support ia32 exec domains without CONFIG_IA32_SUPPORT Message-Id: <1097617824.5178.20.camel@localhost.localdomain> List-Id: References: <41643EC0.1010505@intel.com> <20041007142710.A12688@infradead.org> <4165D4C9.2040804@intel.com> <41671696.1060706@intel.com> <416AF599.2060801@intel.com> In-Reply-To: <416AF599.2060801@intel.com> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: Arun Sharma Cc: Christoph Hellwig , linux-kernel@vger.kernel.org, linux-ia64@vger.kernel.org On Mon, 2004-10-11 at 14:05 -0700, Arun Sharma wrote: > I've prototyped a generic userland solution that covers just open and > stat system calls (for completeness, all path walk related syscalls > need to be covered) using the LD_PRELOAD approach. > > I saw a 16% degradation in system time on this benchmark: > > find /usr/src/linux -name '*.[chS]' | xargs grep fsck > > mainly due to the doubling of the number of calls to open. Also, there > was a slight increase in user time as well, due to malloc/free > overhead. The patch is entirely bogus. This isn't at all ia64-specific, and doesn't live in arch/ia64. It's just as applicable on _all_ systems where we may want to do CPU or OS emulation. If you make it generic so that qemu can use it for emulating i386 even on machines like ppc64, perhaps it would be saner. -- dwmw2