From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jack Steiner Date: Thu, 03 Jun 2004 02:38:01 +0000 Subject: Re: Scalability enhancements for gettimeofday Message-Id: <20040603023800.GA14520@sgi.com> List-Id: References: <200405201205.57930.clameter@sgi.com> In-Reply-To: <200405201205.57930.clameter@sgi.com> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: linux-ia64@vger.kernel.org On Wed, Jun 02, 2004 at 03:58:26PM -0500, John Hesterberg wrote: > On Wed, Jun 02, 2004 at 10:33:39AM -0700, David Mosberger wrote: > > Christoph> Also the fast system call handler for sys_gettimeofday() > > Christoph> can only improve scalability if the ITC is the basis for > > Christoph> gettimeofday which is also never the case on Altix SN2 > > Christoph> systems. > > > > Does this really make a substantial difference? My guess is no and, > > if so, I'd rather not add any more #ifdefs. > > It also only helps SN2 kernels and not generic kernels. > I don't think this is they key part of the patch, > so unless we see some compelling performance evidence, > I'd agree it seems better to leave out the #ifdef. > > John What is the limitation in gettimeofday that prevents us from using a fast system call handler to access the RTC on SN2. I realize that fsys.S does not currently support this, but is there any fundamental restriction that makes this impossible to add? If not, it would seem preferable to support SN2 rather add the #ifdefs. -- Thanks Jack Steiner (steiner@sgi.com) 651-683-5302 Principal Engineer SGI - Silicon Graphics, Inc.