From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andy Lutomirski Subject: Re: [RFC 0/2] __vdso_findsym Date: Sat, 14 Jun 2014 23:49:53 -0700 Message-ID: References: <20140614213000.GO179@brightrain.aerifal.cx> <539CDC6F.6010501@zytor.com> <539D1A28.7040107@zytor.com> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Return-path: In-Reply-To: <539D1A28.7040107-YMNOUZJC4hwAvxtiuMwx3w@public.gmane.org> Sender: linux-api-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: "H. Peter Anvin" Cc: Rich Felker , Russ Cox , Linux API , Ian Taylor , "linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org" , X86 ML List-Id: linux-api@vger.kernel.org On Sat, Jun 14, 2014 at 8:59 PM, H. Peter Anvin wrote: > On 06/14/2014 04:36 PM, H. Peter Anvin wrote: >> >> Now, if we can do this without adding another page then that becomes >> much more compelling IMO. I suspect we can if you are already down to >> 488 bytes. >> > > (If it isn't already the case, was inteded to be part of that...) It's down to 375 bytes now. I'm not sure how to do better than that without starting to hardcode variously nasty assumptions. With CONFIG_PARAVIRT_CLOCK=y I think the 64-bit vdso is two pages anyway. With CONFIG_PARAVIRT_CLOCK=n, it's one page even with my code. (At some point I plan on tackling the paravirt clock code. It's much more complicated and much slower than it needs to be.) --Andy