On Sat, Mar 15, 2008 at 10:06:32PM +0100, Alexander van Heukelum wrote:So does GEODE_LX BTW in my experience (on user-space code), optimizing for "i586" gives good results. However, the cache is small, so everything which can reduce code size (especially loop/jump/function alignment) is worth checking. Regards, Willy --
| Andi Kleen | [PATCH x86] [6/16] Add a new arch_early_alloc() interface for x86-64 |
| Jay L. T. Cornwall | 2.6.22-rc5: pdflush oops under heavy disk load |
| Adrian Bunk | If you want me to quit I will quit |
| H. Willstrand | Re: AMD Quad Core clock problem? |
git: | |
| Pietro Mascagni | GIT vs Other: Need argument |
| Wink Saville | Resolving conflicts |
| Wink Saville | Using git with Eclipse |
| Toby White | Using Filemerge.app as a git-diff viewer |
| GVG GVG | ssh_exchange_identification: Connection closed by remote host |
| Beavis | mutiple pptp pass-through PF |
| Andrei Pirvan | apache 1.3.29 + PHP 5.2.6 on OpenBSD 4.4 |
| STeve Andre' | Re: Perpetually Current |
| Herbert Xu | Re: [PATCH] pkt_sched: Destroy gen estimators under rtnl_lock(). |
| Denys | r8169 crash |
| David Miller | Re: [crash] BUG: unable to handle kernel NULL pointer dereference at 0000000000000... |
| Jianjun Kong | [PATCH 2/6] nets: clean up net/ipv4/ipip.c raw.c tcp.c tcp_minisocks.c tcp_yeah.c ... |
