Re: [PATCH net-next] sparc64: Improve 64-bit constant loading in eBPF JIT.

2017-04-24 Thread David Miller
From: Alexei Starovoitov Date: Mon, 24 Apr 2017 20:18:56 -0700 > all gcc-ism actually gives confidence that most likely > it's all good and battle proven logic :) Yes, this is old code written 15 years ago :) I'll fix up the spaces.

Re: [PATCH net-next] sparc64: Improve 64-bit constant loading in eBPF JIT.

2017-04-24 Thread Alexei Starovoitov
On 4/24/17 7:53 PM, David Miller wrote: Doing a full 64-bit decomposition is really stupid especially for simple values like 0 and -1. But if we are going to optimize this, go all the way and try for all 2 and 3 instruction sequences not requiring a temporary register as well. Signed-off-by: D

[PATCH net-next] sparc64: Improve 64-bit constant loading in eBPF JIT.

2017-04-24 Thread David Miller
Doing a full 64-bit decomposition is really stupid especially for simple values like 0 and -1. But if we are going to optimize this, go all the way and try for all 2 and 3 instruction sequences not requiring a temporary register as well. Signed-off-by: David S. Miller --- This one is dedicated