|
From: | Richard Henderson |
Subject: | Re: [Qemu-devel] [PATCH 1/6] target-ppc: add vmul10[u, eu, cu, ecu]q instructions |
Date: | Wed, 28 Sep 2016 21:00:51 -0700 |
User-agent: | Mozilla/5.0 (X11; Linux x86_64; rv:45.0) Gecko/20100101 Thunderbird/45.3.0 |
On 09/28/2016 07:07 PM, David Gibson wrote:
+ tcg_gen_mulu2_i64(t0, t1, cpu_avrl[rA(ctx->opcode)], val); \Do you really want to be using an actual mul op, rather than (in << 3) + (in << 1)? Obviously working out al the carries correctly will be a bit fiddly.
I think it's fine. Modern hardware will do the double-word multiply in 3-5 cycles, which is probably equal to what we could do by hand with shifts.
r~
[Prev in Thread] | Current Thread | [Next in Thread] |