X87: [turbofan] avoid xchg instruction on Intel.

X87:  [turbofan] avoid xchg instruction on Intel.

  port 7440edae1da2b78a37e6fe3a558e249dc33444aa(r33282)

  original commit message:
  On Intel, xchg stalls the pipeline. We use xchg to implement swap
  moves. In a separate exploration, the presence of xchg in a very hot
  loop, due to a change in register allocation, lead to over 20% regression.

  Simply changing that instruction with push/mov/pop (almost) eliminated
  the regression.

  In light of that, I removed uses of xchg. This leads to more instructions,
  though. That is particularly problematic for long cycles, which, today,
  we translate to successions of swaps.

  I plan to address this cycle issue in a separate change. For now, the
  goal is to unblock the initial work that lead here.

BUG=

[zhengxing.li, 2016-01-15 01:59:10]

zhengxing.li@intel.com changed reviewers:
+ chunyang.dai@intel.com, mtrofin@chromium.org, weiliang.lin@intel.com

[zhengxing.li, 2016-01-15 01:59:11]

PTAL, Thanks!

[Mircea Trofin, 2016-01-22 16:31:31]

On 2016/01/15 01:59:11, zhengxing.li wrote:
> PTAL, Thanks!

LGTM
sorry for the delay