Review Board 2.0.15


cpu: discard nops at decode in o3

Review Request #1804 - Created March 29, 2013 and updated

Information
Mitch Hayenga
gem5
Reviewers
Default 
Changes the o3 cpu to discard nops at decode.

This stops nops from occupying scheduler/rob entries and commit/issue bandwidth.
A fast libquantum run.
Ship it!
Ali Saidi
Posted (July 15, 2013, 8:13 a.m.)
Seems Ok to me as long as all the regressions still run (obviously the stats will change). Anyone else have input?
  1. It needs a change or two.  In it's current form, it would break FS.

Here's a self-quote from an email with Korey a while back.
"I also realized this current patch would break FS mode, since the frontend signals instruction fetch page faults by creating a nop with a fault attached (this patch would just discard that nop).  So, checking for a fault would be required before discarding."

I'm frantically trying to finish a thesis (defending in August), so probably won't get to it for a bit even though its a minor change.