From 1814a85a055732baf98fd030441bb4c5c5db9bdc Mon Sep 17 00:00:00 2001
From: Andreas Sandberg <Andreas.Sandberg@ARM.com>
Date: Mon, 7 Jan 2013 13:05:46 -0500
Subject: cpu: Rewrite O3 draining to avoid stopping in microcode

Previously, the O3 CPU could stop in the middle of a microcode
sequence. This patch makes sure that the pipeline stops when it has
committed a normal instruction or exited from a microcode
sequence. Additionally, it makes sure that the pipeline has no
instructions in flight when it is drained, which should make draining
more robust.

Draining is controlled in the commit stage, which checks if the next
PC after a committed instruction is in microcode. If this isn't the
case, it requests a squash of all instructions after that the
instruction that just committed and immediately signals a drain stall
to the fetch stage. The CPU then continues to execute until the
pipeline and all associated buffers are empty.
---
 src/cpu/o3/inst_queue_impl.hh | 33 ++++++++-------------------------
 1 file changed, 8 insertions(+), 25 deletions(-)

(limited to 'src/cpu/o3/inst_queue_impl.hh')

diff --git a/src/cpu/o3/inst_queue_impl.hh b/src/cpu/o3/inst_queue_impl.hh
index 785f86676..3e3325beb 100644
--- a/src/cpu/o3/inst_queue_impl.hh
+++ b/src/cpu/o3/inst_queue_impl.hh
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2011 ARM Limited
+ * Copyright (c) 2011-2012 ARM Limited
  * All rights reserved.
  *
  * The license below extends only to copyright in the software and shall
@@ -93,8 +93,6 @@ InstructionQueue<Impl>::InstructionQueue(O3CPU *cpu_ptr, IEW *iew_ptr,
 {
     assert(fuPool);
 
-    switchedOut = false;
-
     numThreads = params->numThreads;
 
     // Set the number of physical registers as the number of int + float
@@ -439,29 +437,19 @@ InstructionQueue<Impl>::setTimeBuffer(TimeBuffer<TimeStruct> *tb_ptr)
 
 template <class Impl>
 void
-InstructionQueue<Impl>::switchOut()
+InstructionQueue<Impl>::drainSanityCheck() const
 {
-/*
-    if (!instList[0].empty() || (numEntries != freeEntries) ||
-        !readyInsts[0].empty() || !nonSpecInsts.empty() || !listOrder.empty()) {
-        dumpInsts();
-//        assert(0);
-    }
-*/
-    resetState();
-    dependGraph.reset();
-    instsToExecute.clear();
-    switchedOut = true;
-    for (ThreadID tid = 0; tid < numThreads; ++tid) {
-        memDepUnit[tid].switchOut();
-    }
+    assert(dependGraph.empty());
+    assert(instsToExecute.empty());
+    for (ThreadID tid = 0; tid < numThreads; ++tid)
+        memDepUnit[tid].drainSanityCheck();
 }
 
 template <class Impl>
 void
 InstructionQueue<Impl>::takeOverFrom()
 {
-    switchedOut = false;
+    resetState();
 }
 
 template <class Impl>
@@ -716,14 +704,9 @@ void
 InstructionQueue<Impl>::processFUCompletion(DynInstPtr &inst, int fu_idx)
 {
     DPRINTF(IQ, "Processing FU completion [sn:%lli]\n", inst->seqNum);
+    assert(!cpu->switchedOut());
     // The CPU could have been sleeping until this op completed (*extremely*
     // long latency op).  Wake it if it was.  This may be overkill.
-    if (isSwitchedOut()) {
-        DPRINTF(IQ, "FU completion not processed, IQ is switched out [sn:%lli]\n",
-                inst->seqNum);
-        return;
-    }
-
     iewStage->wakeCPU();
 
     if (fu_idx > -1)
-- 
cgit v1.2.3