Check in of various updates to the CPU. Mainly adds in stats, improves

branch prediction, and makes memory dependence work properly.

SConscript:
    Added return address stack, tournament predictor.
cpu/base_cpu.cc:
    Added debug break and print statements.
cpu/base_dyn_inst.cc:
cpu/base_dyn_inst.hh:
    Comment out possibly unneeded variables.
cpu/beta_cpu/2bit_local_pred.cc:
    2bit predictor no longer speculatively updates itself.
cpu/beta_cpu/alpha_dyn_inst.hh:
    Comment formatting.
cpu/beta_cpu/alpha_full_cpu.hh:
    Formatting
cpu/beta_cpu/alpha_full_cpu_builder.cc:
    Added new parameters for branch predictors, and IQ parameters.
cpu/beta_cpu/alpha_full_cpu_impl.hh:
    Register stats.
cpu/beta_cpu/alpha_params.hh:
    Added parameters for IQ, branch predictors, and store sets.
cpu/beta_cpu/bpred_unit.cc:
    Removed one class.
cpu/beta_cpu/bpred_unit.hh:
    Add in RAS, stats.  Changed branch predictor unit functionality
    so that it holds a history of past branches so it can update, and also
    hold a proper history of the RAS so it can be restored on branch
    mispredicts.
cpu/beta_cpu/bpred_unit_impl.hh:
    Added in stats, history of branches, RAS.  Now bpred unit actually
    modifies the instruction's predicted next PC.
cpu/beta_cpu/btb.cc:
    Add in sanity checks.
cpu/beta_cpu/comm.hh:
    Add in communication where needed, remove it where it's not.
cpu/beta_cpu/commit.hh:
cpu/beta_cpu/rename.hh:
cpu/beta_cpu/rename_impl.hh:
    Add in stats.
cpu/beta_cpu/commit_impl.hh:
    Stats, update what is sent back on branch mispredict.
cpu/beta_cpu/cpu_policy.hh:
    Change the bpred unit being used.
cpu/beta_cpu/decode.hh:
cpu/beta_cpu/decode_impl.hh:
    Stats.
cpu/beta_cpu/fetch.hh:
    Stats, change squash so it can handle squashes from decode differently
    than squashes from commit.
cpu/beta_cpu/fetch_impl.hh:
    Add in stats.  Change how a cache line is fetched.  Update to work with
    caches.  Also have separate functions for different behavior if squash
    is coming from decode vs commit.
cpu/beta_cpu/free_list.hh:
    Remove some old comments.
cpu/beta_cpu/full_cpu.cc:
cpu/beta_cpu/full_cpu.hh:
    Added function to remove instructions from back of instruction list
    until a certain sequence number.
cpu/beta_cpu/iew.hh:
    Stats, separate squashing behavior due to branches vs memory.
cpu/beta_cpu/iew_impl.hh:
    Stats, separate squashing behavior for branches vs memory.
cpu/beta_cpu/inst_queue.cc:
    Debug stuff
cpu/beta_cpu/inst_queue.hh:
    Stats, change how mem dep unit works, debug stuff
cpu/beta_cpu/inst_queue_impl.hh:
    Stats, change how mem dep unit works, debug stuff.  Also add in
    parameters that used to be hardcoded.
cpu/beta_cpu/mem_dep_unit.hh:
cpu/beta_cpu/mem_dep_unit_impl.hh:
    Add in stats, change how memory dependence unit works.  It now holds
    the memory instructions that are waiting for their memory dependences
    to resolve.  It provides which instructions are ready directly to the
    IQ.
cpu/beta_cpu/regfile.hh:
    Fix up sanity checks.
cpu/beta_cpu/rename_map.cc:
    Fix loop variable type.
cpu/beta_cpu/rob_impl.hh:
    Remove intermediate DynInstPtr
cpu/beta_cpu/store_set.cc:
    Add in debugging statements.
cpu/beta_cpu/store_set.hh:
    Reorder function arguments to match the rest of the calls.

--HG--
extra : convert_revision : aabf9b1fecd1d743265dfc3b174d6159937c6f44

1 files changed, 62 insertions, 1 deletions

diff --git a/cpu/beta_cpu/alpha_full_cpu_builder.cc b/cpu/beta_cpu/alpha_full_cpu_builder.cc
index 5fe96d656..f37081232 100644
--- a/cpu/beta_cpu/alpha_full_cpu_builder.cc
+++ b/cpu/beta_cpu/alpha_full_cpu_builder.cc
@@ -81,17 +81,38 @@ Param<unsigned> issueWidth;
 Param<unsigned> executeWidth;
 Param<unsigned> executeIntWidth;
 Param<unsigned> executeFloatWidth;
+Param<unsigned> executeBranchWidth;
+Param<unsigned> executeMemoryWidth;
 
 Param<unsigned> iewToCommitDelay;
 Param<unsigned> renameToROBDelay;
 Param<unsigned> commitWidth;
 Param<unsigned> squashWidth;
 
+#if 0
 Param<unsigned> localPredictorSize;
 Param<unsigned> localPredictorCtrBits;
+#endif
+Param<unsigned> local_predictor_size;
+Param<unsigned> local_ctr_bits;
+Param<unsigned> local_history_table_size;
+Param<unsigned> local_history_bits;
+Param<unsigned> global_predictor_size;
+Param<unsigned> global_ctr_bits;
+Param<unsigned> global_history_bits;
+Param<unsigned> choice_predictor_size;
+Param<unsigned> choice_ctr_bits;
+
 Param<unsigned> BTBEntries;
 Param<unsigned> BTBTagSize;
 
+Param<unsigned> RASSize;
+
+Param<unsigned> LQEntries;
+Param<unsigned> SQEntries;
+Param<unsigned> LFSTSize;
+Param<unsigned> SSITSize;
+
 Param<unsigned> numPhysIntRegs;
 Param<unsigned> numPhysFloatRegs;
 Param<unsigned> numIQEntries;
@@ -168,6 +189,8 @@ BEGIN_INIT_SIM_OBJECT_PARAMS(BaseFullCPU)
     INIT_PARAM(executeWidth, "Execute width"),
     INIT_PARAM(executeIntWidth, "Integer execute width"),
     INIT_PARAM(executeFloatWidth, "Floating point execute width"),
+    INIT_PARAM(executeBranchWidth, "Branch execute width"),
+    INIT_PARAM(executeMemoryWidth, "Memory execute width"),
 
     INIT_PARAM(iewToCommitDelay, "Issue/Execute/Writeback to commit "
                "delay"),
@@ -175,12 +198,30 @@ BEGIN_INIT_SIM_OBJECT_PARAMS(BaseFullCPU)
     INIT_PARAM(commitWidth, "Commit width"),
     INIT_PARAM(squashWidth, "Squash width"),
 
+#if 0
     INIT_PARAM(localPredictorSize, "Size of the local predictor in entries. "
                "Must be a power of 2."),
     INIT_PARAM(localPredictorCtrBits, "Number of bits per counter for bpred"),
+#endif
+    INIT_PARAM(local_predictor_size, "Size of local predictor"),
+    INIT_PARAM(local_ctr_bits, "Bits per counter"),
+    INIT_PARAM(local_history_table_size, "Size of local history table"),
+    INIT_PARAM(local_history_bits, "Bits for the local history"),
+    INIT_PARAM(global_predictor_size, "Size of global predictor"),
+    INIT_PARAM(global_ctr_bits, "Bits per counter"),
+    INIT_PARAM(global_history_bits, "Bits of history"),
+    INIT_PARAM(choice_predictor_size, "Size of choice predictor"),
+    INIT_PARAM(choice_ctr_bits, "Bits of choice counters"),
+
     INIT_PARAM(BTBEntries, "Number of BTB entries"),
     INIT_PARAM(BTBTagSize, "Size of the BTB tags, in bits"),
 
+    INIT_PARAM(RASSize, "RAS size"),
+
+    INIT_PARAM(LQEntries, "Number of load queue entries"),
+    INIT_PARAM(SQEntries, "Number of store queue entries"),
+    INIT_PARAM(LFSTSize, "Last fetched store table size"),
+    INIT_PARAM(SSITSize, "Store set ID table size"),
 
     INIT_PARAM(numPhysIntRegs, "Number of physical integer registers"),
     INIT_PARAM(numPhysFloatRegs, "Number of physical floating point "
@@ -277,17 +318,37 @@ CREATE_SIM_OBJECT(BaseFullCPU)
     params.executeWidth = executeWidth;
     params.executeIntWidth = executeIntWidth;
     params.executeFloatWidth = executeFloatWidth;
+    params.executeBranchWidth = executeBranchWidth;
+    params.executeMemoryWidth = executeMemoryWidth;
 
     params.iewToCommitDelay = iewToCommitDelay;
     params.renameToROBDelay = renameToROBDelay;
     params.commitWidth = commitWidth;
     params.squashWidth = squashWidth;
-
+#if 0
     params.localPredictorSize = localPredictorSize;
     params.localPredictorCtrBits = localPredictorCtrBits;
+#endif
+    params.local_predictor_size = local_predictor_size;
+    params.local_ctr_bits = local_ctr_bits;
+    params.local_history_table_size = local_history_table_size;
+    params.local_history_bits = local_history_bits;
+    params.global_predictor_size = global_predictor_size;
+    params.global_ctr_bits = global_ctr_bits;
+    params.global_history_bits = global_history_bits;
+    params.choice_predictor_size = choice_predictor_size;
+    params.choice_ctr_bits = choice_ctr_bits;
+
     params.BTBEntries = BTBEntries;
     params.BTBTagSize = BTBTagSize;
 
+    params.RASSize = RASSize;
+
+    params.LQEntries = LQEntries;
+    params.SQEntries = SQEntries;
+    params.SSITSize = SSITSize;
+    params.LFSTSize = LFSTSize;
+
     params.numPhysIntRegs = numPhysIntRegs;
     params.numPhysFloatRegs = numPhysFloatRegs;
     params.numIQEntries = numIQEntries;