Files
gem5/cpu/beta_cpu/rob_impl.hh
Kevin Lim e3fb9afa79 Update to make multiple instruction issue and different latencies work.
Also change to ref counted DynInst.

SConscript:
    Add branch predictor, BTB, load store queue, and storesets.
arch/isa_parser.py:
    Specify the template parameter for AlphaDynInst
base/traceflags.py:
    Add load store queue, store set, and mem dependence unit to the
    list of trace flags.
cpu/base_dyn_inst.cc:
    Change formating, add in debug statement.
cpu/base_dyn_inst.hh:
    Change DynInst to be RefCounted, add flag to clear whether or not this
    instruction can commit.  This is likely to be removed in the future.
cpu/beta_cpu/alpha_dyn_inst.cc:
    AlphaDynInst has been changed to be templated, so now this CC file
    is just used to force instantiations of AlphaDynInst.
cpu/beta_cpu/alpha_dyn_inst.hh:
    Changed AlphaDynInst to be templated on Impl.  Removed some unnecessary
    functions.
cpu/beta_cpu/alpha_full_cpu.cc:
    AlphaFullCPU has been changed to be templated, so this CC file is now
    just used to force instantation of AlphaFullCPU.
cpu/beta_cpu/alpha_full_cpu.hh:
    Change AlphaFullCPU to be templated on Impl.
cpu/beta_cpu/alpha_impl.hh:
    Update it to reflect AlphaDynInst and AlphaFullCPU being templated
    on Impl.  Also removed time buffers from here, as they are really
    a part of the CPU and are thus in the CPU policy now.
cpu/beta_cpu/alpha_params.hh:
    Make AlphaSimpleParams inherit from the BaseFullCPU so that it doesn't
    need to specifically declare any parameters that are already in the
    BaseFullCPU.
cpu/beta_cpu/comm.hh:
    Changed the structure of the time buffer communication structs.  Now
    they include the size of the packet of instructions it is sending.
    Added some parameters to the backwards communication struct, mainly
    for squashing.
cpu/beta_cpu/commit.hh:
    Update typenames to reflect change in location of time buffer structs.
    Update DynInst to DynInstPtr (it is refcounted now).
cpu/beta_cpu/commit_impl.hh:
    Formatting changes mainly.  Also sends back proper information
    on branch mispredicts so that the bpred unit can update itself.
    Updated behavior for non-speculative instructions (stores, any
    other non-spec instructions): once they reach the head of the ROB,
    the ROB signals back to the IQ that it can go ahead and issue the
    non-speculative instruction.  The instruction itself is updated so that
    commit won't try to commit it again until it is done executing.
cpu/beta_cpu/cpu_policy.hh:
    Added branch prediction unit, mem dependence prediction unit, load
    store queue.  Moved time buffer structs from AlphaSimpleImpl to here.
cpu/beta_cpu/decode.hh:
    Changed typedefs to reflect change in location of time buffer structs
    and also the change from DynInst to ref counted DynInstPtr.
cpu/beta_cpu/decode_impl.hh:
    Continues to buffer instructions even while unblocking now.  Changed
    how it loops through groups of instructions so it can properly block
    during the middle of a group of instructions.
cpu/beta_cpu/fetch.hh:
    Changed typedefs to reflect change in location of time buffer structs
    and the change to ref counted DynInsts.  Also added in branch
    brediction unit.
cpu/beta_cpu/fetch_impl.hh:
    Add in branch prediction.  Changed how fetch checks inputs and its
    current state to make for easier logic.
cpu/beta_cpu/free_list.cc:
    Changed int regs and float regs to logically use one flat namespace.
    Future change will be moving them to a single scoreboard to conserve
    space.
cpu/beta_cpu/free_list.hh:
    Mostly debugging statements.  Might be removed for performance in future.
cpu/beta_cpu/full_cpu.cc:
    Added in some debugging statements.  Updated BaseFullCPU to take
    a params object.
cpu/beta_cpu/full_cpu.hh:
    Added params class within BaseCPU that other param classes will be
    able to inherit from.  Updated typedefs to reflect change in location
    of time buffer structs and ref counted DynInst.
cpu/beta_cpu/iew.hh:
    Updated typedefs to reflect change in location of time buffer structs
    and use of ref counted DynInsts.
cpu/beta_cpu/iew_impl.hh:
    Added in load store queue, updated iew to be able to execute non-
    speculative instructions, instead of having them execute in commit.
cpu/beta_cpu/inst_queue.hh:
    Updated change to ref counted DynInsts.  Changed inst queue to hold
    non-speculative instructions as well, which are issued only when
    commit signals backwards that a nonspeculative instruction is at
    the head of the ROB.
cpu/beta_cpu/inst_queue_impl.hh:
    Updated to allow for non-speculative instructions to be in the inst
    queue.  Also added some debug functions.
cpu/beta_cpu/regfile.hh:
    Added debugging statements, changed formatting.
cpu/beta_cpu/rename.hh:
    Updated typedefs, added some functions to clean up code.
cpu/beta_cpu/rename_impl.hh:
    Moved some code into functions to make it easier to read.
cpu/beta_cpu/rename_map.cc:
    Changed int and float reg behavior to use a single flat namespace.  In
    the future, the rename maps can be combined to a single rename map to
    save space.
cpu/beta_cpu/rename_map.hh:
    Added destructor.
cpu/beta_cpu/rob.hh:
    Updated it with change from DynInst to ref counted DynInst.
cpu/beta_cpu/rob_impl.hh:
    Formatting, updated to use ref counted DynInst.
cpu/static_inst.hh:
    Updated forward declaration for AlphaDynInst now that it is templated.

--HG--
extra : convert_revision : 1045f240ee9b6a4bd368e1806aca029ebbdc6dd3
2004-09-23 14:06:03 -04:00

289 lines
7.3 KiB
C++

#ifndef __ROB_IMPL_HH__
#define __ROB_IMPL_HH__
#include "cpu/beta_cpu/rob.hh"
template <class Impl>
ROB<Impl>::ROB(unsigned _numEntries, unsigned _squashWidth)
: numEntries(_numEntries),
squashWidth(_squashWidth),
numInstsInROB(0),
squashedSeqNum(0)
{
doneSquashing = true;
}
template <class Impl>
void
ROB<Impl>::setCPU(FullCPU *cpu_ptr)
{
cpu = cpu_ptr;
// Set the tail to the beginning of the CPU instruction list so that
// upon the first instruction being inserted into the ROB, the tail
// iterator can simply be incremented.
tail = cpu->instList.begin();
// Set the squash iterator to the end of the instruction list.
squashIt = cpu->instList.end();
}
template <class Impl>
int
ROB<Impl>::countInsts()
{
// Start at 1; if the tail matches cpu->instList.begin(), then there is
// one inst in the ROB.
int return_val = 1;
// There are quite a few special cases. Do not use this function other
// than for debugging purposes.
if (cpu->instList.begin() == cpu->instList.end()) {
// In this case there are no instructions in the list. The ROB
// must be empty.
return 0;
} else if (tail == cpu->instList.end()) {
// In this case, the tail is not yet pointing to anything valid.
// The ROB must be empty.
return 0;
}
// Iterate through the ROB from the head to the tail, counting the
// entries.
for (InstIt_t i = cpu->instList.begin(); i != tail; ++i)
{
assert(i != cpu->instList.end());
++return_val;
}
return return_val;
// Because the head won't be tracked properly until the ROB gets the
// first instruction, and any time that the ROB is empty and has not
// yet gotten the instruction, this function doesn't work.
// return numInstsInROB;
}
template <class Impl>
void
ROB<Impl>::insertInst(DynInstPtr &inst)
{
// Make sure we have the right number of instructions.
assert(numInstsInROB == countInsts());
// Make sure the instruction is valid.
assert(inst);
DPRINTF(ROB, "ROB: Adding inst PC %#x to the ROB.\n", inst->readPC());
// If the ROB is full then exit.
assert(numInstsInROB != numEntries);
++numInstsInROB;
// Increment the tail iterator, moving it one instruction back.
// There is a special case if the ROB was empty prior to this insertion,
// in which case the tail will be pointing at instList.end(). If that
// happens, then reset the tail to the beginning of the list.
if (tail != cpu->instList.end()) {
++tail;
} else {
tail = cpu->instList.begin();
}
// Make sure the tail iterator is actually pointing at the instruction
// added.
assert((*tail) == inst);
DPRINTF(ROB, "ROB: Now has %d instructions.\n", numInstsInROB);
}
// Whatever calls this function needs to ensure that it properly frees up
// registers prior to this function.
template <class Impl>
void
ROB<Impl>::retireHead()
{
assert(numInstsInROB == countInsts());
assert(numInstsInROB > 0);
DynInstPtr head_inst;
// Get the head ROB instruction.
head_inst = cpu->instList.front();
// Make certain this can retire.
assert(head_inst->readyToCommit());
DPRINTF(ROB, "ROB: Retiring head instruction of the ROB, "
"instruction PC %#x, seq num %i\n", head_inst->readPC(),
head_inst->seqNum);
// Keep track of how many instructions are in the ROB.
--numInstsInROB;
// Tell CPU to remove the instruction from the list of instructions.
// A special case is needed if the instruction being retired is the
// only instruction in the ROB; otherwise the tail iterator will become
// invalidated.
if (tail == cpu->instList.begin()) {
cpu->removeFrontInst(head_inst);
tail = cpu->instList.end();
} else {
cpu->removeFrontInst(head_inst);
}
}
template <class Impl>
bool
ROB<Impl>::isHeadReady()
{
if (numInstsInROB != 0) {
DynInstPtr head_inst = cpu->instList.front();
return head_inst->readyToCommit();
}
return false;
}
template <class Impl>
unsigned
ROB<Impl>::numFreeEntries()
{
assert(numInstsInROB == countInsts());
return numEntries - numInstsInROB;
}
template <class Impl>
void
ROB<Impl>::doSquash()
{
DPRINTF(ROB, "ROB: Squashing instructions.\n");
assert(squashIt != cpu->instList.end());
for (int numSquashed = 0;
numSquashed < squashWidth && (*squashIt)->seqNum != squashedSeqNum;
++numSquashed)
{
// Ensure that the instruction is younger.
assert((*squashIt)->seqNum > squashedSeqNum);
DPRINTF(ROB, "ROB: Squashing instruction PC %#x, seq num %i.\n",
(*squashIt)->readPC(), (*squashIt)->seqNum);
// Mark the instruction as squashed, and ready to commit so that
// it can drain out of the pipeline.
(*squashIt)->setSquashed();
(*squashIt)->setCanCommit();
// Special case for when squashing due to a syscall. It's possible
// that the squash happened after the head instruction was already
// committed, meaning that (*squashIt)->seqNum != squashedSeqNum
// will never be false. Normally the squash would never be able
// to go past the head of the ROB; in this case it might, so it
// must be handled otherwise it will segfault.
#ifndef FULL_SYSTEM
if (squashIt == cpu->instList.begin()) {
DPRINTF(ROB, "ROB: Reached head of instruction list while "
"squashing.\n");
squashIt = cpu->instList.end();
doneSquashing = true;
return;
}
#endif
// Move the tail iterator to the next instruction.
squashIt--;
}
// Check if ROB is done squashing.
if ((*squashIt)->seqNum == squashedSeqNum) {
DPRINTF(ROB, "ROB: Done squashing instructions.\n");
squashIt = cpu->instList.end();
doneSquashing = true;
}
}
template <class Impl>
void
ROB<Impl>::squash(InstSeqNum squash_num)
{
DPRINTF(ROB, "ROB: Starting to squash within the ROB.\n");
doneSquashing = false;
squashedSeqNum = squash_num;
assert(tail != cpu->instList.end());
squashIt = tail;
doSquash();
}
template <class Impl>
uint64_t
ROB<Impl>::readHeadPC()
{
assert(numInstsInROB == countInsts());
DynInstPtr head_inst = cpu->instList.front();
return head_inst->readPC();
}
template <class Impl>
uint64_t
ROB<Impl>::readHeadNextPC()
{
assert(numInstsInROB == countInsts());
DynInstPtr head_inst = cpu->instList.front();
return head_inst->readNextPC();
}
template <class Impl>
InstSeqNum
ROB<Impl>::readHeadSeqNum()
{
// Return the last sequence number that has not been squashed. Other
// stages can use it to squash any instructions younger than the current
// tail.
DynInstPtr head_inst = cpu->instList.front();
return head_inst->seqNum;
}
template <class Impl>
uint64_t
ROB<Impl>::readTailPC()
{
assert(numInstsInROB == countInsts());
assert(tail != cpu->instList.end());
return (*tail)->readPC();
}
template <class Impl>
InstSeqNum
ROB<Impl>::readTailSeqNum()
{
// Return the last sequence number that has not been squashed. Other
// stages can use it to squash any instructions younger than the current
// tail.
return (*tail)->seqNum;
}
#endif // __ROB_IMPL_HH__