gem5/src/cpu/ozone/lw_lsq.hh
Ali Saidi 1b370431d0 sim: Remove FastAlloc
While FastAlloc provides a small performance increase (~1.5%) over regular malloc it isn't thread safe.
After removing FastAlloc and using tcmalloc I've seen a performance increase of 12% over libc malloc
when running twolf for ARM.
2012-06-05 01:23:08 -04:00

698 lines
21 KiB
C++

/*
* Copyright (c) 2006 The Regents of The University of Michigan
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions are
* met: redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer;
* redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution;
* neither the name of the copyright holders nor the names of its
* contributors may be used to endorse or promote products derived from
* this software without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
* "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
* LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
* A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
* OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
* SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
* LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
* DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
* THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
* OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*
* Authors: Kevin Lim
*/
#ifndef __CPU_OZONE_LW_LSQ_HH__
#define __CPU_OZONE_LW_LSQ_HH__
#include <algorithm>
#include <list>
#include <map>
#include <queue>
#include "arch/types.hh"
#include "base/hashmap.hh"
#include "config/the_isa.hh"
#include "cpu/inst_seq.hh"
#include "mem/packet.hh"
#include "mem/port.hh"
//#include "mem/page_table.hh"
#include "sim/debug.hh"
#include "sim/fault_fwd.hh"
#include "sim/sim_object.hh"
class MemObject;
/**
* Class that implements the actual LQ and SQ for each specific thread.
* Both are circular queues; load entries are freed upon committing, while
* store entries are freed once they writeback. The LSQUnit tracks if there
* are memory ordering violations, and also detects partial load to store
* forwarding cases (a store only has part of a load's data) that requires
* the load to wait until the store writes back. In the former case it
* holds onto the instruction until the dependence unit looks at it, and
* in the latter it stalls the LSQ until the store writes back. At that
* point the load is replayed.
*/
template <class Impl>
class OzoneLWLSQ {
public:
typedef typename Impl::Params Params;
typedef typename Impl::OzoneCPU OzoneCPU;
typedef typename Impl::BackEnd BackEnd;
typedef typename Impl::DynInstPtr DynInstPtr;
typedef typename Impl::IssueStruct IssueStruct;
typedef TheISA::IntReg IntReg;
typedef typename std::map<InstSeqNum, DynInstPtr>::iterator LdMapIt;
public:
/** Constructs an LSQ unit. init() must be called prior to use. */
OzoneLWLSQ();
/** Initializes the LSQ unit with the specified number of entries. */
void init(Params *params, unsigned maxLQEntries,
unsigned maxSQEntries, unsigned id);
/** Returns the name of the LSQ unit. */
std::string name() const;
void regStats();
/** Sets the CPU pointer. */
void setCPU(OzoneCPU *cpu_ptr);
/** Sets the back-end stage pointer. */
void setBE(BackEnd *be_ptr)
{ be = be_ptr; }
Port *getDcachePort() { return &dcachePort; }
/** Ticks the LSQ unit, which in this case only resets the number of
* used cache ports.
* @todo: Move the number of used ports up to the LSQ level so it can
* be shared by all LSQ units.
*/
void tick() { usedPorts = 0; }
/** Inserts an instruction. */
void insert(DynInstPtr &inst);
/** Inserts a load instruction. */
void insertLoad(DynInstPtr &load_inst);
/** Inserts a store instruction. */
void insertStore(DynInstPtr &store_inst);
/** Executes a load instruction. */
Fault executeLoad(DynInstPtr &inst);
/** Executes a store instruction. */
Fault executeStore(DynInstPtr &inst);
/** Commits the head load. */
void commitLoad();
/** Commits loads older than a specific sequence number. */
void commitLoads(InstSeqNum &youngest_inst);
/** Commits stores older than a specific sequence number. */
void commitStores(InstSeqNum &youngest_inst);
/** Writes back stores. */
void writebackStores();
/** Completes the data access that has been returned from the
* memory system. */
void completeDataAccess(PacketPtr pkt);
// @todo: Include stats in the LSQ unit.
//void regStats();
/** Clears all the entries in the LQ. */
void clearLQ();
/** Clears all the entries in the SQ. */
void clearSQ();
/** Resizes the LQ to a given size. */
void resizeLQ(unsigned size);
/** Resizes the SQ to a given size. */
void resizeSQ(unsigned size);
/** Squashes all instructions younger than a specific sequence number. */
void squash(const InstSeqNum &squashed_num);
/** Returns if there is a memory ordering violation. Value is reset upon
* call to getMemDepViolator().
*/
bool violation() { return memDepViolator; }
/** Returns the memory ordering violator. */
DynInstPtr getMemDepViolator();
/** Returns if a load became blocked due to the memory system. It clears
* the bool's value upon this being called.
*/
bool loadBlocked()
{ return isLoadBlocked; }
void clearLoadBlocked()
{ isLoadBlocked = false; }
bool isLoadBlockedHandled()
{ return loadBlockedHandled; }
void setLoadBlockedHandled()
{ loadBlockedHandled = true; }
/** Returns the number of free entries (min of free LQ and SQ entries). */
unsigned numFreeEntries();
/** Returns the number of loads ready to execute. */
int numLoadsReady();
/** Returns the number of loads in the LQ. */
int numLoads() { return loads; }
/** Returns the number of stores in the SQ. */
int numStores() { return stores + storesInFlight; }
/** Returns if either the LQ or SQ is full. */
bool isFull() { return lqFull() || sqFull(); }
/** Returns if the LQ is full. */
bool lqFull() { return loads >= (LQEntries - 1); }
/** Returns if the SQ is full. */
bool sqFull() { return (stores + storesInFlight) >= (SQEntries - 1); }
/** Debugging function to dump instructions in the LSQ. */
void dumpInsts();
/** Returns the number of instructions in the LSQ. */
unsigned getCount() { return loads + stores; }
/** Returns if there are any stores to writeback. */
bool hasStoresToWB() { return storesToWB; }
/** Returns the number of stores to writeback. */
int numStoresToWB() { return storesToWB; }
/** Returns if the LSQ unit will writeback on this cycle. */
bool willWB() { return storeQueue.back().canWB &&
!storeQueue.back().completed &&
!isStoreBlocked; }
void switchOut();
void takeOverFrom(ThreadContext *old_tc = NULL);
bool isSwitchedOut() { return switchedOut; }
bool switchedOut;
private:
/** Writes back the instruction, sending it to IEW. */
void writeback(DynInstPtr &inst, PacketPtr pkt);
/** Handles completing the send of a store to memory. */
void storePostSend(PacketPtr pkt, DynInstPtr &inst);
/** Completes the store at the specified index. */
void completeStore(DynInstPtr &inst);
void removeStore(int store_idx);
/** Handles doing the retry. */
void recvRetry();
private:
/** Pointer to the CPU. */
OzoneCPU *cpu;
/** Pointer to the back-end stage. */
BackEnd *be;
class DcachePort : public MasterPort
{
protected:
OzoneLWLSQ *lsq;
public:
DcachePort(OzoneLWLSQ *_lsq)
: lsq(_lsq)
{ }
protected:
virtual Tick recvAtomic(PacketPtr pkt);
virtual void recvFunctional(PacketPtr pkt);
/**
* Is a snooper due to LSQ maintenance
*/
virtual bool isSnooping() const { return true; }
virtual bool recvTiming(PacketPtr pkt);
virtual void recvRetry();
};
/** D-cache port. */
DcachePort dcachePort;
public:
struct SQEntry {
/** Constructs an empty store queue entry. */
SQEntry()
: inst(NULL), req(NULL), size(0), data(0),
canWB(0), committed(0), completed(0), lqIt(NULL)
{ }
/** Constructs a store queue entry for a given instruction. */
SQEntry(DynInstPtr &_inst)
: inst(_inst), req(NULL), size(0), data(0),
canWB(0), committed(0), completed(0), lqIt(NULL)
{ }
/** The store instruction. */
DynInstPtr inst;
/** The memory request for the store. */
RequestPtr req;
/** The size of the store. */
int size;
/** The store data. */
IntReg data;
/** Whether or not the store can writeback. */
bool canWB;
/** Whether or not the store is committed. */
bool committed;
/** Whether or not the store is completed. */
bool completed;
typename std::list<DynInstPtr>::iterator lqIt;
};
/** Derived class to hold any sender state the LSQ needs. */
class LSQSenderState : public Packet::SenderState
{
public:
/** Default constructor. */
LSQSenderState()
: noWB(false)
{ }
/** Instruction who initiated the access to memory. */
DynInstPtr inst;
/** Whether or not it is a load. */
bool isLoad;
/** The LQ/SQ index of the instruction. */
int idx;
/** Whether or not the instruction will need to writeback. */
bool noWB;
};
/** Writeback event, specifically for when stores forward data to loads. */
class WritebackEvent : public Event {
public:
/** Constructs a writeback event. */
WritebackEvent(DynInstPtr &_inst, PacketPtr pkt, OzoneLWLSQ *lsq_ptr);
/** Processes the writeback event. */
void process();
/** Returns the description of this event. */
const char *description() const;
private:
/** Instruction whose results are being written back. */
DynInstPtr inst;
/** The packet that would have been sent to memory. */
PacketPtr pkt;
/** The pointer to the LSQ unit that issued the store. */
OzoneLWLSQ<Impl> *lsqPtr;
};
enum Status {
Running,
Idle,
DcacheMissStall,
DcacheMissSwitch
};
private:
/** The OzoneLWLSQ thread id. */
unsigned lsqID;
/** The status of the LSQ unit. */
Status _status;
/** The store queue. */
std::list<SQEntry> storeQueue;
/** The load queue. */
std::list<DynInstPtr> loadQueue;
typedef typename std::list<SQEntry>::iterator SQIt;
typedef typename std::list<DynInstPtr>::iterator LQIt;
struct HashFn {
size_t operator() (const int a) const
{
unsigned hash = (((a >> 14) ^ ((a >> 2) & 0xffff))) & 0x7FFFFFFF;
return hash;
}
};
m5::hash_map<int, SQIt, HashFn> SQItHash;
std::queue<int> SQIndices;
m5::hash_map<int, LQIt, HashFn> LQItHash;
std::queue<int> LQIndices;
typedef typename m5::hash_map<int, LQIt, HashFn>::iterator LQHashIt;
typedef typename m5::hash_map<int, SQIt, HashFn>::iterator SQHashIt;
// Consider making these 16 bits
/** The number of LQ entries. */
unsigned LQEntries;
/** The number of SQ entries. */
unsigned SQEntries;
/** The number of load instructions in the LQ. */
int loads;
/** The number of store instructions in the SQ (excludes those waiting to
* writeback).
*/
int stores;
int storesToWB;
public:
int storesInFlight;
private:
/// @todo Consider moving to a more advanced model with write vs read ports
/** The number of cache ports available each cycle. */
int cachePorts;
/** The number of used cache ports in this cycle. */
int usedPorts;
//list<InstSeqNum> mshrSeqNums;
/** Tota number of memory ordering violations. */
Stats::Scalar lsqMemOrderViolation;
//Stats::Scalar dcacheStallCycles;
Counter lastDcacheStall;
// Make these per thread?
/** Whether or not the LSQ is stalled. */
bool stalled;
/** The store that causes the stall due to partial store to load
* forwarding.
*/
InstSeqNum stallingStoreIsn;
/** The index of the above store. */
LQIt stallingLoad;
/** The packet that needs to be retried. */
PacketPtr retryPkt;
/** Whehter or not a store is blocked due to the memory system. */
bool isStoreBlocked;
/** Whether or not a load is blocked due to the memory system. It is
* cleared when this value is checked via loadBlocked().
*/
bool isLoadBlocked;
bool loadBlockedHandled;
InstSeqNum blockedLoadSeqNum;
/** The oldest faulting load instruction. */
DynInstPtr loadFaultInst;
/** The oldest faulting store instruction. */
DynInstPtr storeFaultInst;
/** The oldest load that caused a memory ordering violation. */
DynInstPtr memDepViolator;
// Will also need how many read/write ports the Dcache has. Or keep track
// of that in stage that is one level up, and only call executeLoad/Store
// the appropriate number of times.
public:
/** Executes the load at the given index. */
template <class T>
Fault read(RequestPtr req, T &data, int load_idx);
/** Executes the store at the given index. */
template <class T>
Fault write(RequestPtr req, T &data, int store_idx);
/** Returns the sequence number of the head load instruction. */
InstSeqNum getLoadHeadSeqNum()
{
if (!loadQueue.empty()) {
return loadQueue.back()->seqNum;
} else {
return 0;
}
}
/** Returns the sequence number of the head store instruction. */
InstSeqNum getStoreHeadSeqNum()
{
if (!storeQueue.empty()) {
return storeQueue.back().inst->seqNum;
} else {
return 0;
}
}
/** Returns whether or not the LSQ unit is stalled. */
bool isStalled() { return stalled; }
};
template <class Impl>
template <class T>
Fault
OzoneLWLSQ<Impl>::read(RequestPtr req, T &data, int load_idx)
{
//Depending on issue2execute delay a squashed load could
//execute if it is found to be squashed in the same
//cycle it is scheduled to execute
typename m5::hash_map<int, LQIt, HashFn>::iterator
lq_hash_it = LQItHash.find(load_idx);
assert(lq_hash_it != LQItHash.end());
DynInstPtr inst = (*(*lq_hash_it).second);
// Make sure this isn't an uncacheable access
// A bit of a hackish way to get uncached accesses to work only if they're
// at the head of the LSQ and are ready to commit (at the head of the ROB
// too).
// @todo: Fix uncached accesses.
if (req->isUncacheable() &&
(inst != loadQueue.back() || !inst->isAtCommit())) {
DPRINTF(OzoneLSQ, "[sn:%lli] Uncached load and not head of "
"commit/LSQ!\n",
inst->seqNum);
be->rescheduleMemInst(inst);
return TheISA::genMachineCheckFault();
}
// Check the SQ for any previous stores that might lead to forwarding
SQIt sq_it = storeQueue.begin();
int store_size = 0;
DPRINTF(OzoneLSQ, "Read called, load idx: %i addr: %#x\n",
load_idx, req->getPaddr());
while (sq_it != storeQueue.end() && (*sq_it).inst->seqNum > inst->seqNum)
++sq_it;
while (1) {
// End once we've reached the top of the LSQ
if (sq_it == storeQueue.end()) {
break;
}
assert((*sq_it).inst);
store_size = (*sq_it).size;
if (store_size == 0 || (*sq_it).committed) {
sq_it++;
continue;
}
// Check if the store data is within the lower and upper bounds of
// addresses that the request needs.
bool store_has_lower_limit =
req->getVaddr() >= (*sq_it).inst->effAddr;
bool store_has_upper_limit =
(req->getVaddr() + req->getSize()) <= ((*sq_it).inst->effAddr +
store_size);
bool lower_load_has_store_part =
req->getVaddr() < ((*sq_it).inst->effAddr +
store_size);
bool upper_load_has_store_part =
(req->getVaddr() + req->getSize()) > (*sq_it).inst->effAddr;
// If the store's data has all of the data needed, we can forward.
if (store_has_lower_limit && store_has_upper_limit) {
int shift_amt = req->getVaddr() & (store_size - 1);
// Assumes byte addressing
shift_amt = shift_amt << 3;
// Cast this to type T?
data = (*sq_it).data >> shift_amt;
assert(!inst->memData);
inst->memData = new uint8_t[64];
memcpy(inst->memData, &data, req->getSize());
DPRINTF(OzoneLSQ, "Forwarding from store [sn:%lli] to load to "
"[sn:%lli] addr %#x, data %#x\n",
(*sq_it).inst->seqNum, inst->seqNum, req->getVaddr(),
*(inst->memData));
PacketPtr data_pkt = new Packet(req, Packet::ReadReq);
data_pkt->dataStatic(inst->memData);
WritebackEvent *wb = new WritebackEvent(inst, data_pkt, this);
// We'll say this has a 1 cycle load-store forwarding latency
// for now.
// @todo: Need to make this a parameter.
wb->schedule(curTick());
// Should keep track of stat for forwarded data
return NoFault;
} else if ((store_has_lower_limit && lower_load_has_store_part) ||
(store_has_upper_limit && upper_load_has_store_part) ||
(lower_load_has_store_part && upper_load_has_store_part)) {
// This is the partial store-load forwarding case where a store
// has only part of the load's data.
// If it's already been written back, then don't worry about
// stalling on it.
if ((*sq_it).completed) {
sq_it++;
break;
}
// Must stall load and force it to retry, so long as it's the oldest
// load that needs to do so.
if (!stalled ||
(stalled &&
inst->seqNum <
(*stallingLoad)->seqNum)) {
stalled = true;
stallingStoreIsn = (*sq_it).inst->seqNum;
stallingLoad = (*lq_hash_it).second;
}
// Tell IQ/mem dep unit that this instruction will need to be
// rescheduled eventually
be->rescheduleMemInst(inst);
DPRINTF(OzoneLSQ, "Load-store forwarding mis-match. "
"Store [sn:%lli] to load addr %#x\n",
(*sq_it).inst->seqNum, req->getVaddr());
return NoFault;
}
sq_it++;
}
// If there's no forwarding case, then go access memory
DPRINTF(OzoneLSQ, "Doing functional access for inst PC %#x\n",
inst->readPC());
assert(!inst->memData);
inst->memData = new uint8_t[64];
++usedPorts;
DPRINTF(OzoneLSQ, "Doing timing access for inst PC %#x\n",
inst->readPC());
PacketPtr data_pkt =
new Packet(req,
(req->isLLSC() ?
MemCmd::LoadLockedReq : Packet::ReadReq));
data_pkt->dataStatic(inst->memData);
LSQSenderState *state = new LSQSenderState;
state->isLoad = true;
state->idx = load_idx;
state->inst = inst;
data_pkt->senderState = state;
// if we have a cache, do cache access too
if (!dcachePort.sendTiming(data_pkt)) {
// There's an older load that's already going to squash.
if (isLoadBlocked && blockedLoadSeqNum < inst->seqNum)
return NoFault;
// Record that the load was blocked due to memory. This
// load will squash all instructions after it, be
// refetched, and re-executed.
isLoadBlocked = true;
loadBlockedHandled = false;
blockedLoadSeqNum = inst->seqNum;
// No fault occurred, even though the interface is blocked.
return NoFault;
}
if (req->isLLSC()) {
cpu->lockFlag = true;
}
return NoFault;
}
template <class Impl>
template <class T>
Fault
OzoneLWLSQ<Impl>::write(RequestPtr req, T &data, int store_idx)
{
SQHashIt sq_hash_it = SQItHash.find(store_idx);
assert(sq_hash_it != SQItHash.end());
SQIt sq_it = (*sq_hash_it).second;
assert((*sq_it).inst);
DPRINTF(OzoneLSQ, "Doing write to store idx %i, addr %#x data %#x"
" | [sn:%lli]\n",
store_idx, req->getPaddr(), data, (*sq_it).inst->seqNum);
(*sq_it).req = req;
(*sq_it).size = sizeof(T);
(*sq_it).data = data;
/*
assert(!req->data);
req->data = new uint8_t[64];
memcpy(req->data, (uint8_t *)&(*sq_it).data, req->size);
*/
// This function only writes the data to the store queue, so no fault
// can happen here.
return NoFault;
}
#endif // __CPU_OZONE_LW_LSQ_HH__