gem5/src/cpu/base_dyn_inst_impl.hh
Giacomo Gabrielli e2507407b1 O3: Enhance data address translation by supporting hardware page table walkers.
Some ISAs (like ARM) relies on hardware page table walkers.  For those ISAs,
when a TLB miss occurs, initiateTranslation() can return with NoFault but with
the translation unfinished.

Instructions experiencing a delayed translation due to a hardware page table
walk are deferred until the translation completes and kept into the IQ.  In
order to keep track of them, the IQ has been augmented with a queue of the
outstanding delayed memory instructions.  When their translation completes,
instructions are re-executed (only their initiateAccess() was already
executed; their DTB translation is now skipped).  The IEW stage has been
modified to support such a 2-pass execution.
2011-02-11 18:29:35 -06:00

267 lines
6.6 KiB
C++

/*
* Copyright (c) 2011 ARM Limited
* All rights reserved.
*
* The license below extends only to copyright in the software and shall
* not be construed as granting a license to any other intellectual
* property including but not limited to intellectual property relating
* to a hardware implementation of the functionality of the software
* licensed hereunder. You may use the software subject to the license
* terms below provided that you ensure that this notice is replicated
* unmodified and in its entirety in all distributions of the software,
* modified or unmodified, in source code or in binary form.
*
* Copyright (c) 2004-2006 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
*/
#include <iostream>
#include <set>
#include <string>
#include <sstream>
#include "base/cprintf.hh"
#include "base/trace.hh"
#include "config/the_isa.hh"
#include "cpu/base_dyn_inst.hh"
#include "cpu/exetrace.hh"
#include "mem/request.hh"
#include "sim/faults.hh"
#define NOHASH
#ifndef NOHASH
#include "base/hashmap.hh"
unsigned int MyHashFunc(const BaseDynInst *addr)
{
unsigned a = (unsigned)addr;
unsigned hash = (((a >> 14) ^ ((a >> 2) & 0xffff))) & 0x7FFFFFFF;
return hash;
}
typedef m5::hash_map<const BaseDynInst *, const BaseDynInst *, MyHashFunc>
my_hash_t;
my_hash_t thishash;
#endif
template <class Impl>
BaseDynInst<Impl>::BaseDynInst(StaticInstPtr _staticInst,
TheISA::PCState _pc, TheISA::PCState _predPC,
InstSeqNum seq_num, ImplCPU *cpu)
: staticInst(_staticInst), traceData(NULL), cpu(cpu)
{
seqNum = seq_num;
pc = _pc;
predPC = _predPC;
predTaken = false;
initVars();
}
template <class Impl>
BaseDynInst<Impl>::BaseDynInst(TheISA::ExtMachInst inst,
TheISA::PCState _pc, TheISA::PCState _predPC,
InstSeqNum seq_num, ImplCPU *cpu)
: staticInst(inst, _pc.instAddr()), traceData(NULL), cpu(cpu)
{
seqNum = seq_num;
pc = _pc;
predPC = _predPC;
predTaken = false;
initVars();
}
template <class Impl>
BaseDynInst<Impl>::BaseDynInst(StaticInstPtr &_staticInst)
: staticInst(_staticInst), traceData(NULL)
{
seqNum = 0;
initVars();
}
template <class Impl>
void
BaseDynInst<Impl>::initVars()
{
memData = NULL;
effAddr = 0;
effAddrValid = false;
physEffAddr = 0;
translationStarted = false;
translationCompleted = false;
isUncacheable = false;
reqMade = false;
readyRegs = 0;
instResult.integer = 0;
recordResult = true;
status.reset();
eaCalcDone = false;
memOpDone = false;
predicate = true;
lqIdx = -1;
sqIdx = -1;
// Eventually make this a parameter.
threadNumber = 0;
// Also make this a parameter, or perhaps get it from xc or cpu.
asid = 0;
// Initialize the fault to be NoFault.
fault = NoFault;
#ifndef NDEBUG
++cpu->instcount;
if (cpu->instcount > 1500) {
#ifdef DEBUG
cpu->dumpInsts();
dumpSNList();
#endif
assert(cpu->instcount <= 1500);
}
DPRINTF(DynInst,
"DynInst: [sn:%lli] Instruction created. Instcount for %s = %i\n",
seqNum, cpu->name(), cpu->instcount);
#endif
#ifdef DEBUG
cpu->snList.insert(seqNum);
#endif
}
template <class Impl>
BaseDynInst<Impl>::~BaseDynInst()
{
if (memData) {
delete [] memData;
}
if (traceData) {
delete traceData;
}
fault = NoFault;
#ifndef NDEBUG
--cpu->instcount;
DPRINTF(DynInst,
"DynInst: [sn:%lli] Instruction destroyed. Instcount for %s = %i\n",
seqNum, cpu->name(), cpu->instcount);
#endif
#ifdef DEBUG
cpu->snList.erase(seqNum);
#endif
}
#ifdef DEBUG
template <class Impl>
void
BaseDynInst<Impl>::dumpSNList()
{
std::set<InstSeqNum>::iterator sn_it = cpu->snList.begin();
int count = 0;
while (sn_it != cpu->snList.end()) {
cprintf("%i: [sn:%lli] not destroyed\n", count, (*sn_it));
count++;
sn_it++;
}
}
#endif
template <class Impl>
void
BaseDynInst<Impl>::dump()
{
cprintf("T%d : %#08d `", threadNumber, pc.instAddr());
std::cout << staticInst->disassemble(pc.instAddr());
cprintf("'\n");
}
template <class Impl>
void
BaseDynInst<Impl>::dump(std::string &outstring)
{
std::ostringstream s;
s << "T" << threadNumber << " : 0x" << pc.instAddr() << " "
<< staticInst->disassemble(pc.instAddr());
outstring = s.str();
}
template <class Impl>
void
BaseDynInst<Impl>::markSrcRegReady()
{
DPRINTF(IQ, "[sn:%lli] has %d ready out of %d sources. RTI %d)\n",
seqNum, readyRegs+1, numSrcRegs(), readyToIssue());
if (++readyRegs == numSrcRegs()) {
setCanIssue();
}
}
template <class Impl>
void
BaseDynInst<Impl>::markSrcRegReady(RegIndex src_idx)
{
_readySrcRegIdx[src_idx] = true;
markSrcRegReady();
}
template <class Impl>
bool
BaseDynInst<Impl>::eaSrcsReady()
{
// For now I am assuming that src registers 1..n-1 are the ones that the
// EA calc depends on. (i.e. src reg 0 is the source of the data to be
// stored)
for (int i = 1; i < numSrcRegs(); ++i) {
if (!_readySrcRegIdx[i])
return false;
}
return true;
}