summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
-rw-r--r--configs/splash2/cluster.py303
-rw-r--r--configs/splash2/run.py2
-rw-r--r--src/cpu/memtest/memtest.cc10
-rw-r--r--src/cpu/memtest/memtest.hh4
-rw-r--r--src/cpu/o3/fetch.hh2
-rw-r--r--src/cpu/o3/fetch_impl.hh9
-rw-r--r--src/cpu/o3/lsq.hh2
-rw-r--r--src/cpu/o3/lsq_impl.hh10
-rw-r--r--src/cpu/simple/atomic.cc12
-rw-r--r--src/cpu/simple/atomic.hh2
-rw-r--r--src/cpu/simple/timing.cc11
-rw-r--r--src/cpu/simple/timing.hh4
-rw-r--r--src/mem/bus.cc43
-rw-r--r--src/mem/cache/base_cache.hh7
-rw-r--r--src/mem/cache/miss/miss_queue.cc2
15 files changed, 401 insertions, 22 deletions
diff --git a/configs/splash2/cluster.py b/configs/splash2/cluster.py
new file mode 100644
index 000000000..799b85e6c
--- /dev/null
+++ b/configs/splash2/cluster.py
@@ -0,0 +1,303 @@
+# 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: Ron Dreslinski
+
+# Simple test script
+#
+# "m5 test.py"
+
+import m5
+from m5.objects import *
+import os, optparse, sys
+m5.AddToPath('../common')
+
+# --------------------
+# Define Command Line Options
+# ====================
+
+parser = optparse.OptionParser()
+
+parser.add_option("-d", "--detailed", action="store_true")
+parser.add_option("-t", "--timing", action="store_true")
+parser.add_option("-m", "--maxtick", type="int")
+parser.add_option("-c", "--numclusters",
+ help="Number of clusters", type="int")
+parser.add_option("-n", "--numcpus",
+ help="Number of cpus in total", type="int")
+parser.add_option("-f", "--frequency",
+ default = "1GHz",
+ help="Frequency of each CPU")
+parser.add_option("-p", "--protocol",
+ default="moesi",
+ help="The coherence protocol to use for the L1'a (i.e. MOESI, MOSI)")
+parser.add_option("--l1size",
+ default = "32kB")
+parser.add_option("--l1latency",
+ default = 1)
+parser.add_option("--l2size",
+ default = "256kB")
+parser.add_option("--l2latency",
+ default = 10)
+parser.add_option("--rootdir",
+ help="ROot directory of Splash2",
+ default="/dist/splash2/codes/")
+parser.add_option("-b", "--benchmark",
+ help="Splash 2 benchmark to run")
+
+(options, args) = parser.parse_args()
+
+if args:
+ print "Error: script doesn't take any positional arguments"
+ sys.exit(1)
+
+# --------------------
+# Define Splash2 Benchmarks
+# ====================
+class Cholesky(LiveProcess):
+ executable = options.rootdir + '/kernels/cholesky/CHOLESKY'
+ cmd = 'CHOLESKY -p' + str(options.numcpus) + ' '\
+ + options.rootdir + '/kernels/cholesky/inputs/tk23.O'
+
+class FFT(LiveProcess):
+ executable = options.rootdir + 'kernels/fft/FFT'
+ cmd = 'FFT -p' + str(options.numcpus) + ' -m18'
+
+class LU_contig(LiveProcess):
+ executable = options.rootdir + 'kernels/lu/contiguous_blocks/LU'
+ cmd = 'LU -p' + str(options.numcpus)
+
+class LU_noncontig(LiveProcess):
+ executable = options.rootdir + 'kernels/lu/non_contiguous_blocks/LU'
+ cmd = 'LU -p' + str(options.numcpus)
+
+class Radix(LiveProcess):
+ executable = options.rootdir + 'kernels/radix/RADIX'
+ cmd = 'RADIX -n524288 -p' + str(options.numcpus)
+
+class Barnes(LiveProcess):
+ executable = options.rootdir + 'apps/barnes/BARNES'
+ cmd = 'BARNES'
+ input = options.rootdir + 'apps/barnes/input.p' + str(options.numcpus)
+
+class FMM(LiveProcess):
+ executable = options.rootdir + 'apps/fmm/FMM'
+ cmd = 'FMM'
+ input = options.rootdir + 'apps/fmm/inputs/input.2048.p' + str(options.numcpus)
+
+class Ocean_contig(LiveProcess):
+ executable = options.rootdir + 'apps/ocean/contiguous_partitions/OCEAN'
+ cmd = 'OCEAN -p' + str(options.numcpus)
+
+class Ocean_noncontig(LiveProcess):
+ executable = options.rootdir + 'apps/ocean/non_contiguous_partitions/OCEAN'
+ cmd = 'OCEAN -p' + str(options.numcpus)
+
+class Raytrace(LiveProcess):
+ executable = options.rootdir + 'apps/raytrace/RAYTRACE'
+ cmd = 'RAYTRACE -p' + str(options.numcpus) + ' ' \
+ + options.rootdir + 'apps/raytrace/inputs/teapot.env'
+
+class Water_nsquared(LiveProcess):
+ executable = options.rootdir + 'apps/water-nsquared/WATER-NSQUARED'
+ cmd = 'WATER-NSQUARED'
+ input = options.rootdir + 'apps/water-nsquared/input.p' + str(options.numcpus)
+
+class Water_spatial(LiveProcess):
+ executable = options.rootdir + 'apps/water-spatial/WATER-SPATIAL'
+ cmd = 'WATER-SPATIAL'
+ input = options.rootdir + 'apps/water-spatial/input.p' + str(options.numcpus)
+
+
+# --------------------
+# Base L1 Cache Definition
+# ====================
+
+class L1(BaseCache):
+ latency = options.l1latency
+ block_size = 64
+ mshrs = 12
+ tgts_per_mshr = 8
+ protocol = CoherenceProtocol(protocol=options.protocol)
+
+# ----------------------
+# Base L2 Cache Definition
+# ----------------------
+
+class L2(BaseCache):
+ block_size = 64
+ latency = options.l2latency
+ mshrs = 92
+ tgts_per_mshr = 16
+ write_buffers = 8
+
+# ----------------------
+# Define the clusters with their cpus
+# ----------------------
+class Cluster:
+ pass
+
+cpusPerCluster = options.numcpus/options.numclusters
+
+busFrequency = Frequency(options.frequency)
+busFrequency *= cpusPerCluster
+
+all_cpus = []
+all_l1s = []
+all_l1buses = []
+if options.timing:
+ clusters = [ Cluster() for i in xrange(options.numclusters)]
+ for j in xrange(options.numclusters):
+ clusters[j].id = j
+ for cluster in clusters:
+ cluster.clusterbus = Bus(clock=busFrequency)
+ all_l1buses += [cluster.clusterbus]
+ cluster.cpus = [TimingSimpleCPU(cpu_id = i + cluster.id,
+ clock=options.frequency)
+ for i in xrange(cpusPerCluster)]
+ all_cpus += cluster.cpus
+ cluster.l1 = L1(size=options.l1size, assoc = 4)
+ all_l1s += [cluster.l1]
+elif options.detailed:
+ clusters = [ Cluster() for i in xrange(options.numclusters)]
+ for j in xrange(options.numclusters):
+ clusters[j].id = j
+ for cluster in clusters:
+ cluster.clusterbus = Bus(clock=busFrequency)
+ all_l1buses += [cluster.clusterbus]
+ cluster.cpus = [DerivO3CPU(cpu_id = i + cluster.id,
+ clock=options.frequency)
+ for i in xrange(cpusPerCluster)]
+ all_cpus += cluster.cpus
+ cluster.l1 = L1(size=options.l1size, assoc = 4)
+ all_l1s += [cluster.l1]
+else:
+ clusters = [ Cluster() for i in xrange(options.numclusters)]
+ for j in xrange(options.numclusters):
+ clusters[j].id = j
+ for cluster in clusters:
+ cluster.clusterbus = Bus(clock=busFrequency)
+ all_l1buses += [cluster.clusterbus]
+ cluster.cpus = [AtomicSimpleCPU(cpu_id = i + cluster.id,
+ clock=options.frequency)
+ for i in xrange(cpusPerCluster)]
+ all_cpus += cluster.cpus
+ cluster.l1 = L1(size=options.l1size, assoc = 4)
+ all_l1s += [cluster.l1]
+
+# ----------------------
+# Create a system, and add system wide objects
+# ----------------------
+system = System(cpu = all_cpus, l1_ = all_l1s, l1bus_ = all_l1buses, physmem = PhysicalMemory(),
+ membus = Bus(clock = busFrequency))
+
+system.toL2bus = Bus(clock = busFrequency)
+system.l2 = L2(size = options.l2size, assoc = 8)
+
+# ----------------------
+# Connect the L2 cache and memory together
+# ----------------------
+
+system.physmem.port = system.membus.port
+system.l2.cpu_side = system.toL2bus.port
+system.l2.mem_side = system.membus.port
+
+# ----------------------
+# Connect the L2 cache and clusters together
+# ----------------------
+for cluster in clusters:
+ cluster.l1.cpu_side = cluster.clusterbus.port
+ cluster.l1.mem_side = system.toL2bus.port
+ for cpu in cluster.cpus:
+ cpu.icache_port = cluster.clusterbus.port
+ cpu.dcache_port = cluster.clusterbus.port
+ cpu.mem = cluster.l1
+
+# ----------------------
+# Define the root
+# ----------------------
+
+root = Root(system = system)
+
+# --------------------
+# Pick the correct Splash2 Benchmarks
+# ====================
+if options.benchmark == 'Cholesky':
+ root.workload = Cholesky()
+elif options.benchmark == 'FFT':
+ root.workload = FFT()
+elif options.benchmark == 'LUContig':
+ root.workload = LU_contig()
+elif options.benchmark == 'LUNoncontig':
+ root.workload = LU_noncontig()
+elif options.benchmark == 'Radix':
+ root.workload = Radix()
+elif options.benchmark == 'Barnes':
+ root.workload = Barnes()
+elif options.benchmark == 'FMM':
+ root.workload = FMM()
+elif options.benchmark == 'OceanContig':
+ root.workload = Ocean_contig()
+elif options.benchmark == 'OceanNoncontig':
+ root.workload = Ocean_noncontig()
+elif options.benchmark == 'Raytrace':
+ root.workload = Raytrace()
+elif options.benchmark == 'WaterNSquared':
+ root.workload = Water_nsquared()
+elif options.benchmark == 'WaterSpatial':
+ root.workload = Water_spatial()
+else:
+ panic("The --benchmark environment variable was set to something" \
+ +" improper.\nUse Cholesky, FFT, LUContig, LUNoncontig, Radix" \
+ +", Barnes, FMM, OceanContig,\nOceanNoncontig, Raytrace," \
+ +" WaterNSquared, or WaterSpatial\n")
+
+# --------------------
+# Assign the workload to the cpus
+# ====================
+
+for cluster in clusters:
+ for cpu in cluster.cpus:
+ cpu.workload = root.workload
+
+# ----------------------
+# Run the simulation
+# ----------------------
+
+if options.timing or options.detailed:
+ root.system.mem_mode = 'timing'
+
+# instantiate configuration
+m5.instantiate(root)
+
+# simulate until program terminates
+if options.maxtick:
+ exit_event = m5.simulate(options.maxtick)
+else:
+ exit_event = m5.simulate(m5.MaxTick)
+
+print 'Exiting @ tick', m5.curTick(), 'because', exit_event.getCause()
+
diff --git a/configs/splash2/run.py b/configs/splash2/run.py
index 7d56cb830..b162e0cc7 100644
--- a/configs/splash2/run.py
+++ b/configs/splash2/run.py
@@ -262,7 +262,7 @@ m5.instantiate(root)
if options.maxtick:
exit_event = m5.simulate(options.maxtick)
else:
- exit_event = m5.simulate(1000000000000)
+ exit_event = m5.simulate(m5.MaxTick)
print 'Exiting @ tick', m5.curTick(), 'because', exit_event.getCause()
diff --git a/src/cpu/memtest/memtest.cc b/src/cpu/memtest/memtest.cc
index 91e073cf0..180f41541 100644
--- a/src/cpu/memtest/memtest.cc
+++ b/src/cpu/memtest/memtest.cc
@@ -81,8 +81,13 @@ MemTest::CpuPort::recvFunctional(PacketPtr pkt)
void
MemTest::CpuPort::recvStatusChange(Status status)
{
- if (status == RangeChange)
+ if (status == RangeChange) {
+ if (!snoopRangeSent) {
+ snoopRangeSent = true;
+ sendStatusChange(Port::RangeChange);
+ }
return;
+ }
panic("MemTest doesn't expect recvStatusChange callback!");
}
@@ -145,6 +150,9 @@ MemTest::MemTest(const string &name,
// thread = new SimpleThread(NULL, 0, NULL, 0, mainMem);
curTick = 0;
+ cachePort.snoopRangeSent = false;
+ funcPort.snoopRangeSent = true;
+
// Needs to be masked off once we know the block size.
traceBlockAddr = _traceAddr;
baseAddr1 = 0x100000;
diff --git a/src/cpu/memtest/memtest.hh b/src/cpu/memtest/memtest.hh
index 2694efd39..7bf34d827 100644
--- a/src/cpu/memtest/memtest.hh
+++ b/src/cpu/memtest/memtest.hh
@@ -100,6 +100,8 @@ class MemTest : public MemObject
: Port(_name, _memtest), memtest(_memtest)
{ }
+ bool snoopRangeSent;
+
protected:
virtual bool recvTiming(PacketPtr pkt);
@@ -120,6 +122,8 @@ class MemTest : public MemObject
CpuPort cachePort;
CpuPort funcPort;
+ bool snoopRangeSent;
+
class MemTestSenderState : public Packet::SenderState
{
public:
diff --git a/src/cpu/o3/fetch.hh b/src/cpu/o3/fetch.hh
index cc9a8abf5..04016347a 100644
--- a/src/cpu/o3/fetch.hh
+++ b/src/cpu/o3/fetch.hh
@@ -83,6 +83,8 @@ class DefaultFetch
: Port(_fetch->name() + "-iport"), fetch(_fetch)
{ }
+ bool snoopRangeSent;
+
protected:
/** Atomic version of receive. Panics. */
virtual Tick recvAtomic(PacketPtr pkt);
diff --git a/src/cpu/o3/fetch_impl.hh b/src/cpu/o3/fetch_impl.hh
index 25faa407e..63d22b293 100644
--- a/src/cpu/o3/fetch_impl.hh
+++ b/src/cpu/o3/fetch_impl.hh
@@ -70,8 +70,13 @@ template<class Impl>
void
DefaultFetch<Impl>::IcachePort::recvStatusChange(Status status)
{
- if (status == RangeChange)
+ if (status == RangeChange) {
+ if (!snoopRangeSent) {
+ snoopRangeSent = true;
+ sendStatusChange(Port::RangeChange);
+ }
return;
+ }
panic("DefaultFetch doesn't expect recvStatusChange callback!");
}
@@ -287,6 +292,8 @@ DefaultFetch<Impl>::setCPU(O3CPU *cpu_ptr)
// Name is finally available, so create the port.
icachePort = new IcachePort(this);
+ icachePort->snoopRangeSent = false;
+
#if USE_CHECKER
if (cpu->checker) {
cpu->checker->setIcachePort(icachePort);
diff --git a/src/cpu/o3/lsq.hh b/src/cpu/o3/lsq.hh
index 6b12d75b4..7559a36d5 100644
--- a/src/cpu/o3/lsq.hh
+++ b/src/cpu/o3/lsq.hh
@@ -298,6 +298,8 @@ class LSQ {
: lsq(_lsq)
{ }
+ bool snoopRangeSent;
+
protected:
/** Atomic version of receive. Panics. */
virtual Tick recvAtomic(PacketPtr pkt);
diff --git a/src/cpu/o3/lsq_impl.hh b/src/cpu/o3/lsq_impl.hh
index 5e7945c1c..6758e51c8 100644
--- a/src/cpu/o3/lsq_impl.hh
+++ b/src/cpu/o3/lsq_impl.hh
@@ -53,9 +53,13 @@ template <class Impl>
void
LSQ<Impl>::DcachePort::recvStatusChange(Status status)
{
- if (status == RangeChange)
+ if (status == RangeChange) {
+ if (!snoopRangeSent) {
+ snoopRangeSent = true;
+ sendStatusChange(Port::RangeChange);
+ }
return;
-
+ }
panic("O3CPU doesn't expect recvStatusChange callback!");
}
@@ -97,6 +101,8 @@ LSQ<Impl>::LSQ(Params *params)
{
DPRINTF(LSQ, "Creating LSQ object.\n");
+ dcachePort.snoopRangeSent = false;
+
//**********************************************/
//************ Handle SMT Parameters ***********/
//**********************************************/
diff --git a/src/cpu/simple/atomic.cc b/src/cpu/simple/atomic.cc
index 58dc1fe5f..325260609 100644
--- a/src/cpu/simple/atomic.cc
+++ b/src/cpu/simple/atomic.cc
@@ -94,7 +94,7 @@ Tick
AtomicSimpleCPU::CpuPort::recvAtomic(PacketPtr pkt)
{
//Snooping a coherence request, just return
- return curTick;
+ return 0;
}
void
@@ -107,8 +107,13 @@ AtomicSimpleCPU::CpuPort::recvFunctional(PacketPtr pkt)
void
AtomicSimpleCPU::CpuPort::recvStatusChange(Status status)
{
- if (status == RangeChange)
+ if (status == RangeChange) {
+ if (!snoopRangeSent) {
+ snoopRangeSent = true;
+ sendStatusChange(Port::RangeChange);
+ }
return;
+ }
panic("AtomicSimpleCPU doesn't expect recvStatusChange callback!");
}
@@ -127,6 +132,9 @@ AtomicSimpleCPU::AtomicSimpleCPU(Params *p)
{
_status = Idle;
+ icachePort.snoopRangeSent = false;
+ dcachePort.snoopRangeSent = false;
+
ifetch_req = new Request();
ifetch_req->setThreadContext(p->cpu_id, 0); // Add thread ID if we add MT
ifetch_pkt = new Packet(ifetch_req, Packet::ReadReq, Packet::Broadcast);
diff --git a/src/cpu/simple/atomic.hh b/src/cpu/simple/atomic.hh
index 166a18127..0df6fe079 100644
--- a/src/cpu/simple/atomic.hh
+++ b/src/cpu/simple/atomic.hh
@@ -90,6 +90,8 @@ class AtomicSimpleCPU : public BaseSimpleCPU
: Port(_name, _cpu), cpu(_cpu)
{ }
+ bool snoopRangeSent;
+
protected:
virtual bool recvTiming(PacketPtr pkt);
diff --git a/src/cpu/simple/timing.cc b/src/cpu/simple/timing.cc
index db2c940c0..1ea2df894 100644
--- a/src/cpu/simple/timing.cc
+++ b/src/cpu/simple/timing.cc
@@ -82,8 +82,13 @@ TimingSimpleCPU::CpuPort::recvFunctional(PacketPtr pkt)
void
TimingSimpleCPU::CpuPort::recvStatusChange(Status status)
{
- if (status == RangeChange)
+ if (status == RangeChange) {
+ if (!snoopRangeSent) {
+ snoopRangeSent = true;
+ sendStatusChange(Port::RangeChange);
+ }
return;
+ }
panic("TimingSimpleCPU doesn't expect recvStatusChange callback!");
}
@@ -101,6 +106,10 @@ TimingSimpleCPU::TimingSimpleCPU(Params *p)
cpu_id(p->cpu_id)
{
_status = Idle;
+
+ icachePort.snoopRangeSent = false;
+ dcachePort.snoopRangeSent = false;
+
ifetch_pkt = dcache_pkt = NULL;
drainEvent = NULL;
fetchEvent = NULL;
diff --git a/src/cpu/simple/timing.hh b/src/cpu/simple/timing.hh
index 408fa315e..fe5d03666 100644
--- a/src/cpu/simple/timing.hh
+++ b/src/cpu/simple/timing.hh
@@ -82,6 +82,8 @@ class TimingSimpleCPU : public BaseSimpleCPU
: Port(_name, _cpu), cpu(_cpu), lat(_lat)
{ }
+ bool snoopRangeSent;
+
protected:
virtual Tick recvAtomic(PacketPtr pkt);
@@ -166,8 +168,6 @@ class TimingSimpleCPU : public BaseSimpleCPU
PacketPtr ifetch_pkt;
PacketPtr dcache_pkt;
-
-
int cpu_id;
Tick previousTick;
diff --git a/src/mem/bus.cc b/src/mem/bus.cc
index 8ea67a0e4..b97a7ddb9 100644
--- a/src/mem/bus.cc
+++ b/src/mem/bus.cc
@@ -290,7 +290,10 @@ Bus::findPort(Addr addr, int id)
// we shouldn't be sending this back to where it came from
- assert(dest_id != id);
+ // only on a functional access and then we should terminate
+ // the cyclical call.
+ if (dest_id == id)
+ return 0;
return interfaces[dest_id];
}
@@ -307,7 +310,18 @@ Bus::findSnoopPorts(Addr addr, int id)
if (portSnoopList[i].range == addr && portSnoopList[i].portId != id) {
//Careful to not overlap ranges
//or snoop will be called more than once on the port
- ports.push_back(portSnoopList[i].portId);
+
+ //@todo Fix this hack because ranges are overlapping
+ //need to make sure we dont't create overlapping ranges
+ bool hack_overlap = false;
+ int size = ports.size();
+ for (int j=0; j < size; j++) {
+ if (ports[j] == portSnoopList[i].portId)
+ hack_overlap = true;
+ }
+
+ if (!hack_overlap)
+ ports.push_back(portSnoopList[i].portId);
// DPRINTF(Bus, " found snoop addr %#llx on device%d\n", addr,
// portSnoopList[i].portId);
}
@@ -339,10 +353,13 @@ Bus::functionalSnoop(PacketPtr pkt)
{
std::vector<int> ports = findSnoopPorts(pkt->getAddr(), pkt->getSrc());
+ //The packet may be changed by another bus on snoops, restore the id after each
+ int id = pkt->getSrc();
while (!ports.empty() && pkt->result != Packet::Success)
{
interfaces[ports.back()]->sendFunctional(pkt);
ports.pop_back();
+ pkt->setSrc(id);
}
}
@@ -392,8 +409,11 @@ Bus::recvFunctional(PacketPtr pkt)
functionalSnoop(pkt);
// If the snooping found what we were looking for, we're done.
- if (pkt->result != Packet::Success)
- findPort(pkt->getAddr(), pkt->getSrc())->sendFunctional(pkt);
+ if (pkt->result != Packet::Success) {
+ Port* port = findPort(pkt->getAddr(), pkt->getSrc());
+ if (port)
+ port->sendFunctional(pkt);
+ }
}
/** Function called by the port when the bus is receiving a status change.*/
@@ -451,6 +471,7 @@ Bus::recvStatusChange(Port::Status status, int id)
dm.portId = id;
dm.range = *iter;
+ //@todo, make sure we don't overlap ranges
DPRINTF(BusAddrRanges, "Adding snoop range %#llx - %#llx for id %d\n",
dm.range.start, dm.range.end, id);
portSnoopList.push_back(dm);
@@ -493,7 +514,7 @@ Bus::addressRanges(AddrRangeList &resp, AddrRangeList &snoop, int id)
for (dflt_iter = defaultRange.begin(); dflt_iter != defaultRange.end();
dflt_iter++) {
resp.push_back(*dflt_iter);
- DPRINTF(BusAddrRanges, " -- %#llx : %#llx\n",dflt_iter->start,
+ DPRINTF(BusAddrRanges, " -- Dflt: %#llx : %#llx\n",dflt_iter->start,
dflt_iter->end);
}
for (portIter = portList.begin(); portIter != portList.end(); portIter++) {
@@ -519,6 +540,18 @@ Bus::addressRanges(AddrRangeList &resp, AddrRangeList &snoop, int id)
portIter->range.start, portIter->range.end);
}
}
+
+ for (portIter = portSnoopList.begin();
+ portIter != portSnoopList.end(); portIter++)
+ {
+ if (portIter->portId != id) {
+ snoop.push_back(portIter->range);
+ DPRINTF(BusAddrRanges, " -- Snoop: %#llx : %#llx\n",
+ portIter->range.start, portIter->range.end);
+ //@todo We need to properly insert snoop ranges
+ //not overlapping the ranges (multiple)
+ }
+ }
}
unsigned int
diff --git a/src/mem/cache/base_cache.hh b/src/mem/cache/base_cache.hh
index 584c2d5df..ef4955432 100644
--- a/src/mem/cache/base_cache.hh
+++ b/src/mem/cache/base_cache.hh
@@ -144,8 +144,6 @@ class BaseCache : public MemObject
protected:
CachePort *memSidePort;
- bool snoopRangesSent;
-
public:
virtual Port *getPort(const std::string &if_name, int idx = -1);
@@ -171,10 +169,6 @@ class BaseCache : public MemObject
if (status == Port::RangeChange){
if (!isCpuSide) {
cpuSidePort->sendStatusChange(Port::RangeChange);
- if (!snoopRangesSent) {
- snoopRangesSent = true;
- memSidePort->sendStatusChange(Port::RangeChange);
- }
}
else {
memSidePort->sendStatusChange(Port::RangeChange);
@@ -358,7 +352,6 @@ class BaseCache : public MemObject
//Start ports at null if more than one is created we should panic
cpuSidePort = NULL;
memSidePort = NULL;
- snoopRangesSent = false;
}
~BaseCache()
diff --git a/src/mem/cache/miss/miss_queue.cc b/src/mem/cache/miss/miss_queue.cc
index fe467a8ea..3c4586272 100644
--- a/src/mem/cache/miss/miss_queue.cc
+++ b/src/mem/cache/miss/miss_queue.cc
@@ -612,6 +612,8 @@ MissQueue::handleResponse(PacketPtr &pkt, Tick time)
if (mshr->hasTargets()) {
// Didn't satisfy all the targets, need to resend
Packet::Command cmd = mshr->getTarget()->cmd;
+ mshr->pkt->setDest(Packet::Broadcast);
+ mshr->pkt->result = Packet::Unknown;
mq.markPending(mshr, cmd);
mshr->order = order++;
cache->setMasterRequest(Request_MSHR, time);