MEM: Separate requests and responses for timing accesses

This patch moves send/recvTiming and send/recvTimingSnoop from the
Port base class to the MasterPort and SlavePort, and also splits them
into separate member functions for requests and responses:
send/recvTimingReq, send/recvTimingResp, and send/recvTimingSnoopReq,
send/recvTimingSnoopResp. A master port sends requests and receives
responses, and also receives snoop requests and sends snoop
responses. A slave port has the reciprocal behaviour as it receives
requests and sends responses, and sends snoop requests and receives
snoop responses.

For all MemObjects that have only master ports or slave ports (but not
both), e.g. a CPU, or a PIO device, this patch merely adds more
clarity to what kind of access is taking place. For example, a CPU
port used to call sendTiming, and will now call
sendTimingReq. Similarly, a response previously came back through
recvTiming, which is now recvTimingResp. For the modules that have
both master and slave ports, e.g. the bus, the behaviour was
previously relying on branches based on pkt->isRequest(), and this is
now replaced with a direct call to the apprioriate member function
depending on the type of access. Please note that send/recvRetry is
still shared by all the timing accessors and remains in the Port base
class for now (to maintain the current bus functionality and avoid
changing the statistics of all regressions).

The packet queue is split into a MasterPort and SlavePort version to
facilitate the use of the new timing accessors. All uses of the
PacketQueue are updated accordingly.

With this patch, the type of packet (request or response) is now well
defined for each type of access, and asserts on pkt->isRequest() and
pkt->isResponse() are now moved to the appropriate send member
functions. It is also worth noting that sendTimingSnoopReq no longer
returns a boolean, as the semantics do not alow snoop requests to be
rejected or stalled. All these assumptions are now excplicitly part of
the port interface itself.

7 files changed, 21 insertions, 29 deletions

diff --git a/src/cpu/o3/cpu.cc b/src/cpu/o3/cpu.cc
index fe70c3fcf..e8fc968b7 100644
--- a/src/cpu/o3/cpu.cc
+++ b/src/cpu/o3/cpu.cc
@@ -87,9 +87,8 @@ BaseO3CPU::regStats()
 
 template<class Impl>
 bool
-FullO3CPU<Impl>::IcachePort::recvTiming(PacketPtr pkt)
+FullO3CPU<Impl>::IcachePort::recvTimingResp(PacketPtr pkt)
 {
-    assert(pkt->isResponse());
     DPRINTF(O3CPU, "Fetch unit received timing\n");
     // We shouldn't ever get a block in ownership state
     assert(!(pkt->memInhibitAsserted() && !pkt->sharedAsserted()));
@@ -107,18 +106,16 @@ FullO3CPU<Impl>::IcachePort::recvRetry()
 
 template <class Impl>
 bool
-FullO3CPU<Impl>::DcachePort::recvTiming(PacketPtr pkt)
+FullO3CPU<Impl>::DcachePort::recvTimingResp(PacketPtr pkt)
 {
-    assert(pkt->isResponse());
-    return lsq->recvTiming(pkt);
+    return lsq->recvTimingResp(pkt);
 }
 
 template <class Impl>
-bool
-FullO3CPU<Impl>::DcachePort::recvTimingSnoop(PacketPtr pkt)
+void
+FullO3CPU<Impl>::DcachePort::recvTimingSnoopReq(PacketPtr pkt)
 {
-    assert(pkt->isRequest());
-    return lsq->recvTimingSnoop(pkt);
+    lsq->recvTimingSnoopReq(pkt);
 }
 
 template <class Impl>
diff --git a/src/cpu/o3/cpu.hh b/src/cpu/o3/cpu.hh
index be51f415f..41128110b 100644
--- a/src/cpu/o3/cpu.hh
+++ b/src/cpu/o3/cpu.hh
@@ -148,8 +148,8 @@ class FullO3CPU : public BaseO3CPU
 
         /** Timing version of receive.  Handles setting fetch to the
          * proper status to start fetching. */
-        virtual bool recvTiming(PacketPtr pkt);
-        virtual bool recvTimingSnoop(PacketPtr pkt) { return true; }
+        virtual bool recvTimingResp(PacketPtr pkt);
+        virtual void recvTimingSnoopReq(PacketPtr pkt) { }
 
         /** Handles doing a retry of a failed fetch. */
         virtual void recvRetry();
@@ -176,8 +176,8 @@ class FullO3CPU : public BaseO3CPU
         /** Timing version of receive.  Handles writing back and
          * completing the load or store that has returned from
          * memory. */
-        virtual bool recvTiming(PacketPtr pkt);
-        virtual bool recvTimingSnoop(PacketPtr pkt);
+        virtual bool recvTimingResp(PacketPtr pkt);
+        virtual void recvTimingSnoopReq(PacketPtr pkt);
 
         /** Handles doing a retry of the previous send. */
         virtual void recvRetry();
diff --git a/src/cpu/o3/fetch_impl.hh b/src/cpu/o3/fetch_impl.hh
index 2480211e4..f4ce77f22 100644
--- a/src/cpu/o3/fetch_impl.hh
+++ b/src/cpu/o3/fetch_impl.hh
@@ -621,7 +621,7 @@ DefaultFetch<Impl>::finishTranslation(Fault fault, RequestPtr mem_req)
         fetchedCacheLines++;
 
         // Access the cache.
-        if (!cpu->getInstPort().sendTiming(data_pkt)) {
+        if (!cpu->getInstPort().sendTimingReq(data_pkt)) {
             assert(retryPkt == NULL);
             assert(retryTid == InvalidThreadID);
             DPRINTF(Fetch, "[tid:%i] Out of MSHRs!\n", tid);
@@ -1356,7 +1356,7 @@ DefaultFetch<Impl>::recvRetry()
         assert(retryTid != InvalidThreadID);
         assert(fetchStatus[retryTid] == IcacheWaitRetry);
 
-        if (cpu->getInstPort().sendTiming(retryPkt)) {
+        if (cpu->getInstPort().sendTimingReq(retryPkt)) {
             fetchStatus[retryTid] = IcacheWaitResponse;
             retryPkt = NULL;
             retryTid = InvalidThreadID;
diff --git a/src/cpu/o3/lsq.hh b/src/cpu/o3/lsq.hh
index dac5fab18..026033539 100644
--- a/src/cpu/o3/lsq.hh
+++ b/src/cpu/o3/lsq.hh
@@ -297,9 +297,9 @@ class LSQ {
      *
      * @param pkt Response packet from the memory sub-system
      */
-    bool recvTiming(PacketPtr pkt);
+    bool recvTimingResp(PacketPtr pkt);
 
-    bool recvTimingSnoop(PacketPtr pkt);
+    void recvTimingSnoopReq(PacketPtr pkt);
 
     /** The CPU pointer. */
     O3CPU *cpu;
diff --git a/src/cpu/o3/lsq_impl.hh b/src/cpu/o3/lsq_impl.hh
index c2f410e37..72ffdd58b 100644
--- a/src/cpu/o3/lsq_impl.hh
+++ b/src/cpu/o3/lsq_impl.hh
@@ -319,9 +319,8 @@ LSQ<Impl>::recvRetry()
 
 template <class Impl>
 bool
-LSQ<Impl>::recvTiming(PacketPtr pkt)
+LSQ<Impl>::recvTimingResp(PacketPtr pkt)
 {
-    assert(pkt->isResponse());
     if (pkt->isError())
         DPRINTF(LSQ, "Got error packet back for address: %#X\n",
                 pkt->getAddr());
@@ -330,10 +329,9 @@ LSQ<Impl>::recvTiming(PacketPtr pkt)
 }
 
 template <class Impl>
-bool
-LSQ<Impl>::recvTimingSnoop(PacketPtr pkt)
+void
+LSQ<Impl>::recvTimingSnoopReq(PacketPtr pkt)
 {
-    assert(pkt->isRequest());
     DPRINTF(LSQ, "received pkt for addr:%#x %s\n", pkt->getAddr(),
             pkt->cmdString());
 
@@ -345,9 +343,6 @@ LSQ<Impl>::recvTimingSnoop(PacketPtr pkt)
             thread[tid].checkSnoop(pkt);
         }
     }
-
-    // to provide stronger consistency model
-    return true;
 }
 
 template<class Impl>
diff --git a/src/cpu/o3/lsq_unit.hh b/src/cpu/o3/lsq_unit.hh
index 44c3df0bf..ad1e26d2f 100644
--- a/src/cpu/o3/lsq_unit.hh
+++ b/src/cpu/o3/lsq_unit.hh
@@ -801,7 +801,7 @@ LSQUnit<Impl>::read(Request *req, Request *sreqLow, Request *sreqHigh,
             state->mainPkt = data_pkt;
         }
 
-        if (!dcachePort->sendTiming(fst_data_pkt)) {
+        if (!dcachePort->sendTimingReq(fst_data_pkt)) {
             // Delete state and data packet because a load retry
             // initiates a pipeline restart; it does not retry.
             delete state;
@@ -830,7 +830,7 @@ LSQUnit<Impl>::read(Request *req, Request *sreqLow, Request *sreqHigh,
             // The first packet will return in completeDataAccess and be
             // handled there.
             ++usedPorts;
-            if (!dcachePort->sendTiming(snd_data_pkt)) {
+            if (!dcachePort->sendTimingReq(snd_data_pkt)) {
 
                 // The main packet will be deleted in completeDataAccess.
                 delete snd_data_pkt->req;
diff --git a/src/cpu/o3/lsq_unit_impl.hh b/src/cpu/o3/lsq_unit_impl.hh
index f4182e30d..4f82ad9e3 100644
--- a/src/cpu/o3/lsq_unit_impl.hh
+++ b/src/cpu/o3/lsq_unit_impl.hh
@@ -1180,7 +1180,7 @@ template <class Impl>
 bool
 LSQUnit<Impl>::sendStore(PacketPtr data_pkt)
 {
-    if (!dcachePort->sendTiming(data_pkt)) {
+    if (!dcachePort->sendTimingReq(data_pkt)) {
         // Need to handle becoming blocked on a store.
         isStoreBlocked = true;
         ++lsqCacheBlocked;
@@ -1203,7 +1203,7 @@ LSQUnit<Impl>::recvRetry()
         LSQSenderState *state =
             dynamic_cast<LSQSenderState *>(retryPkt->senderState);
 
-        if (dcachePort->sendTiming(retryPkt)) {
+        if (dcachePort->sendTimingReq(retryPkt)) {
             // Don't finish the store unless this is the last packet.
             if (!TheISA::HasUnalignedMemAcc || !state->pktToSend ||
                     state->pendingPacket == retryPkt) {