/* * Copyright (c) 2010-2012 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) 2002-2005 The Regents of The University of Michigan * Copyright (c) 2010 Advanced Micro Devices, Inc. * 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: Erik Hallnor * Dave Greene * Nathan Binkert * Steve Reinhardt * Ron Dreslinski */ /** * @file * Cache definitions. */ #include "base/misc.hh" #include "base/range.hh" #include "base/types.hh" #include "debug/Cache.hh" #include "debug/CachePort.hh" #include "mem/cache/prefetch/base.hh" #include "mem/cache/blk.hh" #include "mem/cache/cache.hh" #include "mem/cache/mshr.hh" #include "sim/sim_exit.hh" template Cache::Cache(const Params *p, TagStore *tags) : BaseCache(p), tags(tags), prefetcher(p->prefetcher), doFastWrites(true), prefetchOnAccess(p->prefetch_on_access) { tempBlock = new BlkType(); tempBlock->data = new uint8_t[blkSize]; cpuSidePort = new CpuSidePort(p->name + ".cpu_side", this, "CpuSidePort"); memSidePort = new MemSidePort(p->name + ".mem_side", this, "MemSidePort"); tags->setCache(this); if (prefetcher) prefetcher->setCache(this); } template void Cache::regStats() { BaseCache::regStats(); tags->regStats(name()); } template void Cache::cmpAndSwap(BlkType *blk, PacketPtr pkt) { uint64_t overwrite_val; bool overwrite_mem; uint64_t condition_val64; uint32_t condition_val32; int offset = tags->extractBlkOffset(pkt->getAddr()); uint8_t *blk_data = blk->data + offset; assert(sizeof(uint64_t) >= pkt->getSize()); overwrite_mem = true; // keep a copy of our possible write value, and copy what is at the // memory address into the packet pkt->writeData((uint8_t *)&overwrite_val); pkt->setData(blk_data); if (pkt->req->isCondSwap()) { if (pkt->getSize() == sizeof(uint64_t)) { condition_val64 = pkt->req->getExtraData(); overwrite_mem = !std::memcmp(&condition_val64, blk_data, sizeof(uint64_t)); } else if (pkt->getSize() == sizeof(uint32_t)) { condition_val32 = (uint32_t)pkt->req->getExtraData(); overwrite_mem = !std::memcmp(&condition_val32, blk_data, sizeof(uint32_t)); } else panic("Invalid size for conditional read/write\n"); } if (overwrite_mem) { std::memcpy(blk_data, &overwrite_val, pkt->getSize()); blk->status |= BlkDirty; } } template void Cache::satisfyCpuSideRequest(PacketPtr pkt, BlkType *blk, bool deferred_response, bool pending_downgrade) { assert(blk && blk->isValid()); // Occasionally this is not true... if we are a lower-level cache // satisfying a string of Read and ReadEx requests from // upper-level caches, a Read will mark the block as shared but we // can satisfy a following ReadEx anyway since we can rely on the // Read requester(s) to have buffered the ReadEx snoop and to // invalidate their blocks after receiving them. // assert(!pkt->needsExclusive() || blk->isWritable()); assert(pkt->getOffset(blkSize) + pkt->getSize() <= blkSize); // Check RMW operations first since both isRead() and // isWrite() will be true for them if (pkt->cmd == MemCmd::SwapReq) { cmpAndSwap(blk, pkt); } else if (pkt->isWrite()) { if (blk->checkWrite(pkt)) { pkt->writeDataToBlock(blk->data, blkSize); blk->status |= BlkDirty; } } else if (pkt->isRead()) { if (pkt->isLLSC()) { blk->trackLoadLocked(pkt); } pkt->setDataFromBlock(blk->data, blkSize); if (pkt->getSize() == blkSize) { // special handling for coherent block requests from // upper-level caches if (pkt->needsExclusive()) { // if we have a dirty copy, make sure the recipient // keeps it marked dirty if (blk->isDirty()) { pkt->assertMemInhibit(); } // on ReadExReq we give up our copy unconditionally tags->invalidateBlk(blk); } else if (blk->isWritable() && !pending_downgrade && !pkt->sharedAsserted()) { // we can give the requester an exclusive copy (by not // asserting shared line) on a read request if: // - we have an exclusive copy at this level (& below) // - we don't have a pending snoop from below // signaling another read request // - no other cache above has a copy (otherwise it // would have asseretd shared line on request) if (blk->isDirty()) { // special considerations if we're owner: if (!deferred_response && !isTopLevel) { // if we are responding immediately and can // signal that we're transferring ownership // along with exclusivity, do so pkt->assertMemInhibit(); blk->status &= ~BlkDirty; } else { // if we're responding after our own miss, // there's a window where the recipient didn't // know it was getting ownership and may not // have responded to snoops correctly, so we // can't pass off ownership *or* exclusivity pkt->assertShared(); } } } else { // otherwise only respond with a shared copy pkt->assertShared(); } } } else { // Not a read or write... must be an upgrade. it's OK // to just ack those as long as we have an exclusive // copy at this level. assert(pkt->isUpgrade()); tags->invalidateBlk(blk); } } ///////////////////////////////////////////////////// // // MSHR helper functions // ///////////////////////////////////////////////////// template void Cache::markInService(MSHR *mshr, PacketPtr pkt) { markInServiceInternal(mshr, pkt); #if 0 if (mshr->originalCmd == MemCmd::HardPFReq) { DPRINTF(HWPrefetch, "%s:Marking a HW_PF in service\n", name()); //Also clear pending if need be if (!prefetcher->havePending()) { deassertMemSideBusRequest(Request_PF); } } #endif } template void Cache::squash(int threadNum) { bool unblock = false; BlockedCause cause = NUM_BLOCKED_CAUSES; if (noTargetMSHR && noTargetMSHR->threadNum == threadNum) { noTargetMSHR = NULL; unblock = true; cause = Blocked_NoTargets; } if (mshrQueue.isFull()) { unblock = true; cause = Blocked_NoMSHRs; } mshrQueue.squash(threadNum); if (unblock && !mshrQueue.isFull()) { clearBlocked(cause); } } ///////////////////////////////////////////////////// // // Access path: requests coming in from the CPU side // ///////////////////////////////////////////////////// template bool Cache::access(PacketPtr pkt, BlkType *&blk, int &lat, PacketList &writebacks) { if (pkt->req->isUncacheable()) { if (pkt->req->isClearLL()) { tags->clearLocks(); } else if (pkt->isWrite()) { blk = tags->findBlock(pkt->getAddr()); if (blk != NULL) { tags->invalidateBlk(blk); } } blk = NULL; lat = hitLatency; return false; } int id = pkt->req->hasContextId() ? pkt->req->contextId() : -1; blk = tags->accessBlock(pkt->getAddr(), lat, id); DPRINTF(Cache, "%s%s %x %s\n", pkt->cmdString(), pkt->req->isInstFetch() ? " (ifetch)" : "", pkt->getAddr(), (blk) ? "hit" : "miss"); if (blk != NULL) { if (pkt->needsExclusive() ? blk->isWritable() : blk->isReadable()) { // OK to satisfy access incHitCount(pkt); satisfyCpuSideRequest(pkt, blk); return true; } } // Can't satisfy access normally... either no block (blk == NULL) // or have block but need exclusive & only have shared. // Writeback handling is special case. We can write the block // into the cache without having a writeable copy (or any copy at // all). if (pkt->cmd == MemCmd::Writeback) { assert(blkSize == pkt->getSize()); if (blk == NULL) { // need to do a replacement blk = allocateBlock(pkt->getAddr(), writebacks); if (blk == NULL) { // no replaceable block available, give up. // writeback will be forwarded to next level. incMissCount(pkt); return false; } int id = pkt->req->masterId(); tags->insertBlock(pkt->getAddr(), blk, id); blk->status = BlkValid | BlkReadable; } std::memcpy(blk->data, pkt->getPtr(), blkSize); blk->status |= BlkDirty; if (pkt->isSupplyExclusive()) { blk->status |= BlkWritable; } // nothing else to do; writeback doesn't expect response assert(!pkt->needsResponse()); incHitCount(pkt); return true; } incMissCount(pkt); if (blk == NULL && pkt->isLLSC() && pkt->isWrite()) { // complete miss on store conditional... just give up now pkt->req->setExtraData(0); return true; } return false; } class ForwardResponseRecord : public Packet::SenderState { Packet::SenderState *prevSenderState; PortID prevSrc; #ifndef NDEBUG BaseCache *cache; #endif public: ForwardResponseRecord(Packet *pkt, BaseCache *_cache) : prevSenderState(pkt->senderState), prevSrc(pkt->getSrc()) #ifndef NDEBUG , cache(_cache) #endif {} void restore(Packet *pkt, BaseCache *_cache) { assert(_cache == cache); pkt->senderState = prevSenderState; pkt->setDest(prevSrc); } }; template bool Cache::timingAccess(PacketPtr pkt) { //@todo Add back in MemDebug Calls // MemDebug::cacheAccess(pkt); /// @todo temporary hack to deal with memory corruption issue until /// 4-phase transactions are complete for (int x = 0; x < pendingDelete.size(); x++) delete pendingDelete[x]; pendingDelete.clear(); // we charge hitLatency for doing just about anything here Tick time = curTick() + hitLatency; if (pkt->isResponse()) { // must be cache-to-cache response from upper to lower level ForwardResponseRecord *rec = dynamic_cast(pkt->senderState); if (rec == NULL) { assert(pkt->cmd == MemCmd::HardPFResp); // Check if it's a prefetch response and handle it. We shouldn't // get any other kinds of responses without FRRs. DPRINTF(Cache, "Got prefetch response from above for addr %#x\n", pkt->getAddr()); handleResponse(pkt); return true; } rec->restore(pkt, this); delete rec; memSidePort->respond(pkt, time); return true; } assert(pkt->isRequest()); if (pkt->memInhibitAsserted()) { DPRINTF(Cache, "mem inhibited on 0x%x: not responding\n", pkt->getAddr()); assert(!pkt->req->isUncacheable()); // Special tweak for multilevel coherence: snoop downward here // on invalidates since there may be other caches below here // that have shared copies. Not necessary if we know that // supplier had exclusive copy to begin with. if (pkt->needsExclusive() && !pkt->isSupplyExclusive()) { Packet *snoopPkt = new Packet(pkt, true); // clear flags snoopPkt->setExpressSnoop(); snoopPkt->assertMemInhibit(); memSidePort->sendTimingReq(snoopPkt); // main memory will delete snoopPkt } // since we're the official target but we aren't responding, // delete the packet now. /// @todo nominally we should just delete the packet here, /// however, until 4-phase stuff we can't because sending /// cache is still relying on it pendingDelete.push_back(pkt); return true; } if (pkt->req->isUncacheable()) { if (pkt->req->isClearLL()) { tags->clearLocks(); } else if (pkt->isWrite()) { BlkType *blk = tags->findBlock(pkt->getAddr()); if (blk != NULL) { tags->invalidateBlk(blk); } } // writes go in write buffer, reads use MSHR if (pkt->isWrite() && !pkt->isRead()) { allocateWriteBuffer(pkt, time, true); } else { allocateUncachedReadBuffer(pkt, time, true); } assert(pkt->needsResponse()); // else we should delete it here?? return true; } int lat = hitLatency; BlkType *blk = NULL; PacketList writebacks; bool satisfied = access(pkt, blk, lat, writebacks); #if 0 /** @todo make the fast write alloc (wh64) work with coherence. */ // If this is a block size write/hint (WH64) allocate the block here // if the coherence protocol allows it. if (!blk && pkt->getSize() >= blkSize && coherence->allowFastWrites() && (pkt->cmd == MemCmd::WriteReq || pkt->cmd == MemCmd::WriteInvalidateReq) ) { // not outstanding misses, can do this MSHR *outstanding_miss = mshrQueue.findMatch(pkt->getAddr()); if (pkt->cmd == MemCmd::WriteInvalidateReq || !outstanding_miss) { if (outstanding_miss) { warn("WriteInv doing a fastallocate" "with an outstanding miss to the same address\n"); } blk = handleFill(NULL, pkt, BlkValid | BlkWritable, writebacks); ++fastWrites; } } #endif // track time of availability of next prefetch, if any Tick next_pf_time = 0; bool needsResponse = pkt->needsResponse(); if (satisfied) { if (prefetcher && (prefetchOnAccess || (blk && blk->wasPrefetched()))) { if (blk) blk->status &= ~BlkHWPrefetched; next_pf_time = prefetcher->notify(pkt, time); } if (needsResponse) { pkt->makeTimingResponse(); cpuSidePort->respond(pkt, curTick()+lat); } else { /// @todo nominally we should just delete the packet here, /// however, until 4-phase stuff we can't because sending /// cache is still relying on it pendingDelete.push_back(pkt); } } else { // miss Addr blk_addr = blockAlign(pkt->getAddr()); MSHR *mshr = mshrQueue.findMatch(blk_addr); if (mshr) { // MSHR hit //@todo remove hw_pf here assert(pkt->req->masterId() < system->maxMasters()); mshr_hits[pkt->cmdToIndex()][pkt->req->masterId()]++; if (mshr->threadNum != 0/*pkt->req->threadId()*/) { mshr->threadNum = -1; } mshr->allocateTarget(pkt, time, order++); if (mshr->getNumTargets() == numTarget) { noTargetMSHR = mshr; setBlocked(Blocked_NoTargets); // need to be careful with this... if this mshr isn't // ready yet (i.e. time > curTick()_, we don't want to // move it ahead of mshrs that are ready // mshrQueue.moveToFront(mshr); } } else { // no MSHR assert(pkt->req->masterId() < system->maxMasters()); mshr_misses[pkt->cmdToIndex()][pkt->req->masterId()]++; // always mark as cache fill for now... if we implement // no-write-allocate or bypass accesses this will have to // be changed. if (pkt->cmd == MemCmd::Writeback) { allocateWriteBuffer(pkt, time, true); } else { if (blk && blk->isValid()) { // If we have a write miss to a valid block, we // need to mark the block non-readable. Otherwise // if we allow reads while there's an outstanding // write miss, the read could return stale data // out of the cache block... a more aggressive // system could detect the overlap (if any) and // forward data out of the MSHRs, but we don't do // that yet. Note that we do need to leave the // block valid so that it stays in the cache, in // case we get an upgrade response (and hence no // new data) when the write miss completes. // As long as CPUs do proper store/load forwarding // internally, and have a sufficiently weak memory // model, this is probably unnecessary, but at some // point it must have seemed like we needed it... assert(pkt->needsExclusive() && !blk->isWritable()); blk->status &= ~BlkReadable; } allocateMissBuffer(pkt, time, true); } if (prefetcher) { next_pf_time = prefetcher->notify(pkt, time); } } } if (next_pf_time != 0) requestMemSideBus(Request_PF, std::max(time, next_pf_time)); // copy writebacks to write buffer while (!writebacks.empty()) { PacketPtr wbPkt = writebacks.front(); allocateWriteBuffer(wbPkt, time, true); writebacks.pop_front(); } return true; } // See comment in cache.hh. template PacketPtr Cache::getBusPacket(PacketPtr cpu_pkt, BlkType *blk, bool needsExclusive) { bool blkValid = blk && blk->isValid(); if (cpu_pkt->req->isUncacheable()) { //assert(blk == NULL); return NULL; } if (!blkValid && (cpu_pkt->cmd == MemCmd::Writeback || cpu_pkt->isUpgrade())) { // Writebacks that weren't allocated in access() and upgrades // from upper-level caches that missed completely just go // through. return NULL; } assert(cpu_pkt->needsResponse()); MemCmd cmd; // @TODO make useUpgrades a parameter. // Note that ownership protocols require upgrade, otherwise a // write miss on a shared owned block will generate a ReadExcl, // which will clobber the owned copy. const bool useUpgrades = true; if (blkValid && useUpgrades) { // only reason to be here is that blk is shared // (read-only) and we need exclusive assert(needsExclusive && !blk->isWritable()); cmd = cpu_pkt->isLLSC() ? MemCmd::SCUpgradeReq : MemCmd::UpgradeReq; } else { // block is invalid cmd = needsExclusive ? MemCmd::ReadExReq : MemCmd::ReadReq; } PacketPtr pkt = new Packet(cpu_pkt->req, cmd, blkSize); pkt->allocate(); return pkt; } template Tick Cache::atomicAccess(PacketPtr pkt) { int lat = hitLatency; // @TODO: make this a parameter bool last_level_cache = false; if (pkt->memInhibitAsserted()) { assert(!pkt->req->isUncacheable()); // have to invalidate ourselves and any lower caches even if // upper cache will be responding if (pkt->isInvalidate()) { BlkType *blk = tags->findBlock(pkt->getAddr()); if (blk && blk->isValid()) { tags->invalidateBlk(blk); DPRINTF(Cache, "rcvd mem-inhibited %s on 0x%x: invalidating\n", pkt->cmdString(), pkt->getAddr()); } if (!last_level_cache) { DPRINTF(Cache, "forwarding mem-inhibited %s on 0x%x\n", pkt->cmdString(), pkt->getAddr()); lat += memSidePort->sendAtomic(pkt); } } else { DPRINTF(Cache, "rcvd mem-inhibited %s on 0x%x: not responding\n", pkt->cmdString(), pkt->getAddr()); } return lat; } // should assert here that there are no outstanding MSHRs or // writebacks... that would mean that someone used an atomic // access in timing mode BlkType *blk = NULL; PacketList writebacks; if (!access(pkt, blk, lat, writebacks)) { // MISS PacketPtr bus_pkt = getBusPacket(pkt, blk, pkt->needsExclusive()); bool is_forward = (bus_pkt == NULL); if (is_forward) { // just forwarding the same request to the next level // no local cache operation involved bus_pkt = pkt; } DPRINTF(Cache, "Sending an atomic %s for %x\n", bus_pkt->cmdString(), bus_pkt->getAddr()); #if TRACING_ON CacheBlk::State old_state = blk ? blk->status : 0; #endif lat += memSidePort->sendAtomic(bus_pkt); DPRINTF(Cache, "Receive response: %s for addr %x in state %i\n", bus_pkt->cmdString(), bus_pkt->getAddr(), old_state); assert(!bus_pkt->wasNacked()); // If packet was a forward, the response (if any) is already // in place in the bus_pkt == pkt structure, so we don't need // to do anything. Otherwise, use the separate bus_pkt to // generate response to pkt and then delete it. if (!is_forward) { if (pkt->needsResponse()) { assert(bus_pkt->isResponse()); if (bus_pkt->isError()) { pkt->makeAtomicResponse(); pkt->copyError(bus_pkt); } else if (bus_pkt->isRead() || bus_pkt->cmd == MemCmd::UpgradeResp) { // we're updating cache state to allow us to // satisfy the upstream request from the cache blk = handleFill(bus_pkt, blk, writebacks); satisfyCpuSideRequest(pkt, blk); } else { // we're satisfying the upstream request without // modifying cache state, e.g., a write-through pkt->makeAtomicResponse(); } } delete bus_pkt; } } // Note that we don't invoke the prefetcher at all in atomic mode. // It's not clear how to do it properly, particularly for // prefetchers that aggressively generate prefetch candidates and // rely on bandwidth contention to throttle them; these will tend // to pollute the cache in atomic mode since there is no bandwidth // contention. If we ever do want to enable prefetching in atomic // mode, though, this is the place to do it... see timingAccess() // for an example (though we'd want to issue the prefetch(es) // immediately rather than calling requestMemSideBus() as we do // there). // Handle writebacks if needed while (!writebacks.empty()){ PacketPtr wbPkt = writebacks.front(); memSidePort->sendAtomic(wbPkt); writebacks.pop_front(); delete wbPkt; } // We now have the block one way or another (hit or completed miss) if (pkt->needsResponse()) { pkt->makeAtomicResponse(); } return lat; } template void Cache::functionalAccess(PacketPtr pkt, bool fromCpuSide) { Addr blk_addr = blockAlign(pkt->getAddr()); BlkType *blk = tags->findBlock(pkt->getAddr()); MSHR *mshr = mshrQueue.findMatch(blk_addr); pkt->pushLabel(name()); CacheBlkPrintWrapper cbpw(blk); // Note that just because an L2/L3 has valid data doesn't mean an // L1 doesn't have a more up-to-date modified copy that still // needs to be found. As a result we always update the request if // we have it, but only declare it satisfied if we are the owner. // see if we have data at all (owned or otherwise) bool have_data = blk && blk->isValid() && pkt->checkFunctional(&cbpw, blk_addr, blkSize, blk->data); // data we have is dirty if marked as such or if valid & ownership // pending due to outstanding UpgradeReq bool have_dirty = have_data && (blk->isDirty() || (mshr && mshr->inService && mshr->isPendingDirty())); bool done = have_dirty || cpuSidePort->checkFunctional(pkt) || mshrQueue.checkFunctional(pkt, blk_addr) || writeBuffer.checkFunctional(pkt, blk_addr) || memSidePort->checkFunctional(pkt); DPRINTF(Cache, "functional %s %x %s%s%s\n", pkt->cmdString(), pkt->getAddr(), (blk && blk->isValid()) ? "valid " : "", have_data ? "data " : "", done ? "done " : ""); // We're leaving the cache, so pop cache->name() label pkt->popLabel(); if (done) { pkt->makeResponse(); } else { // if it came as a request from the CPU side then make sure it // continues towards the memory side if (fromCpuSide) { memSidePort->sendFunctional(pkt); } else if (forwardSnoops && cpuSidePort->isSnooping()) { // if it came from the memory side, it must be a snoop request // and we should only forward it if we are forwarding snoops cpuSidePort->sendFunctionalSnoop(pkt); } } } ///////////////////////////////////////////////////// // // Response handling: responses from the memory side // ///////////////////////////////////////////////////// template void Cache::handleResponse(PacketPtr pkt) { Tick time = curTick() + hitLatency; MSHR *mshr = dynamic_cast(pkt->senderState); bool is_error = pkt->isError(); assert(mshr); if (pkt->wasNacked()) { //pkt->reinitFromRequest(); warn("NACKs from devices not connected to the same bus " "not implemented\n"); return; } if (is_error) { DPRINTF(Cache, "Cache received packet with error for address %x, " "cmd: %s\n", pkt->getAddr(), pkt->cmdString()); } DPRINTF(Cache, "Handling response to %x\n", pkt->getAddr()); MSHRQueue *mq = mshr->queue; bool wasFull = mq->isFull(); if (mshr == noTargetMSHR) { // we always clear at least one target clearBlocked(Blocked_NoTargets); noTargetMSHR = NULL; } // Initial target is used just for stats MSHR::Target *initial_tgt = mshr->getTarget(); BlkType *blk = tags->findBlock(pkt->getAddr()); int stats_cmd_idx = initial_tgt->pkt->cmdToIndex(); Tick miss_latency = curTick() - initial_tgt->recvTime; PacketList writebacks; if (pkt->req->isUncacheable()) { assert(pkt->req->masterId() < system->maxMasters()); mshr_uncacheable_lat[stats_cmd_idx][pkt->req->masterId()] += miss_latency; } else { assert(pkt->req->masterId() < system->maxMasters()); mshr_miss_latency[stats_cmd_idx][pkt->req->masterId()] += miss_latency; } bool is_fill = !mshr->isForward && (pkt->isRead() || pkt->cmd == MemCmd::UpgradeResp); if (is_fill && !is_error) { DPRINTF(Cache, "Block for addr %x being updated in Cache\n", pkt->getAddr()); // give mshr a chance to do some dirty work mshr->handleFill(pkt, blk); blk = handleFill(pkt, blk, writebacks); assert(blk != NULL); } // First offset for critical word first calculations int initial_offset = 0; if (mshr->hasTargets()) { initial_offset = mshr->getTarget()->pkt->getOffset(blkSize); } while (mshr->hasTargets()) { MSHR::Target *target = mshr->getTarget(); switch (target->source) { case MSHR::Target::FromCPU: Tick completion_time; if (is_fill) { satisfyCpuSideRequest(target->pkt, blk, true, mshr->hasPostDowngrade()); // How many bytes past the first request is this one int transfer_offset = target->pkt->getOffset(blkSize) - initial_offset; if (transfer_offset < 0) { transfer_offset += blkSize; } // If critical word (no offset) return first word time completion_time = tags->getHitLatency() + (transfer_offset ? pkt->finishTime : pkt->firstWordTime); assert(!target->pkt->req->isUncacheable()); assert(target->pkt->req->masterId() < system->maxMasters()); missLatency[target->pkt->cmdToIndex()][target->pkt->req->masterId()] += completion_time - target->recvTime; } else if (pkt->cmd == MemCmd::UpgradeFailResp) { // failed StoreCond upgrade assert(target->pkt->cmd == MemCmd::StoreCondReq || target->pkt->cmd == MemCmd::StoreCondFailReq || target->pkt->cmd == MemCmd::SCUpgradeFailReq); completion_time = tags->getHitLatency() + pkt->finishTime; target->pkt->req->setExtraData(0); } else { // not a cache fill, just forwarding response completion_time = tags->getHitLatency() + pkt->finishTime; if (pkt->isRead() && !is_error) { target->pkt->setData(pkt->getPtr()); } } target->pkt->makeTimingResponse(); // if this packet is an error copy that to the new packet if (is_error) target->pkt->copyError(pkt); if (target->pkt->cmd == MemCmd::ReadResp && (pkt->isInvalidate() || mshr->hasPostInvalidate())) { // If intermediate cache got ReadRespWithInvalidate, // propagate that. Response should not have // isInvalidate() set otherwise. target->pkt->cmd = MemCmd::ReadRespWithInvalidate; } cpuSidePort->respond(target->pkt, completion_time); break; case MSHR::Target::FromPrefetcher: assert(target->pkt->cmd == MemCmd::HardPFReq); if (blk) blk->status |= BlkHWPrefetched; delete target->pkt->req; delete target->pkt; break; case MSHR::Target::FromSnoop: // I don't believe that a snoop can be in an error state assert(!is_error); // response to snoop request DPRINTF(Cache, "processing deferred snoop...\n"); assert(!(pkt->isInvalidate() && !mshr->hasPostInvalidate())); handleSnoop(target->pkt, blk, true, true, mshr->hasPostInvalidate()); break; default: panic("Illegal target->source enum %d\n", target->source); } mshr->popTarget(); } if (blk) { if (pkt->isInvalidate() || mshr->hasPostInvalidate()) { tags->invalidateBlk(blk); } else if (mshr->hasPostDowngrade()) { blk->status &= ~BlkWritable; } } if (mshr->promoteDeferredTargets()) { // avoid later read getting stale data while write miss is // outstanding.. see comment in timingAccess() if (blk) { blk->status &= ~BlkReadable; } MSHRQueue *mq = mshr->queue; mq->markPending(mshr); requestMemSideBus((RequestCause)mq->index, pkt->finishTime); } else { mq->deallocate(mshr); if (wasFull && !mq->isFull()) { clearBlocked((BlockedCause)mq->index); } } // copy writebacks to write buffer while (!writebacks.empty()) { PacketPtr wbPkt = writebacks.front(); allocateWriteBuffer(wbPkt, time, true); writebacks.pop_front(); } // if we used temp block, clear it out if (blk == tempBlock) { if (blk->isDirty()) { allocateWriteBuffer(writebackBlk(blk), time, true); } blk->status &= ~BlkValid; tags->invalidateBlk(blk); } delete pkt; } template PacketPtr Cache::writebackBlk(BlkType *blk) { assert(blk && blk->isValid() && blk->isDirty()); writebacks[Request::wbMasterId]++; Request *writebackReq = new Request(tags->regenerateBlkAddr(blk->tag, blk->set), blkSize, 0, Request::wbMasterId); PacketPtr writeback = new Packet(writebackReq, MemCmd::Writeback); if (blk->isWritable()) { writeback->setSupplyExclusive(); } writeback->allocate(); std::memcpy(writeback->getPtr(), blk->data, blkSize); blk->status &= ~BlkDirty; return writeback; } template typename Cache::BlkType* Cache::allocateBlock(Addr addr, PacketList &writebacks) { BlkType *blk = tags->findVictim(addr, writebacks); if (blk->isValid()) { Addr repl_addr = tags->regenerateBlkAddr(blk->tag, blk->set); MSHR *repl_mshr = mshrQueue.findMatch(repl_addr); if (repl_mshr) { // must be an outstanding upgrade request on block // we're about to replace... assert(!blk->isWritable()); assert(repl_mshr->needsExclusive()); // too hard to replace block with transient state // allocation failed, block not inserted return NULL; } else { DPRINTF(Cache, "replacement: replacing %x with %x: %s\n", repl_addr, addr, blk->isDirty() ? "writeback" : "clean"); if (blk->isDirty()) { // Save writeback packet for handling by caller writebacks.push_back(writebackBlk(blk)); } } } return blk; } // Note that the reason we return a list of writebacks rather than // inserting them directly in the write buffer is that this function // is called by both atomic and timing-mode accesses, and in atomic // mode we don't mess with the write buffer (we just perform the // writebacks atomically once the original request is complete). template typename Cache::BlkType* Cache::handleFill(PacketPtr pkt, BlkType *blk, PacketList &writebacks) { Addr addr = pkt->getAddr(); #if TRACING_ON CacheBlk::State old_state = blk ? blk->status : 0; #endif if (blk == NULL) { // better have read new data... assert(pkt->hasData()); // need to do a replacement blk = allocateBlock(addr, writebacks); if (blk == NULL) { // No replaceable block... just use temporary storage to // complete the current request and then get rid of it assert(!tempBlock->isValid()); blk = tempBlock; tempBlock->set = tags->extractSet(addr); tempBlock->tag = tags->extractTag(addr); DPRINTF(Cache, "using temp block for %x\n", addr); } else { int id = pkt->req->masterId(); tags->insertBlock(pkt->getAddr(), blk, id); } // starting from scratch with a new block blk->status = 0; } else { // existing block... probably an upgrade assert(blk->tag == tags->extractTag(addr)); // either we're getting new data or the block should already be valid assert(pkt->hasData() || blk->isValid()); // don't clear block status... if block is already dirty we // don't want to lose that } blk->status |= BlkValid | BlkReadable; if (!pkt->sharedAsserted()) { blk->status |= BlkWritable; // If we got this via cache-to-cache transfer (i.e., from a // cache that was an owner) and took away that owner's copy, // then we need to write it back. Normally this happens // anyway as a side effect of getting a copy to write it, but // there are cases (such as failed store conditionals or // compare-and-swaps) where we'll demand an exclusive copy but // end up not writing it. if (pkt->memInhibitAsserted()) blk->status |= BlkDirty; } DPRINTF(Cache, "Block addr %x moving from state %i to %i\n", addr, old_state, blk->status); // if we got new data, copy it in if (pkt->isRead()) { std::memcpy(blk->data, pkt->getPtr(), blkSize); } blk->whenReady = pkt->finishTime; return blk; } ///////////////////////////////////////////////////// // // Snoop path: requests coming in from the memory side // ///////////////////////////////////////////////////// template void Cache:: doTimingSupplyResponse(PacketPtr req_pkt, uint8_t *blk_data, bool already_copied, bool pending_inval) { // timing-mode snoop responses require a new packet, unless we // already made a copy... PacketPtr pkt = already_copied ? req_pkt : new Packet(req_pkt); assert(req_pkt->isInvalidate() || pkt->sharedAsserted()); pkt->allocate(); pkt->makeTimingResponse(); if (pkt->isRead()) { pkt->setDataFromBlock(blk_data, blkSize); } if (pkt->cmd == MemCmd::ReadResp && pending_inval) { // Assume we defer a response to a read from a far-away cache // A, then later defer a ReadExcl from a cache B on the same // bus as us. We'll assert MemInhibit in both cases, but in // the latter case MemInhibit will keep the invalidation from // reaching cache A. This special response tells cache A that // it gets the block to satisfy its read, but must immediately // invalidate it. pkt->cmd = MemCmd::ReadRespWithInvalidate; } memSidePort->respond(pkt, curTick() + hitLatency); } template void Cache::handleSnoop(PacketPtr pkt, BlkType *blk, bool is_timing, bool is_deferred, bool pending_inval) { // deferred snoops can only happen in timing mode assert(!(is_deferred && !is_timing)); // pending_inval only makes sense on deferred snoops assert(!(pending_inval && !is_deferred)); assert(pkt->isRequest()); // the packet may get modified if we or a forwarded snooper // responds in atomic mode, so remember a few things about the // original packet up front bool invalidate = pkt->isInvalidate(); bool M5_VAR_USED needs_exclusive = pkt->needsExclusive(); if (forwardSnoops) { // first propagate snoop upward to see if anyone above us wants to // handle it. save & restore packet src since it will get // rewritten to be relative to cpu-side bus (if any) bool alreadyResponded = pkt->memInhibitAsserted(); if (is_timing) { Packet snoopPkt(pkt, true); // clear flags snoopPkt.setExpressSnoop(); snoopPkt.senderState = new ForwardResponseRecord(pkt, this); cpuSidePort->sendTimingSnoopReq(&snoopPkt); if (snoopPkt.memInhibitAsserted()) { // cache-to-cache response from some upper cache assert(!alreadyResponded); pkt->assertMemInhibit(); } else { delete snoopPkt.senderState; } if (snoopPkt.sharedAsserted()) { pkt->assertShared(); } } else { cpuSidePort->sendAtomicSnoop(pkt); if (!alreadyResponded && pkt->memInhibitAsserted()) { // cache-to-cache response from some upper cache: // forward response to original requester assert(pkt->isResponse()); } } } if (!blk || !blk->isValid()) { return; } // we may end up modifying both the block state and the packet (if // we respond in atomic mode), so just figure out what to do now // and then do it later bool respond = blk->isDirty() && pkt->needsResponse(); bool have_exclusive = blk->isWritable(); if (pkt->isRead() && !invalidate) { assert(!needs_exclusive); pkt->assertShared(); int bits_to_clear = BlkWritable; const bool haveOwnershipState = true; // for now if (!haveOwnershipState) { // if we don't support pure ownership (dirty && !writable), // have to clear dirty bit here, assume memory snarfs data // on cache-to-cache xfer bits_to_clear |= BlkDirty; } blk->status &= ~bits_to_clear; } DPRINTF(Cache, "snooped a %s request for addr %x, %snew state is %i\n", pkt->cmdString(), blockAlign(pkt->getAddr()), respond ? "responding, " : "", invalidate ? 0 : blk->status); if (respond) { assert(!pkt->memInhibitAsserted()); pkt->assertMemInhibit(); if (have_exclusive) { pkt->setSupplyExclusive(); } if (is_timing) { doTimingSupplyResponse(pkt, blk->data, is_deferred, pending_inval); } else { pkt->makeAtomicResponse(); pkt->setDataFromBlock(blk->data, blkSize); } } else if (is_timing && is_deferred) { // if it's a deferred timing snoop then we've made a copy of // the packet, and so if we're not using that copy to respond // then we need to delete it here. delete pkt; } // Do this last in case it deallocates block data or something // like that if (invalidate) { tags->invalidateBlk(blk); } } template void Cache::snoopTiming(PacketPtr pkt) { // Note that some deferred snoops don't have requests, since the // original access may have already completed if ((pkt->req && pkt->req->isUncacheable()) || pkt->cmd == MemCmd::Writeback) { //Can't get a hit on an uncacheable address //Revisit this for multi level coherence return; } BlkType *blk = tags->findBlock(pkt->getAddr()); Addr blk_addr = blockAlign(pkt->getAddr()); MSHR *mshr = mshrQueue.findMatch(blk_addr); // Let the MSHR itself track the snoop and decide whether we want // to go ahead and do the regular cache snoop if (mshr && mshr->handleSnoop(pkt, order++)) { DPRINTF(Cache, "Deferring snoop on in-service MSHR to blk %x\n", blk_addr); if (mshr->getNumTargets() > numTarget) warn("allocating bonus target for snoop"); //handle later return; } //We also need to check the writeback buffers and handle those std::vector writebacks; if (writeBuffer.findMatches(blk_addr, writebacks)) { DPRINTF(Cache, "Snoop hit in writeback to addr: %x\n", pkt->getAddr()); //Look through writebacks for any non-uncachable writes, use that if (writebacks.size()) { // We should only ever find a single match assert(writebacks.size() == 1); mshr = writebacks[0]; assert(!mshr->isUncacheable()); assert(mshr->getNumTargets() == 1); PacketPtr wb_pkt = mshr->getTarget()->pkt; assert(wb_pkt->cmd == MemCmd::Writeback); assert(!pkt->memInhibitAsserted()); pkt->assertMemInhibit(); if (!pkt->needsExclusive()) { pkt->assertShared(); // the writeback is no longer the exclusive copy in the system wb_pkt->clearSupplyExclusive(); } else { // if we're not asserting the shared line, we need to // invalidate our copy. we'll do that below as long as // the packet's invalidate flag is set... assert(pkt->isInvalidate()); } doTimingSupplyResponse(pkt, wb_pkt->getPtr(), false, false); if (pkt->isInvalidate()) { // Invalidation trumps our writeback... discard here markInService(mshr); delete wb_pkt; } } // writebacks.size() } // If this was a shared writeback, there may still be // other shared copies above that require invalidation. // We could be more selective and return here if the // request is non-exclusive or if the writeback is // exclusive. handleSnoop(pkt, blk, true, false, false); } template bool Cache::CpuSidePort::recvTimingSnoopResp(PacketPtr pkt) { // Express snoop responses from master to slave, e.g., from L1 to L2 cache->timingAccess(pkt); return true; } template Tick Cache::snoopAtomic(PacketPtr pkt) { if (pkt->req->isUncacheable() || pkt->cmd == MemCmd::Writeback) { // Can't get a hit on an uncacheable address // Revisit this for multi level coherence return hitLatency; } BlkType *blk = tags->findBlock(pkt->getAddr()); handleSnoop(pkt, blk, false, false, false); return hitLatency; } template MSHR * Cache::getNextMSHR() { // Check both MSHR queue and write buffer for potential requests MSHR *miss_mshr = mshrQueue.getNextMSHR(); MSHR *write_mshr = writeBuffer.getNextMSHR(); // Now figure out which one to send... some cases are easy if (miss_mshr && !write_mshr) { return miss_mshr; } if (write_mshr && !miss_mshr) { return write_mshr; } if (miss_mshr && write_mshr) { // We have one of each... normally we favor the miss request // unless the write buffer is full if (writeBuffer.isFull() && writeBuffer.inServiceEntries == 0) { // Write buffer is full, so we'd like to issue a write; // need to search MSHR queue for conflicting earlier miss. MSHR *conflict_mshr = mshrQueue.findPending(write_mshr->addr, write_mshr->size); if (conflict_mshr && conflict_mshr->order < write_mshr->order) { // Service misses in order until conflict is cleared. return conflict_mshr; } // No conflicts; issue write return write_mshr; } // Write buffer isn't full, but need to check it for // conflicting earlier writeback MSHR *conflict_mshr = writeBuffer.findPending(miss_mshr->addr, miss_mshr->size); if (conflict_mshr) { // not sure why we don't check order here... it was in the // original code but commented out. // The only way this happens is if we are // doing a write and we didn't have permissions // then subsequently saw a writeback (owned got evicted) // We need to make sure to perform the writeback first // To preserve the dirty data, then we can issue the write // should we return write_mshr here instead? I.e. do we // have to flush writes in order? I don't think so... not // for Alpha anyway. Maybe for x86? return conflict_mshr; } // No conflicts; issue read return miss_mshr; } // fall through... no pending requests. Try a prefetch. assert(!miss_mshr && !write_mshr); if (prefetcher && !mshrQueue.isFull()) { // If we have a miss queue slot, we can try a prefetch PacketPtr pkt = prefetcher->getPacket(); if (pkt) { Addr pf_addr = blockAlign(pkt->getAddr()); if (!tags->findBlock(pf_addr) && !mshrQueue.findMatch(pf_addr) && !writeBuffer.findMatch(pf_addr)) { // Update statistic on number of prefetches issued // (hwpf_mshr_misses) assert(pkt->req->masterId() < system->maxMasters()); mshr_misses[pkt->cmdToIndex()][pkt->req->masterId()]++; // Don't request bus, since we already have it return allocateMissBuffer(pkt, curTick(), false); } else { // free the request and packet delete pkt->req; delete pkt; } } } return NULL; } template PacketPtr Cache::getTimingPacket() { MSHR *mshr = getNextMSHR(); if (mshr == NULL) { return NULL; } // use request from 1st target PacketPtr tgt_pkt = mshr->getTarget()->pkt; PacketPtr pkt = NULL; if (tgt_pkt->cmd == MemCmd::SCUpgradeFailReq || tgt_pkt->cmd == MemCmd::StoreCondFailReq) { // SCUpgradeReq or StoreCondReq saw invalidation while queued // in MSHR, so now that we are getting around to processing // it, just treat it as if we got a failure response pkt = new Packet(tgt_pkt); pkt->cmd = MemCmd::UpgradeFailResp; pkt->senderState = mshr; pkt->firstWordTime = pkt->finishTime = curTick(); handleResponse(pkt); return NULL; } else if (mshr->isForwardNoResponse()) { // no response expected, just forward packet as it is assert(tags->findBlock(mshr->addr) == NULL); pkt = tgt_pkt; } else { BlkType *blk = tags->findBlock(mshr->addr); if (tgt_pkt->cmd == MemCmd::HardPFReq) { // It might be possible for a writeback to arrive between // the time the prefetch is placed in the MSHRs and when // it's selected to send... if so, this assert will catch // that, and then we'll have to figure out what to do. assert(blk == NULL); // We need to check the caches above us to verify that they don't have // a copy of this block in the dirty state at the moment. Without this // check we could get a stale copy from memory that might get used // in place of the dirty one. PacketPtr snoop_pkt = new Packet(tgt_pkt, true); snoop_pkt->setExpressSnoop(); snoop_pkt->senderState = mshr; cpuSidePort->sendTimingSnoopReq(snoop_pkt); if (snoop_pkt->memInhibitAsserted()) { markInService(mshr, snoop_pkt); DPRINTF(Cache, "Upward snoop of prefetch for addr %#x hit\n", tgt_pkt->getAddr()); delete snoop_pkt; return NULL; } delete snoop_pkt; } pkt = getBusPacket(tgt_pkt, blk, mshr->needsExclusive()); mshr->isForward = (pkt == NULL); if (mshr->isForward) { // not a cache block request, but a response is expected // make copy of current packet to forward, keep current // copy for response handling pkt = new Packet(tgt_pkt); pkt->allocate(); if (pkt->isWrite()) { pkt->setData(tgt_pkt->getPtr()); } } } assert(pkt != NULL); pkt->senderState = mshr; return pkt; } template Tick Cache::nextMSHRReadyTime() { Tick nextReady = std::min(mshrQueue.nextMSHRReadyTime(), writeBuffer.nextMSHRReadyTime()); if (prefetcher) { nextReady = std::min(nextReady, prefetcher->nextPrefetchReadyTime()); } return nextReady; } template void Cache::serialize(std::ostream &os) { warn("*** Creating checkpoints with caches is not supported. ***\n"); warn(" Remove any caches before taking checkpoints\n"); warn(" This checkpoint will not restore correctly and dirty data in " "the cache will be lost!\n"); // Since we don't write back the data dirty in the caches to the physical // memory if caches exist in the system we won't be able to restore // from the checkpoint as any data dirty in the caches will be lost. bool bad_checkpoint = true; SERIALIZE_SCALAR(bad_checkpoint); } template void Cache::unserialize(Checkpoint *cp, const std::string §ion) { bool bad_checkpoint; UNSERIALIZE_SCALAR(bad_checkpoint); if (bad_checkpoint) { fatal("Restoring from checkpoints with caches is not supported in the " "classic memory system. Please remove any caches before taking " "checkpoints.\n"); } } /////////////// // // CpuSidePort // /////////////// template AddrRangeList Cache::CpuSidePort::getAddrRanges() const { return cache->getAddrRanges(); } template bool Cache::CpuSidePort::recvTimingReq(PacketPtr pkt) { // always let inhibited requests through even if blocked if (!pkt->memInhibitAsserted() && blocked) { DPRINTF(Cache,"Scheduling a retry while blocked\n"); mustSendRetry = true; return false; } cache->timingAccess(pkt); return true; } template Tick Cache::CpuSidePort::recvAtomic(PacketPtr pkt) { // atomic request return cache->atomicAccess(pkt); } template void Cache::CpuSidePort::recvFunctional(PacketPtr pkt) { // functional request cache->functionalAccess(pkt, true); } template Cache:: CpuSidePort::CpuSidePort(const std::string &_name, Cache *_cache, const std::string &_label) : BaseCache::CacheSlavePort(_name, _cache, _label), cache(_cache) { } /////////////// // // MemSidePort // /////////////// template bool Cache::MemSidePort::recvTimingResp(PacketPtr pkt) { // this needs to be fixed so that the cache updates the mshr and sends the // packet back out on the link, but it probably won't happen so until this // gets fixed, just panic when it does if (pkt->wasNacked()) panic("Need to implement cache resending nacked packets!\n"); cache->handleResponse(pkt); return true; } // Express snooping requests to memside port template void Cache::MemSidePort::recvTimingSnoopReq(PacketPtr pkt) { // handle snooping requests cache->snoopTiming(pkt); } template Tick Cache::MemSidePort::recvAtomicSnoop(PacketPtr pkt) { // atomic snoop return cache->snoopAtomic(pkt); } template void Cache::MemSidePort::recvFunctionalSnoop(PacketPtr pkt) { // functional snoop (note that in contrast to atomic we don't have // a specific functionalSnoop method, as they have the same // behaviour regardless) cache->functionalAccess(pkt, false); } template void Cache::MemSidePacketQueue::sendDeferredPacket() { // if we have a response packet waiting we have to start with that if (deferredPacketReady()) { // use the normal approach from the timing port trySendTiming(); } else { // check for request packets (requests & writebacks) PacketPtr pkt = cache.getTimingPacket(); if (pkt == NULL) { // can happen if e.g. we attempt a writeback and fail, but // before the retry, the writeback is eliminated because // we snoop another cache's ReadEx. waitingOnRetry = false; } else { MSHR *mshr = dynamic_cast(pkt->senderState); waitingOnRetry = !masterPort.sendTimingReq(pkt); if (waitingOnRetry) { DPRINTF(CachePort, "now waiting on a retry\n"); if (!mshr->isForwardNoResponse()) { // we are awaiting a retry, but we // delete the packet and will be creating a new packet // when we get the opportunity delete pkt; } // note that we have now masked any requestBus and // schedSendEvent (we will wait for a retry before // doing anything), and this is so even if we do not // care about this packet and might override it before // it gets retried } else { cache.markInService(mshr, pkt); } } } // if we succeeded and are not waiting for a retry, schedule the // next send, not only looking at the response transmit list, but // also considering when the next MSHR is ready if (!waitingOnRetry) { scheduleSend(cache.nextMSHRReadyTime()); } } template Cache:: MemSidePort::MemSidePort(const std::string &_name, Cache *_cache, const std::string &_label) : BaseCache::CacheMasterPort(_name, _cache, _queue), _queue(*_cache, *this, _label), cache(_cache) { }