summaryrefslogtreecommitdiff
path: root/src
AgeCommit message (Collapse)Author
2018-07-24cpu-o3: Missing freeing the heads of DepGraph in IQ squashingHanhwi Jang
Free the squahsed instructions' heads of DepGraph in IQ squashing In a system with large register file (ex.2048), the number of DynInst hits the hardcoded limit (1500). This is caused by missing freeing the heads of DepGraph in IQ. IQ only clears out the heads when instructions reach writeback stage. If a instruction is squashed before writeback stage, its head of dependency graph, which holds the instruction's DynInstPtr, would not be cleared out. This prevents freeing the DynInst of the squahsed instruction even after it is committed. Change-Id: I05b3db93cb6ad8960183d7ae765149c7f292e5b3 Reviewed-on: https://gem5-review.googlesource.com/7481 Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Maintainer: Jason Lowe-Power <jason@lowepower.com>
2018-07-23systemc: Add a stubbed out sc_event_finder class.Gabe Black
The standard defines this class very loosely, and so there isn't much in the stub definition. Change-Id: I2f8d07927a4eb087235e345a09d5a4d4891413b5 Reviewed-on: https://gem5-review.googlesource.com/10831 Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-07-23systemc: Implement a stub version of the sc_prim class.Gabe Black
Change-Id: Iad32f6e385e62dc10491783c1e5cdd5d9bfcc3e6 Reviewed-on: https://gem5-review.googlesource.com/10830 Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-07-23systemc: Add stubbed out versions of sc_port and sc_export.Gabe Black
Change-Id: I04edb5da995212f9179eeb7a97486031eea71aff Reviewed-on: https://gem5-review.googlesource.com/10829 Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-07-23systemc: Add stubbed out versions of sc_event and related classes.Gabe Black
Change-Id: Id45c80cbb8774d8469d4df6ce7915161df977de0 Reviewed-on: https://gem5-review.googlesource.com/10828 Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-07-23systemc: Add stubbed out versions of the sc_time functions.Gabe Black
Change-Id: Ie7e3eac0382dc2ed861eaa9ea53ab11069812db8 Reviewed-on: https://gem5-review.googlesource.com/10827 Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-07-23systemc: Add the sc_nbdefs.hh header from Accellera.Gabe Black
This header defines the uint64 type alias needed for the sc_time class. Change-Id: I7793dbfb98001796c8c8fe24f69fe7868249ff85 Reviewed-on: https://gem5-review.googlesource.com/10826 Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-07-23systemc: Add a stub version of the sc_interface class.Gabe Black
Change-Id: Iad1da472e13b0e16ad4de03f456ca0a001e69b51 Reviewed-on: https://gem5-review.googlesource.com/10825 Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-07-23systemc: Hook up sc_main.Gabe Black
sc_main is exported as a python method on the SystemC_Kernel class and takes a series of string arguments. The internal c++ implementation converts those arguments into the standard argc and argv and uses them to call the standard SystemC version of that function. A weak SystemC version of sc_main is provided so that systemc will compile with or without a simulation provided version of that function. The weak version just complains and dies. Change-Id: Iad735536c37c8bc85d06cf24779f607ae4309b8b Reviewed-on: https://gem5-review.googlesource.com/10824 Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-07-23systemc: Partially implement the sc_module_name class.Gabe Black
This class is mostly implemented as defined by the spec, except that it doesn't maintain the module name stack (which doesn't yet exist). Change-Id: I05fdc4aa40fb0497b0165824baee87ebf01a7821 Reviewed-on: https://gem5-review.googlesource.com/10823 Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-07-23mem: Rename Packet::checkFunctional to trySatisfyFunctionalRobert Kovacsics
Packet::checkFunctional also wrote data to/from the packet depending on if it was read/write, respectively, which the 'check' in the name would suggest otherwise. This renames it to doFunctional, which is more suggestive. It also renames any function called checkFunctional which calls Packet::checkFunctional. These are - Bridge::BridgeMasterPort::checkFunctional - calls Packet::checkFunctional - MSHR::checkFunctional - calls Packet::checkFunctional - MSHR::TargetList::checkFunctional - calls Packet::checkFunctional - Queue<>::checkFunctional (of src/mem/cache/queue.hh, not src/cpu/minor/buffers.h) - Instantiated with Queue<WriteQueueEntry> and Queue<MSHR> - WriteQueueEntry - calls Packet::checkFunctional - WriteQueueEntry::TargetList - calls Packet::checkFunctional - MemDelay::checkFunctional - calls QueuedSlavePort/QueuedMasterPort::checkFunctional - Packet::checkFunctional - PacketQueue::checkFunctional - calls Packet::checkFunctional - QueuedSlavePort::checkFunctional - calls PacketQueue::doFunctional - QueuedMasterPort::checkFunctional - calls PacketQueue::doFunctional - SerialLink::SerialLinkMasterPort::checkFunctional - calls Packet::doFunctional Change-Id: Ieca2579c020c329040da053ba8e25820801b62c5 Reviewed-on: https://gem5-review.googlesource.com/11810 Reviewed-by: Daniel Carvalho <odanrc@yahoo.com.br> Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Maintainer: Jason Lowe-Power <jason@lowepower.com>
2018-07-20mem: Removed "using namespace std;" from src/mem/packet.ccRobert Kovacsics
To avoid unintentional variable capture, all std calls must be prefixed. These are the identifiers which are in the std namespace (according to https://en.cppreference.com/w/cpp/symbol_index), but that will remain unprefixed with this change: int8_t int16_t int32_t int64_t uint8_t uint16_t uint32_t uint64_t The (u)int types are included from the packet header file, which includes <inttypes.h>, where they occur in the global namespace. They are in the std namespace in <cinttypes>/<cstdint>. There is an occurrence of "set" in this file, which is "Packet::set" and not "std::set", so it is not prefixed with the std namespace Change-Id: I7f6c0b61b09658e224fe31a9f73150b81861d6f8 Reviewed-on: https://gem5-review.googlesource.com/11809 Reviewed-by: Daniel Carvalho <odanrc@yahoo.com.br> Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com> Maintainer: Jason Lowe-Power <jason@lowepower.com>
2018-07-19mem: Fix off-by-one error in checkFunctional, and simplify itRobert Kovacsics
There was an off-by-one error in the isRead() case, as `val_end` and `func_end` pointed to the last byte to write to (not one past the last byte), and thus `*_end - *_start` was not the length of the data to memcpy. This was correct in the case of val_start >= func_start && val_end <= func_end where `overlap_size = size`, but if it were (as the other cases suggest) `overlap_size = val_end - val_start`, then it would also be off by one. Also, the isWrite() case catered for this. I simplified the four ifs into one case which uses min/max (this is how I spotted the inconsistency). Change-Id: Ib5c5da084652e752f6baf1eec56b51b4f0f5c95c Reviewed-on: https://gem5-review.googlesource.com/11750 Reviewed-by: Daniel Carvalho <odanrc@yahoo.com.br> Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com> Maintainer: Nikos Nikoleris <nikos.nikoleris@arm.com>
2018-07-19mem-cache: Typo in comment: 'proceed' -> 'precede'Robert Kovacsics
The writebacks happen before anything below, not after. Change-Id: I7eaefbbf33aa17c496255dedd964a56118a28741 Reviewed-on: https://gem5-review.googlesource.com/11749 Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com> Maintainer: Jason Lowe-Power <jason@lowepower.com>
2018-07-17dev, arm: accept and ignore writes to GIC APRn registersCiro Santilli
Otherwise the Linux kernel v4.17 boot fails with error: Tried to write Gic cpu at offset 0xd0 Change-Id: Ie8063212c9e2b29e2e4766801b4b9538e9eccbf8 Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-on: https://gem5-review.googlesource.com/11590 Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
2018-07-16systemc: Add a stub kernel SimObject.Gabe Black
The guts of this class will be added in later changes. Change-Id: I3582c40f88f7d9ba6028a6f0a8ee5c32924a65bf Reviewed-on: https://gem5-review.googlesource.com/10822 Maintainer: Gabe Black <gabeblack@google.com> Reviewed-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
2018-07-16systemc: Add a stubbed out sc_object class.Gabe Black
Also add a SConsopt variable USE_SYSTEMC to hide systemc support until it's usable. Change-Id: Ibb37483432b147ee690a36bb5c8dd74f1c4c7ae4 Reviewed-on: https://gem5-review.googlesource.com/10821 Reviewed-by: Giacomo Travaglini <giacomo.travaglini@arm.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-07-16arch-arm: Introduce ARMv8.1 Virtual Timer System RegistersGiacomo Travaglini
Adding CNTHV_CTL_EL2, CNTHV_CVAL_EL2, CNTHV_TVAL_EL2 System Registers into the decode tree. They are currently implemented as a generic timer and produces a warning if accessed. Change-Id: I1a23035d67f95eeac49d890283e9a0d58426d504 Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com> Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-on: https://gem5-review.googlesource.com/11592 Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
2018-07-16arch-arm: Introduce RAS System RegistersGiacomo Travaglini
Adding RAS Extension System Registers into the decode tree. They are currently unimplemented and produce a warning (not failure) if accessed. Change-Id: I4baeded822c9582a2cb9d5277409b029eb00a962 Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com> Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-on: https://gem5-review.googlesource.com/11591 Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
2018-07-13cpu: Add a Python-enabled traffic generatorAndreas Sandberg
The current traffic generator relies on a configuration file that describes a small machine to generate stimuli. This configuration file is usually generated by the gem5 Python configuration. This creates an unnecessary and fragile step. This changeset introduces a Python-based trace module. When instantiated, the module exposes a start method that takes an iterable object as a parameter (e.g., a generator). The iterable object is expected to represent a list of generators that will be run one after the other. For example: system.tgen = PyTrafficGen() m5.instantiate() def trace(): yield system.tgen.createIdle(1000) yield system.tgen.createExit(0) system.tgen.start(trace()) Change-Id: I58e60ca517e86c197859f4daaa67750066abdc1c Signed-off-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com> Reviewed-on: https://gem5-review.googlesource.com/11518 Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
2018-07-13cpu: Support trace termination in BaseTrafficGenAndreas Sandberg
Make the BaseTrafficGen handle cases where getNextPacket() can't find a new packet and returns NULL. In that case, assume the generator has run out of packets and switch to the next generator. Change-Id: I5ca6ead550005812fb849ed9ce6b5007a65ddfa7 Signed-off-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com> Reviewed-on: https://gem5-review.googlesource.com/11517 Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
2018-07-13cpu: Unify error handling for address generatorsAndreas Sandberg
Unify error handling and create factory methods for address generators. Change-Id: Ic3ab705e1bb58affd498a7db176536ebc721b904 Signed-off-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com> Reviewed-on: https://gem5-review.googlesource.com/11516 Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
2018-07-13cpu: Split the traffic generator into two classesAndreas Sandberg
The traffic generator currently assumes that it is always driven from a configuration file. Split it into a base class (BaseTrafficGen) that handles basic packet generation and a derived class that implements the config handling (TrafficGen). Change-Id: I9407f04c40ad7e40a263c8d1ef29d37ff8e6f1b4 Signed-off-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com> Reviewed-on: https://gem5-review.googlesource.com/11515
2018-07-10misc: Fix BaseCPU doxygenJason Lowe-Power
Doxygen was stopping with #include "arch/null/cpu_dummy.hh" so the html for BaseCPU was only the three functions defined in the dummy CPU. This forces doxygen to skip this #include correctly. Note: The file references for base_cpu still aren't quite right, but it's better than it was. Change-Id: Ifafe247df2511caee2569d534bd29348a5ce9e8e Signed-off-by: Jason Lowe-Power <jason@lowepower.com> Reviewed-on: https://gem5-review.googlesource.com/11649 Reviewed-by: Brandon Potter <Brandon.Potter@amd.com>
2018-07-09arch-riscv: enable rudimentary fs simulationRobert
These changes enable a simple binary to be simulated in full system mode. Additionally, a new fault was implemented. It is executed once the CPU is initialized. This fault clears all interrupts and sets the pc to a reset vector. Change-Id: I50cfac91a61ba39a6ef3d38caca8794073887c88 Reviewed-on: https://gem5-review.googlesource.com/9061 Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Maintainer: Jason Lowe-Power <jason@lowepower.com>
2018-07-09arch-riscv: Fix the srlw and srliw instructions.Austin Harris
Change-Id: I14ccb0655819887db2306fee1188e1c83a991743 Signed-off-by: Austin Harris <austinharris@utexas.edu> Reviewed-on: https://gem5-review.googlesource.com/11669 Reviewed-by: Alec Roelke <alec.roelke@gmail.com> Maintainer: Alec Roelke <alec.roelke@gmail.com>
2018-06-29base: Add a M5_PUBLIC and M5_LOCAL attribute macroAndreas Sandberg
There are cases where we need to limit the symbol visibility to avoid compilation errors. This is a problem for Python code that relies on PyBind11 since recent versions enforce hidden symbols. As a consequence, classes that have member variables from PyBind11 need to be declared with the hidden attribute (or gem5 needs to be compiled with -fvisibility=hidden). Change-Id: I30e582fde494ff61ab7a596a595efc26a2952a5f Signed-off-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-by: Giacomo Travaglini <giacomo.travaglini@arm.com> Reviewed-on: https://gem5-review.googlesource.com/11513 Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
2018-06-28python: Fix call bug in @cxxMethod when override is TrueAndreas Sandberg
Change-Id: Ifa9efbd329fd01eb13100bc6690e651df2c12294 Signed-off-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-by: Javier Setoain <javier.setoain@arm.com> Reviewed-by: Giacomo Travaglini <giacomo.travaglini@arm.com> Reviewed-on: https://gem5-review.googlesource.com/11514 Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Maintainer: Jason Lowe-Power <jason@lowepower.com>
2018-06-28cpu: Remove reduntant protobuf includesAndreas Sandberg
Change-Id: Ic34b94b3a2ea951bc023cfce2d09ce304a602e41 Signed-off-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com> Reviewed-on: https://gem5-review.googlesource.com/11512 Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
2018-06-28python: Fixup incorrect syntax in PyBind argument handlerAndreas Sandberg
Change-Id: Ie81104d89b554795ec1020d5ce4edcf28795eda8 Signed-off-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com> Reviewed-on: https://gem5-review.googlesource.com/11511 Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Maintainer: Jason Lowe-Power <jason@lowepower.com>
2018-06-28mem: Add a memory delay simulatorAndreas Sandberg
Add a memory system component that delays traffic. The base functionality to delay packets is implemented in the abstract MemDelay class. This class exposes three methods that control packet delays: * delayReq(pkt) * delayResp(pkt) * delaySnoopResp(pkt) These methods should be specialized to implement delays for specific packet types. The class SimpleMemDelay uses the MemDelay base class to implement constant delays for read/write requests and responses. The intention is that these classes can be used for rapid prototyping of components that add a small fixed delay and the same throughput as the interconnect. I.e., any buffering done in the base class will be small and proportional to the introduced delay. Change-Id: I158cb85f20e32bfdbcbfed66a785b4b2dd47b628 Signed-off-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-by: Nicholas Lindsey <nicholas.lindsay@arm.com> Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com> Reviewed-on: https://gem5-review.googlesource.com/11521 Reviewed-by: Daniel Carvalho <odanrc@yahoo.com.br> Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Maintainer: Nikos Nikoleris <nikos.nikoleris@arm.com>
2018-06-28arch-arm: Fix incorrect t{0,1}sz field in TTBCRAndreas Sandberg
The t0sz and t1sz fields in TTBCR only are only three bits wide unlike aarch64 which has a 6-bit wide field. The higher bits of the aarch64-equivalent should be treated as RES0. Change-Id: I60df73105c34500c0348a44a491c117e9b28f18f Signed-off-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-by: Giacomo Travaglini <giacomo.travaglini@arm.com> Reviewed-on: https://gem5-review.googlesource.com/11589 Maintainer: Giacomo Travaglini <giacomo.travaglini@arm.com>
2018-06-28base: Add an asymmetrical Coroutine classGiacomo Travaglini
This patch is providing gem5 a Coroutine class to be used for instantiating asymmetrical coroutines. Coroutines are built on top of gem5 fibers, which makes them ucontext based. Change-Id: I7bb673a954d4a456997afd45b696933534f3e239 Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com> Reviewed-on: https://gem5-review.googlesource.com/11195 Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-06-26gpu-compute: Remove unneeded Request::setVirt callAlexandru Dutu
This sets the members of a Request object to the values they already hold, except the atomicOpFunctor which is set to nullptr. This call introduces a bug for atomics and is not useful for non-atomic requests. This changeset is also adding the wave PC and instruction sequence number to the Request object. Change-Id: I62f7b4a597483b0aa848a0cfbc72181e1063f56a Reviewed-on: https://gem5-review.googlesource.com/11549 Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Reviewed-by: Anthony Gutierrez <anthony.gutierrez@amd.com> Reviewed-by: Matt Sinclair <mattdsinclair@gmail.com> Maintainer: Anthony Gutierrez <anthony.gutierrez@amd.com>
2018-06-26python: Add support for multiplying proxies to compatible ParamNikos Nikoleris
Previously we allowed multiplications between proxy Param and compatible constants (int, long, float). This change extends this functionality and adds support for multiplying with between proxy Param and compatible proxy Param. Change-Id: I23a083881ae4d770e818895b893534767cd2472d Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-on: https://gem5-review.googlesource.com/11510 Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Maintainer: Jason Lowe-Power <jason@lowepower.com>
2018-06-26scons: Generalize building binaries.Gabe Black
Building gem5 binaries or regression test binaries needs to be done from within the make_env function which builds an environment for each flavor of build (opt, fast, debug, etc.). That makes it impossible to add new types of binaries without modifying the central SConscript. This change refactors how binaries are set up so that the class that represents them handles the details of how the binary should be built. Also, a metaclass and some lists track types of binaries and individual instances of binaries so that they can be iterated over automatically in make_env. Each new executable class can define a declare_all class function which calls declare() on individual instances. declare_all is a place to do any processing that only has to happen once (for instance specializing the environment) for a particular family of executables. Change-Id: I8a6ee9438280cd67e6c0b92ca28738a53cb16950 Reviewed-on: https://gem5-review.googlesource.com/10915 Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
2018-06-25syscall_emul: adding symlink system callMatt Sinclair
Change-Id: Iebda05c130b4d2ee8434cad1e703933bfda486c8 Reviewed-on: https://gem5-review.googlesource.com/11490 Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Maintainer: Jason Lowe-Power <jason@lowepower.com>
2018-06-25syscall_emul: adding link system callMatt Sinclair
Change-Id: If8922c2233bbe1f6fce35f64d1a44b91d2cfeed2 Reviewed-on: https://gem5-review.googlesource.com/11489 Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Maintainer: Brandon Potter <Brandon.Potter@amd.com>
2018-06-22mem-cache: Promote deferred targets on cache clean responsesNikos Nikoleris
While a cache clean operation is pending, all requests to the corresponding block get deferred. When the response of a cache clean operation is received, if the block is present and the response is not invalidating, we can service all deferred targets that didn't require writable. This change implements this functionality. Change-Id: Ief47e74d07749a6a9736ab450eb46eefa53464a2 Reviewed-on: https://gem5-review.googlesource.com/11018 Reviewed-by: Daniel Carvalho <odanrc@yahoo.com.br> Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Maintainer: Jason Lowe-Power <jason@lowepower.com>
2018-06-22mem-cache: Promote targets that don't require writableNikos Nikoleris
Until now, all deferred targets of an MSHR would be promoted together as soon as the targets were serviced. Due to the way we handle cache clean operations we might need to promote only deferred targets that don't require writable, leaving some targets as deferred. This change adds support for this selective promotion. Change-Id: I502e523dc9adbaf394955cbacea8286ab6a9b6bc Reviewed-on: https://gem5-review.googlesource.com/11017 Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Maintainer: Jason Lowe-Power <jason@lowepower.com>
2018-06-22mem-cache: Fix promoting of targets that need writableNikos Nikoleris
There are cases where a request which does not need a writable copy gets an response upgraded reponse and fills in a writable copy. When this happens, we promote deferred MSHR targets that were deferred because they needed a writable copy to service them immediately. Previously, we would uncoditionally promote deferred targets. Since the deferred targets might contain a cache invalidation operation, we have to make sure that any targets following the cache invalidation is not promoted. Change-Id: I1f7b28f7d35f84329e065c8f63117db21852365a Reviewed-on: https://gem5-review.googlesource.com/11016 Reviewed-by: Daniel Carvalho <odanrc@yahoo.com.br> Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Maintainer: Jason Lowe-Power <jason@lowepower.com>
2018-06-22mem-cache: Selectively clear downstream pendingNikos Nikoleris
Until now, all deferred targets of an MSHR would be promoted together as soon as the targets were serviced. When we promote deferred targets we also clear the downstreamPending flag. Due to the way we handle cache clean operations we might need to promote only deferred targets that don't require writable, leaving some targets as deferred. To allow for partial target promotion, this change adds support for clearing the downstreamPending only for a subset of a TargetsList. Change-Id: Id06953643ba9a975ebacc76ac10215441e264e74 Reviewed-on: https://gem5-review.googlesource.com/11015 Reviewed-by: Daniel Carvalho <odanrc@yahoo.com.br> Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Maintainer: Jason Lowe-Power <jason@lowepower.com>
2018-06-22arch-arm: AArch32 execution triggering AArch64 SW BreakGiacomo Travaglini
AArch32 Software Breakpoint (BKPT) can trigger an AArch64 fault when interprocessing if the trapping conditions are met. Change-Id: I485852ed19429f9cd928a6447a95eb6f471f189c Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com> Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-on: https://gem5-review.googlesource.com/11197 Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
2018-06-22arch-arm: BadMode checking if corresponding EL is implementedGiacomo Travaglini
The old utility function called badMode was only checking if the mode passed as an argument was a recognized mode. It was not checking if the corresponding mode/EL was implemented. That function has been renamed to unknownMode and a new badMode has been introduced. This is used by the cpsrWriteByInstruction function. In this way any try to change the execution mode won't succeed if the mode hasn't been implemented. Change-Id: Ibfe385c5465b904acc0d2eb9647710891d72c9df Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com> Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-on: https://gem5-review.googlesource.com/11196 Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
2018-06-21base: Add a class which encapsulates Fibers.Gabe Black
This class encapsulates the idea of a Fiber in such a way that other implementations can be substituted in in the future. This implementation uses the ucontext family of functions. This change also adds a new unit test which exercises the new class. It creates three new fibers which accept a sequence of other fibers to switch to, one after the other. The main test function switches to the these fibers which switch with each other and occasionally back to the main fiber. Each time a test fiber is activated, it checks against a list which shows the correct order for the fibers to run in. When the main fiber gets control, it makes sure that list has been progressed through by the correct amount. Change-Id: I1fc2afa414b51baaa91e350a4ebc791d989f0b8a Reviewed-on: https://gem5-review.googlesource.com/10935 Reviewed-by: Giacomo Travaglini <giacomo.travaglini@arm.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-06-21sim: Use the canonical way of iterating over a dictionaryAndreas Sandberg
Instead of using a convoluted getattr call, use the conventional iteritems() interface. Change-Id: I6d6bbccf865f8a0e8ff0767914157a7460099b09 Signed-off-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-on: https://gem5-review.googlesource.com/10782 Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
2018-06-21dev-arm: Use recurseDeviceTree instead of custom in platformAndreas Sandberg
The platform code uses a custom mechanism to traverse the object hierarchy when generating device trees. This is highly undesirable since this breaks for common cases such as when SimObjects are stored in a list. Change-Id: I1b968e5fa1db62f1456e3c0ac3de47ab1299e58d Signed-off-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-on: https://gem5-review.googlesource.com/10781 Reviewed-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
2018-06-21cpu: Fix bug introduced by RequestPtr type changeGiacomo Travaglini
Missing buffer allocation in mwaitAtomic. Change-Id: Ifccb6df2427df8b0daac5ee6a99e5cca0b20825e Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com> Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com> Reviewed-on: https://gem5-review.googlesource.com/11469 Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com> Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
2018-06-20base: Fix includes in AddrRangeMap header fileNikos Nikoleris
Change-Id: I4bdd6cf7c8d22219c0582ab206ec8372a4357759 Reviewed-on: https://gem5-review.googlesource.com/11429 Reviewed-by: Daniel Carvalho <odanrc@yahoo.com.br> Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Maintainer: Jason Lowe-Power <jason@lowepower.com>
2018-06-20mem-cache: Fix TempCacheBlock insertJason Lowe-Power
TempCacheBlock insert() had a different signature than the parent class which caused an error on clang. This matches the signature with default zero values. Change-Id: Ic096914497f3d17e88295c9e65a04d76fdddf365 Signed-off-by: Jason Lowe-Power <jason@lowepower.com> Reviewed-on: https://gem5-review.googlesource.com/11349 Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com> Maintainer: Nikos Nikoleris <nikos.nikoleris@arm.com>