summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2018-09-11systemc: Implement most of the sc_report_handler mechanism.Gabe Black
This doesn't include support for the deprecated integer message ids. Change-Id: I309d58df1cdc464428189eb0b7180edf41ca4f67 Reviewed-on: https://gem5-review.googlesource.com/12048 Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-09-11systemc: Ignore direct requests to mark a process as ready.Gabe Black
That might happen when a process is being marked as ready at the start of simulation. Because the process might not end up on the ready list, displacing it from the init list, excplicitly pop it off the init list as well. Change-Id: Iebf972e3e1baedec17b9b99b4da9dd44cd8e6957 Reviewed-on: https://gem5-review.googlesource.com/12047 Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-09-11systemc: Fill out process handle kill and reset mechanisms.Gabe Black
Some flags were being updated too early, making the functions think what they were about to do had already been done. Also, actually check for and throw the exception installed in a process when it's next supposed to run, and when injecting an exception schedule that other process to run immediately. Change-Id: I0856b69903699b2c66f9dc7f44942bbfe3cfdcc4 Reviewed-on: https://gem5-review.googlesource.com/12046 Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-09-11systemc: Fix a "problem" with kill/reset exceptions.Gabe Black
Despite what it says in the spec, the proc_ctrl compliance test throws a copy of the reset exception it catches, not the original. Because of that, the code in the kernel which catches the exception gets the base class, not the derived class with overridden virtual methods, etc. This happens to work for the Accellera implementation because they manipulate members of the base class itself which are preserved despite this bug. To make the test work, we imitate their implementation, even though it exposes more implementation details through the header files. Change-Id: I7ed9818c0552869ec790cb7f7bfbe365ade5e49c Reviewed-on: https://gem5-review.googlesource.com/12045 Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-09-11systemc: Implement much of sc_spawn.Gabe Black
This doesn't implement reset signals, although those aren't implemented for static processes either yet. Change-Id: I748a7f75b9b91774c91d969bc1ff5b07e1711aa3 Reviewed-on: https://gem5-review.googlesource.com/12044 Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-09-11systemc: Promote some functions to sc_export_base.Gabe Black
There are a few functions which return a generic sc_interface pointer which were (in the spec) defined to be in the interface type specific sc_export class. They don't need to be and aren't in the Accellera implementation, and without having them in the base class there's no good way to get at a generic interface pointer from an export. Change-Id: Iba692c79bf1d4f7684f28447d8b22c88ef4b804d Reviewed-on: https://gem5-review.googlesource.com/12043 Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-09-11systemc: Add a sensitivity type for exports.Gabe Black
Dynamic processes can be sensitive to exports, so we need a pending sensitivity to represent them. Change-Id: I87952f8bbb7847ad1f1d98d35cd5dd488e7b28ab Reviewed-on: https://gem5-review.googlesource.com/12042 Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-09-11systemc: Make some functions of the kernel static.Gabe Black
This makes it possible to call them without having to have a kernel instance available. The kernel is a singleton anyway, so there should only ever be a single instance of any of these values. Change-Id: I3610d60cc72e9f3114997fe63db94b96ccaac3cd Reviewed-on: https://gem5-review.googlesource.com/12041 Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-09-11systemc: Implement + and - for sc_time, and sc_max_time.Gabe Black
Change-Id: I294b63840e42e2afdef198229adc60ddbb60b9a1 Reviewed-on: https://gem5-review.googlesource.com/12040 Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-09-11systemc: Fix a couple memory errors.Gabe Black
Because events are held in vectors, it doesn't make sense to keep an iterator into the parent to keep track of where that event is for easy removal since the iterator becomes invalid when the vector is changed. The events need to be stored in a vector because systemc defines an accessor which returns that vector, and building a vector on the fly would be cumbersome. Also, make sure the Event parent pointer is set to nullptr if there isn't a parent. Change-Id: I63a676190e7747e60baaca50009161d47bfc1c54 Reviewed-on: https://gem5-review.googlesource.com/12039 Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-09-11systemc: Implement the SC_EXIT_ON_STARVATION exit mode.Gabe Black
This mode implies checking whether there's any activity left either before starting a delta cycle, or processing delta or timed notification or timeout. Change-Id: I0780a1f720cf63f3d2907b8dd28685266b52d6b4 Reviewed-on: https://gem5-review.googlesource.com/12038 Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-09-11systemc: Fix the time resolution when constructing an sc_time.Gabe Black
This is (sort of) mandated by the spec. More specifically the spec says that the systemc API for changing the time resolution can only be called once, and can only be called before a non-zero sc_time is constructed. Because sc_time can be constructed during elaboration and the gem5 version of time resolution is generally not locked down until the actual simulation starts (after elaboration), the sc_time constructor needs to call the fixing function itself to ensure that, for instance, the scaling factors for various real life time units within gem5 are initialized. Change-Id: Ied4b43659834761b55b5ae49ea62779af891d9e3 Reviewed-on: https://gem5-review.googlesource.com/12037 Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-09-11systemc: Handle event notifications scheduled before sc_start.Gabe Black
After sc_start is called, gem5 has run far enough to have an event queue to schedule the notification events on. Before then, it's still legal to request a timed notification. The scheduler should keep track of those requests, and once an event queue is available it should add them to it. Change-Id: Ie7445b1f2e616f4bd36044a09dbef9e1d12d7350 Reviewed-on: https://gem5-review.googlesource.com/12036 Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-09-11systemc: Make orphans top level objects instead of panic-ing.Gabe Black
When a simulation ends, the sc_objects it contains are destroyed one by one, not necessarily in hierarchy order. That means that a parent object can legitimately be destroyed before its children. Instead of panic-ing when that inevitably happens, this change makes gem5 turn those children into top level objects. Change-Id: Icad9c99310fbc3ddcadbbb4f8a990b4fbfe35bdf Reviewed-on: https://gem5-review.googlesource.com/12035 Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-09-11systemc: Implement some of the basics of sc_time.Gabe Black
This is a very incomplete implementation, but is enough to unblock implementing some other dependent features. Change-Id: Ibd2f3476fe01389e277c6956d48337f551d14acd Reviewed-on: https://gem5-review.googlesource.com/12034 Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-09-11systemc: Implement the various sc_module stage callbacks.Gabe Black
This change also gets rid of the SystemC namespace which was deprecated in favor of sc_gem5. A few utility functions which check whether certain callbacks have finished were also implemented. status tracking moved from a global variable in sc_main.cc to a member of the kernel simobject. Change-Id: I50967fae9c576fbe45b1faff587aaa824857a289 Reviewed-on: https://gem5-review.googlesource.com/12033 Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-09-11base: Avoid dividing by zero in Time::[gs]etTick.Gabe Black
Those functions used the value SimClock::Int::ns which, if the time resolution is larger than 1ns, can/will be zero. That will make getTick always return zero, and setTick divide by zero. This change modifies those functions so that the math they do avoids using any integer Ticks per time unit value except for Frequency. It seems unlikely that the Ticks will increment at less than 1Hz. Change-Id: I5cc9db14699c00dcbff48e4593b98522b13b4ccd Reviewed-on: https://gem5-review.googlesource.com/12573 Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-09-11base: Correct a small typo in sim/core.(hh|cc).Gabe Black
The value GHz with a small z was spelled GHZ with a large z, which was inconsistent with the other frequency-like values in that namespace. Change-Id: I55dfc447a5811ae584e46769cd9cadd08bd1e716 Reviewed-on: https://gem5-review.googlesource.com/12572 Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
2018-09-11base: Save and restore the width setting of streams used with cprintf.Gabe Black
The flags, precision, and fill character were all being saved and restored, but cprintf might also change the width setting of the stream, and that wasn't being saved or restored and could leak from the cprintf statement. This change adds the code to save and restore that value. Change-Id: Ibedb26f7f538cd3be4fe0462d2ee4e5efd62bc59 Reviewed-on: https://gem5-review.googlesource.com/12571 Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
2018-09-11net: Fix a bug when receiving fragamented packetsPin-Yen Lin
In the previous implementation, the function EtherTap::recvReal will only read one packet when received some ``interrupt'' (explicitly, when async_IO set to true). When someone tries to send a large message to the simulated device, the message will be divided to several packets due to packet fragmentation. In this situation recvReal will only read one packet and left the other packets in the buffer. This significantly increases the networking latency. So before reading from socket, I change the socket into non-blocking mode and keep reading from it until there's no packet left. Change-Id: Ieb94a8532cd3994862b6f3eb9474caf7ccf617da Reviewed-on: https://gem5-review.googlesource.com/12338 Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Maintainer: Jason Lowe-Power <jason@lowepower.com>
2018-09-10base: Fix isSubset() for addr ranges with interleavingNikos Nikoleris
This change extends isSubset() which checks whether the range is a subset of an input range to support address ranges with interleaving and hashing. Change-Id: I3dc9ceccb189b7c8665de0355f0555fc2c37d872 Reviewed-on: https://gem5-review.googlesource.com/12319 Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com> Maintainer: Nikos Nikoleris <nikos.nikoleris@arm.com>
2018-09-10configs: Use the same address ranges for dir and mem_ctrlsNikos Nikoleris
In Ruby, for every directory we create one memory controller for every range in the memory ranges. Previously the memory controllers and the directories created their address ranges independently and as a result a mismatch was possible. In fact, we assinged an interleaved address range with hasing for the memory controllers while the corresponding directories would be assigned the same interleaved address range without hashing. This change uses the address range of the memory controllers to populate the list of address ranges for the corresponding directory and avoid bugs due to code duplication. Change-Id: I1e321c81a254199e5aaa9f3b81f4a4642c60a67a Reviewed-on: https://gem5-review.googlesource.com/12318 Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Maintainer: Jason Lowe-Power <jason@lowepower.com>
2018-09-10dev-arm: Make GenericTimer use standard ArmInterruptPinGiacomo Travaglini
This patch is deleting the custom ArchTimer::Interrupt implementation in favour of the standard ArmInterruptPin. Change-Id: I5aa5661e48834398bd7aae15df9578b8db5c8da3 Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com> Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-on: https://gem5-review.googlesource.com/12402 Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
2018-09-10dev-arm: Factory SimObject for generating ArmInterruptPinGiacomo Travaglini
With this patch the python ArmInterruptPin SimObject matches to the C++ ArmInterruptPinGen. The latter is in charge of generating the ArmInterruptPin (which is not a SimObject anymore). This is meant to ease the generation of ArmInterruptPins: by not being SimObjects we are not forced to instantiate them in the configuration script; we can generate them dynamically instead throughout simulation. Change-Id: I917d73a26168447221f5993c8ae975ee3771e3bf Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com> Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-on: https://gem5-review.googlesource.com/12401 Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
2018-09-10arm: Use the interrupt adaptor in the PMUAndreas Sandberg
Rewrite interrupt handling in the PMU model to use the new interrupt adaptor. Change-Id: I2cbb99580c46d3e21a1335b897843b7b6e41f10c Signed-off-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com> Reviewed-on: https://gem5-review.googlesource.com/12400 Reviewed-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
2018-09-10arm: Add support for tracking TCs in ISA devicesAndreas Sandberg
ISA devices typically need to keep track of the thread context they are associated with. Among other things, this is required for interrupt delivery. Add a BaseISADevice:setThreadContext() method to wire such models to the right thread context. Change-Id: Iad354d176c0c4c4e34c6ab8b5acaee0b69da0406 Signed-off-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com> Reviewed-on: https://gem5-review.googlesource.com/12399 Reviewed-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
2018-09-10dev, arm: Add misc reg tracing to the generic timerAndreas Sandberg
Change-Id: Ice9376b8eb42423679b0191910e8c980f8017f88 Signed-off-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-by: Giacomo Travaglini <giacomo.travaglini@arm.com> Reviewed-by: Curtis Dunham <curtis.dunham@arm.com> Reviewed-on: https://gem5-review.googlesource.com/12398
2018-09-10dev-arm: Create a getter for ArmInterruptPin ID numberGiacomo Travaglini
A pin owner might want to know which is the irq number associated with the pin. Change-Id: I095393d4d25efe13eb2a75a0b0b055d386c2c126 Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com> Reviewed-by: Ciro Santilli <ciro.santilli@arm.com> Reviewed-on: https://gem5-review.googlesource.com/12298 Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
2018-09-07mem: Make DRAMCtrl a QoS-aware Memory ControllerMatteo Andreozzi
This patch is turning DRAMCtrl a QoS-aware Memory Controller with "no policy" as a default policy. Change-Id: I48163da8c8208498cf0398b07094cb840272507f Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com> Reviewed-on: https://gem5-review.googlesource.com/11973 Maintainer: Nikos Nikoleris <nikos.nikoleris@arm.com> Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com>
2018-09-07mem: Implement base QoS Policies.Giacomo Travaglini
This patch implements a base fixed priority policy and an ideal turnaround policy for the QoS memory controller. Change-Id: I38ce16f845fc0ec86d6fc4cc5dc5406f213a465e Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com> Reviewed-on: https://gem5-review.googlesource.com/11972 Maintainer: Nikos Nikoleris <nikos.nikoleris@arm.com> Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com>
2018-09-07mem: Add a simple QoS-aware Memory ControllerMatteo Andreozzi
This patch implements QoSMemorySink: a simple generic QoS-aware memory controller which inherits from QoS::MemCtrl. Change-Id: I537a4e2d4cb8f54fa0002eb088b2c6957afb9973 Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com> Reviewed-on: https://gem5-review.googlesource.com/11971 Maintainer: Nikos Nikoleris <nikos.nikoleris@arm.com> Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com> Reviewed-by: Matthew Poremba <porembam@gmail.com>
2018-09-07mem: Add a QoS-aware Memory Controller typeMatteo Andreozzi
This is the implementation of QoS algorithms support for gem5 memory objects. This change-list provides a framework for specifying QoS algorithm which can be used to prioritise service to specific masters in the memory controller. The QoS support implemented here is designed to be extendable so that new QoS algorithms can be easily plugged into the memory controller as "QoS Policies". Change-Id: I0b611f13fce54dd1dd444eb806f8e98afd248bd5 Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com> Reviewed-on: https://gem5-review.googlesource.com/11970 Maintainer: Nikos Nikoleris <nikos.nikoleris@arm.com> Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com>
2018-09-07sim: Add System method for MasterID lookupGiacomo Travaglini
A new method (lookupMasterId) has been added to the System. A client should use it when querying the System for the MasterID of a particular master. It changes from getMasterId since it is not registering a new MasterID if the master is not found in the master's list. Change-Id: I701158d22e235085bba9ab91154fbb702cae1467 Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com> Reviewed-on: https://gem5-review.googlesource.com/11969 Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Maintainer: Jason Lowe-Power <jason@lowepower.com>
2018-09-07mem-cache: Create Skewed Assoc placement policyDaniel R. Carvalho
Create a class that implements the skewed associative placement policy. It uses the hash function and expansions of the skewing functions described in "Skewed-Associative caches", by Seznec. Only 8 skewing functions are implemented, and therefore if more are needed a hash function will be recursively applied on top of the output of one of these functions to generate different values. This is not optimal, and if more functions are needed it might be more effective to implement them. Change-Id: Ibc77edffd8128114a8b200cec5d8deedfb5105cb Reviewed-on: https://gem5-review.googlesource.com/8886 Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com> Maintainer: Nikos Nikoleris <nikos.nikoleris@arm.com>
2018-09-07net: Fix a bug when handling IPv6 packetsPin-Yen Lin
When gem5 runs with the networking support, it will run into an assertion fail and aborted. This is because it tries to calculate checksum for IPv6 packet and this makes the IpPtr pointer ``ip'' become NULL. For that there is functions and classes for IPv6 in base/inet.cc, I added IPv6 support for i8254xGBe.cc. Because IPv6 header does not have identification number, I ignored some of the debug messages using ip->id(). Change-Id: Ida5e36aefd2c5c26053f8152a0aac24191e7757c Reviewed-on: https://gem5-review.googlesource.com/12339 Reviewed-by: Earl Ou <shunhsingou@google.com> Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-09-05systemc: Implement pending activity related functionsGabe Black
Track the number of notifications/timeouts that are scheduled at any given time. This lets us implement sc_pending_activity_at_current_time, sc_pending_activity_at_future_time, and sc_time_to_pending_activity. Change-Id: Ia3fcd29bdbfe1a6c77eb52ce4836982d4705263c Reviewed-on: https://gem5-review.googlesource.com/12032 Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-09-05systemc: Implement sc_pause, sc_stop, and re-sc_start-ing.Gabe Black
This change further modifies the scheduler to implement the sc_pause and sc_stop functions, and to ensure that calling sc_start again works. Also, some small changes were made to how processes and contexts are hooked up. Now, rather than checking whether a process is running to determine wether it started on its own or needs to be started manually, there's a bool which explicitly tracks whether it needs this step. The problem was that once a thread finished, it wasn't considered running any more. In that case it had run but finished, but that was indistinguishable from it needing to run but not having been started. Change-Id: I3aefb5493f91d9efa1a1382586196339b67925fe Reviewed-on: https://gem5-review.googlesource.com/12031 Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-09-05systemc: Implement sc_get_current_process_handle's other personality.Gabe Black
When run during simulation, sc_get_current_process_handle returns a handle for the currently running process or a invalid handle if no process is running (ie sc_main is running). When run during elaboration, it returns a handle to the most recently created process. This second context is what this change handles. Change-Id: I3fb247b9b7bf83891c782966cfef474753159158 Reviewed-on: https://gem5-review.googlesource.com/12030 Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-09-05systemc: Handle suspended processes and handle sensitivity overload.Gabe Black
This change keeps track of whether a process would have become ready but was suspended so that it can become ready when the process is resumed. Also, this makes a process ignore its static sensitivity while a dynamic sensitivity is in place. Change-Id: If3f6c62f370051e574f81bf227746db8c43527e2 Reviewed-on: https://gem5-review.googlesource.com/11715 Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-09-05systemc: Implement next_trigger and wait.Gabe Black
These set up dynamic sensitivity. Change-Id: I6cf2c2e4f93fe001eb7df8377753d51846b57d32 Reviewed-on: https://gem5-review.googlesource.com/11714 Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-09-05systemc: Implement the sensitivity mechanism.Gabe Black
This change lets processes be sensitive to events, timeouts, etc. Change-Id: If30a256dfa8a2e92192c1f9c96b48e2aa28ec27e Reviewed-on: https://gem5-review.googlesource.com/11713 Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-09-05systemc: Hook up sc_time_stamp sc_delta_count.Gabe Black
sc_time_stamp reports the current simulation time. sc_delta_count was hooked up to a dummy value. This change hooks it up to the scheduler so that it returns the real value. Change-Id: I354c4be32161eabeea86af653f5cb0a5d384645b Reviewed-on: https://gem5-review.googlesource.com/11712 Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-09-05systemc: Implement much of events, event lists and event exprs.Gabe Black
Three things aren't yet implemented, waking up processes which are sensitive to the event, triggering of events, and garbage collecting list objects which came from expression objects. The garbage collection aspect is problematic since there doesn't seem to be a correct way to implement it given the constraints in the spec, including the way that's implemented by Accellera. It's something that will need to be dealt with at some point, but in the interest of forward progress it's being ignored for now. Change-Id: Ic4e3c219ff482729f1f1302ab10181a798d48041 Reviewed-on: https://gem5-review.googlesource.com/11711 Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-09-05systemc: Implement channel updates and rework the scheduler.Gabe Black
This change implements channel updates, and also reworks the scheduler to delegate more to the gem5 event queue by taking advantage of event priorities to ensure things happen in the right order. There's a lengthy comment in scheduler.hh describes how that all works. Change-Id: I5dee71b86b2e612bb720a4429f3a72e4b7c6d01f Reviewed-on: https://gem5-review.googlesource.com/11710 Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-09-05systemc: Partially implement the scheduler.Gabe Black
This change implements the "evaluate" part of the delta cycles, and sketches out a function to run delta cycles and the initialization phase. The kernel object now schedules an event at time zero which runs the initialization phase. Also, some small places which were stubbed out pending a way to check the currently running process have been filled in now that that's being tracked. Change-Id: I6899569eb0195ff1c059fa4e68e90ef162b2f2df Reviewed-on: https://gem5-review.googlesource.com/11709 Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-09-05systemc: Fill out sc_process_handle and create Process classes.Gabe Black
The sc_process_handle class now primarily delegates to a Process object it points at. The Process object does book keeping as far as its internal state, but doesn't yet have a way to run its target function or to schedule itself or inject exceptions into its context of execution. Change-Id: I98389778abe29aa26e3e3a91bf02e6721acc8a9c Reviewed-on: https://gem5-review.googlesource.com/11613 Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-09-05systemc: Implement some basic plumbing in sc_module.hh.Gabe Black
Take care of some low hanging fruit as far as wrapper methods and the sc_bind_proxy class. Change-Id: I7f55a37eeaa82338bd608218c0261fbc39e65fc2 Reviewed-on: https://gem5-review.googlesource.com/11612 Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-09-05systemc: Implement most of sc_object.Gabe Black
To avoid making it hard to change sc_object's implementation in the future, this change keeps most of the data members out of sc_object and keeps them in a seperate Object which is managed independently but still matches to the sc_objects one to one. This change also moves away from the SystemC/sc_gem5 namespace pair in favor of sc_gem5. Having two namespaces with classes, etc, living in both was complicating things. Having to use a namespace that doesn't fit in one scheme or the other isn't great, but it's the lesser of two evils. Change-Id: Ib59c3c515ca98c7fe519c59e9fe9270304b71cc0 Reviewed-on: https://gem5-review.googlesource.com/11611 Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-09-05systemc: Get sc_main to work even when dynamically linked.Gabe Black
Unfortunately multiply defined weak symbols don't work like they're supposed to when the different versions are across dynamic linking boundaries. Fortunately, a weak symbol with no definition at all will still consistently evaluate as 0, and a singularly defined weak symbol will still resolve correctly. Instead of relying on a weak version of sc_main being overridden by a strong/non-default definition, this change leaves it as undefined and detects at run time whether the symbol resolves to 0 or to an actual code location. Change-Id: I31c4ff769b0c52277c6cf4845ca3f85000009583 Reviewed-on: https://gem5-review.googlesource.com/11610 Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Gabe Black <gabeblack@google.com>
2018-09-05systemc: Implement the sc_attr classes.Gabe Black
Change-Id: Ibbe6da957b1b36687178f226e80718adc0f4ab81 Reviewed-on: https://gem5-review.googlesource.com/11609 Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Gabe Black <gabeblack@google.com>
generated by cgit v1.2.3 (git 2.25.1) at 2024-11-17 16:58:23 +0000