This patch adds some more functionality to the cpu model and the arch to
interface with the vector register file.
This change consists mainly of augmenting ThreadContexts and ExecContexts
with calls to get/set full vectors, underlying microarchitectural elements
or lanes. Those are meant to interface with the vector register file. All
classes that implement this interface also get an appropriate implementation.
This requires implementing the vector register file for the different
models using the VecRegContainer class.
This change set also updates the Result abstraction to contemplate the
possibility of having a vector as result.
The changes also affect how the remote_gdb connection works.
There are some (nasty) side effects, such as the need to define dummy
numPhysVecRegs parameter values for architectures that do not implement
vector extensions.
Nathanael Premillieu's work with an increasing number of fixes and
improvements of mine.
Change-Id: Iee65f4e8b03abfe1e94e6940a51b68d0977fd5bb
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
[ Fix RISCV build issues and CC reg free list initialisation ]
Signed-off-by: Andreas Sandberg <andreas.sandberg@arm.com>
Reviewed-on: https://gem5-review.googlesource.com/2705
When importing the cores.arm package, we currently throw an exception
if a timing model can't be imported due to a missing dependency (e.g.,
the required CPU model wasn't included in the build). This is
undesirable since it prevents other, working, timing models from being
added to the package. Wrap the import_module call in a try-except
block and skip timing models that have missing dependencies.
Change-Id: I92bab62c989f433a8a4a7bf59207d9d81b3d19e1
Signed-off-by: Andreas Sandberg <andreas.sandberg@arm.com>
Reviewed-on: https://gem5-review.googlesource.com/3946
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Maintainer: Jason Lowe-Power <jason@lowepower.com>
The ex5_LITTLE and ex5_big configs currently depend on Caches.py and
O3_ARM_v7a.py. These aren't actual dependencies since all of the
params from the caches and the old O3 model are overridden. This
changeset updates the ex5 models to derive from the base SimObjects
instead.
Change-Id: I999e73bb9cc21ad96865c1bc0dd5973faa48ab61
Signed-off-by: Andreas Sandberg <andreas.sandberg@arm.com>
Reviewed-by: Gabor Dozsa <gabor.dozsa@arm.com>
Reviewed-on: https://gem5-review.googlesource.com/3942
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Maintainer: Jason Lowe-Power <jason@lowepower.com>
MemConfig currently assumes that all callers include the its full set
of options in the command line parser. This is unnecessary and
sometimes confusing. Make most of the options optional to avoid having
to add all of them to example scripts.
Change-Id: I2d73be2454427b00db16716edcfd96a47133c888
Signed-off-by: Andreas Sandberg <andreas.sandberg@arm.com>
Reviewed-by: Gabor Dozsa <gabor.dozsa@arm.com>
Reviewed-on: https://gem5-review.googlesource.com/3940
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Maintainer: Jason Lowe-Power <jason@lowepower.com>
Ruby for ARM systems is not fully supported but certain configurations
are expected to work. This change removes the more general fail
statement and warns or fails depending on the particular
configuration.
Change-Id: Ic24799aff966ba15858b93482e0f24a8672d9483
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
Reviewed-on: https://gem5-review.googlesource.com/2905
Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
ARM systems require local exclusive monitors for the implementation of
synchronization primitives between processors. A ruby memory system
needs to forward invalidations to the local exclusive monitors to
to correctly determine their state.
Change-Id: I7bc4d0f2a5be0f4e36a25c87aa4a81a3f086fb3c
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
Reviewed-on: https://gem5-review.googlesource.com/2904
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Maintainer: Jason Lowe-Power <jason@lowepower.com>
Previously the directory covered a flat address range that always
started from address 0. This change adds a vector of address ranges
with interleaving and hashing that each directory keeps track of and
the necessary flexibility to support systems with non continuous
memory ranges.
Change-Id: I6ea1c629bdf4c5137b7d9c89dbaf6c826adfd977
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
Reviewed-on: https://gem5-review.googlesource.com/2903
Reviewed-by: Bradford Beckmann <brad.beckmann@amd.com>
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Maintainer: Jason Lowe-Power <jason@lowepower.com>
This patch enables using calibrated big and LITTLE cores, ex5_big and
ex5_LITTLE instead of the default 'arm_detailed' and 'minor' cpus. The ex5
model is based on the Samsung Exynos 5 Octa (5422) SoC. Operation and memory
hierarchy latencies have been calibrated using the lmbench micro-benchmark
suite. The preliminary validation results have been published as: 'Full-System
Simulation of big.LITTLE Multicore Architecture for Performance and Energy
Exploration', in International Symposium on Embedded Multicore/Many-core
Systems-on-Chip (MCSoC'16), Lyon, France (Sep, 2016).
From http://reviews.gem5.org/r/3666
Change-Id: I4935dee0a9222bd1bf7adfccb9443014945bb2d7
Signed-off-by: Anastasiia Butko <abutko@lbl.gov>
Signed-off-by: Pierre-Yves Péneau <pierre-yves.peneau@lirmm.fr>
Reviewed-on: https://gem5-review.googlesource.com/2464
Reviewed-by: Gabor Dozsa <gabor.dozsa@arm.com>
Maintainer: Jason Lowe-Power <jason@lowepower.com>
This was added for backwards compatability, but it adds a decent amount
of complexity.
The table below shows what CPU class name to use in place of a given
alias.
+==========+========================================================+
| Alias | CPU class |
+==========+========================================================+
| timing | TimingSimpleCPU |
| atomic | AtomicSimpleCPU |
| minor | MinorCPU |
| detailed | DrivO3CPU |
| kvm | ArmKvmCPU, ArmV8KvmCPU or X86KvmCPU, depending on arch |
| trace | TraceCPU |
+==========+========================================================+
Change-Id: I251c4f64b7869c6b64dd25b36967ae240f01ef08
Reviewed-on: https://gem5-review.googlesource.com/2940
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Maintainer: Jason Lowe-Power <jason@lowepower.com>
Add a script to demonstrate how power models can be wired to gem5
models. The script is meant as an example only and does not correlate
with any realistic implementation.
Change-Id: Ib95a74b2cb4af77a7816e3e8e89c89f3460775a1
Signed-off-by: Andreas Sandberg <andreas.sandberg@arm.com>
Reviewed-on: https://gem5-review.googlesource.com/2721
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Maintainer: Jason Lowe-Power <jason@lowepower.com>
When the change below removed the hard coded disk name for the SPARC FS
configuration, it broke the regression which had not specified a disk name.
This change adds a default disk name so that the regression will continue to
work like it used to, but preserving the effect of this other change.
commit 86a25bbcee
Author: Jakub Jermar <jakub@jermar.eu>
Date: Tue Jul 19 09:52:46 2016 -0500
config: Allow SPARC FS image to be specified on the command line
Change-Id: Ieb317b2bf573a4f2fc435d34cccd1f246c28d84c
Reviewed-on: https://gem5-review.googlesource.com/2645
Maintainer: Jason Lowe-Power <jason@lowepower.com>
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Add support for KVM in the big.LITTLE(tm) example configuration. This
replaces the --atomic option with a --cpu-type option that can be used
to switch between atomic, kvm, and timing simulation.
When running in KVM mode, the simulation script automatically assigns
separate event queues (threads) to each of the simulated CPUs. All
simulated devices, including CPU child devices (e.g., interrupt
controllers and caches), are assigned to event queue 0.
Change-Id: Ic9a3f564db91f5a3d3cb754c5a02fdd5c17d5fdf
Signed-off-by: Andreas Sandberg <andreas.sandberg@arm.com>
Reviewed-by: Curtis Dunham <curtis.dunham@arm.com>
Reviewed-by: Sascha Bischoff <sascha.bischoff@arm.com>
Reviewed-by: Gabor Dozsa <gabor.dozsa@arm.com>
Reviewed-on: https://gem5-review.googlesource.com/2561
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Reviewed-by: Weiping Liao <weipingliao@google.com>
The vanilla bL configuration file and the dist-gem5 configuration file
use slightly different code paths when restoring from
checkpoints. Unify this by passing the parsed options to the
instantiate() method and adding an optional checkpoint keyword
argument for checkpoint directories (only used by the dist-gem5
script).
Change-Id: I9943ec10bd7a256465e29c8de571142ec3fbaa0e
Signed-off-by: Andreas Sandberg <andreas.sandberg@arm.com>
Reviewed-by: Curtis Dunham <curtis.dunham@arm.com>
Reviewed-by: Sascha Bischoff <sascha.bischoff@arm.com>
Reviewed-by: Gabor Dozsa <gabor.dozsa@arm.com>
Reviewed-on: https://gem5-review.googlesource.com/2560
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Reviewed-by: Weiping Liao <weipingliao@google.com>
The EIOProcess class was removed recently and it was the only other class
which derived from Process. Since every Process invocation is also a
LiveProcess invocation, it makes sense to simplify the organization by
combining the fields from LiveProcess into Process.
This patch extends the example big.LITTLE configuration to enable
dist-gem5 simulations of big.LITTLE systems.
Change-Id: I49c095ab3c737b6a082f7c6f15f514c269217756
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
This patch prevents the body of the script getting executed when
the script is imported as a module.
Change-Id: I70a50f6295f1e7a088398017f5fa9d06fe90476a
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
This patch prepares future extensions and customisation of the example
big.LITTLE configuration script. It breaks out the major phases into
functions so they can be called from other python scripts.
Change-Id: I2cb7c207c410fe14602cf17af7482719abba6c24
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
A KVM VM is typically a child of the System object already, but for
solving future issues with configuration graph resolution, the most
logical way to keep track of this object is for it to be an actual
parameter of the System object.
Change-Id: I965ded22203ff8667db9ca02de0042ff1c772220
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
Names of DRAM configurations were updated to reflect both
the channel and device data width.
Previous naming format was:
<DEVICE_TYPE>_<DATA_RATE>_<CHANNEL_WIDTH>
The following nomenclature is now used:
<DEVICE_TYPE>_<DATA_RATE>_<n>x<w>
where n = The number of devices per rank on the channel
x = Device width
Total channel width can be calculated by n*w
Example:
A 64-bit DDR4, 2400 channel consisting of 4-bit devices:
n = 16
w = 4
The resulting configuration name is:
DDR4_2400_16x4
Updated scripts to match new naming convention.
Added unique configurations for DDR4 for:
1) 16x4
2) 8x8
3) 4x16
Change-Id: Ibd7f763b7248835c624309143cb9fc29d56a69d1
Reviewed-by: Radhika Jagtap <radhika.jagtap@arm.com>
Reviewed-by: Curtis Dunham <curtis.dunham@arm.com>
The current TLM bridge only provides a Slave Port that allows the gem5
world to send request to the SystemC world. This patch series refractors
and cleans up the existing code, and adds a Master Port that allows the
SystemC world to send requests to the gem5 world.
This patch:
* Restructure the existing sources in preparation of the addition of the
* new
Master Port.
* Refractor names to allow for distinction of the slave and master port.
* Replace the Makefile by a SConstruct.
Testing Done: The examples provided in util/tlm (now
util/tlm/examples/slave_port) still compile and run error free.
Reviewed at http://reviews.gem5.org/r/3527/
Signed-off-by: Jason Lowe-Power <jason@lowepower.com>
Bugfix for Elastic Traces
This patch fixes the bug when elastic traces are used:
build/ARM/gem5.opt \
configs/example/fs.py \
--cpu-type=arm_detailed \
--num-cpu=1 \
--mem-type=SimpleMemory \
--mem-size=512MB \
--mem-channels=1 \
--caches \
--elastic-trace-en \
--data-trace-file=data.proto.gz \
--inst-trace-file=inst.proto.gz \
--machine-type=VExpress_EMM \
--dtb-filename=vexpress.aarch32.ll_20131205.0-gem5.1cpu.dtb \
--kernel=vmlinux.aarch32.ll_20131205.0-gem5 \
--disk-image=linux-aarch32-ael.img
NameError: global name 'CpuConfig' is not defined
Signed-off by: Jason Lowe-Power <jason@lowepower.com>
Some configuration scripts need periodic stat dumps. One of the ways
this can be achieved is by using the pariodicStatDump helper
function. This function was previously only exported in the internal
name space. Export it as a normal function in m5.stat instead.
Change-Id: Ic88bf1fd33042a62ab436d5944d8ed778264ac98
Signed-off-by: Andreas Sandberg <andreas.sandberg@arm.com>
Reviewed-by: Sascha Bischoff <sascha.bischoff@arm.com>
This patch detects garnet network deadlock by monitoring
network interfaces. If a network interface continuously
fails to allocate virtual channels for a message, a
possible deadlock is detected.
This patch adds an IOCache to the example bigLITTLE
configuration. An IOCache is required for correct DMA
transfers when we have caches in the system.
Change-Id: Ifeddc1b360aacbb16b1393f361dd98873c834012
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
This change adds the option to use the memcheck with random memory
hierarchies at the moment limited to a maximum depth of 3 allowing
testing with uncommon topologies.
Change-Id: Id2c2fe82a8175d9a67eb4cd7f3d2e2720a809b60
Reviewed-by: Andreas Hansson <andreas.hansson@arm.com>
If the cache access mode is parallel, i.e. "sequential_access" parameter
is set to "False", tags and data are accessed in parallel. Therefore,
the hit_latency is the maximum latency between tag_latency and
data_latency. On the other hand, if the cache access mode is
sequential, i.e. "sequential_access" parameter is set to "True",
tags and data are accessed sequentially. Therefore, the hit_latency
is the sum of tag_latency plus data_latency.
Signed-off-by: Jason Lowe-Power <jason@lowepower.com>
This patch adds the ability for an application to request dist-gem5 to begin/
end synchronization using an m5 op. When toggling on sync, all nodes agree
on the next sync point based on the maximum of all nodes' ticks. CPUs are
suspended until the sync point to avoid sending network messages until sync has
been enabled. Toggling off sync acts like a global execution barrier, where
all CPUs are disabled until every node reaches the toggle off point. This
avoids tricky situations such as one node hitting a toggle off followed by a
toggle on before the other nodes hit the first toggle off.
this patch adds an ordered response buffer to the GM pipeline
to ensure in-order data delivery. the buffer is implemented as
a stl ordered map, which sorts the request in program order by
using their sequence ID. when requests return to the GM pipeline
they are marked as done. only the oldest request may be serviced
from the ordered buffer, and only if is marked as done.
the FIFO response buffers are kept and used in OoO delivery mode
This patch breaks out the most basic configuration options into a set
of base options, to allow them to be used also by scripts that do not
involve any ISA, and thus no actual CPUs or devices.
The patch also fixes a few modules so that they can be imported in a
NULL build, and avoid dragging in FSConfig every time Options is
imported.
Modify the opClass assigned to AArch64 FP instructions from SimdFloat* to
Float*. Also create the FloatMemRead and FloatMemWrite opClasses, which
distinguishes writes to the INT and FP register banks.
Change the latency of (Simd)FloatMultAcc to 5, based on the Cortex-A72,
where the "latency" of FMADD is 3 if the next instruction is a FMADD and
has only the augend to destination dependency, otherwise it's 7 cycles.
Signed-off-by: Jason Lowe-Power <jason@lowepower.com>
Continue along the same line as the recent patch that made the
Ruby-related config scripts Python packages and make also the
configs/common directory a package.
All affected config scripts are updated (hopefully).
Note that this change makes it apparent that the current organisation
and naming of the config directory and its subdirectories is rather
chaotic. We mix scripts that are directly invoked with scripts that
merely contain convenience functions. While it is not addressed in
this patch we should follow up with a re-organisation of the
config structure, and renaming of some of the packages.
This patch moves the addition of network options into the Ruby module
to avoid the regressions all having to add it explicitly. Doing this
exposes an issue in our current config system though, namely the fact
that addtoPath is relative to the Python script being executed. Since
both example and regression scripts use the Ruby module we would end
up with two different (relative) paths being added. Instead we take a
first step at turning the config modules into Python packages, simply
by adding a __init__.py in the configs/ruby, configs/topologies and
configs/network subdirectories.
As a result, we can now add the top-level configs directory to the
Python search path, and then use the package names in the various
modules. The example scripts are also updated, and the messy
path-deducing variations in the scripts are unified.
This patch adds port direction names to the links during topology
creation, which can be used for better printed names for the links
or for users to code up their own adaptive routing algorithms.
It also adds support for every router to have an independent latency
value to support heterogeneous topologies with the subsequent
garnet2.0 patch.
This patch makes the internal links within the network topology
unidirectional, thus allowing any deadlock-free routing algorithms to
be specified from the topology itself using weights.
This patch also renames Mesh.py and MeshDirCorners.py to
Mesh_XY.py and MeshDirCorners_XY.py (Mesh with XY routing).
It also adds a Mesh_westfirst.py and CrossbarGarnet.py topologies.
networktest is essentially a collection of synthetic traffic patterns
for the network. The protocol name and the tester having the same name
led to multiple python configuration files with the same name, adding
confusion. This patch renames networktest to garnet_synthetic_traffic,
and also adds more synthetic traffic patterns.
