Changes:
1. RiscvBareMetal
The RiscvBareMetal class and API are preserved for backwards
compatibility, but the base class RiscvFSWorkload is removed
as it inherits from the Workload class. However, most needed
functionalities are already implemented in the KernelWorkload
class
2. RiscvLinux
The RiscvLinux class is added. A dtb filename can be specified
to be loaded to the corresponding memory address.
3. HiFive, Clint, Plic, Uart8250, VirtIOMMIO
Devicetree node generation function is added.
4. tlb, faults
Unnecessary includes of arch/riscv/fs_workload are removed.
Change-Id: Ia239b5614bd93d8e794330ead266f6121a4d13cb
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/42053
Maintainer: Bobby R. Bruce <bbruce@ucdavis.edu>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Ayaz Akram <yazakram@ucdavis.edu>
Reviewed-by: Bobby R. Bruce <bbruce@ucdavis.edu>
Changed format from yaml to plain python. The new py configuration
file, when provided, must specialize the CHI node types defined in
configs/ruby/CHI_config.py (moved from configs/ruby/CHI.py). This
is required in order to setup the node->router bindings when the
CustomMesh topology is used.
See configs/example/noc_config/2x4.py (replaces
configs/example/noc_config/2x4.yaml) for an example.
--noc-config was also renamed to --chi-config, since the CHI node types
can be fully specialized in the configuration file.
Change-Id: Ic0c5407dba3d2483d5c30634c115b5410a5228fd
Signed-off-by: Tiago Mück <tiago.muck@arm.com>
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/43123
Reviewed-by: Jason Lowe-Power <power.jg@gmail.com>
Maintainer: Jason Lowe-Power <power.jg@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
This patch add a new Ruby cache coherence protocol based on Arm' AMBA5
CHI specification. The CHI protocol defines and implements two state
machine types:
- Cache_Controller: generic cache controller that can be configured as:
- Top-level L1 I/D cache
- A intermediate level (L2, L3, ...) private or shared cache
- A CHI home node (i.e. the point of coherence of the system and
has the global directory)
- A DMA requester
- Memory_Controller: implements a CHI slave node and interfaces with
gem5 memory controller. This controller has the functionality of a
Directory_Controller on the other Ruby protocols, except it doesn't
have a directory.
The Cache_Controller has multiple cache allocation/deallocation
parameters to control the clusivity with respect to upstream caches.
Allocation can be completely disabled to use Cache_Controller as a
DMA requester or as a home node without a shared LLC.
The standard configuration file configs/ruby/CHI.py provides a
'create_system' compatible with configs/example/fs.py and
configs/example/se.py and creates a system with private L1/L2 caches
per core and a shared LLC at the home nodes. Different cache topologies
can be defined by modifying 'create_system' or by creating custom
scripts using the structures defined in configs/ruby/CHI.py.
This patch also includes the 'CustomMesh' topology script to be used
with CHI. CustomMesh generates a 2D mesh topology with the placement
of components manually defined in a separate configuration file using
the --noc-config parameter.
The example in configs/example/noc_config/2x4.yaml creates a simple 2x4
mesh. For example, to run a SE mode simulation, with 4 cores,
4 mem ctnrls, and 4 home nodes (L3 caches):
build/ARM/gem5.opt configs/example/se.py \
--cmd 'tests/test-progs/hello/bin/arm/linux/hello' \
--ruby --num-cpus=4 --num-dirs=4 --num-l3caches=4 \
--topology=CustomMesh --noc-config=configs/example/noc_config/2x4.yaml
If one doesn't care about the component placement on the interconnect,
the 'Crossbar' and 'Pt2Pt' may be used and they do not require the
--noc-config option.
Additional authors:
Joshua Randall <joshua.randall@arm.com>
Pedro Benedicte <pedro.benedicteillescas@arm.com>
Tuan Ta <tuan.ta2@arm.com>
JIRA: https://gem5.atlassian.net/browse/GEM5-908
Change-Id: I856524b0afd30842194190f5bd69e7e6ded906b0
Signed-off-by: Tiago Mück <tiago.muck@arm.com>
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/42563
Reviewed-by: Jason Lowe-Power <power.jg@gmail.com>
Maintainer: Jason Lowe-Power <power.jg@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
The WalkCache stats are wrongly using the legacy framework.
With this patch we are registering those to the hierarchical structure.
As we need to pass the Stats::Group parent at construction time,
we are replacing 2d arrays with Vector2d for count stats and using a flat
vector of pointers for the Formula stats
Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Change-Id: I8992bc262a376e4e81a4d608c11dff6902e0a01d
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/42984
Maintainer: Jason Lowe-Power <power.jg@gmail.com>
Reviewed-by: Jason Lowe-Power <power.jg@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Added takeover methods for PMA Checker and RiscvTLB to ensure
that checkpoint restoration works. Also added logic in PLIC
to prevent posting interrupts to a CPU that has yet to complete
the current interrupt. PLIC's behaviour when a CPU claims another
interrupt before completion is also changed. Now PLIC will return
the uncompleted interrupt ID instead of return 0. This behaviour
is not documented in the specs but is designed this way to avoid
issues from CPU side (especially MinorCPU).
Change-Id: I68eaaf56d2c4d76cc1e0a1e2160f5abe184c2cd5
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/41933
Maintainer: Bobby R. Bruce <bbruce@ucdavis.edu>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Ayaz Akram <yazakram@ucdavis.edu>
We were originally generating default create() methods along side the
pybind definitions, but unfortunately those are only included when
python support is included. Since the SimObject Param structs are
unconditionally provided even if the thing calling their create()
methods is not, we need to also unconditionally provide the default
create() definitions. We do that by putting them in their own new .cc
files.
Change-Id: I29d1573d578794b3fe7ec2bc16ef5c8c58e56d0e
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/42589
Maintainer: Gabe Black <gabe.black@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Jason Lowe-Power <power.jg@gmail.com>
Reviewed-by: Earl Ou <shunhsingou@google.com>
This change:
systemc: remove pipe through flag in TLM extension
applied a change from gem5 proper in the gem5/TLM bridge improperly
here, adding a reference to a member variable that didn't exist. This
change removes the extra and invalid level of indirection to get things
to build again.
Change-Id: I77ffdb5408525e116d414df2095a944c58e40b4d
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/42586
Reviewed-by: Bobby R. Bruce <bbruce@ucdavis.edu>
Reviewed-by: Earl Ou <shunhsingou@google.com>
Maintainer: Bobby R. Bruce <bbruce@ucdavis.edu>
Tested-by: kokoro <noreply+kokoro@google.com>
This is in the Accellera version of systemc, used when gem5 is built
into systemc as a black box. std::gets was depdecated in c++11, and
removed entirely in c++14. Since gem5 has moved to c++14, this can't be
brought in with "using" in that header. More recent versions of systemc
from Accellera may have this fixed, and it's pretty bad practice to have
a universal header file squash unrelated namespaces like that in the
first place.
Rather tha update all of this copy of systemc, this change just makes
the smallest adjustment possible to get things to work again.
Change-Id: I8a95665d4b5d49ffc014097714cf2e602bf9b937
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/42583
Reviewed-by: Bobby R. Bruce <bbruce@ucdavis.edu>
Maintainer: Bobby R. Bruce <bbruce@ucdavis.edu>
Tested-by: kokoro <noreply+kokoro@google.com>
Due the compute unit pipeline being executed in reverse order, there
exists a scenario where a compute unit will execute an extra
instruction when it's supposed to be stopped at a barrier. It occurs
as follows:
* The ScheduleStage sets a barrier instruction ready to execute.
* The ScoreboardCheckStage adds another instruction to the readyList.
This is where the barrier is checked, but because the barrier isn't
executing yet, the instruction can be passed along to ScheduleStage
* The barrier executes, and stalls
* The ScheduleStage sees that there's a new instruction and schedules
it to be executed.
* Only now will the ScoreboardCheckStage realize a barrier is active
and stall accordingly
* The subsequent instruction executes
This patch sets the wavefront status to be S_BARRIER in ScheduleStage
instead of in the barrier instruction execution in order to have
ScoreboardCheckStage realize that we're going to execute a barrier,
preventing it from marking another instruciton as ready.
Change-Id: Ib683e2c68f361d7ee60a3beaf53b4b6c888c9f8d
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/41573
Reviewed-by: Matt Sinclair <mattdsinclair@gmail.com>
Reviewed-by: Alexandru Duțu <alexandru.dutu@amd.com>
Maintainer: Matt Sinclair <mattdsinclair@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
This change adds three functions, a `children` function which will
iterate through all of the children of group based (optionally) on some
predicate. Then, it implements a `find` function and a `find_re`
function using the `children` function.
The `find` function allows users to match statistics or groups
within a group. For instance, you might want to find all of the groups
within the system which have the name "cpu{i}". This is useful for
aggregate statistic values across multiple components.
Example:
total_instruuctions = sum([cpu.exec_context.thread_0.numInsts.value
for cpu in simstat.system.find('cpu')])
The find function matches based on substring. If the name given the find
function is a substring of the stat name or the group name the
stat/group will be returned.
The `find_re` function is the same as find, but matches a regular
expression instead of a simple substring match.
Note: this was originally reviewed on
https://gem5-review.googlesource.com/c/public/gem5/+/41603 was rebased
incorrectly before merging. This change fixes the rebase and adds back
the children() and re_find() functions.
Change-Id: Idaa1e9efc56fd26de3285d3fa505087ddd78ac8a
Signed-off-by: Jason Lowe-Power <jason@lowepower.com>
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/42014
Maintainer: Jason Lowe-Power <power.jg@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
This change adds three functions, a `children` function which will
iterate through all of the children of group based (optionally) on some
predicate. Then, it implements a `find` function and a `find_re`
function using the `children` function.
The `find` function allows users to match statistics or groups
within a group. For instance, you might want to find all of the groups
within the system which have the name "cpu{i}". This is useful for
aggregate statistic values across multiple components.
Example:
total_instruuctions = sum([cpu.exec_context.thread_0.numInsts.value
for cpu in simstat.system.find('cpu')])
The find function matches based on substring. If the name given the find
function is a substring of the stat name or the group name the
stat/group will be returned.
The `find_re` function is the same as find, but matches a regular
expression instead of a simple substring match.
Change-Id: I31c2a029d8a6b1d97225ab4efa34a4d13147ea32
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/41603
Reviewed-by: Bobby R. Bruce <bbruce@ucdavis.edu>
Maintainer: Bobby R. Bruce <bbruce@ucdavis.edu>
Tested-by: kokoro <noreply+kokoro@google.com>
Splitting hw_prefetches into prefetch_hits and prefetch_misses so both
events can be tracked separately. Also added appropriate functions to
increment stats. Renamed m_prefetches for consistency.
sw_prefetches is not used and has been removed. The sequencer converts
SW prefetch requests into a RubyRequestType_LD/RubyRequestType_ST
which are handled as demand requests by the all current protocols.
Change-Id: Iafa6b31c84843ddd1fad98fa7e5afed02b8c4b4d
Signed-off-by: Tiago Mück <tiago.muck@arm.com>
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/41816
Reviewed-by: Jason Lowe-Power <power.jg@gmail.com>
Maintainer: Jason Lowe-Power <power.jg@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
We have a fail_if in attachDriver to prevent driver from being
overwritten. However, the fail_if only checks for if the driver
is not nullptr.
Previously, in some cases driver was set to garbage, which made
the fail_if trip the first time we were assigning the driver.
This patch explicitly sets driver to nullptr in the constructor, thus
ensuring that it will be nullptr the first time we call attachDriver
Change-Id: I325f6033e785025a912e3af3888c66cee0332f40
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/41973
Reviewed-by: Matt Sinclair <mattdsinclair@gmail.com>
Maintainer: Matt Sinclair <mattdsinclair@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
This patch adds the source code for a mode of traffic generator to
generate strided access pattern to the memory. The main difference
between a stridedGen and linearGen are in the way startAddr and
nextAddr are set. In stridedGen instead of increasing the current
address by blocksize to generate nextAddr, it is increased by
strideSize. Also, the offset param is used to indicate the order
of any instances of traffic generator in an array (similar to
threadId.x in CUDA)
Change-Id: I80df414faf1c73f68e87400654675a553de0caa5
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/40515
Maintainer: Jason Lowe-Power <power.jg@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Jason Lowe-Power <power.jg@gmail.com>
When requestor id is requested, it is stripped of the System name via
the `stripSystemName` function in `system.cc`. However, there is a bug
in this code that leaves a leading ".". E.g.:
`system.cpu.mmu.dtb.walker` is stripped to `.cpu.mmu.dtb.walker`.
This patch fixes this issue.
Change-Id: I825cbc60c7f7eaa84c8a0150c30e9f2902cff6cb
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/41513
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Bobby R. Bruce <bbruce@ucdavis.edu>
Maintainer: Bobby R. Bruce <bbruce@ucdavis.edu>
This model is used to store and represent the "new" hierarchical stats
at the Python level. Over time these classes may be extended with
functions to ease in the analysis of gem5 stats. Though, for this
commit, such functions have been kept to a minimum.
`m5/pystats/loader.py` contains functions for translating the gem5 `_m5.stats`
statistics exposed via Pybind11 to the Python Stats model. For example:
```
import m5.pystats.gem5stats as gem5stats
simstat = gem5stats.get_simstat(root)
```
All the python Stats model classes inherit from JsonSerializable meaning
they can be translated to JSON. For example:
```
import m5.pystats.gem5stats as gem5stats
simstat = gem5stats.get_simstat(root)
with open('test.json', 'w') as f:
simstat.dump(f)
```
The stats have also been exposed via the python statistics API. Via
command line, a JSON output may be specified with the argument
`--stats-file json://<file path>`.
Change-Id: I253a869f6b6d8c0de4dbed708892ee0cc33c5665
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/38615
Reviewed-by: Jason Lowe-Power <power.jg@gmail.com>
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
Maintainer: Jason Lowe-Power <power.jg@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
