This patch fixes the problem during checkpoing where the mempool is not
restored, but using only the one specified in the config file as a new
execution.
In order to fix that this changes modifyies the serialize/unserialize
functions for mempools and create new funcionts on se_workload to make
sure mempools ends up in the m5.cpt.
We change as well the unserialize mempool function to update
according the checkpoint file so the execution starts with the same
free pages and free pointers.
JIRA: https://gem5.atlassian.net/browse/GEM5-1191
Change-Id: I289bf91eb4f01d9c01a31a39b968e30f8b8d2bdc
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/56969
Reviewed-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Maintainer: Giacomo Travaglini <giacomo.travaglini@arm.com>
Tested-by: kokoro <noreply+kokoro@google.com>
This makes what are configuration and what are internal SCons variables
explicit and separate, and makes it unnecessary to call out what
variables to export to C++.
These variables will also be plumbed into and out of kconfiglib in later
changes.
Change-Id: Iaf5e098d7404af06285c421dbdf8ef4171b3f001
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/56892
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
Maintainer: Gabe Black <gabe.black@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Add a new option `auto_unlink_shared_backstore` to System so it will
remove the shared backstore used in physical memories when the System is
getting destructed. This will prevent unintended memory leak.
If the shared memory is designed to live through multiple round of
simulations, you may set the option to false to prevent the removal.
Test: Run a simulation with shared_backstore set, and see whether there
is anything left in /dev/shm/ after simulation ends.
Change-Id: I0267b643bd24e62cb7571674fe98f831c13a586d
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/57469
Reviewed-by: Daniel Carvalho <odanrc@yahoo.com.br>
Maintainer: Daniel Carvalho <odanrc@yahoo.com.br>
Tested-by: kokoro <noreply+kokoro@google.com>
When the clone syscall is called, a new process is created which
allocates a new page table. If clone was called with CLONE_THREAD, the
page table of that new process is then marked as shared. Next, initState
is called on the process which calls the page table's initState. For the
multi level page table, initState only sets the base pointer if shared
is false. This means that in this order the base pointer of the new page
table is not currently initialized causing spurious errors.
To fix this, the page table is explicitly initialized after the new
process and new page table are created but before the page table is
marked as shared. The process initState continues as normal and the new
page table's base pointer is not modified by further calls to initState
as it is already marked shared.
Change-Id: I4a533e13565fa572fb9153a926f70958bc7488b7
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/56366
Reviewed-by: Jason Lowe-Power <power.jg@gmail.com>
Maintainer: Jason Lowe-Power <power.jg@gmail.com>
Reviewed-by: Matt Sinclair <mattdsinclair@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
The contextId is generally treated as (and should be) an opaque index
into the System objects threadContext array. When forcing it to
particular values, that introduces gaps in the threadContext array which
trips up other code which is expecting the array to have only valid
entries.
Change-Id: I4997e989b436a3008f65f348722dfb843b2f110a
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/57089
Maintainer: Gabe Black <gabe.black@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Yu-hsin Wang <yuhsingw@google.com>
The KvmVM will declare itself to the System object, instead of the other
way around. This way the System object can just keep an opaque KvmVM
pointer which does not depend on the KvmVM code even being compiled into
gem5. If there is a KvmVM object, that can more safely assume there is a
corresponding System object to attach itself to.
Also move use of the KvmVM pointer out of constructors, since the VM may
not have registered itself with the System object yet. Those uses can
happen in the init() method instead.
Change-Id: Ia0842612b101315bc1af0232d7f5ae2b55a15922
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/56187
Reviewed-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Maintainer: Gabe Black <gabe.black@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
These registers used to be accessed with a two dimensional index, with
one dimension specifying the register, and the second index specifying
the element within that register. This change linearizes that index down
to one dimension, where the elements of each register are laid out one
after the other in sequence.
Change-Id: I41110f57b505679a327108369db61c826d24922e
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/49148
Reviewed-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Maintainer: Giacomo Travaglini <giacomo.travaglini@arm.com>
Tested-by: kokoro <noreply+kokoro@google.com>
This file populated the "cxx_config_directory" map from type names to
directory entry generating types. It used a comprehensive list of
includes of all SimObject headers, and a comprehensive list of the
generating types to fill everything in.
Instead, this change creates a new singleton helper class which, when
instantiated as a static member of a CxxConfigParams subclass, will
install a pointer to a CxxConfigDirectoryEntry to that map during global
object construction time.
Also, this change renames the map to cxxConfigDirectory which is in
compliance with the style guide, and puts it behind an accessor which
returns a static variable which is the actual map. This avoids any
problems that might come from global object construction order.
Change-Id: Iaa913fbe5af1b11d90ca618e29420eeb7cb0faed
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/49455
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
Tested-by: kokoro <noreply+kokoro@google.com>
The previous incarnation of this support used faults to make the CPU
reexecute the system call instruction again and again to prevent
emulating/passing through blocking system calls from blocking gem5 as
a whole. That support was accidentally removed a while ago. This new
version suspends the thread context executing the system call, and
periodically wakes it up to retry using a periodically scheduled event.
Jira Issue: https://gem5.atlassian.net/browse/GEM5-1123
Change-Id: I155fa8205d7ea45e3d102216aeca6ee1979a522f
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/54205
Reviewed-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Maintainer: Giacomo Travaglini <giacomo.travaglini@arm.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Rather than make each ISA include boilerplate to ignore a
SyscallReturn's value when it's marked as suppressed or needing a retry,
put that code into the SyscallDesc::doSyscall method instead.
That has two benefits. First, it removes a decent amount of code
duplication which is nice from a maintenance perspective. Second, it
puts the SyscallDesc in charge of figuring out what to do once a system
call implementation finishes. That will let it schedule a retry of the
system call for instance, without worrying about what the ISA is doing
with the SyscallReturn behind its back.
Jira Issue: https://gem5.atlassian.net/browse/GEM5-1123
Change-Id: I76760cba75fd23e6e3357f6169c0140bee3f01b6
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/54204
Reviewed-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Maintainer: Giacomo Travaglini <giacomo.travaglini@arm.com>
Tested-by: kokoro <noreply+kokoro@google.com>
This method depended on all of the EmbeddedPyBind objects having all
been constructed already so that it would have a complete list. This
would only be true if it was called after static intialization was
complete, which is not true if python is ready to go as soon as gem5
(in library form) is loaded.
This change makes EmbeddedPyBind able to defer initialization of a
module more generically than before, so that they can wait for either
another module to be initiailized, or the _m5 package itself.
Change-Id: I6b636524f969bd9882d8c1a7594dc36eb4e78037
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/54005
Maintainer: Bobby Bruce <bbruce@ucdavis.edu>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Bobby Bruce <bbruce@ucdavis.edu>
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/54265
For most system calls, it doesn't matter if the PC is advanced to the
instruction after the system call instruction before or after the system
call itself is invoked, because the system call itself doesn't interact
with it.
For some system calls however, specifically "clone" and "execve",
advancing the PC *after* the system call complicates things, because it
means the system call needs to set the PC to something that will equal
the desired value only *after* it's advanced.
By setting the PC *before* the system call, the system call can set the
PC to whatever it needs to. This means the new cloned context doesn't
need to advance the PC because it's already advanced, and execve doesn't
need to set NPC, it can leave the PC set to the correct value from the
entry point set during Process initialization.
Change-Id: I830607c2e9adcc22e738178fd3663417512e2e56
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/53983
Reviewed-by: Jason Lowe-Power <power.jg@gmail.com>
Maintainer: Jason Lowe-Power <power.jg@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Matt Sinclair <mattdsinclair@gmail.com>
Maintainer: Matt Sinclair <mattdsinclair@gmail.com>
Reviewed-by: Kyle Roarty <kyleroarty1716@gmail.com>
Remove the line "For use for simulation and test purposes only" in files
were AMD is the only copyright holder listed in the header. This happens
to be the case for all files where this line exists, removing it
completely from gem5.
Change-Id: I623f266b002f564301b28774f49081099cfc60fd
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/53943
Reviewed-by: Jason Lowe-Power <power.jg@gmail.com>
Maintainer: Jason Lowe-Power <power.jg@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
By separating out this utility class, we make it possible to build
embedded python modules into other binarys without dragging along lots
of other, unrelated gem5 dependencies.
Also, move the class from sim/init.hh (which is a largely unrelated
name) to python/embedded.hh which much more directly describes what that
file contains.
Change-Id: Ia83439144893ad8401a5d51003e2686d9c9b2d7b
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/49418
Reviewed-by: Gabe Black <gabe.black@gmail.com>
Maintainer: Gabe Black <gabe.black@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Put that code into a singleton class in src/proto, so that it gets
called during initialization and teardown of gem5 without cluttering up
gem5Main. This also removes the need to use #ifdefs to guard for
actualling having protobuf support.
Change-Id: I93b5d994eee478a9c159a3f3d02b3e996af02a3e
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/49416
Maintainer: Gabe Black <gabe.black@gmail.com>
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Use pybind11 to avoid having to use the python C API directly, which is
simpler, easier to read, and less error prone. Also, use its
PYBIND11_EMBEDDED_MODULE macro to set up the _m5 module instead of a
callback which has to be proactively called from main().
Change-Id: I9c8bcebea934844d16a1fdd88f66a5e66ef0486f
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/49413
Maintainer: Bobby Bruce <bbruce@ucdavis.edu>
Reviewed-by: Gabe Black <gabe.black@gmail.com>
Reviewed-by: Jason Lowe-Power <power.jg@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
m5MainCommands had been a way to override the python code which would
get the python side of gem5 started, used by some "unit" tests which
were really tests of all of gem5 but focus on a particular area. Those
tests have either been converted into real unit tests or eliminated,
and so that mechanism is no longer needed.
This change eliminates that mechanism, and also uses pybind11 to
significantly simplify the code that calls m5.main().
Change-Id: I553ae17074cd5389708f1b7313630a13a6946d76
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/49412
Reviewed-by: Gabe Black <gabe.black@gmail.com>
Maintainer: Gabe Black <gabe.black@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Previously, the importer module was built into gem5 as a compressed
bytecode blob like all the other code, and it had to be singled out and
installed manually so that it could help bring in all the other modules.
That adds some amount of complexity since it has to be identified and
treated as a special case.
Instead, this change builds it into gem5 using pybind11's
PYBIND11_EMBEDDED_MODULE macro, and a string that gets evaluated into
the new module's __dict__. This means the importer module is
automatically available just by building in that .cc, and it can just be
imported to start using it.
Theoretically all the embedded python could be handled this way, but
that would mean building it into gem5 as raw strings which wouldn't even
be compiled into byte code until run time. That would take more space in
the binary, and also delay catching simple errors.
Change-Id: Ic600bf6bce41a53289a2833484a655dd5a226e03
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/49410
Reviewed-by: Jason Lowe-Power <power.jg@gmail.com>
Maintainer: Jason Lowe-Power <power.jg@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
When the "path" argument is empty, use the file name of the node
referred to by the fd file descriptor. This matches the behavior of
"at" system calls when the TGT_AT_EMPTY_PATH flag is set. The system
calls themselves are responsible for checking for that flag, and
returning an error if an empty "path" is not allowed.
Change-Id: Ib48d91ff983b3edb6f65e83686b90d79d74f3471
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/53683
Reviewed-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Maintainer: Giacomo Travaglini <giacomo.travaglini@arm.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Instead, get the byte order from the workload. The workload has a better
idea what the byte order should be, for instance based on what software
it's loaded or how the hardware was configured, and this gets rid of a
use of TARGET_ISA which was setting a default endianness.
Change-Id: Ic5d8a6f69a664957c4f837e3799ff93397ccfc64
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/52106
Tested-by: kokoro <noreply+kokoro@google.com>
Maintainer: Gabe Black <gabe.black@gmail.com>
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
The workload would have a better idea of what it's endianness is than
the system object that holds it. This is the first step towards getting
rid of the getByteOrder method on the system object, which currently
checks TARGET_ISA to determine what the default endianness should be.
If it makes sense for a Workload, it could determine the endianness
dynamically by, for instance, reading it out of a binary image before
putting it into memory.
This does assume that the workload has a consistent endianness
throughout which may not be true, but this is not a new assumption.
Also, mark the SEWorkload SimObject class as "abstract", since it isn't
useful until they get subclassed by some arch specific version.
Change-Id: I8d4ba8382f22236a81f9738cc3506cdb97bdbfb2
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/52104
Reviewed-by: Gabe Black <gabe.black@gmail.com>
Maintainer: Gabe Black <gabe.black@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
This way there will always be a workload object, even if nothing needs
to be set up. This default can also be used in low_power_sweep.py,
where the workload object was just a placeholder.
This will allow required functionality like determining endianness of a
system into the workload, rather than (for instance) in the more generic
System object. This also makes accessing less essential functionality
simpler, because we don't have to detect whether the workload is there,
and can just return default, placeholder values from the StubWorkload.
Change-Id: Idfc3e75c65318d75a3eae6a19944ae1f79a2d111
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/52103
Reviewed-by: Gabe Black <gabe.black@gmail.com>
Maintainer: Gabe Black <gabe.black@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
When calling a method in a superclass, you can/should use the super()
method to get a reference to that class. The python 2 version of that
method takes two parameters, the current class name, and the "self"
instance. The python 3 version takes no arguments. This is better for a
at least three reasons.
First, this version is less verbose because you don't have to specify
any arguments.
Second, you don't have to remember which argument goes where (I always
have to look it up), and you can't accidentally use the wrong class
name, or forget to update it if you copy code from a different class.
Third, this version will work correctly if you use a class decorator.
I don't know exactly how the mechanics of this work, but it is referred
to in a comment on this stackoverflow question:
https://stackoverflow.com/questions/681953/how-to-decorate-a-class
Change-Id: I427737c8f767e80da86cd245642e3b057121bc3b
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/52224
Reviewed-by: Gabe Black <gabe.black@gmail.com>
Maintainer: Gabe Black <gabe.black@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
