This patch addresses multiple cases:
- When a controller has read/write permissions while others have read
only permissions, the one with r/w permissions performs the read as
the others may have stale data
- When controllers only have lines with stale or busy access permissions,
a valid copy of the line may be in a message in transit in the network
or in a message buffer (not seen by the controller yet). In this case,
we forward the functional request accordingly.
- Sequencer messages should not accept functional reads
- Functional writes also update the packet data on the sequencer
outstanding request lists and the cpu-side response queue.
Change-Id: I6b0656f1a2b81d41bdcf6c783dfa522a77393981
Signed-off-by: Tiago Mück <tiago.muck@arm.com>
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/22022
Tested-by: Gem5 Cloud Project GCB service account <345032938727@cloudbuild.gserviceaccount.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: John Alsop <johnathan.alsop@amd.com>
Reviewed-by: Jason Lowe-Power <power.jg@gmail.com>
Maintainer: Jason Lowe-Power <power.jg@gmail.com>
The components in base/loader were moved into a namespace called
Loader. This will make it easier to add loader components with fairly
short natural names which don't invite name collisions.
gem5 should use namespaces more in general for that reason and to make
it easier to write independent components without having to worry about
name collisions being added in the future.
Unfortunately this namespace has the same name as a class used to load
an object file into a process object. These names can be disambiguated
because the Process loader is inside the Process scope and the Loader
namespace is at global scope, but it's still confusing to read.
Fortunately, this shouldn't last for very long since the responsibility
for loading Processes is going to move to a fake OS object which will
expect to load a particular type of Process, for instance, fake 64 bit
x86 linux will load either 32 or 64 bit x86 processes.
That means that the capability to feed any binary that matches the
current build into gem5 and have gem5 figure out what to do with it
will likely be going away in the future. That's likely for the best,
since it will force users to be more explicit about what they're trying
to do, ie what OS they want to try to load a given binary, and also
will prevent loading two or more Processes which are for different OSes
to the same system, something that's possible today as far as I know
since there are no consistency checks.
Change-Id: Iea0012e98f39f5e20a7c351b78cdff9401f5e326
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/24783
Reviewed-by: Gabe Black <gabeblack@google.com>
Maintainer: Gabe Black <gabeblack@google.com>
Tested-by: kokoro <noreply+kokoro@google.com>
This is specialized per arch, and the Workload class is the only thing
actually using it. It doesn't make any sense to dispatch those calls
over to the System object, especially since that was, in most cases,
the only reason an ISA specific system class even still existed.
After this change, only ARM still has an architecture specific System
class.
Change-Id: I81b6c4db14b612bff8840157cfc56393370095e2
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/24287
Maintainer: Gabe Black <gabeblack@google.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
This generalized Workload SimObject is not geared towards FS or SE
simulations, although currently it's only used in FS. This gets rid
of the ARM specific highestELIs64 property (from the workload, not the
system) and replaces it with a generic getArch.
The old globally accessible kernel symtab has been replaced with a
symtab accessor which takes a ThreadContext *. The parameter isn't used
for anything for now, but in cases where there might be multiple
symbol tables to choose from (kernel vs. current user space?) the
method will now be able to distinguish which to use. This also makes
it possible for the workload to manage its symbol table with whatever
policy makes sense for it.
That method returns a const SymbolTable * since most of the time the
symbol table doesn't need to be modified. In the one case where an
external entity needs to modify the table, two pseudo instructions,
the table to modify isn't necessarily the one that's currently active.
For instance, the pseudo instruction will likely execute in user space,
but might be intended to add a symbol to the kernel in case something
like a module was loaded.
To support that usage, the workload has a generic "insertSymbol" method
which will insert the symbol in the table that "makes sense". There is
a lot of ambiguity what that means, but it's no less ambiguous than
today where we're only saved by the fact that there is generally only
one active symbol table to worry about.
This change also introduces a KernelWorkload SimObject class which
inherits from Workload and adds in kernel related members for cases
where the kernel is specified in the config and loaded by gem5 itself.
That's the common case, but the base Workload class would be used
directly when, for instance, doing a baremetal simulation or if the
kernel is loaded by software within the simulation as is the case for
SPARC FS.
Because a given architecture specific workload class needs to inherit
from either Workload or KernelWorkload, this change removes the
ability to boot ARM without a kernel. This ability should be restored
in the future.
To make having or not having a kernel more flexible, the kernel
specific members of the KernelWorkload should be factored out into
their own object which can then be attached to a workload through a
(potentially unused) property rather than inheritance.
Change-Id: Idf72615260266d7b4478d20d4035ed5a1e7aa241
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/24283
Reviewed-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Maintainer: Gabe Black <gabeblack@google.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Two python Enum parameter types had some very generic elements which
both include one named "none". When headers for both are included that
creates a conflict which breaks the build. Enums which such extremely
generic names need to be scoped so that they don't invite these sorts
of collisions.
This change converts them from Enum to ScopedEnum in python, and also
makes a few small changes to where they're used in c++ to match.
Issue-on: https://gem5.atlassian.net/browse/GEM5-447
Change-Id: Ibda6e6cfcd700a618f8c68d174f33ec1e178b9ac
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/27950
Reviewed-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com>
Maintainer: Gabe Black <gabeblack@google.com>
Tested-by: kokoro <noreply+kokoro@google.com>
A new Prefetcher namespace was added which holds the gem5 prefetchers
and means they don't all need a "Prefetcher" in their name. Unfortunately
that means that there is now both a Prefetcher namespace and a
Prefetcher class which conflict with each other.
This change tries to resolve the conflict with as little disruption as
possible by simply renaming the c++ ruby Pretcher class RubyPrefetcher,
leaving the python name alone so that configs aren't affected.
Issue-on: https://gem5.atlassian.net/browse/GEM5-447
Change-Id: I7afdf5dbc57dbf46d82552113c52f3a9207870f2
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/27949
Reviewed-by: Daniel Carvalho <odanrc@yahoo.com.br>
Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com>
Maintainer: Gabe Black <gabeblack@google.com>
Tested-by: kokoro <noreply+kokoro@google.com>
It is assumed that the semihosting configuration uses the semihosting
number which includes gem5's pseudo insts.
Given the complexity and likely limitted value of letting the user
arbitrarily configure fast model's semihosting, and the fact that that
semihosting implementation would compete with gem5's own, those
parameters should be removed from python and set purely within C++.
Also note that if this semihosting support is used, the System object
needs to have an ArmSemihosting object installed to handle the calls.
Change-Id: I8e1de7717c9784dc7873795acd0a06389ec527b1
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/25623
Tested-by: Gem5 Cloud Project GCB service account <345032938727@cloudbuild.gserviceaccount.com>
Reviewed-by: Gabe Black <gabeblack@google.com>
Maintainer: Gabe Black <gabeblack@google.com>
The GuestABI used to call the system-calls infers the size of values
read from the registers based on the function signature of the system
call. For mmap this was causing offset to be truncated to a 32-bit
value. In the GPUComputeDriver mmap, the offset must be a 64-bit
value. This fixes a bug where the doorbell memory was not setup and
causing GPU applications to fail.
Change-Id: I75d9b32c0470d1907c68826ef81cf6cd46f60ea7
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/27367
Tested-by: Gem5 Cloud Project GCB service account <345032938727@cloudbuild.gserviceaccount.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Matthew Poremba <matthew.poremba@amd.com>
Reviewed-by: Jason Lowe-Power <power.jg@gmail.com>
Maintainer: Jason Lowe-Power <power.jg@gmail.com>
The main applications are to run baremetal programs and initramfs Linux
kernel.
Before this patch, disks() calls in makeArmSystem would throw:
IOError: Can't find file 'linux-aarch32-ael.img' on M5_PATH.
In order to achieve this, this commit also removes the default hardcoded
disk image basenames.
For example, before this commit, running without a --disk-image in X86
would automatically search for an image with basename x86root.img in
M5_PATH, which means we would either have to ignore any disk image error,
or else running without disk images would fail.
After this commit, you would have to pass --disk-image x86root.img to
achieve the old behaviour.
Change-Id: I0ae8c4b3b93d0074fd4fca0d5ed52181c50b6c04
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/27867
Tested-by: Gem5 Cloud Project GCB service account <345032938727@cloudbuild.gserviceaccount.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Reviewed-by: Jason Lowe-Power <power.jg@gmail.com>
Reviewed-by: Gabe Black <gabeblack@google.com>
Maintainer: Jason Lowe-Power <power.jg@gmail.com>
This simplifies the SPARC FS workload significantly, and removes
assumptions about what ROMs exist, where they go, etc. It removes
other components from the loop which don't have anything to contribute
as far as setting up the ROMs.
One side effect of this is that there isn't specialized support for
adding PC based events which would fire in the ROMs, but that was never
done and the files that were being used were flat binary blobs with no
symbols in the first place.
This also necessitates building a unified image which goes into the single
8MB ROM that is located at address 0xfff0000000. That is simply done
with the following commands:
dd if=/dev/zero of=t1000_rom.bin bs=1024 count=8192
dd if=reset_new.bin of=t1000_rom.bin
dd if=q_new.bin of=t1000_rom.bin bs=1024 seek=64
dd if=openboot_new.bin of=t1000_rom.bin bs=1024 seek=512
This results in an 8MB blob which can be loaded verbatim into the ROM.
Alternatively, and with some extra effort, an ELF file could be
constructed which had each of these components as segments, offset to the
right location in the ELF header. That would be slightly more work to set up,
but wouldn't waste space on regions of the image that are all zeroes.
Change-Id: Id4e08f4e047e7bd36a416c197a36be841eba4a15
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/27268
Reviewed-by: Gabe Black <gabeblack@google.com>
Maintainer: Gabe Black <gabeblack@google.com>
Tested-by: Gem5 Cloud Project GCB service account <345032938727@cloudbuild.gserviceaccount.com>
This patch is updating the arm regression configs so that the newer
VExpress_GEM5_Foundation platform is used instead VExpress_GEM5_V1
for running regressions.
The platform has the same memory map as V1, except for a different
position of PCI regions in the memory map and more importantly
the use of GICv3 instead of GICv2
Change-Id: I48ff56dce38d482be7f1acf162f17725286f54fe
Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/27714
Tested-by: Gem5 Cloud Project GCB service account <345032938727@cloudbuild.gserviceaccount.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Ciro Santilli <ciro.santilli@arm.com>
This is fixing switcheroo tests when using a PMU/GICv3. When you switch
cpus you usually instantiate multiple cpus at the beginning and you
switch them at runtime with the m5.switchCpus function.
Every cpu will have its own set of ThreadContexts/ISAs.
When you switch cpu/tc/isa, you need to update the tc/isa pointer
cached in the device model otherwise those will still reference
the switched out cpu.
Change-Id: I3aeee890286851189c3a8a4d378c83f32e973361
Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/27713
Reviewed-by: Ciro Santilli <ciro.santilli@arm.com>
Tested-by: kokoro <noreply+kokoro@google.com>
This forces a test in the gcc Makefile to pick the right answer, where
its own check will not. Without this fix, installing libsanitizer fails
to build while installing the c++ headers because it can't find a
definition for PATH_MAX. Disabling building libsanitizer seems to work
around the problem, but other problems crop up later when using the
cross compiler, specifically when trying to build the googletest
library.
The chrome authors apparently ran into a similar problem when building
the native client tool chain as described in this bug:
https://bugs.chromium.org/p/nativeclient/issues/detail?id=3190
The CL which fixed the issue is here:
https://codereview.chromium.org/11462002/patch/1/2.
With a similar fix applied to build_cross_gcc.py, the cross compilers
build without issue, and are then able to build the googletest library
without issue.
Change-Id: Ia6869d3dc523cb0d964e82bb300f8b092693739b
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/27489
Tested-by: Gem5 Cloud Project GCB service account <345032938727@cloudbuild.gserviceaccount.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Ciro Santilli <ciro.santilli@arm.com>
Maintainer: Bobby R. Bruce <bbruce@ucdavis.edu>
The revamp of the GenericTimer was not taking into account:
* The name of the variable will be printed on the checkpoint to label the
data. It is not possible to use different variable names when
serializing/unserializing, and it is not possible to use the same
temporary variable to serialize/unserialize different values.
* the serializeSection is creating a new sub section in the
checkpoint. Doing the following:
void
GenericTimerFrame::serialize(CheckpointOut &cp) const
{
physTimer.serializeSection(cp, "phys_timer");
virtTimer.serializeSection(cp, "virt_timer");
SERIALIZE_SCALAR(accessBits);
}
will serialize the accessBits under the virt_timer subsection
rather than the parent generic_timer_frame.
JIRA: https://gem5.atlassian.net/projects/GEM5/issues/GEM5-426
Change-Id: I7676309965a33156789d2ef13e966c7a4ad88a71
Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Reviewed-by: Ciro Santilli <ciro.santilli@arm.com>
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/27708
Tested-by: kokoro <noreply+kokoro@google.com>
These are currently specific to aarch64, but will be expanded to cover
all other versions of the utility as well.
The intention of these new files is to centralize the build mechanism
for the different versions of the utility so that they have consistent
features, mechanisms, and targets, and so that new features will
automatically be shared by all versions without having to be implemented
in each.
This also sets up a separate build directory which will keep the source
tree clean, and will (with some more development) make it possible to
build multiple versions of the m5 utility at the same time without them
running into each other.
Change-Id: I10018eef6beb4af30a8d3bbab8b82cabd2b3f22c
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/27213
Reviewed-by: Bobby R. Bruce <bbruce@ucdavis.edu>
Maintainer: Gabe Black <gabeblack@google.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Faults in the TLB ought to cause a page walk. Force that by removing
the fixup in X86 TLB.
This fixes rare race conditions where a timing page walk is
intercepted by a TLB miss which fixes up the fault resulting in
double calls to allocateMem in Process class.
Change-Id: Iaef4d636cd2997144d8bc5012cd7c2a0a97102e5
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/27507
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Jason Lowe-Power <power.jg@gmail.com>
Maintainer: Jason Lowe-Power <power.jg@gmail.com>
This change includes:
1) Verify available command bandwidth
2) Add support for multi-cycle commands
3) Add new timing parameters
4) Add ability to interleave bursts
5) Add LPDDR5 configurations
The DRAM controller historically does not verify contention on the
command bus and if there is adaquate command bandwidth to issue a
new command. As memory technologies evolve, multiple cycles are becoming
a requirement for some commands. Depending on the burst length, this
can stress the command bandwidth. A check was added to verify command
issue does not exceed a maximum value within a defined window. The
default window is a burst, with the maximum value defined based on the
burst length and media clocking characteristics. When the command bandwidth
is exceeded, commands will be shifted to subsequent burst windows.
Added support for multi-cycle commands, specifically Activate, which
requires a larger address width as capacities grow. Additionally,
added support for multi-cycle Read / Write bursts for low power
DRAM cases in which additional CLK synchronization may be required
to run at higher speeds.
To support emerging memories, added the following new timing parameters.
1) tPPD -- Precharge-to-Precharge delay
2) tAAD -- Max delay between Activate-1 and Activate-2 commands
I/O data rates are continuing to increase for DRAM but the core frequency
is still fairly stagnant for many technologies. As we increase the burst
length, either the core prefetch needs to increase (for a seamless burst)
or the burst will be transferred with gaps on the data bus. To support
the latter case, added the ability to interleave 2 bursts across bank
groups.
Using the changes above, added an initial set of LPDDR5 configurations.
Change-Id: I1b14fed221350e6e403f7cbf089fe6c7f033c181
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/26236
Reviewed-by: Matthew Poremba <matthew.poremba@amd.com>
Reviewed-by: Jason Lowe-Power <power.jg@gmail.com>
Maintainer: Jason Lowe-Power <power.jg@gmail.com>
Tested-by: Gem5 Cloud Project GCB service account <345032938727@cloudbuild.gserviceaccount.com>
Tested-by: kokoro <noreply+kokoro@google.com>
A new VExpress_GEM5_Foundation platform has been added in order to match
the FVP Armv8-A Foundation Platform described in:
Armv8-A Foundation Platform - User Guide - Version 11.8
The VExpress_GEM5_V1/V2 are already loosely based on the Foundation
platform, however there are some differences in the PCI regions (V1/V2)
and the GICv3 regions (V2).
We hence introduce the VExpress_GEM5_Foundation to match closely the
FVP Foundation Platform
Change-Id: I1604c64ce566308d888c3a630019494b9fae7acf
Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com>
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/27388
Tested-by: Gem5 Cloud Project GCB service account <345032938727@cloudbuild.gserviceaccount.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Ciro Santilli <ciro.santilli@arm.com>
When accessing m5_mem and building PIC code, we need to get the address
of m5_mem out of the global offset table, and then load the value from
there. If we try to load from m5_mem directly, the assembled code has a
relocation type the linker can't handle when building a shared object.
Change-Id: Ieb19c3d17c37ef810559ee24b68886b18ddcc869
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/27212
Reviewed-by: Bobby R. Bruce <bbruce@ucdavis.edu>
Maintainer: Gabe Black <gabeblack@google.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Calls to queueMemoryRead and queueMemoryWrite do not consider the size
of the queue between ruby directories and DRAMCtrl which causes infinite
buffering in the queued port between the two. This adds a MessageBuffer
in between which uses enqueues in SLICC and is therefore size checked
before any SLICC transaction pushing to the buffer can occur, removing
the infinite buffering between the two.
Change-Id: Iedb9070844e4f6c8532a9c914d126105ec98d0bc
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/27427
Tested-by: Gem5 Cloud Project GCB service account <345032938727@cloudbuild.gserviceaccount.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Bradford Beckmann <brad.beckmann@amd.com>
Reviewed-by: Jason Lowe-Power <power.jg@gmail.com>
Reviewed-by: Matt Sinclair <mattdsinclair@gmail.com>
Maintainer: Bradford Beckmann <brad.beckmann@amd.com>
The DictImporter's __del__ method calls unload, and that imports
sys.modules so that it can remove the modules that the DictImporter had
set up as the importer goes away.
Unfortunately, the importer only goes away when python is shutting down,
and at that time some aspects of the system, namely sys.meta_path, have
been cleaned up. When unload tries to import sys, that causes an
exception which scons/python reports but which doesn't do anything bad
otherwise.
In all of the examples of this older style of import object online, none
had a __del__ method, and none worried about cleaning up sys.modules
when they went away. In light of that, I've removed the __del__ method
entirely.
Another reason I think it's safe to remove __del__ is that the importer
was not actually being deleted even when it was removed from
sys.meta_path, and all the modules it had loaded where removed from
sys.modules. I think that was because the SimObject classes that it had
set up still had references (they are used later in the SConscript), and
those would, either directly or indirectly, refer back to the modules
and the importer. Those remaining references kept the importer alive,
preventing __del__ from being called before all those other objects were
cleaned up.
I think in python 2, the order things were cleaned up just so happened
to avoid trying to import sys when it was no longer possible, but in
python 3 that changed and resulted in this exception being thrown.
I've tried building gem5 with scons running under python 2 and python 3,
and with this change there is no error at shutdown. Both also produce a
gem5 binary which can run hello world without problems.
Change-Id: Ib1f5c7403df57fc420cec7ec0fef20a164a06991
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/27247
Tested-by: Gem5 Cloud Project GCB service account <345032938727@cloudbuild.gserviceaccount.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Bobby R. Bruce <bbruce@ucdavis.edu>
Maintainer: Bobby R. Bruce <bbruce@ucdavis.edu>
