Add logic to collect pointers to all GPU TLBs in full system. Implement
the invalid TLBs PM4 packet. The invalidate is done functionally since
there is really no benefit to simulate it with timing and there is no
support in the TLB to do so. This allow application with much larger
data sets which may reuse device memory pages to work in gem5 without
possibly crashing due to a stale translation being leftover in the TLB.
Change-Id: Ia30cce02154d482d8f75b2280409abb8f8375c24
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/58470
Reviewed-by: Jason Lowe-Power <power.jg@gmail.com>
Maintainer: Jason Lowe-Power <power.jg@gmail.com>
Reviewed-by: Matt Sinclair <mattdsinclair@gmail.com>
Maintainer: Matt Sinclair <mattdsinclair@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
The downloader briefly caches the resources.json file in "/tmp". This
works fine if the system only has one user, but in a system supporting
many users, where "/tmp" is shared, there were permission issues when a
user tried to access a resources.json cache from another user. To
resolve this the uid of the user is appended to the filename.
Change-Id: Iba8ab12cd7054c2f5636e444ac0f1f8f73fedc89
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/58489
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 supported version of ROCm in GPUFS only supports the DKMS build of
the amdgpu driver. However, since a gem5 user can potentially pass in
any Linux kernel as a parameter, it is possible that the DKMS package
for that kernel was not installed on the disk image. This would result
in the simulation appearing to work when in reality it is just spinning
waiting for commands from the driver. This check exits gem5 early in the
simulation and outputs an error on the console to sanity check the
correct driver is being used.
Change-Id: I708912e5625e47eba15dcb2f722772a3b2928b98
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/58129
Reviewed-by: Matt Sinclair <mattdsinclair@gmail.com>
Maintainer: Matt Sinclair <mattdsinclair@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Requests in GPUFS which go to system memory will not generate the
WriteCompleteResp packets that the VIPER protocol would normally created
for device requests which go through the caches. Therefore, we need to
callback the GM pipe handleResponse to complete the access and make
forward progress.
Change-Id: Ic00c430ce420a591fe5743f758b780d93afd2a38
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/57989
Reviewed-by: Matt Sinclair <mattdsinclair@gmail.com>
Maintainer: Matt Sinclair <mattdsinclair@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
In full system mode, the AMDKernelCode object can reside in either the
system memory or in the dGPU device memory. Currently only reading from
the host/system memory is supported. This adds the necessary code to
read from the dGPU device memory.
Change-Id: I887fc706b3f9834db14e40f36fd29dd3d4602925
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/57710
Reviewed-by: Matt Sinclair <mattdsinclair@gmail.com>
Maintainer: Matt Sinclair <mattdsinclair@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
The emulated driver is currently called in a few locations
unconditionally. This changeset adds checks that we are not in full
system before calling any emulated driver function. In full system the
amdgpu driver running on the disk image handles these functions.
Change-Id: Iea3546b574e29c649351c0fce9154530be89e9b1
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/57712
Reviewed-by: Matt Sinclair <mattdsinclair@gmail.com>
Maintainer: Matt Sinclair <mattdsinclair@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
The amdgpu driver supports reading and writing scalar and vector memory
addresses that reside in system memory. This is commonly used for things
like blit kernels that perform host-to-device or device-to-host copies
using GPU load/store instructions.
This is done by utilizing the system hub device added in a prior
changeset. Memory packets translated by the Scalar or VMEM TLBs will
have the correspoding system request field set from the PTE in the TLB
which can be used in the compute unit to determine if a request is for
system memory or not.
Another important change is to return global memory tokens for system
requests. Since these do not flow through the GPU coalescer where the
token is returned, the token can be returned once the request is known
to be a system request.
Change-Id: I35030e0b3698f10c63a397f96b81267271e3130e
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/57711
Reviewed-by: Matt Sinclair <mattdsinclair@gmail.com>
Maintainer: Matt Sinclair <mattdsinclair@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
This path fix build failed when gem5's path contains Non-ASCII
characters.
build_tools/marshal.py embed "abspath" and "modpath" into the
generated c++ file. However, if either of them contains Non-ASCII
characters, marshal.py will throw an exception.
This can be fixed if we set a suitable $LANG for gem5py. For
example, we can simply set $LANG=en_US.UTF-8 in src/SConscript
to fix this.
This patch fixes this bug by passing the host's $LANG environment
variable to gem5py. Because for users who may use Non-ASCII
characters, their $LANG is generally set to their native language
(e.g. zh_CN.UTF-8, en_US.UTF-8, etc.), which contains Non-ASCII
characters. Also, it is more reasonable to use the host $LANG than
to directly hard-code the $LANG used by gem5py into a language
that supports non-ASCII characters.
Change-Id: Ia0343cbd6995af3a9e43bf217ad3f186d5633fed
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/58369
Reviewed-by: Gabe Black <gabe.black@gmail.com>
Maintainer: Gabe Black <gabe.black@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
The two functions are `get_cpu_types_str_set()` which returns a set of
valid CPUTypes as strings, and `get_cpu_type_from_str()` which will
return a CPUType enum given an input string.
The purpose of these functions is to aid and standardize user input
parameters or environment variables.
Test scripts are updated accordingly.
Change-Id: I7cb9263321fe36bc8a7530edfd0d8e8bbd329e0e
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/58491
Reviewed-by: Bobby Bruce <bbruce@ucdavis.edu>
Maintainer: Bobby Bruce <bbruce@ucdavis.edu>
Tested-by: kokoro <noreply+kokoro@google.com>
The main restriction with this design is it results in one ISA target
per board. The ISA is declared per core. To make the design simpler it's
assumed a Processor (a collection of cores) are all of the same ISA. As
each board has one processor, this also means a board is typically tied
to one ISA per simulation.
In order to remain backwards compatible and maintain the standard
library APIs, this patch adds a `--main-isa` parameter which will
determine what `gem5.runtime.get_runtime_isa` returns in cases where
mutliple ISAs are compiled in. When setting the ISA in a simulation (via
the Processor or Cores), the user may, as before, choose not to and, in
this case, the `gem5.runtime.get_runtime_isa` function is used.
The `gem5.runtime.get_runtime_isa` function is an intermediate step
which should be removed in future versions of gem5 (users should specify
precisely what ISA they want via configuration scripts). For this reason
it throws a warning when used and should not be heavily relied upon. It
is deprecated.
Change-Id: Ia76541bfa9a5a4b6b86401309281849b49dc724b
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/55423
Reviewed-by: Gabe Black <gabe.black@gmail.com>
Maintainer: Gabe Black <gabe.black@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
This helper extracts config variables and sets them up for use in
fast model tools. Rework this function for two purposes. First, ensure
that the ['CONF'] subdict is used consistently. It was used in one spot
in that function, but not anywhere else. Avoid those mistakes by
accessing the configuration dict only once, and then reusing the value.
Second, only print the values of those variables if they aren't empty.
That avoids extra log output for values which aren't set up and aren't
being used, while still exposing those values if they are set up. This
will print them even if they aren't being actively used, but are set up
and could be with other changes in the config.
Change-Id: Ia3ea76e6cb6b3347fd082112a1eaa8d594e06ca2
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/58354
Reviewed-by: Gabe Black <gabe.black@gmail.com>
Maintainer: Gabe Black <gabe.black@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
If a fast model static library can't be found, we should treat that as
a warning instead of an error, and pass back the original library name
so that it can at least be added and potentially come from somewhere
else.
In practice, this is important because gem5 will be configured by SCons
indirectly in the future, using kconfig based tools that SCons runs on
the user's behalf. If SCons is misconfigured or not configured, this
error can trip, preventing those tools from starting. That creates a
catch 22, since you'd need SCons to fix the config, and SCons can't
run because of the config.
We can avoid that problem by making SCons more lenient, so that it can
still run even if it doesn't find static libraries where it might have
expected to.
Change-Id: Iadfd823b61fe96b937c2650250487d290492f265
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/58352
Maintainer: Gabe Black <gabe.black@gmail.com>
Reviewed-by: Yu-hsin Wang <yuhsingw@google.com>
Tested-by: kokoro <noreply+kokoro@google.com>
This held the location of the license file, assuming it actually was a
license file and not a license server URL. If this variable was unset
because the fast model configuration parameters hadn't been set, then
the "File" would resolve to the local directory, and SCons would get
upset that you'd called the same path a directory and a file.
We can avoid this problem by just getting rid of this variable, since
it isn't used anyway.
Change-Id: I2ccda90b85f2c83c73816967d145e6bf45733f89
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/58351
Maintainer: Gabe Black <gabe.black@gmail.com>
Reviewed-by: Yu-hsin Wang <yuhsingw@google.com>
Tested-by: kokoro <noreply+kokoro@google.com>
This helps specify ordering for libraries that need it. Libraries with
a higher priority will be sorted earlier in the list, which can be
necessary when working with static libraries/archives.
The default value for "priority" is zero. It's only really necessary
to ensure relative ordering of particular pairings of libraries, so
it should be ok to use an absolute integer value for this. If you
need to order relative to a library, there is a good chance you're
adding it, or the place it's added is well known and you can easily
find its priority value. It's also unlikely that there would be a
complex series of interactions between libraries that would make a
more complicated system warranted.
Change-Id: Ie94a35e6563c07f8d462a4a52d0173ea3cf4f8de
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/58350
Maintainer: Gabe Black <gabe.black@gmail.com>
Reviewed-by: Jui-min Lee <fcrh@google.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Yu-hsin Wang <yuhsingw@google.com>
In Python, "+=" operator for list acts more like append and list assign
doesn't make a copy. This will cause unexpected append to the orignal
list. Since we have multiple env to "declare", these functions will be
called multiple times and could wrongly append duplicated entries in
later calls.
Make a copy before appending the entries from filter to avoid this
problem.
Change-Id: I144d5054e4d93191ebc94b93291ff9a3f8a6c429
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/58409
Reviewed-by: Yu-hsin Wang <yuhsingw@google.com>
Maintainer: Gabe Black <gabe.black@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
ProtoBuf files can (and should) be declared to SCons unconditionally,
but then also use a "protobuf" tag. This lets SCons still include them
conditionally when it gathers sources to build certain binaries without
having additional python logic in SConscripts, or manually interpreting
config state.
Change-Id: I9426d647bc2d6c9d7eee7dadba77f75ed83bcaad
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/58349
Maintainer: Giacomo Travaglini <giacomo.travaglini@arm.com>
Reviewed-by: Jason Lowe-Power <power.jg@gmail.com>
Reviewed-by: Bobby Bruce <bbruce@ucdavis.edu>
Tested-by: kokoro <noreply+kokoro@google.com>
If we create abstract memories with a sub-page size on a system with
shared backstore, the offset of next mmap might become non-page-align
and cause an invalid argument error.
In this CL, we always upscale the range size to multiple of page before
updating the offset, so the offset is always on page boundary.
Change-Id: I3a6adf312f2cb5a09ee6a24a87adc62b630eac66
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/58289
Reviewed-by: Gabe Black <gabe.black@gmail.com>
Maintainer: Gabe Black <gabe.black@gmail.com>
Reviewed-by: Boris Shingarov <shingarov@labware.com>
Tested-by: kokoro <noreply+kokoro@google.com>
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>
The QCntxt is reused when a queue is unmapped and mapped again. This is
fairly common in GPU full system. If this is not done the readIndex on
the queue context is reset to 1, causing getCommandsFromHost to read
from the wrong slot which is typically an old dispatch packet or an
invalid packet. This causes simulation to stall as the incorrect
completion signal is eventually written.
Change-Id: I65541e559fe04f5eb44b936ca37e3f802262fe6a
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/57670
Reviewed-by: Matt Sinclair <mattdsinclair@gmail.com>
Maintainer: Matt Sinclair <mattdsinclair@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
