SimplePoolManager doesn't allow mapping of two WGs
simultaneously on the same Compute Unit (provided
the previous WG has been mapped to all the SIMDs)
even if there is sufficient VRF and SRF space
available.
DynPoolManager takes care of that by dynamically
allocating and deallocating register file space
to wavefronts
Change-Id: I2255c68d4b421615d7b231edc05d3ebb27cbd66c
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/32034
Maintainer: Matt Sinclair <mattdsinclair@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Jason Lowe-Power <power.jg@gmail.com>
Reviewed-by: Alexandru Duțu <alexandru.dutu@amd.com>
The create() method on Params structs usually instantiate SimObjects
using a constructor which takes the Params struct as a parameter
somehow. There has been a lot of needless variation in how that was
done, making it annoying to pass Params down to base classes. Some of
the different forms were:
const Params &
Params &
Params *
const Params *
Params const*
This change goes through and fixes up every constructor and every
create() method to use the const Params & form. We use a reference
because the Params struct should never be null. We use const because
neither the create method nor the consuming object should modify the
record of the parameters as they came in from the config. That would
make consuming them not idempotent, and make it impossible to tell what
the actual simulation configuration was since it would change from any
user visible form (config script, config.ini, dot pdf output).
Change-Id: I77453cba52fdcfd5f4eec92dfb0bddb5a9945f31
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/35938
Reviewed-by: Gabe Black <gabeblack@google.com>
Reviewed-by: Daniel Carvalho <odanrc@yahoo.com.br>
Maintainer: Gabe Black <gabeblack@google.com>
Tested-by: kokoro <noreply+kokoro@google.com>
When the TokenPort was moved from the GCN3 staging branch to develop the
TokenPort was changed from being the port connecting the ComputeUnit to
Ruby's vector memory port to a sideband port which inhibits requests to
Ruby's vector memory port. As such, it needs to be explicitly connected
as a new port. This changes the getPort method in ComputeUnit to be
aware of the port as well as modifying the example config to connect to
TCPs.
The iteration to connect in the config file was modified since it was
not properly connecting to TCPs each time and Ruby.py does not
explicitly return a list of each MachineType.
Change-Id: Ia70a6756b2af54d95e94d19bec5d8aadd3c2d5c0
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/35096
Reviewed-by: Matt Sinclair <mattdsinclair@gmail.com>
Reviewed-by: Jason Lowe-Power <power.jg@gmail.com>
Maintainer: Matt Sinclair <mattdsinclair@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
The CU would initialize its ports in getMasterPort(), which
is not desirable as getMasterPort() may be called several
times for the same port. This can lead to a fatal if the CU
expects to only create a single port of a given type, and may
lead to other issues where stat names are duplicated.
This change instantiates and initializes the CU's ports in the
CU constructor using the CU params.
The index field is also removed from the CU's ports because the
base class already has an ID field, which will be set to the
default value in the base class's constructor for scalar ports.
It doesn't make sense for scalar port's to take an index because
they are scalar, so we let the base class initialize the ID to
the invalid port ID.
Change-Id: Id18386f5f53800a6447d968380676d8fd9bac9df
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/32836
Reviewed-by: Anthony Gutierrez <anthony.gutierrez@amd.com>
Maintainer: Anthony Gutierrez <anthony.gutierrez@amd.com>
Tested-by: kokoro <noreply+kokoro@google.com>
This change separates the pipeline stage interfaces
for the GPU's compute unit into their own classes
with a well-defined interface. This helps to create
a cleaner interface for users to extend the CU
pipeline's capabilities and also helps consolidate
all the pipeline communication code in one place
in the source.
Change-Id: I569d52bce84dc1b9fbf8f0f96d53a81a2b6773c6
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/29972
Reviewed-by: Anthony Gutierrez <anthony.gutierrez@amd.com>
Maintainer: Anthony Gutierrez <anthony.gutierrez@amd.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Barriers were not modeled properly. Firstly, barriers were
allocated to each WG that was launched, which is not
correct, and the CU would provide an infinite number
of barrier slots. There are a limited number of barrier slots
per CU in reality. In addition, the CU will not allocate
barrier slots to WGs with a single WF (nothing to sync if
only one WF).
Beyond modeling problems, there also the issue of deadlock.
The barrier could deadlock because not all WFs are freed
from the barrier once it has been satisfied. Instead, we
relied on the scoreboard stage to release them lazily,
one-by-one.
Under this implementation the scoreboard may not fully release
all WFs participating in a barrier; this happens because the
first WF to be freed from the barrier could reach an s_barrier
instruction again, forever causing the barrier counts across
WFs to be out-of-sync.
This change refactors the barrier logic to:
1) Create a proper barrier slot implementation
2) Enforce (via a parameter) the number of barrier
slots on the CU.
3) Simplify the logic and cleanup the code (i.e., we
no longer iterate through the entire WF list each
time we check if a barrier is satisfied).
4) Fix deadlock issues.
Change-Id: If53955b54931886baaae322640a7b9da7a1595e0
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/29943
Reviewed-by: Anthony Gutierrez <anthony.gutierrez@amd.com>
Maintainer: Anthony Gutierrez <anthony.gutierrez@amd.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Remove the read/write tables and coalescing table and introduce a two
levels of tables for uncoalesced and coalesced packets. Tokens are
granted to GPU instructions to place in uncoalesced table. If tokens
are available, the operation always succeeds such that the 'Aliased'
status is never returned. Coalesced accesses are placed in the
coalesced table while requests are outstanding. Requests to the same
address are added as targets to the table similar to how MSHRs
operate.
Change-Id: I44983610307b638a97472db3576d0a30df2de600
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/27429
Reviewed-by: Bradford Beckmann <brad.beckmann@amd.com>
Reviewed-by: Jason Lowe-Power <power.jg@gmail.com>
Maintainer: Bradford Beckmann <brad.beckmann@amd.com>
Tested-by: kokoro <noreply+kokoro@google.com>
MemObject doesn't provide anything beyond its base ClockedObject any
more, so this change removes it from most inheritance hierarchies.
Occasionally MemObject is replaced with SimObject when I was fairly
confident that the extra functionality of ClockedObject wasn't needed.
Change-Id: Ic014ab61e56402e62548e8c831eb16e26523fdce
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/18289
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Anthony Gutierrez <anthony.gutierrez@amd.com>
Maintainer: Gabe Black <gabeblack@google.com>
this patch removes the GPUStaticInst enums that were defined in GPU.py.
instead, a simple set of attribute flags that can be set in the base
instruction class are used. this will help unify the attributes of HSAIL
and machine ISA instructions within the model itself.
because the static instrution now carries the attributes, a GPUDynInst
must carry a pointer to a valid GPUStaticInst so a new static kernel launch
instruction is added, which carries the attributes needed to perform a
the kernel launch.
This patch adds a method to the Wavefront class to compute the actual workgroup
size. This can be different from the maximum workgroup size specified when
launching the kernel through the NDRange object. Current solution is still not
optimal, as we are computing these for each wavefront and the dispatcher also
needs to have this information and can't actually call
Wavefront::computeActuallWgSz before the wavefronts are being created. A long
term solution would be to have a Workgroup class that deals with all these
details.
WFContext struct is currently unused and it has been rendered not useful in
saving and restoring the context of a Wavefront. Wavefront class should be
sufficient for that purpose and the runtime can figure out the memory size
it will need to allocate for a Wavefront through an IOCTL.
Eliminate the VSZ constant that defined the Wavefront size (in numbers of work
items); replaced it with a parameter in the GPU.py configuration script.
Changed all data structures dependent on the Wavefront size to be dynamically
sized. Legal values of Wavefront size are 16, 32, 64 for now and checked at
initialization time.
