Age | Commit message (Collapse) | Author |
|
Change-Id: I9f47a20a869553515a759d9a29c05f6ce4b42d64
Signed-off-by: Sean Wilson <spwilson2@wisc.edu>
Reviewed-on: https://gem5-review.googlesource.com/3930
Maintainer: Jason Lowe-Power <jason@lowepower.com>
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
|
|
Over the past 6 years, we realized that the protocol is essentially used
to run the garnet network in a standalone manner, and feed standard synthetic
traffic patterns through it.
|
|
|
|
This patch moves away from using M5_ATTR_OVERRIDE and the m5::hashmap
(and similar) abstractions, as these are no longer needed with gcc 4.7
and clang 3.1 as minimum compiler versions.
|
|
|
|
|
|
This patch eliminates the type Address defined by the ruby memory system.
This memory system would now use the type Addr that is in use by the
rest of the system.
|
|
The RubyCache (CacheMemory) latency parameter is only used for top-level caches
instantiated for Ruby coherence protocols. However, the top-level cache hit
latency is assessed by the Sequencer as accesses flow through to the cache
hierarchy. Further, protocol state machines should be enforcing these cache hit
latencies, but RubyCaches do not expose their latency to any existng state
machines through the SLICC/C++ interface. Thus, the RubyCache latency parameter
is superfluous for all caches. This is confusing for users.
As a step toward pushing L0/L1 cache hit latency into the top-level cache
controllers, move their latencies out of the RubyCache declarations and over to
their Sequencers. Eventually, these Sequencer parameters should be exposed as
parameters to the top-level cache controllers, which should assess the latency.
NOTE: Assessing these latencies in the cache controllers will require modifying
each to eliminate instantaneous Ruby hit callbacks in transitions that finish
accesses, which is likely a large undertaking.
|
|
The directory ruby/system is crowded and unorganized. Hence, the files the
hold actual physical structures, are being moved to the directory
ruby/structures. This includes Cache Memory, Directory Memory,
Memory Controller, Wire Buffer, TBE Table, Perfect Cache Memory, Timer Table,
Bank Array.
The directory ruby/systems has the glue code that holds these structures
together.
--HG--
rename : src/mem/ruby/system/MachineID.hh => src/mem/ruby/common/MachineID.hh
rename : src/mem/ruby/buffers/MessageBuffer.cc => src/mem/ruby/network/MessageBuffer.cc
rename : src/mem/ruby/buffers/MessageBuffer.hh => src/mem/ruby/network/MessageBuffer.hh
rename : src/mem/ruby/buffers/MessageBufferNode.cc => src/mem/ruby/network/MessageBufferNode.cc
rename : src/mem/ruby/buffers/MessageBufferNode.hh => src/mem/ruby/network/MessageBufferNode.hh
rename : src/mem/ruby/system/AbstractReplacementPolicy.hh => src/mem/ruby/structures/AbstractReplacementPolicy.hh
rename : src/mem/ruby/system/BankedArray.cc => src/mem/ruby/structures/BankedArray.cc
rename : src/mem/ruby/system/BankedArray.hh => src/mem/ruby/structures/BankedArray.hh
rename : src/mem/ruby/system/Cache.py => src/mem/ruby/structures/Cache.py
rename : src/mem/ruby/system/CacheMemory.cc => src/mem/ruby/structures/CacheMemory.cc
rename : src/mem/ruby/system/CacheMemory.hh => src/mem/ruby/structures/CacheMemory.hh
rename : src/mem/ruby/system/DirectoryMemory.cc => src/mem/ruby/structures/DirectoryMemory.cc
rename : src/mem/ruby/system/DirectoryMemory.hh => src/mem/ruby/structures/DirectoryMemory.hh
rename : src/mem/ruby/system/DirectoryMemory.py => src/mem/ruby/structures/DirectoryMemory.py
rename : src/mem/ruby/system/LRUPolicy.hh => src/mem/ruby/structures/LRUPolicy.hh
rename : src/mem/ruby/system/MemoryControl.cc => src/mem/ruby/structures/MemoryControl.cc
rename : src/mem/ruby/system/MemoryControl.hh => src/mem/ruby/structures/MemoryControl.hh
rename : src/mem/ruby/system/MemoryControl.py => src/mem/ruby/structures/MemoryControl.py
rename : src/mem/ruby/system/MemoryNode.cc => src/mem/ruby/structures/MemoryNode.cc
rename : src/mem/ruby/system/MemoryNode.hh => src/mem/ruby/structures/MemoryNode.hh
rename : src/mem/ruby/system/MemoryVector.hh => src/mem/ruby/structures/MemoryVector.hh
rename : src/mem/ruby/system/PerfectCacheMemory.hh => src/mem/ruby/structures/PerfectCacheMemory.hh
rename : src/mem/ruby/system/PersistentTable.cc => src/mem/ruby/structures/PersistentTable.cc
rename : src/mem/ruby/system/PersistentTable.hh => src/mem/ruby/structures/PersistentTable.hh
rename : src/mem/ruby/system/PseudoLRUPolicy.hh => src/mem/ruby/structures/PseudoLRUPolicy.hh
rename : src/mem/ruby/system/RubyMemoryControl.cc => src/mem/ruby/structures/RubyMemoryControl.cc
rename : src/mem/ruby/system/RubyMemoryControl.hh => src/mem/ruby/structures/RubyMemoryControl.hh
rename : src/mem/ruby/system/RubyMemoryControl.py => src/mem/ruby/structures/RubyMemoryControl.py
rename : src/mem/ruby/system/SparseMemory.cc => src/mem/ruby/structures/SparseMemory.cc
rename : src/mem/ruby/system/SparseMemory.hh => src/mem/ruby/structures/SparseMemory.hh
rename : src/mem/ruby/system/TBETable.hh => src/mem/ruby/structures/TBETable.hh
rename : src/mem/ruby/system/TimerTable.cc => src/mem/ruby/structures/TimerTable.cc
rename : src/mem/ruby/system/TimerTable.hh => src/mem/ruby/structures/TimerTable.hh
rename : src/mem/ruby/system/WireBuffer.cc => src/mem/ruby/structures/WireBuffer.cc
rename : src/mem/ruby/system/WireBuffer.hh => src/mem/ruby/structures/WireBuffer.hh
rename : src/mem/ruby/system/WireBuffer.py => src/mem/ruby/structures/WireBuffer.py
rename : src/mem/ruby/recorder/CacheRecorder.cc => src/mem/ruby/system/CacheRecorder.cc
rename : src/mem/ruby/recorder/CacheRecorder.hh => src/mem/ruby/system/CacheRecorder.hh
|
|
|
|
|
|
The patch started of with removing the global variables from the profiler for
profiling the miss latency of requests made to the cache. The corrresponding
histograms have been moved to the Sequencer. These are combined together when
the histograms are printed. Separate histograms are now maintained for
tracking latency of all requests together, of hits only and of misses only.
A particular set of histograms used to use the type GenericMachineType defined
in one of the protocol files. This patch removes this type. Now, everything
that relied on this type would use MachineType instead. To do this, SLICC has
been changed so that multiple machine types can be declared by a controller
in its preamble.
|
|
The histogram for tracking outstanding counts per cycle is maintained
in the profiler. For a parallel implementation of the memory system, we
need that this histogram is maintained locally. Hence it will now be
kept in the sequencer itself. The resulting histograms will be merged
when the stats are printed.
|
|
The MESI CMP directory coherence protocol, while transitioning from SM to IM,
did not invalidate the lock that it might have taken on a cache line. This
patch adds an action for doing so.
The problem was found by Dibakar, but I was not happy with his proposed
solution. So I implemented a different solution.
Committed by: Nilay Vaish <nilay@cs.wisc.edu>
|
|
This patch is as of now the final patch in the series of patches that replace
Time with Cycles.This patch further replaces Time with Cycles in Sequencer,
Profiler, different protocols and related entities.
Though Time has not been completely removed, the places where it is in use
seem benign as of now.
|
|
|
|
This patch is a first step to using Cycles as a parameter type. The
main affected modules are the CPUs and the Ruby caches. There are
definitely plenty more places that are affected, but this patch serves
as a starting point to making the transition.
An important part of this patch is to actually enable parameters to be
specified as Param.Cycles which involves some changes to params.py.
|
|
This patch removes RubyEventQueue. Consumer objects now rely on RubySystem
or themselves for scheduling events.
|
|
This patch removes printConfig() functions from all structures in Ruby.
Most of the information is already part of config.ini, and where ever it
is not, it would become in due course.
|
|
Updates to Ruby to support statistics counting of cache accesses. This feature
serves multiple purposes beyond simple stats collection. It provides the
foundation for ruby to model the cache tag and data arrays as physical
resources, as well as provide the necessary input data for McPAT power
modeling.
|
|
This patch adds the necessary flags to the SConstruct and SConscript
files for compiling using clang 2.9 and later (on Ubuntu et al and OSX
XCode 4.2), and also cleans up a bunch of compiler warnings found by
clang. Most of the warnings are related to hidden virtual functions,
comparisons with unsigneds >= 0, and if-statements with empty
bodies. A number of mismatches between struct and class are also
fixed. clang 2.8 is not working as it has problems with class names
that occur in multiple namespaces (e.g. Statistics in
kernel_stats.hh).
clang has a bug (http://llvm.org/bugs/show_bug.cgi?id=7247) which
causes confusion between the container std::set and the function
Packet::set, and this is currently addressed by not including the
entire namespace std, but rather selecting e.g. "using std::vector" in
the appropriate places.
|
|
This patch implements the functionality for forwarding invalidations and
replacements from the L1 cache of the Ruby memory system to the O3 CPU. The
implementation adds a list of ports to RubyPort. Whenever a replacement or an
invalidation is performed, the L1 cache forwards this to all the ports, which
is the LSQ in case of the O3 CPU.
|
|
This patch resurrects ruby's cache warmup capability. It essentially
makes use of all the infrastructure that was added to the controllers,
memories and the cache recorder.
|
|
This patch changes the implementation of Ruby's recvTiming() function so
that it pushes a packet in to the Sequencer instead of a RubyRequest. This
requires changes in the Sequencer's makeRequest() and issueRequest()
functions, as they also need to operate on a Packet instead of RubyRequest.
|
|
|
|
|
|
The tester code is in testers/networktest.
The tester can be invoked by configs/example/ruby_network_test.py.
A dummy coherence protocol called Network_test is also addded for network-only simulations and testing. The protocol takes in messages from the tester and just pushes them into the network in the appropriate vnet, without storing any state.
|
|
This patch converts CacheRequestType to RubyRequestType so that both the
protocol dependent and independent code makes use of the same request type.
|
|
This patch converts AccessModeType to RubyAccessMode so that both the
protocol dependent and independent code uses the same access mode.
|
|
|
|
This patch tracks the number of cycles a transaction is delayed at different
points of the request-forward-response loop.
|
|
|
|
Added support so that ruby can determine the outcome of store conditional
operations and reflect that outcome to M5 physical memory and cpus.
|
|
This patch adds back to ruby the capability to understand the response time
for messages that hit in different levels of the cache heirarchy.
Specifically add support for the MI_example, MOESI_hammer, and MOESI_CMP_token
protocols.
|
|
|
|
In addition to obvious changes, this required a slight change to the slicc
grammar to allow types with :: in them. Otherwise slicc barfs on std::string
which we need for the headers that slicc generates.
|
|
|
|
Removed static members in RubyPort and removed the ruby request unique id.
|
|
|
|
|
|
This patch includes the necessary changes to connect ruby objects using
the python configuration system. Mainly it consists of removing
unnecessary ruby object pointers and connecting the necessary object
pointers using the generated param objects. This patch includes the
slicc changes necessary to connect generated ruby objects together using
the python configuraiton system.
|
|
The necessary companion conversion of Ruby objects generated by SLICC
are converted to M5 SimObjects in the following patch, so this patch
alone does not compile.
Conversion of Garnet network models is also handled in a separate
patch; that code is temporarily disabled from compiling to allow
testing of interim code.
|
|
This patch changes the way that Ruby handles atomic RMW instructions. This implementation, unlike the prior one, is protocol independent. It works by locking an address from the sequencer immediately after the read portion of an RMW completes. When that address is locked, the coherence controller will only satisfy requests coming from one port (e.g., the mandatory queue) and will ignore all others. After the write portion completed, the line is unlocked. This should also work with multi-line atomics, as long as the blocks are always acquired in the same order.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
don't allow enquing from other threads if servicing and atomic for a thread
|