summaryrefslogtreecommitdiff
path: root/src/mem/protocol
AgeCommit message (Collapse)Author
2010-08-24MOESI_hammer: fixed bug for dma reads in single cpu systemsBrad Beckmann
2010-08-20ruby: Added merge GETS optimization to hammerBrad Beckmann
Added an optimization that merges multiple pending GETS requests into a single request to the owner node.
2010-08-20ruby: Stall and wait input messages instead of recyclingBrad Beckmann
This patch allows messages to be stalled in their input buffers and wait until a corresponding address changes state. In order to make this work, all in_ports must be ranked in order of dependence and those in_ports that may unblock an address, must wake up the stalled messages. Alot of this complexity is handled in slicc and the specification files simply annotate the in_ports. --HG-- rename : src/mem/slicc/ast/CheckAllocateStatementAST.py => src/mem/slicc/ast/StallAndWaitStatementAST.py rename : src/mem/slicc/ast/CheckAllocateStatementAST.py => src/mem/slicc/ast/WakeUpDependentsStatementAST.py
2010-08-20ruby: Recycle latency fix for hammerBrad Beckmann
Patch allows each individual message buffer to have different recycle latencies and allows the overall recycle latency to be specified at the cmd line. The patch also adds profiling info to make sure no one processor's requests are recycled too much.
2010-08-20MOESI_hammer: break down miss latency stalled cyclesBrad Beckmann
This patch tracks the number of cycles a transaction is delayed at different points of the request-forward-response loop.
2010-08-20ruby: added probe filter support to hammerBrad Beckmann
2010-08-20ruby: Disable migratory sharing for token and hammerBrad Beckmann
This patch allows one to disable migratory sharing for those cache blocks that are accessed by atomic requests. While the implementations are different between the token and hammer protocols, the motivation is the same. For Alpha, LLSC semantics expect that normal loads do not unlock cache blocks that have been locked by LL accesses. Therefore, locked blocks should not transfer write permissions when responding to these load requests. Instead, only they only transfer read permissions so that the subsequent SC access can possibly succeed.
2010-08-20ruby: fixed token bugs associated with owner token countsBrad Beckmann
This patch fixes several bugs related to previous inconsistent assumptions on how many tokens the Owner had. Mike Marty should have fixes these bugs years ago. :)
2010-08-20ruby: MOESI_CMP_token dma fixesBrad Beckmann
This patch fixes various protocol bugs regarding races between dma requests and persistent requests.
2010-08-20ruby: Resurrected Ruby's deterministic testsBrad Beckmann
Added the request series and invalidate deterministic tests as new cpu models and removed the no longer needed ruby tests --HG-- rename : configs/example/rubytest.py => configs/example/determ_test.py rename : src/mem/ruby/tester/DetermGETXGenerator.cc => src/cpu/directedtest/DirectedGenerator.cc rename : src/mem/ruby/tester/DetermGETXGenerator.hh => src/cpu/directedtest/DirectedGenerator.hh rename : src/mem/ruby/tester/DetermGETXGenerator.cc => src/cpu/directedtest/InvalidateGenerator.cc rename : src/mem/ruby/tester/DetermGETXGenerator.hh => src/cpu/directedtest/InvalidateGenerator.hh rename : src/cpu/rubytest/RubyTester.cc => src/cpu/directedtest/RubyDirectedTester.cc rename : src/cpu/rubytest/RubyTester.hh => src/cpu/directedtest/RubyDirectedTester.hh rename : src/mem/ruby/tester/DetermGETXGenerator.cc => src/cpu/directedtest/SeriesRequestGenerator.cc rename : src/mem/ruby/tester/DetermGETXGenerator.hh => src/cpu/directedtest/SeriesRequestGenerator.hh
2010-08-20ruby: Updated MOESI_hammer L2 latency behaviorBrad Beckmann
Previously, the MOESI_hammer protocol calculated the same latency for L1 and L2 hits. This was because the protocol was written using the old ruby assumption that L1 hits used the sequencer fast path. Since ruby no longer uses the fast-path, the protocol delays L2 hits by placing them on the trigger queue.
2010-08-20ruby: Reduced ruby latenciesBrad Beckmann
The previous slower ruby latencies created a mismatch between the faster M5 cpu models and the much slower ruby memory system. Specifically smp interrupts were much slower and infrequent, as well as cpus moving in and out of spin locks. The result was many cpus were idle for large periods of time. These changes fix the latency mismatch.
2010-08-20ruby: fix ruby llsc support to sync sc outcomesBrad Beckmann
Added support so that ruby can determine the outcome of store conditional operations and reflect that outcome to M5 physical memory and cpus.
2010-08-20ruby: Fixed L2 cache miss profilingBrad Beckmann
Fixed L2 cache miss profiling for the MOESI_CMP_token protocol
2010-08-20ruby: Added bcast msg profiling to hammer and tokenBrad Beckmann
2010-08-20ruby: Reincarnated the responding machine profilingBrad Beckmann
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.
2010-08-20MOESI_CMP_token: Fixed dma persistent lockdown bugsBrad Beckmann
2010-08-20MESI_CMP_directory: bug fix for old PUTX requestsBrad Beckmann
2010-04-02ruby: get "using namespace" out of headersNathan Binkert
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.
2010-03-21ruby: Removed the unnecessary MachineType message fieldsBrad Beckmann
2010-03-21ruby: Reordered protocol buffersBrad Beckmann
Reordered vnet priorities to agree with PerfectSwitch for protocols MI_example, MOESI_CMP_token, and MOESI_hammer
2010-03-21ruby: Ruby support for sparse memoryBrad Beckmann
The patch includes direct support for the MI example protocol.
2010-03-21ruby: Minor dma latency initialization fixBrad Beckmann
2010-03-21ruby: Fixed small data msg bug in MOESI_hammer-dirBrad Beckmann
2010-03-21ruby: Fix MOESI_hammer cache profiler calls for L2 missesBrad Beckmann
2010-03-10slicc: have a central mechanism for creating a code_formatter.Nathan Binkert
This makes it easier to add global variables like protocol
2010-02-10ruby: Initialize sender in MI_example-dirBrad Beckmann
2010-02-01ruby: Set default protocol back to MI_exampleBrad Beckmann
2010-01-29m5: Regression Tester UpdateBrad Beckmann
This patch includes the necessary regression updates to test the new ruby configuration system. The patch includes support for multiple ruby protocols and adds the ruby random tester. The patch removes atomic mode test for ruby since ruby does not support atomic mode acceses. These tests can be added back in when ruby supports atomic mode for real. --HG-- rename : tests/quick/50.memtest/test.py => tests/quick/60.rubytest/test.py
2010-01-29ruby: removed last level cache supportBrad Beckmann
Removed the last level cache support and MOESI_hammer's dependency on it. Replaces the LLC support with the more generic MachineType count.
2010-01-29ruby: Added a Scons option to prevent HTML file creationBrad Beckmann
2010-01-29ruby: Removed static members in RubyPort including hitcallbackBrad Beckmann
Removed static members in RubyPort and removed the ruby request unique id.
2010-01-29ruby: MESI_CMP_directory updated to the new config systemBrad Beckmann
2010-01-29ruby: Sorted the file includes to maintain consistencyBrad Beckmann
2010-01-29ruby: Renamed the MESI directory sm fileBrad Beckmann
Renamed the MESI directory file to be consistent with all other protocols. --HG-- rename : src/mem/protocol/MESI_CMP_directory-mem.sm => src/mem/protocol/MESI_CMP_directory-dir.sm
2010-01-29ruby: Removed the GPL header in MESI_CMP_directory-msgBrad Beckmann
I'm not sure how this got past our initial ruby code import, but this obviously needed to be removed.
2010-01-29ruby: MOESI_CMP_directory updated to the new config systemBrad Beckmann
2010-01-29ruby: Added atomic support to MOESI_CMP_tokenBrad Beckmann
2010-01-29ruby: fixed memory fetch bug for persistent requestsBrad Beckmann
2010-01-29ruby: MOESI_CMP_token updates to use the new config systemBrad Beckmann
2010-01-29ruby: MI_example updates to use the new config systemBrad Beckmann
2010-01-29ruby: Added atomic support to MOESI_hammerBrad Beckmann
2010-01-29ruby: fixed MOESI_hammer data writebacks to the directoryBrad Beckmann
2010-01-29ruby: reorganized ruby python configurationBrad Beckmann
Reorganized ruby python configuration so that protocol and ruby memory system configuration code can be shared by multiple front-end configuration files (i.e. memory tester, full system, and hopefully the regression tester). This code works for memory tester, but have not tested fs mode.
2010-01-29ruby: Converted MOESI_hammer dma cntrl to new config systemBrad Beckmann
2010-01-29ruby: Ruby changes required to use the python config systemBrad Beckmann
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.
2010-01-29scons: ignore blank lines in .slicc filesSteve Reinhardt
2010-01-29ruby: Make SLICC-generated objects SimObjects.Steve Reinhardt
Also add SLICC support for state-machine parameter defaults (passed through to Python as SimObject Param defaults).
2010-01-29ruby: fix out_port declarationBrad Beckmann
2010-01-19ruby: new atomics implementationDerek Hower
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.