Steve Reinhardt [Thu, 26 Aug 2010 04:55:42 +0000 (21:55 -0700)]
memtest: scale associativity and mshrs according to config
Use the actual fanouts in the tree specification to scale
cache associativity and mshrs instead of dumb constants.
Ali Saidi [Thu, 26 Aug 2010 00:10:43 +0000 (19:10 -0500)]
CPU: Print out traces for faluting inst when the flag ExecFaulting is set
Min Kyu Jeong [Thu, 26 Aug 2010 00:10:43 +0000 (19:10 -0500)]
ARM: Support unaligned memory access.
Without this flag set, page-crossing requests were not split into two mem
request.
Depending on the alignment bit in the SCTLR, misaligned access could
raise a fault. However it seems unnecessary to implement that.
Gene WU [Thu, 26 Aug 2010 00:10:43 +0000 (19:10 -0500)]
ARM: Seperate the queues of L1 and L2 walker states.
Min Kyu Jeong [Thu, 26 Aug 2010 00:10:43 +0000 (19:10 -0500)]
ARM: Adding a bogus fault that does nothing.
This fault can used to flush the pipe, not including the faulting instruction.
The particular case I needed this was for a self-modifying code. It needed to
drain the store queue and force the following instruction to refetch from
icache. DCCMVAC cp15 mcr instruction is modified to raise this fault.
William Wang [Thu, 26 Aug 2010 00:10:43 +0000 (19:10 -0500)]
ARM: Remove ALPHA KSeg functions.
These were erronously copied years ago into the ARM directory.
Ali Saidi [Thu, 26 Aug 2010 00:10:43 +0000 (19:10 -0500)]
ARM: Limited implementation of dprintk.
Does not work with vfp arguments or arguments passed on the stack.
Min Kyu Jeong [Thu, 26 Aug 2010 00:10:43 +0000 (19:10 -0500)]
ARM: Fixed register flattening logic (FP_Base_DepTag was set too low)
When decoding a srs instruction, invalid mode encoding returns invalid instruction.
This can happen when garbage instructions are fetched from mispredicted path
Ali Saidi [Thu, 26 Aug 2010 00:10:43 +0000 (19:10 -0500)]
ARM: Make VMSR, RFE PC/LR etc non speculative, and serializing
Ali Saidi [Thu, 26 Aug 2010 00:10:42 +0000 (19:10 -0500)]
ARM: Update regression tests for ldr/str microcode changes.
Gene WU [Thu, 26 Aug 2010 00:10:42 +0000 (19:10 -0500)]
ARM: Use fewer micro-ops for register update loads if possible.
Allow some loads that update the base register to use just two micro-ops. three
micro-ops are only used if the destination register matches the offset register
or the PC is the destination regsiter. If the PC is updated it needs to be
the last micro-op otherwise O3 will mispredict.
Ali Saidi [Thu, 26 Aug 2010 00:10:42 +0000 (19:10 -0500)]
ARM: Set the high bits in the part number so it's considered new by some code.
Ali Saidi [Thu, 26 Aug 2010 00:10:42 +0000 (19:10 -0500)]
ARM: Fix VFP enabled checks for mem instructions
Gabe Black [Thu, 26 Aug 2010 00:10:42 +0000 (19:10 -0500)]
ARM: Seperate out the renamable bits in the FPSCR.
Gabe Black [Thu, 26 Aug 2010 00:10:42 +0000 (19:10 -0500)]
ARM: Eliminate some unused enums.
Gabe Black [Thu, 26 Aug 2010 00:10:42 +0000 (19:10 -0500)]
ARM: Fix type comparison warnings in Neon.
Gabe Black [Thu, 26 Aug 2010 00:10:42 +0000 (19:10 -0500)]
ARM: Implement CPACR register and return Undefined Instruction when FP access is disabled.
Gabe Black [Thu, 26 Aug 2010 00:10:42 +0000 (19:10 -0500)]
ARM: Implement all ARM SIMD instructions.
Gabe Black [Thu, 26 Aug 2010 00:10:41 +0000 (19:10 -0500)]
ARM: Expand the mode checking utility functions.
inUserMode now can take either a threadcontext or a CPSR value directly. If
given a thread context it just extracts the CPSR and calls the other version.
An inPrivelegedMode function was also implemented which just returns the
opposite of inUserMode.
Ali Saidi [Thu, 26 Aug 2010 00:10:41 +0000 (19:10 -0500)]
Tracing: Fix trace so 'Predicated False' doesn't show up
Steve Reinhardt [Wed, 25 Aug 2010 21:08:27 +0000 (14:08 -0700)]
mem: fix dumb typo in copyrights
Brad Beckmann [Tue, 24 Aug 2010 21:08:23 +0000 (14:08 -0700)]
config: changed ruby config file names to be consistent
Brad Beckmann [Tue, 24 Aug 2010 20:20:32 +0000 (13:20 -0700)]
config: remove ruby's requirement on the timing cmd line param
Since ruby only works in timing mode, explicitly requiring the timing cmd line
param to be specified is not necessary.
Brad Beckmann [Tue, 24 Aug 2010 20:20:31 +0000 (13:20 -0700)]
config: fixed ruby dma device connections
Brad Beckmann [Tue, 24 Aug 2010 19:07:22 +0000 (12:07 -0700)]
testers: move testers to a new directory
This patch moves the testers to a new subdirectory under src/cpu and includes
the necessary fixes to work with latest m5 initialization patches.
--HG--
rename : configs/example/determ_test.py => configs/example/ruby_direct_test.py
rename : src/cpu/directedtest/DirectedGenerator.cc => src/cpu/testers/directedtest/DirectedGenerator.cc
rename : src/cpu/directedtest/DirectedGenerator.hh => src/cpu/testers/directedtest/DirectedGenerator.hh
rename : src/cpu/directedtest/InvalidateGenerator.cc => src/cpu/testers/directedtest/InvalidateGenerator.cc
rename : src/cpu/directedtest/InvalidateGenerator.hh => src/cpu/testers/directedtest/InvalidateGenerator.hh
rename : src/cpu/directedtest/RubyDirectedTester.cc => src/cpu/testers/directedtest/RubyDirectedTester.cc
rename : src/cpu/directedtest/RubyDirectedTester.hh => src/cpu/testers/directedtest/RubyDirectedTester.hh
rename : src/cpu/directedtest/RubyDirectedTester.py => src/cpu/testers/directedtest/RubyDirectedTester.py
rename : src/cpu/directedtest/SConscript => src/cpu/testers/directedtest/SConscript
rename : src/cpu/directedtest/SeriesRequestGenerator.cc => src/cpu/testers/directedtest/SeriesRequestGenerator.cc
rename : src/cpu/directedtest/SeriesRequestGenerator.hh => src/cpu/testers/directedtest/SeriesRequestGenerator.hh
rename : src/cpu/memtest/MemTest.py => src/cpu/testers/memtest/MemTest.py
rename : src/cpu/memtest/SConscript => src/cpu/testers/memtest/SConscript
rename : src/cpu/memtest/memtest.cc => src/cpu/testers/memtest/memtest.cc
rename : src/cpu/memtest/memtest.hh => src/cpu/testers/memtest/memtest.hh
rename : src/cpu/rubytest/Check.cc => src/cpu/testers/rubytest/Check.cc
rename : src/cpu/rubytest/Check.hh => src/cpu/testers/rubytest/Check.hh
rename : src/cpu/rubytest/CheckTable.cc => src/cpu/testers/rubytest/CheckTable.cc
rename : src/cpu/rubytest/CheckTable.hh => src/cpu/testers/rubytest/CheckTable.hh
rename : src/cpu/rubytest/RubyTester.cc => src/cpu/testers/rubytest/RubyTester.cc
rename : src/cpu/rubytest/RubyTester.hh => src/cpu/testers/rubytest/RubyTester.hh
rename : src/cpu/rubytest/RubyTester.py => src/cpu/testers/rubytest/RubyTester.py
rename : src/cpu/rubytest/SConscript => src/cpu/testers/rubytest/SConscript
Brad Beckmann [Tue, 24 Aug 2010 19:06:53 +0000 (12:06 -0700)]
MOESI_hammer: fixed bug for dma reads in single cpu systems
Gabe Black [Mon, 23 Aug 2010 23:23:47 +0000 (16:23 -0700)]
Faults: Get rid of some commented out code in sim/faults.hh.
Gabe Black [Mon, 23 Aug 2010 23:14:24 +0000 (16:14 -0700)]
X86: Create a directory for files that define register indexes.
This is to help tidy up arch/x86. These files should not be used external to
the ISA.
--HG--
rename : src/arch/x86/apicregs.hh => src/arch/x86/regs/apic.hh
rename : src/arch/x86/floatregs.hh => src/arch/x86/regs/float.hh
rename : src/arch/x86/intregs.hh => src/arch/x86/regs/int.hh
rename : src/arch/x86/miscregs.hh => src/arch/x86/regs/misc.hh
rename : src/arch/x86/segmentregs.hh => src/arch/x86/regs/segment.hh
Gabe Black [Mon, 23 Aug 2010 23:14:23 +0000 (16:14 -0700)]
Power: Get rid of unused checkFpEnableFault.
This function was brought in from another ISA and doesn't actually do anything
or get used.
Gabe Black [Mon, 23 Aug 2010 23:14:20 +0000 (16:14 -0700)]
ISA: Get rid of old, unused utility functions cluttering up the ISAs.
Gabe Black [Mon, 23 Aug 2010 16:44:19 +0000 (09:44 -0700)]
X86: Get rid of the flagless microop constructor.
This will reduce clutter in the source and hopefully speed up compilation.
Gabe Black [Mon, 23 Aug 2010 16:44:19 +0000 (09:44 -0700)]
X86: Make the TLB fault instead of panic when something is unmapped in SE mode.
The fault object, if invoked, would then panic. This is a bit less direct, but
it means speculative execution won't panic the simulator.
Gabe Black [Mon, 23 Aug 2010 16:44:19 +0000 (09:44 -0700)]
X86: Make the x86 ExtMachInst serializable with (UN)SERIALIZE_SCALAR.
--HG--
rename : src/arch/x86/types.hh => src/arch/x86/types.cc
Gabe Black [Mon, 23 Aug 2010 16:44:19 +0000 (09:44 -0700)]
X86: Define a noop ExtMachInst.
Gabe Black [Mon, 23 Aug 2010 16:44:19 +0000 (09:44 -0700)]
X86: Mark serializing macroops and regular instructions as such.
Gabe Black [Mon, 23 Aug 2010 16:44:19 +0000 (09:44 -0700)]
X86: Add a .serializing directive that makes a macroop serializing.
This directive really just tells the macroop to set IsSerializing and
IsSerializeAfter on its final microop.
Gabe Black [Mon, 23 Aug 2010 16:44:19 +0000 (09:44 -0700)]
X86: Consolidate extra microop flags into one parameter.
This single parameter replaces the collection of bools that set up various
flavors of microops. A flag parameter also allows other flags to be set like
the serialize before/after flags, etc., without having to change the
constructor.
Gabe Black [Mon, 23 Aug 2010 16:44:19 +0000 (09:44 -0700)]
CPU: Make the constants for StaticInst flags visible outside the class.
Ali Saidi [Mon, 23 Aug 2010 16:18:42 +0000 (11:18 -0500)]
BUILD: GCC 4.4.1/2 have a bug in their auto-vectorizer that we trip on
Ali Saidi [Mon, 23 Aug 2010 16:18:42 +0000 (11:18 -0500)]
ALPHA: The previous O3 patch causes a slight stats change with fullsys.
Min Kyu Jeong [Mon, 23 Aug 2010 16:18:42 +0000 (11:18 -0500)]
O3: Skipping mem-order violation check for uncachable loads.
Uncachable load is not executed until it reaches the head of the ROB,
hence cannot cause one.
Min Kyu Jeong [Mon, 23 Aug 2010 16:18:42 +0000 (11:18 -0500)]
ARM: Improve printing of uop disassembly.
Min Kyu Jeong [Mon, 23 Aug 2010 16:18:41 +0000 (11:18 -0500)]
ARM: Clean up flattening for SPSR adding
Gene Wu [Mon, 23 Aug 2010 16:18:41 +0000 (11:18 -0500)]
ARM: Implement DBG instruction that doesn't do much for now.
Gene Wu [Mon, 23 Aug 2010 16:18:41 +0000 (11:18 -0500)]
MEM: Make CLREX a first class request operation and clear locks in caches when it in received
Gene Wu [Mon, 23 Aug 2010 16:18:41 +0000 (11:18 -0500)]
ARM: Make sure that software prefetch instructions can't change the state of the TLB
Gene Wu [Mon, 23 Aug 2010 16:18:41 +0000 (11:18 -0500)]
ARM: Don't write tracedata on writes, it might have been freed already.
Gene Wu [Mon, 23 Aug 2010 16:18:41 +0000 (11:18 -0500)]
ARM: Implement CLREX init/complete acc methods
Gene Wu [Mon, 23 Aug 2010 16:18:41 +0000 (11:18 -0500)]
ARM: Fix Uncachable TLB requests and decoding of xn bit
Gene Wu [Mon, 23 Aug 2010 16:18:41 +0000 (11:18 -0500)]
Devices: Allow a device to specify that a request is uncachable.
Gene Wu [Mon, 23 Aug 2010 16:18:41 +0000 (11:18 -0500)]
ARM: For non-cachable accesses set the UNCACHABLE flag
Gene Wu [Mon, 23 Aug 2010 16:18:41 +0000 (11:18 -0500)]
ARM: Implement DSB, DMB, ISB
Gene Wu [Mon, 23 Aug 2010 16:18:41 +0000 (11:18 -0500)]
ARM: Get SCTLR TE bit from reset SCTLR
Gene Wu [Mon, 23 Aug 2010 16:18:41 +0000 (11:18 -0500)]
ARM: Implement CLREX
Gene Wu [Mon, 23 Aug 2010 16:18:41 +0000 (11:18 -0500)]
ARM: BX instruction can be contitional if last instruction in a IT block
Branches are allowed to be the last instuction in an IT block. Before it was
assumed that they could not. So Branches in thumb2 were Uncond.
Min Kyu Jeong [Mon, 23 Aug 2010 16:18:41 +0000 (11:18 -0500)]
CPU: Print out flatten-out register index as with IntRegs/FloatRegs traceflag
Min Kyu Jeong [Mon, 23 Aug 2010 16:18:41 +0000 (11:18 -0500)]
CPU: Make Exec trace to print predication result (if false) for memory instructions
Min Kyu Jeong [Mon, 23 Aug 2010 16:18:41 +0000 (11:18 -0500)]
ARM: mark msr/mrs instructions as SerializeBefore/After
Since miscellaneous registers bypass wakeup logic, force serialization
to resolve data dependencies through them
* * *
ARM: adding non-speculative/serialize flags for instructions change CPSR
Min Kyu Jeong [Mon, 23 Aug 2010 16:18:40 +0000 (11:18 -0500)]
O3: Handle loads when the destination is the PC.
For loads that PC is the destination, check if the load
was mispredicted again when the value being loaded returns from memory
Min Kyu Jeong [Mon, 23 Aug 2010 16:18:40 +0000 (11:18 -0500)]
ARM/O3: store the result of the predicate evaluation in DynInst or Threadstate.
THis allows the CPU to handle predicated-false instructions accordingly.
This particular patch makes loads that are predicated-false to be sent
straight to the commit stage directly, not waiting for return of the data
that was never requested since it was predicated-false.
Min Kyu Jeong [Mon, 23 Aug 2010 16:18:40 +0000 (11:18 -0500)]
ARM: adding genMachineCheckFault() stub for ARM that doesn't panic
Gene Wu [Mon, 23 Aug 2010 16:18:40 +0000 (11:18 -0500)]
ARM: DFSR status value for sync external data abort is expected to be 0x8 in ARMv7
Gene Wu [Mon, 23 Aug 2010 16:18:40 +0000 (11:18 -0500)]
ARM: Temporary local variables can't conflict with isa parser operands.
PC is an operand, so we can't have a temp called PC
Ali Saidi [Mon, 23 Aug 2010 16:18:40 +0000 (11:18 -0500)]
ARM: Exclusive accesses must be double word aligned
Ali Saidi [Mon, 23 Aug 2010 16:18:40 +0000 (11:18 -0500)]
ARM: Add some registers for big loads/stores to support neon.
Ali Saidi [Mon, 23 Aug 2010 16:18:40 +0000 (11:18 -0500)]
ARM: Decode neon memory instructions.
Gabe Black [Mon, 23 Aug 2010 16:18:40 +0000 (11:18 -0500)]
ARM: Clean up the ISA desc portion of the ARM memory instructions.
Ali Saidi [Mon, 23 Aug 2010 16:18:40 +0000 (11:18 -0500)]
Loader: Don't insert symbols into the symbol table that begin wiht '$'.
Ali Saidi [Mon, 23 Aug 2010 16:18:40 +0000 (11:18 -0500)]
ARM: We don't currently support ThumbEE exceptions, so don't report that we do
Ali Saidi [Mon, 23 Aug 2010 16:18:40 +0000 (11:18 -0500)]
ARM: Change how the AMBA device ID checking is done to make it more generic
Ali Saidi [Mon, 23 Aug 2010 16:18:40 +0000 (11:18 -0500)]
ARM: Add configuration for Linux/Full System
Ali Saidi [Mon, 23 Aug 2010 16:18:40 +0000 (11:18 -0500)]
ARM: Add system for ARM/Linux and bootstrapping
Ali Saidi [Mon, 23 Aug 2010 16:18:40 +0000 (11:18 -0500)]
ARM: Add I/O devices for booting linux
--HG--
rename : src/dev/arm/Versatile.py => src/dev/arm/RealView.py
rename : src/dev/arm/versatile.cc => src/dev/arm/realview.cc
rename : src/dev/arm/versatile.hh => src/dev/arm/realview.hh
Ali Saidi [Mon, 23 Aug 2010 16:18:40 +0000 (11:18 -0500)]
ARM: Implement some more misc registers
Ali Saidi [Mon, 23 Aug 2010 16:18:39 +0000 (11:18 -0500)]
ARM: Fix an un-initialized variable bug
Ali Saidi [Mon, 23 Aug 2010 16:18:39 +0000 (11:18 -0500)]
Loader: Use address mask provided to load*Symbols when loading the symbols from the symbol table.
Ali Saidi [Mon, 23 Aug 2010 16:18:39 +0000 (11:18 -0500)]
Loader: Make the load address mask be a parameter of the system rather than a constant.
This allows one two different OS requirements for the same ISA to be handled.
Some OSes are compiled for a virtual address and need to be loaded into physical
memory that starts at address 0, while other bare metal tools generate
images that start at address 0.
Min Kyu Jeong [Mon, 23 Aug 2010 16:18:39 +0000 (11:18 -0500)]
ARM: Finish the timing translation when taking a fault.
Dam Sunwoo [Mon, 23 Aug 2010 16:18:39 +0000 (11:18 -0500)]
ARM: Use a stl queue for the table walker state
Ali Saidi [Mon, 23 Aug 2010 16:18:39 +0000 (11:18 -0500)]
CPU: Set a default value when readBytes faults.
This was being done in read(), but if readBytes was called directly it
wouldn't happen. Also, instead of setting the memory blob being read to -1
which would (I believe) require using memset with -1 as a parameter, this now
uses bzero. It's hoped that it's more specialized behavior will make it
slightly faster.
Ali Saidi [Mon, 23 Aug 2010 16:18:39 +0000 (11:18 -0500)]
Compiler: Fixes for GCC 4.5.
Ali Saidi [Mon, 23 Aug 2010 16:18:39 +0000 (11:18 -0500)]
BASE: Fix genrand to generate both 0s and 1s when max equals one.
previously was only generating 0s.
Ali Saidi [Mon, 23 Aug 2010 16:18:39 +0000 (11:18 -0500)]
stats: Fix off-by-one error in distributions.
bkt size isn't evenly divisible by max-min and it would round down,
it's possible to sample a distribution and have no place to put the sample.
When this case occured the simulator would assert.
Gabe Black [Mon, 23 Aug 2010 01:42:23 +0000 (18:42 -0700)]
X86: Get rid of unused file arguments.hh.
Gabe Black [Mon, 23 Aug 2010 01:39:39 +0000 (18:39 -0700)]
SPARC: Fix some style issues in utility.hh.
Gabe Black [Mon, 23 Aug 2010 01:24:09 +0000 (18:24 -0700)]
X86: Get rid of the unused getAllocator on the python base microop class.
This function is always overridden, and doesn't actually have the right
signature.
Brad Beckmann [Sat, 21 Aug 2010 00:44:26 +0000 (17:44 -0700)]
regress: Regression tester updates
Regression tester updates required by the following patches:
brad/moved_python_protocol_files: config: moved python protocol config files
brad/ruby_options_movement: config: reorganized how ruby specifies command-line options
brad/config_token_bcast: ruby: added token broadcast config params to cmd options
brad/topology_name: config: Added the topology description to m5 config.ini
brad/ruby_system_names: config: Improve ruby simobject names
brad/consolidated_protocol_stats: slicc: Consolidated the protocol stats printing
brad/ruby_request_type_ostream_fix: ruby: Added ruby_request_type ostream def to libruby.hh
brad/memtest_dma_extension: memtest: Memtester support for DMA
brad/token_dma_lockdown_fix: MOESI_CMP_token: Fixed dma persistent lockdown bugs
brad/profile_generic_mach_type: ruby: Reincarnated the responding machine profiling
brad/network_msg_consolidated_stats: ruby: Added consolidated network msg stats
brad/bcast_msg_profiling: ruby: Added bcast msg profiling to hammer and token
brad/l2cache_profiling_fix: ruby: Fixed L2 cache miss profiling
brad/llsc_ruby_m5_fix: ruby: fix ruby llsc support to sync sc outcomes
brad/ruby_latency_fixes: ruby: Reduced ruby latencies
brad/hammer_l2_cache_latency: ruby: Updated MOESI_hammer L2 latency behavior
brad/deterministic_resurrection: ruby: Resurrected Ruby's deterministic tests
brad/token_dma_fixes: ruby: MOESI_CMP_token dma fixes
brad/ruby_cmd_options: config: added cmd options to control ruby debug
brad/token_owner_fixes: ruby: fixed token bugs associated with owner token counts
brad/ruby_remove_try_except: ruby: Improved try except blocks in ruby creation
brad/ruby_port_callback_fix: ruby: Fixed RubyPort sendTiming callbacks
brad/interrupt_drain_fix: devices: Fixed periodic interrupts to work with draining
brad/llsc_trace_profile: ruby: Added SC fail indication to trace profiling
brad/no_migrate_atomic: ruby: Disable migratory sharing for token and hammer
brad/ruby_start_time_fix: ruby: Reset ruby stats in RubySystem unserialize
brad/numa_bit_select_fix: ruby: fixed DirectoryMemory's numa_high_bit configuration
brad/hammer_probe_filter: ruby: added probe filter support to hammer
brad/miss_latency_detail_profile: MOESI_hammer: break down miss latency stalled cycles
brad/recycle_latency_fix: ruby: Recycle latency fix for hammer
brad/stall_and_wait: ruby: Stall and wait input messages instead of recycling
brad/rubytest_request_flag_fix: ruby: Fixed minor bug in ruby test for setting the request type
brad/hammer_merge_gets: ruby: Added merge GETS optimization to hammer
brad/regress_updates: regress: Regression tester updates
Brad Beckmann [Fri, 20 Aug 2010 18:46:14 +0000 (11:46 -0700)]
ruby: Added merge GETS optimization to hammer
Added an optimization that merges multiple pending GETS requests into a
single request to the owner node.
Brad Beckmann [Fri, 20 Aug 2010 18:46:14 +0000 (11:46 -0700)]
ruby: Fixed minor bug in ruby test for setting the request type
Brad Beckmann [Fri, 20 Aug 2010 18:46:14 +0000 (11:46 -0700)]
ruby: Stall and wait input messages instead of recycling
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
Brad Beckmann [Fri, 20 Aug 2010 18:46:14 +0000 (11:46 -0700)]
ruby: Recycle latency fix for hammer
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.
Brad Beckmann [Fri, 20 Aug 2010 18:46:14 +0000 (11:46 -0700)]
MOESI_hammer: break down miss latency stalled cycles
This patch tracks the number of cycles a transaction is delayed at different
points of the request-forward-response loop.
Brad Beckmann [Fri, 20 Aug 2010 18:46:14 +0000 (11:46 -0700)]
ruby: added probe filter support to hammer
Brad Beckmann [Fri, 20 Aug 2010 18:46:13 +0000 (11:46 -0700)]
ruby: fixed DirectoryMemory's numa_high_bit configuration
This fix includes the off-by-one bit selection bug for numa mapping.
Brad Beckmann [Fri, 20 Aug 2010 18:46:13 +0000 (11:46 -0700)]
ruby: Reset ruby stats in RubySystem unserialize
The main purpose for clearing stats in the unserialize process is so
that the profiler can correctly set its start time to the unserialized
value of curTick.
Brad Beckmann [Fri, 20 Aug 2010 18:46:13 +0000 (11:46 -0700)]
ruby: Disable migratory sharing for token and hammer
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.
Brad Beckmann [Fri, 20 Aug 2010 18:46:13 +0000 (11:46 -0700)]
ruby: Added SC fail indication to trace profiling
Brad Beckmann [Fri, 20 Aug 2010 18:46:13 +0000 (11:46 -0700)]
devices: Fixed periodic interrupts to work with draining
Added drain functions to the RTC and 8254 timer so that periodic interrupts
stop when the system is draining. This patch is needed to checkpoint in
timing mode. Otherwise under certain situations, the event queue will never
be completely empty.
Brad Beckmann [Fri, 20 Aug 2010 18:46:13 +0000 (11:46 -0700)]
ruby: Fixed RubyPort sendTiming callbacks
Fixed RubyPort schedSendTiming calls to match ruby frequency.
Brad Beckmann [Fri, 20 Aug 2010 18:46:13 +0000 (11:46 -0700)]
ruby: Improved try except blocks in ruby creation
Replaced the sys.exit in the try-except blocks with raise so that the python
call stack will be printed