CPU cycle ticks. This allows the user to have CPUs of different
frequencies, and also allows frequencies and latencies that are
not evenly divisible by the CPU frequency. For now, the CPU
frequency is still set to the global frequency, but soon, we'll
hopefully make the global frequency fixed at something like 1THz
and set all other frequencies independently.
arch/alpha/ev5.cc:
The cycles counter is based on the current cpu cycle.
cpu/base_cpu.cc:
frequency isn't the cpu parameter anymore, cycleTime is.
cpu/base_cpu.hh:
frequency isn't the cpu parameter anymore, cycleTime is.
create several public functions for getting the cpu frequency
and the numbers of ticks for a given number of cycles, etc.
cpu/memtest/memtest.cc:
cpu/simple_cpu/simple_cpu.cc:
cpu/simple_cpu/simple_cpu.hh:
cpu/trace/trace_cpu.cc:
Now that ticks aren't cpu cycles, fixup code to advance
by the proper number of ticks.
cpu/memtest/memtest.hh:
cpu/trace/trace_cpu.hh:
Provide a function to get the number of ticks for a given
number of cycles.
dev/alpha_console.cc:
Update for changes in the way that frequencies and latencies are
accessed. Move some stuff to init()
dev/alpha_console.hh:
Need a pointer to the system and the cpu to get the frequency
so we can pass the info to the console code.
dev/etherbus.cc:
dev/etherbus.hh:
dev/etherlink.cc:
dev/etherlink.hh:
dev/ethertap.cc:
dev/ide_disk.hh:
dev/ns_gige.cc:
dev/ns_gige.hh:
update for changes in the way bandwidths are passed from
python to C++ to accomidate the new way that ticks works.
dev/ide_disk.cc:
update for changes in the way bandwidths are passed from
python to C++ to accomidate the new way that ticks works.
Add some extra debugging printfs
dev/platform.cc:
dev/sinic.cc:
dev/sinic.hh:
outline the constructor and destructor
dev/platform.hh:
outline the constructor and destructor.
don't keep track of the interrupt frequency. Only provide the
accessor function.
dev/tsunami.cc:
dev/tsunami.hh:
outline the constructor and destructor
Don't set the interrupt frequency here. Get it from the actual device
that does the interrupting.
dev/tsunami_io.cc:
dev/tsunami_io.hh:
Make the interrupt interval a configuration parameter. (And convert
the interval to the new latency/frequency stuff in the python)
kern/linux/linux_system.cc:
update for changes in the way bandwidths are passed from
python to C++ to accomidate the new way that ticks works.
For now, we must get the boot cpu's frequency as a parameter
since allowing the system to have a pointer to the boot cpu would
cause a cycle.
kern/tru64/tru64_system.cc:
For now, we must get the boot cpu's frequency as a parameter
since allowing the system to have a pointer to the boot cpu would
cause a cycle.
python/m5/config.py:
Fix support for cycle_time relative latencies and frequencies.
Add support for getting a NetworkBandwidth or a MemoryBandwidth.
python/m5/objects/BaseCPU.mpy:
All CPUs now have a cycle_time. The default is the global frequency,
but it is now possible to set the global frequency to some large value
(like 1THz) and set each CPU frequency independently.
python/m5/objects/BaseCache.mpy:
python/m5/objects/Ide.mpy:
Make this a Latency parameter
python/m5/objects/BaseSystem.mpy:
We need to pass the boot CPU's frequency to the system
python/m5/objects/Ethernet.mpy:
Update parameter types to use latency and bandwidth types
python/m5/objects/Platform.mpy:
this frequency isn't needed. We get it from the clock interrupt.
python/m5/objects/Tsunami.mpy:
The clock generator should hold the frequency
sim/eventq.hh:
Need to remove this assertion because the writeback event
queue is different from the CPU's event queue which can cause
this assertion to fail.
sim/process.cc:
Fix comment.
sim/system.hh:
Struct member to hold the boot CPU's frequency.
sim/universe.cc:
remove unneeded variable.
--HG--
extra : convert_revision : 51efe4041095234bf458d9b3b0d417f4cae16fdc
cpu/base_cpu.cc:
cpu/base_cpu.hh:
Convert the CPU stuff to use a params struct
cpu/memtest/memtest.cc:
The memory tester is really not a cpu, so don't derive from
BaseCPU since it just makes things a pain in the butt. Keep
track of max loads in the memtest class now that the base class
doesn't do it for us.
Don't have any default parameters.
cpu/memtest/memtest.hh:
The memory tester is really not a cpu, so don't derive from
BaseCPU since it just makes things a pain in the butt. Keep
track of max loads in the memtest class now that the base class
doesn't do it for us.
cpu/simple_cpu/simple_cpu.cc:
Convert to use a params struct.
remove default parameters
cpu/simple_cpu/simple_cpu.hh:
convert to use a params struct
cpu/trace/opt_cpu.cc:
cpu/trace/opt_cpu.hh:
cpu/trace/trace_cpu.cc:
cpu/trace/trace_cpu.hh:
this isn't really a cpu. don't derive from BaseCPU
objects/MemTest.mpy:
we only need one max_loads parameter
sim/main.cc:
Don't check for the number of CPUs since we may be doing something
else going on. If we don't have anything to simulate, the
simulator will exit anyway.
--HG--
extra : convert_revision : 2195a34a9ec90b5414324054ceb3bab643540dd5
Increase the default number of CSHR's, we should really fix this or make it a parameter
Use a setBlocked call to tell the bus it should block
New technique for sampling and switchover:
1) Sampler switchover event happens
2) All cpus in the current phase of sampling associated with this sampler are signaled to switchover
3) Each cpu drains it's pipe of things being executed (stops fetching and waits for empty pipe)
4) Once the pipe is empty the cpu calls back to the sampler to signal it has finished, and moves into the switchedout state (continues not to fetch)
5) The sampler collects all the signals, once all cpus are drained it calls the new cpu's in the next phase to takeover from the correct cpu
6) The statistics are reset and the next switchover time is calculated from this point
cpu/base_cpu.cc:
cpu/base_cpu.hh:
cpu/simple_cpu/simple_cpu.cc:
cpu/simple_cpu/simple_cpu.hh:
Reconfigure the way the sampling switchover works
cpu/pc_event.cc:
More debugging information on kernel panic's
kern/linux/linux_system.cc:
More debug info for Kernel Die events
kern/linux/linux_system.hh:
More debug info for kernel die events
--HG--
extra : convert_revision : 61cc42e43ba738705aa1f1d167b65d4d6dee51ae
every time the committed PC changes from one symbol scope
to another.
Set function_trace=y on target CPU to enable.
To defer start, use function_trace_start=<tick>
(in addition to setting function_trace=y).
cpu/base_cpu.cc:
cpu/base_cpu.hh:
Add simple function-tracing support.
cpu/simple_cpu/simple_cpu.cc:
Add function_trace, function_trace_start params
Call traceFunctions() on instruction completion
cpu/simple_cpu/simple_cpu.hh:
Add function_trace, function_trace_start params
--HG--
extra : convert_revision : 8a7f84028ccbaee585253629007f32fc8eae35e1
checkInterrupts variable and use that to determine whether an interrupt
can occur on a given cycle.
arch/alpha/ev5.cc:
XC -> CPU (and xc -> CPU) since we're really talking about a CPU here
Don't use the global check_interrupts variable. Add a per-cpu
checkInterrupts variable and use that to determine whether an interrupt
can occur on a given cycle.
--HG--
extra : convert_revision : be4c0247e5834005c60a45796a222cffd327b64e
the registration stuff all moves into BaseCPU
cpu/base_cpu.cc:
Move the registration stuff into the BaseCPU since all
other CPUs use it.
cpu/base_cpu.hh:
Move the defer registration stuff into the BaseCPU since all
other CPUs use it.
cpu/simple_cpu/simple_cpu.cc:
cpu/simple_cpu/simple_cpu.hh:
registration stuff moved to base class
sim/system.cc:
the activation of exec contexts should happen at startup, not
when they are registered.
sim/system.hh:
the system now has a startup function
--HG--
extra : convert_revision : bb6a7c2da5a1ecf5fe7ede1078200bfe5245f8ef
proxies for a real C/C++ scalar value or scalar functor.
This replaces the scalar() and functor() terms that were
previously used in formulas. This helps when dumping
statistics because the formulas are not supposed to change.
cpu/base_cpu.cc:
Add a number of cycles stat to the cpu object that tracks the
number of cycles that the cpu has executed. This starts to pave
the way for cpu cycles being different from event ticks.
cpu/base_cpu.hh:
provide a functor for calculating all simulated instructions
of all CPUs and a virtual function for determining that number.
To deal with the change from functor() to Value::functor()
cpu/simple_cpu/simple_cpu.cc:
simTicks -> numCycles
numInsts is now a real Scalar stat, not a Formula
cpu/simple_cpu/simple_cpu.hh:
numInsts is now a real Scalar stat, not a Formula
count all instructions
sim/stat_control.cc:
simInsts, simTicks, hostMemory, and hostSeconds are no
longer Statistics::Formula but rather Statistics::Value
add new stat for tick frequency
sim/stats.hh:
don't need everything to be extern.
test/Makefile:
Make stuff work a tad bit better
test/stattest.cc:
test out Statistics::Value
--HG--
extra : convert_revision : c812e8baa2b17c08abf3a68ed1e1125dc6f2cfb4
in adaptec_ctrl.hh
cpu/base_cpu.cc:
changed index to 64bits
cpu/base_cpu.hh:
changed index to 64 bits
--HG--
extra : convert_revision : e70d5f09f6066b90fca82cae22bb7d7eb705d65e
and SimpleCPU::setStatus() into separate functions. For example,
setStatus(Active) is now activate().
--HG--
extra : convert_revision : 4392e07caf6c918db0b535f613175109681686fe
This fixes detailed-mpboot, which was broken as of my last change.
Also clean up some of the ExecContext status initialization.
cpu/base_cpu.hh:
CPU::execCtxStatusChg() now takes thread_num as an arg so the CPU knows
which execContext had the status change.
BaseCPU::registerExecContexts() no longer needs to be virtual.
cpu/exec_context.cc:
Initialize _status directly... don't use setStatus() as this will notify the CPU
of the change before it is ready.
CPU::execCtxStatusChg() now takes thread_num as an arg so the CPU knows
which execContext had the status change.
cpu/exec_context.hh:
Don't need initStatus() any more.
cpu/simple_cpu/simple_cpu.cc:
Move execCtxStatusChg() from header to .cc file.
No longer need specialized version of registerExecContexts to schedule TickEvent.
cpu/simple_cpu/simple_cpu.hh:
Move execCtxStatusChg() from header to .cc file.
CPU::execCtxStatusChg() now takes thread_num as arg (must be 0 for SimpleCPU).
No longer need specialized version of registerExecContexts to schedule TickEvent.
kern/tru64/tru64_system.cc:
Don't need initRegs; the PC etc. get initialized in the CPU constructor.
ExecContexts start out as Unallocated, so no need to set them to Unallocated here.
kern/tru64/tru64_system.hh:
Don't need initRegs; the PC etc. get initialized in the CPU constructor.
sim/prog.cc:
ExecContexts start out as Unallocated, so no need to set them to Unallocated here.
--HG--
extra : convert_revision : e960ebbeb845960344633798e251b6c8bf1c0378
of CPUs that get switched round-robin (though currently we're only shooting for
two CPUs and one switch event, and even that doesn't quite work yet). Registration
of ExecContexts with System/Process object factored out so we can create two CPUs
but only register one of them at a time. Also worked at making behavior and naming
in System and Process objects more consistent.
arch/alpha/ev5.cc:
Rename ipr_init to initIPRs and get rid of unused mem arg.
arch/alpha/fake_syscall.cc:
Process:numCpus is now a function (not a data member).
base/remote_gdb.hh:
Support for ExecContext switching.
cpu/base_cpu.cc:
cpu/base_cpu.hh:
cpu/exec_context.cc:
cpu/exec_context.hh:
cpu/simple_cpu/simple_cpu.hh:
Support for ExecContext switching.
Renamed contexts array to execContexts to be consistent with Process.
CPU ID now auto-assigned by system object.
cpu/simple_cpu/simple_cpu.cc:
Support for ExecContext switching.
Renamed contexts array to execContexts to be consistent with Process.
CPU ID now auto-assigned by system object.
Cleaned up MP full-system initialization a bit.
dev/alpha_console.cc:
Renamed xcvec array to execContexts to be consistent with Process.
kern/tru64/tru64_system.cc:
kern/tru64/tru64_system.hh:
Support for ExecContext switching.
CPU ID now auto-assigned by system object.
sim/prog.cc:
sim/prog.hh:
Support for ExecContext switching.
Process:numCpus is now a function (not a data member).
sim/system.cc:
sim/system.hh:
Support for ExecContext switching.
Renamed xcvec array to execContexts to be consistent with Process.
--HG--
extra : convert_revision : 79649cffad5bf3e83de8df44236941907926d791
Two new parameters for the CPU models, max_loads_any_thread and max_loads_all_threads.
cpu/memtest/memtest.cc:
cpu/memtest/memtest.hh:
Swap out maxReads for the new commited loads model.
--HG--
extra : convert_revision : 35031329bbc476122b2203104537a9f8b46addfa
