through the following phases.
1) Construct all param contexts
2) Call the checkParams() on each context
3) Build the configuration hierarchy
4) Construct all SimObjects
5) Initialize all SimObjects by calling init() on each one
6) Unserialize the checkpoint
7) Register all statisitcs
8) Check validity of all statistics (after that, no new stats)
9) Reset all stats.
10) Call SimStartup() which calls startup() on all SimObjects,
ParamContexts, and any other object deriving from StartupCallback
SConscript:
no more SimInit() we have SimStartup() now
sim/param.hh:
Make all params have a startup callback.
sim/sim_events.cc:
the init callbacks no longer exist. We can simplify code by
using startup().
sim/sim_object.hh:
Make all SimObjects derive from StartupCallback
--HG--
extra : convert_revision : ab81e259eb5510cc597f7bacb2bfb619fb4cc15f
sim/main.cc:
Get rid of default.ini processing... it's kind of a pain and nobody uses it.
util/tracediff:
Add comments on usage.
--HG--
extra : convert_revision : b811288b2945585d60685684ea88c99d1913fbf3
No more non-intuitive behavior shifts depending on whether
outputDirectory is set (at the expense of backwards compatibility).
outputDirectory is now always valid, defaults to ".".
dev/etherdump.cc:
Use makeOutputStream() to create output file.
New behavior: actually complain if dump file can't
be opened, instead of quietly ignoring the problem.
dev/etherdump.hh:
dev/simconsole.cc:
dev/simconsole.hh:
Use makeOutputStream() to create output file.
sim/builder.cc:
sim/builder.hh:
sim/main.cc:
builderStream() is now *configStream.
sim/serialize.cc:
outputDirectory is now always valid, no need to check.
sim/universe.cc:
Clean up/standardize handling of various output files.
No more non-intuitive behavior shifts depending on whether
outputDirectory is set (at the expense of backwards compatibility).
outputDirectory is now always valid, defaults to ".".
New function makeOutputStream() does "the right thing" to
associate a stream with a filename.
--HG--
extra : convert_revision : a03c58c547221b3906e0d6f55e4a569843f2d646
arch/alpha/vtophys.cc:
PGOFSET -> ALPHA_PGOFSET to avoid include file problems
base/callback.hh:
Added a class to create a callback from a function
base/intmath.hh:
make FloorLog2 inlined
dev/pcidev.cc:
more work in getting pciconfig space happy with different endiannesses
dev/uart.cc:
used an incorrect size for write uint64_t instead of uint8_t
sim/system.cc:
when writing things into system data structures we need to pay
attention to endianness
--HG--
extra : convert_revision : 52f441b5789c45db30ef2f6fd4975cbc7323a381
code
arch/alpha/alpha_memory.cc:
Fixed unaligned trap faults
arch/alpha/ev5.cc:
little more verbose faulting information
kern/linux/linux_system.cc:
more descriptive errors, and the correct offsets from symbols
sim/system.cc:
load local pal symbols
--HG--
extra : convert_revision : 0c81badf77321d5e1a060dcae2d42204e5a1fc84
arch/alpha/ev5.cc:
set the mode explictly rather than having a bool user/notuser
cpu/simple_cpu/simple_cpu.hh:
there is no class Kernel
kern/kernel_stats.cc:
use cpu_mode_num
kern/kernel_stats.hh:
add interrupt mode and use cpu_mode_num rather than constant
kern/linux/linux_system.cc:
kern/linux/linux_system.hh:
kern/system_events.cc:
kern/system_events.hh:
add events to change the mode to/from interrupt
sim/system.cc:
sim/system.hh:
add a pal symbol table
--HG--
extra : convert_revision : 9d30e826b72122062a5ea12d094f94760e75c66a
dev/tsunami_cchip.cc:
updates to ipi handling chipset code
sim/system.cc:
debugSymbolTable, now has symbols from pal, console, and linux
--HG--
extra : convert_revision : c981d857f7e3d75f4c46172809e6d14e5f0a1238
single place so it's easier to work with.
- Add support for binning kernel/user/idle time separately from
lisa's binning stuff, but make the two compatible.
- KernelStats used to directly implement the pImpl idiom, but
it makes more sense to just remove the level of indirection and
make the exec context have a pointer to the stats.
- Factor common code out of LinuxSystem and Tru64System and put
it into the System base class. While doing that, make all
constructors take a pointer to a parameter struct instead of
naming the parameters individually to make it much easier to add
parameters to these classes.
SConscript:
Move the function tracking and binning stuff around.
arch/alpha/ev5.cc:
kernelStats is now a pointer
arch/alpha/pseudo_inst.cc:
kernelStats is now a pointer
the parameters to the system have been moved into their own
struct
base/trace.hh:
provide a little functor class for wrapping a string that
can allow you to define name() in any scope very simply
for use with DPRINTF
cpu/base_cpu.cc:
New order of arguments for consistency.
cpu/exec_context.cc:
kernelStats no longer has the level of indirection in it,
execContext has the indirection now. so, kernelStats is a pointer.
We also need a pointer to the kernelBinning stuff from the system
and we need to figure out if we want to do binning or not.
Move a whole bunch of code into kern_binning.cc so it's all
in the same place.
cpu/exec_context.hh:
We want pointers to the kernel binning/stats stuff and we'll
have the exec_context and system have the level of indirection
instead of having the extra layer in the kernel stats class.
cpu/simple_cpu/simple_cpu.cc:
call through the exec context to do the special binning
stuff.
kern/kernel_stats.cc:
kern/kernel_stats.hh:
Re-organize the stats stuff and remove the level of indirection
(that was there to simplify building) and move the binning stuff
into its own class/file.
kern/linux/linux_system.cc:
kern/linux/linux_system.hh:
kern/tru64/tru64_system.cc:
kern/tru64/tru64_system.hh:
sim/system.cc:
sim/system.hh:
move lots of common system code into the base system class so
that it can be shared between linux, tru64, and whatever else
we decide to support in the future.
Make the constructor take a pointer to a parameter struct so that
it is easier to pass parameters to the parent.
kern/system_events.cc:
move the majority of the binning code into the Kernel::Binning class
in the kern_binning file
kern/system_events.hh:
FnEvents only need to know the bin
create the Idle start event to find the PCBB of the idle
process when it starts.
kern/tru64/tru64_events.cc:
memCtrl -> memctrl
sim/process.cc:
sim/process.hh:
re-order args for consistency
--HG--
extra : convert_revision : 86cb39738c41fcd680f2aad125c9dde000227b2b
in at least on entry in a VectorParam even if the string was
empty.
sim/param.cc:
if there are no parameters, leave the vector empty.
--HG--
extra : convert_revision : d0d41baa8c77651a27ac6a4d02cc388711f47dbe
SConscript:
Add pyconfig/{pyconfig,code}.cc
Add list of object description (.od) files.
Include pyconfig/SConscript.
base/inifile.cc:
Get rid of CPP_PIPE... it never really worked anyway.
base/inifile.hh:
Make load(ifstream&) method public so pyconfig
code can call it.
sim/main.cc:
Handle Python config scripts (end in '.py' instead of '.ini').
sim/pyconfig/m5configbase.py:
Add license.
Fix minor __setattr__ problem (2.3 related?)
--HG--
rename : util/config/m5configbase.py => sim/pyconfig/m5configbase.py
extra : convert_revision : 5e004922f950bfdefced333285584b80ad7ffb83
be more easily re-used. This currently uses some cooked up matching
function that I wrote a while ago, but should probably be changed
to use regular expressions in the future.
add doDebugBreak to control breakpoints on a per SimObject basis
SConscript:
add match
base/stats/events.cc:
base/trace.cc:
Move the object matching code into a separate file so it can be
more easily shared
base/trace.hh:
the object matching code was wrapped up and moved. adapt.
sim/sim_object.cc:
add the doDebugBreak flag that can be set on a per-SimObject
basis. This will be used in the future to control whether or
not debug_break() will actually break for a given object.
provide a function interface that can be called from the debugger.
sim/sim_object.hh:
add the doDebugBreak flag that can be set on a per-SimObject
basis. This will be used in the future to control whether or
not debug_break() will actually break for a given object.
--HG--
extra : convert_revision : 6bf7924de63d41f5ba6b80d579efdf26ba265a8f
kern/linux/linux_system.cc:
Formatting
sim/host.hh:
When we say ULL and UL we really just want a 64-bit value, not
necessarily a long long, so cast the integer to (u)int64_t
this fixes a problem with byte swapping.
--HG--
extra : convert_revision : ad25ace5a7b92a7c928f2d52e531193c91f3f8c8
a realworld file.
arch/alpha/isa_desc:
arch/alpha/pseudo_inst.hh:
implement the readfile pseudo instruction that will read a
chunk of a realworld file.
arch/alpha/pseudo_inst.cc:
implement the readfile pseudo instruction that will read a
chunk of a realworld file. The filename is a per system
parameter and comes from the system itself.
kern/linux/linux_system.cc:
sim/system.hh:
Create a per-system readfile parameter for use by the readfile
pseudo instruction. That way each system can get its own file.
--HG--
extra : convert_revision : 941b3a3e20702a6252b219ca66a6d90da2944c50
arch/alpha/ev5.cc:
cpu/simple_cpu/simple_cpu.cc:
update for new event interface
base/stats/events.cc:
implement the ignore event function which matches sim objects from which
to ignore events.
Make insert event like insert data and make it able to insert many
events in a single transaction with the database.
base/stats/events.hh:
Make it possible to ignore events
sim/sim_object.cc:
make recordEvent a member function of SimObject to implement
the ignore function easily
sim/sim_object.hh:
implement the ignore event stuff in the sim object. This is a
bit of a hack, but an easy place to put it.
--HG--
extra : convert_revision : ba3f25a14ad03662c53fb35514860d69be8cd4f0
base/misc.cc:
Don't dump trace in panic(), SIGABRT handler will do it now.
sim/main.cc:
Add SIGABRT handler that prints curTick and dumps buffered trace (if any).
This doesn't work as well as I would like since the buffered trace records
often contain stale references to stack-resident temporary std::string objects.
Someday we'll have to put in a fix for that.
--HG--
extra : convert_revision : 67576efbf5c10e63e255fc9a9ec520326fd3567b
dev/ide_disk.cc:
Fix to PIO writes and also add command needed for shutdown
dev/pcidev.cc:
Change the panic on write to read-only registers to a debug print. The
kernel tries to write back over all of the PCI registers to restore
the saved SRM state, so we need to let it do this without panicing.
sim/system.cc:
Add back increment of number of running systems to allow trap of halt
work correctly.
--HG--
extra : convert_revision : 84aba4effbec14545f3610c19a8e321d7e7f7cf2
kern/tru64/tru64_system.cc:
make binned_fns a parameter for System in addition to Tru64System. Do all the fnEvents setting at the System level, since that is system-independent.
kern/tru64/tru64_system.hh:
deal with FnEvents in the System, and move some fns over to System.
sim/system.cc:
sim/system.hh:
lift binning stuff into System out of Tru64System
--HG--
extra : convert_revision : 591dee6f2013f5c43037726c529a00682b5cf82e
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
Add support for generic visitors for stats and use them
to implement independent output functions.
Support for mysql output and some initial code for hacking
on mysql output with python
arch/alpha/pseudo_inst.cc:
base/hybrid_pred.cc:
base/hybrid_pred.hh:
base/sat_counter.cc:
base/sat_counter.hh:
cpu/simple_cpu/simple_cpu.cc:
kern/tru64/tru64_events.cc:
sim/main.cc:
sim/process.cc:
sim/process.hh:
sim/sim_events.cc:
sim/sim_object.cc:
sim/system.hh:
update for changes in stats package
base/statistics.cc:
move the python output code to base/stats/puthon.(cc|hh)
and reimplement it as a visitor.
move the text output code to base/stats/text.(cc|hh) and
reimplement it as a visitor.
move the database stuff into base/stats/statdb.(cc|hh) and
get rid of the class. Put everything as globals in the
Statistics::Database namespace.
allocate unique ids for all stats.
directly implement the check routine and get rid of the
various dumping routines since they're now in separate files.
make sure that no two stats have the same name
clean up some loops
base/statistics.hh:
major changes to the statistics package again
lots of code was factored out of statistics.hh into several
separate files in base/stats/ (this will continue)
There are now two Stat package types Result and Counter that
are specified to allow the user to keep the counted type
separate from the result type. They are currently both doubles
but that's an experiment. There is no more per stat ability to
set the type. Statistics::Counter is not the same as Counter!
Implement a visitor for statistics output so that new output
types can be implemented independently from the stats package
itself.
Add a unique id to each stat so that it can be used to keep
track of stats more simply. This number can also be used in
debugging problems with stats.
Tweak the bucket size stuff a bit to make it work better.
fixed VectorDist size bug
cpu/memtest/memtest.cc:
Fix up for changes in stats package
Don't use value() since it doesn't work with binning. If you
want a number as a stat, and to use it in the program itself,
you really want two separate variables, one that's a stat,
and one that's not.
cpu/memtest/memtest.hh:
Fix up for changes in stats package
test/Makefile:
Try to build stuff now that directories matter
test/stattest.cc:
test all new output types
choose which one with command line options
--HG--
extra : convert_revision : e3a3f5f0828c67c0e2de415d936ad240adaddc89
and 'type') to be specified via instance name and not just
config class. Old code only did instance-name lookup for
SimObject parameters. This feature makes life easier for
transitioning to the Python script-based config.
sim/builder.cc:
Use ConfigNode::find to look for "type" parameter so it can
be found if set under instance name (not config class).
sim/param.cc:
Make Param<bool> accept "1" for true and "0" for false.
--HG--
extra : convert_revision : f40d0878d0f03b2e216f0506c05d0e52db608cca
configuration unnecessarily awkward. Biggest changes are:
- External and internal object names now match in all cases. The
macros still allow them to be different; the only reason I didn't
get rid of that is that the macros themselves should be going away
soon. In the few conflicting cases, I sometimes renamed the C++ object
and sometimes renamed the config object. The latter sets of substitions
are:
s/BaseBus/Bus/;
s/MemoryObject/FunctionalMemory/;
s/MemoryControl/MemoryController/;
s/FUPool/FuncUnitPool/;
- SamplingCPU is temporarily broken... we need to change the model
of how this works in the .ini file. Having it as a CPU proxy is
really awkward.
arch/alpha/alpha_memory.cc:
arch/alpha/alpha_memory.hh:
cpu/simple_cpu/simple_cpu.cc:
sim/process.cc:
Rename objects to match config name.
cpu/base_cpu.cc:
Uncomment SimObject define since SamplingCPU no longer
does this for us.
dev/ethertap.cc:
Use unsigned instead of uint16_t for params.
kern/tru64/tru64_system.cc:
Use unsigned instead of uint64_t for init_param param.
test/paramtest.cc:
Fix old SimObjectParam.
--HG--
extra : convert_revision : 378ebbc6a71ad0694501d09979a44d111a59e8dc
on various platforms.
base/hashmap.hh:
gcc on Alpha doesn't always define __LP64__,
even though it arguably should.
cpu/exec_context.cc:
Clear register file on non-full-system too (even though
it typically gets overwritten by the initial regs from
the Process object).
sim/process.cc:
Clear initial register copy in Process object.
Not all regs get initialized when the executable is loaded.
--HG--
extra : convert_revision : f1fe4734a5ea81331d70994cb5284b1e9db0dceb
sim object(depends on which platform tsunami/tlaser is being used)
dev/alpha_console.cc:
Use dynamice casting once during build to get at the proper device to calculate the frequency.
It is either a tsunami_IO or a tlaser_clock depending on platform (tsunami/tlaser).
If the simobject is not of either of those types then panic
dev/alpha_console.hh:
Pass in a SimObject * that is to either a tlaser_clock or a tsunami_IO
sim/sim_object.cc:
Make it so that you can have a SimObjectParam of type SimObject:
example:
SimObjectParam<SimObject *> so;
--HG--
extra : convert_revision : 08239ef23762b8ea11311630b73fe885d939402e
cpu/exec_context.cc:
you can't delete an element of an array that you newed. oops.
kern/tru64/tru64_events.cc:
changes to reflect .ini changes, and also b/c es_intr and ipintr can happen at ANY point, even within a current calling path being tracked.
sim/system.cc:
can't delete an element of a newed array. must new them separately.
--HG--
extra : convert_revision : 21573327b7b7f20bf9a3fcfb5854526433e17e17
cpu/exec_context.cc:
nother little bug...forgot to pop off stack as i read off it
sim/system.cc:
forgot to pop off stack as i read off it
--HG--
extra : convert_revision : d1f691c0a9f0fa22281c717ee465d8a5f1e45c13
to do it:
in .ini file set binned_fns=foo null bar foo
what this says is, track foo when it is called by null (i.e. anything). bin bar only when it is called by foo.
essentially, if you have a path of functions to track, the 0th, 2nd, 4th fn listed are the fns to track, and the 1st, 3rd, 5th are their respective callers.
base/statistics.hh:
turn it back to FS_MEASURE since we already have a build in place
kern/tru64/tru64_events.cc:
remove FS_MEASURE #defs. add more DPRINTF's. manage an anomaly that happens when tracking idle_thread.
kern/tru64/tru64_events.hh:
remove FS_MEASURE #defs
kern/tru64/tru64_system.cc:
make DumpState print all the time, but only with DPRINTF. add a new parameter to tru64System a vector<string> binned_fns, to read in from .ini file. now all this binning stuff is dynamically generated.
kern/tru64/tru64_system.hh:
remove all static binning stuff, add support for dynamic
sim/system.cc:
change nonPath bin name to Kernel, remove FS_MEASURE
sim/system.hh:
change nonPath to Kernel
--HG--
extra : convert_revision : 9ee813c0a64273bab4125815b7bc8145c5897ec1
code.
base/statistics.hh:
We're getting rid of FS_MEASURE, but for now, we're going
to still use a compile time flag to turn on and off binning
of statistics. (The flag is STATS_BINNING)
cpu/exec_context.cc:
cpu/exec_context.hh:
kern/tru64/tru64_system.cc:
get rid of FS_MEASURE
cpu/simple_cpu/simple_cpu.cc:
yank the function call tracking code out of the cpu and move
it into the software context class itself.
kern/tru64/tru64_system.hh:
get rid of FS_MEASURE
move all of the tacking stuff to the same place.
sim/system.hh:
cleanup
--HG--
extra : convert_revision : 73d3843afe1b3ba0d5445421c39c1148d3f4e7c0
