These classes are all basically empty now that Alpha has been deleted,
except in cases where the arch versions had copied versions of the Alpha
code.
This change pulls all the generic logic out of the arch versions, making
the arch versions much simpler and making it clearer what the core
functionality of the class is, and what parts are architecture specific
details.
In the future, the way the StackTrace class is instantiated should be
delegated to the Workload class so that ISA agnostic code doesn't need
to know about a particular ISA's StackTrace class, and so that
StackTrace logic can, at least theoretically, be specialized for a
particular workload. The way a stack trace is collected could vary from
OS to OS, for example.
Change-Id: Id8108f94e9fe8baf9b4056f2b6404571e9fa52f1
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/30961
Reviewed-by: Gabe Black <gabeblack@google.com>
Maintainer: Gabe Black <gabeblack@google.com>
Tested-by: kokoro <noreply+kokoro@google.com>
This generalized Workload SimObject is not geared towards FS or SE
simulations, although currently it's only used in FS. This gets rid
of the ARM specific highestELIs64 property (from the workload, not the
system) and replaces it with a generic getArch.
The old globally accessible kernel symtab has been replaced with a
symtab accessor which takes a ThreadContext *. The parameter isn't used
for anything for now, but in cases where there might be multiple
symbol tables to choose from (kernel vs. current user space?) the
method will now be able to distinguish which to use. This also makes
it possible for the workload to manage its symbol table with whatever
policy makes sense for it.
That method returns a const SymbolTable * since most of the time the
symbol table doesn't need to be modified. In the one case where an
external entity needs to modify the table, two pseudo instructions,
the table to modify isn't necessarily the one that's currently active.
For instance, the pseudo instruction will likely execute in user space,
but might be intended to add a symbol to the kernel in case something
like a module was loaded.
To support that usage, the workload has a generic "insertSymbol" method
which will insert the symbol in the table that "makes sense". There is
a lot of ambiguity what that means, but it's no less ambiguous than
today where we're only saved by the fact that there is generally only
one active symbol table to worry about.
This change also introduces a KernelWorkload SimObject class which
inherits from Workload and adds in kernel related members for cases
where the kernel is specified in the config and loaded by gem5 itself.
That's the common case, but the base Workload class would be used
directly when, for instance, doing a baremetal simulation or if the
kernel is loaded by software within the simulation as is the case for
SPARC FS.
Because a given architecture specific workload class needs to inherit
from either Workload or KernelWorkload, this change removes the
ability to boot ARM without a kernel. This ability should be restored
in the future.
To make having or not having a kernel more flexible, the kernel
specific members of the KernelWorkload should be factored out into
their own object which can then be attached to a workload through a
(potentially unused) property rather than inheritance.
Change-Id: Idf72615260266d7b4478d20d4035ed5a1e7aa241
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/24283
Reviewed-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Maintainer: Gabe Black <gabeblack@google.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Information about what kernel to load and how to load it was built
into the System object and its subclasses. That overloaded the System
object and made it responsible for too many things, and also was
somewhat awkward when working with SE mode which doesn't have a kernel.
This change extracts the kernel and information related to it from the
System object and puts into into a OsKernel or Workload object.
Currently the idea of a "Workload" to run and a kernel are a bit
muddled, an unfortunate carry-over from the original code. It's also an
implication of trying not to make too sweeping of a change, and to
minimize the number of times configs need to change, ie avoiding
creating a "kernel" parameter which would shortly thereafter be
renamed to "workload".
In future changes, the ideas of a kernel and a workload will be
disentangled, and workloads will be expanded to include emulated
operating systems which shephard and contain Process-es for syscall
emulation.
This change was originally split into pieces to make reviewing it
easier. Those reviews are here:
https: //gem5-review.googlesource.com/c/public/gem5/+/22243
https: //gem5-review.googlesource.com/c/public/gem5/+/24144
https: //gem5-review.googlesource.com/c/public/gem5/+/24145
https: //gem5-review.googlesource.com/c/public/gem5/+/24146
https: //gem5-review.googlesource.com/c/public/gem5/+/24147
https: //gem5-review.googlesource.com/c/public/gem5/+/24286
Change-Id: Ia3d863db276a023b6a2c7ee7a656d8142ff75589
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/26466
Reviewed-by: Gabe Black <gabeblack@google.com>
Maintainer: Gabe Black <gabeblack@google.com>
Tested-by: kokoro <noreply+kokoro@google.com>
This is mostly only a superficial change since the isa parameter is
then dynamic cast to the ISA specific version inside the various
consumers, currently the SimpleThread, O3CPU and Decoder classes. If
those aren't being used, for instance in the fast model CPUs, then you
can use a different ISA implementation without any type clashes.
Change-Id: I2226ef60f9a471ae51b8bfce8683033f7854197a
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/25009
Reviewed-by: Gabe Black <gabeblack@google.com>
Maintainer: Gabe Black <gabeblack@google.com>
Tested-by: kokoro <noreply+kokoro@google.com>
This switches to letting the ThreadContexts use a thread based/local
comInstEventQueue instead of falling back to the CPU's array. Because
the implementation is no longer shared and it's not given where the
comInstEventQueue (or other implementation) should be accessed, the
default implementation has been removed.
Also, because nobody is using the CPU's array of event queues, those
have been removed.
Change-Id: I515e6e00a2174067a928c33ef832bc5c840bdf7f
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/22110
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
Tested-by: kokoro <noreply+kokoro@google.com>
This changeset adds support for partial (or masked) loads/stores, i.e.
loads/stores that can disable accesses to individual bytes within the
target address range. In addition, this changeset extends the code to
crack memory accesses across most CPU models (TimingSimpleCPU still
TBD), so that arbitrarily wide memory accesses are supported. These
changes are required for supporting ISAs with wide vectors.
Additional authors:
- Gabor Dozsa <gabor.dozsa@arm.com>
- Tiago Muck <tiago.muck@arm.com>
Change-Id: Ibad33541c258ad72925c0b1d5abc3e5e8bf92d92
Signed-off-by: Giacomo Gabrielli <giacomo.gabrielli@arm.com>
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/13518
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com>
Maintainer: Nikos Nikoleris <nikos.nikoleris@arm.com>
CPUs have historically instantiated the architecture specific version
of the TLBs to avoid a virtual function call, making them a little bit
more dependent on what the current ISA is. Some simple performance
measurement, the x86 twolf regression on the atomic CPU, shows that
there isn't actually any performance benefit, and if anything the
simulator goes slightly faster (although still within margin of error)
when the TLB functions are virtual.
This change switches everything outside of the architectures themselves
to use the generic BaseTLB type, and then inside the ISA for them to
cast that to their architecture specific type to call into architecture
specific interfaces.
The ARM TLB needed the most adjustment since it was using non-standard
translation function signatures. Specifically, they all took an extra
"type" parameter which defaulted to normal, and translateTiming
returned a Fault. translateTiming actually doesn't need to return a
Fault because everywhere that consumed it just stored it into a
structure which it then deleted(?), and the fault is stored in the
Translation object when the translation is done.
A little more work is needed to fully obviate the arch/tlb.hh header,
so the TheISA::TLB type is still visible outside of the ISAs.
Specifically, the TlbEntry type is used in the generic PageTable which
lives in src/mem.
Change-Id: I51b68ee74411f9af778317eff222f9349d2ed575
Reviewed-on: https://gem5-review.googlesource.com/6921
Maintainer: Gabe Black <gabeblack@google.com>
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
The quiesce family of magic ops can be simplified by the inclusion of
quiesceTick() and quiesce() functions on ThreadContext. This patch also
gets rid of the FS guards, since suspending a CPU is also a valid
operation for SE mode.
This changeset adds support for changing the simulator output
directory. This can be useful when the simulation goes through several
stages (e.g., a warming phase, a simulation phase, and a verification
phase) since it allows the output from each stage to be located in a
different directory. Relocation is done by calling core.setOutputDir()
from Python or simout.setOutputDirectory() from C++.
This change affects several parts of the design of the gem5's output
subsystem. First, files returned by an OutputDirectory instance (e.g.,
simout) are of the type OutputStream instead of a std::ostream. This
allows us to do some more book keeping and control re-opening of files
when the output directory is changed. Second, new subdirectories are
OutputDirectory instances, which should be used to create files in
that sub-directory.
Signed-off-by: Andreas Sandberg <andreas@sandberg.pp.se>
[sascha.bischoff@arm.com: Rebased patches onto a newer gem5 version]
Signed-off-by: Sascha Bischoff <sascha.bischoff@arm.com>
Signed-off-by: Andreas Sandberg <andreas.sandberg@arm.com>
Objects that are can be serialized are supposed to inherit from the
Serializable class. This class is meant to provide a unified API for
such objects. However, so far it has mainly been used by SimObjects
due to some fundamental design limitations. This changeset redesigns
to the serialization interface to make it more generic and hide the
underlying checkpoint storage. Specifically:
* Add a set of APIs to serialize into a subsection of the current
object. Previously, objects that needed this functionality would
use ad-hoc solutions using nameOut() and section name
generation. In the new world, an object that implements the
interface has the methods serializeSection() and
unserializeSection() that serialize into a named /subsection/ of
the current object. Calling serialize() serializes an object into
the current section.
* Move the name() method from Serializable to SimObject as it is no
longer needed for serialization. The fully qualified section name
is generated by the main serialization code on the fly as objects
serialize sub-objects.
* Add a scoped ScopedCheckpointSection helper class. Some objects
need to serialize data structures, that are not deriving from
Serializable, into subsections. Previously, this was done using
nameOut() and manual section name generation. To simplify this,
this changeset introduces a ScopedCheckpointSection() helper
class. When this class is instantiated, it adds a new /subsection/
and subsequent serialization calls during the lifetime of this
helper class happen inside this section (or a subsection in case
of nested sections).
* The serialize() call is now const which prevents accidental state
manipulation during serialization. Objects that rely on modifying
state can use the serializeOld() call instead. The default
implementation simply calls serialize(). Note: The old-style calls
need to be explicitly called using the
serializeOld()/serializeSectionOld() style APIs. These are used by
default when serializing SimObjects.
* Both the input and output checkpoints now use their own named
types. This hides underlying checkpoint implementation from
objects that need checkpointing and makes it easier to change the
underlying checkpoint storage code.
Another churn to clean up undefined behaviour, mostly ARM, but some
parts also touching the generic part of the code base.
Most of the fixes are simply ensuring that proper intialisation. One
of the more subtle changes is the return type of the sign-extension,
which is changed to uint64_t. This is to avoid shifting negative
values (undefined behaviour) in the ISA code.
activate(), suspend(), and halt() used on thread contexts had an optional
delay parameter. However this parameter was often ignored. Also, when used,
the delay was seemily arbitrarily set to 0 or 1 cycle (no other delays were
ever specified). This patch removes the delay parameter and 'Events'
associated with them across all ISAs and cores. Unused activate logic
is also removed.
We currently generate and compile one version of the ISA code per CPU
model. This is obviously wasting a lot of resources at compile
time. This changeset factors out the interface into a separate
ExecContext class, which also serves as documentation for the
interface between CPUs and the ISA code. While doing so, this
changeset also fixes up interface inconsistencies between the
different CPU models.
The main argument for using one set of ISA code per CPU model has
always been performance as this avoid indirect branches in the
generated code. However, this argument does not hold water. Booting
Linux on a simulated ARM system running in atomic mode
(opt/10.linux-boot/realview-simple-atomic) is actually 2% faster
(compiled using clang 3.4) after applying this patch. Additionally,
compilation time is decreased by 35%.
The changes made by the changeset 270c9a75e91f do not work well with switching
of cpus. The problem is that decoder for the old thread context holds state
that is not taken over by the new decoder.
This patch adds a takeOverFrom() function to Decoder class in each ISA. Except
for x86, functions in other ISAs are blank. For x86, the function copies state
from the old decoder to the new decoder.
The changes made by the changeset 9376 were not quite correct. The patch made
changes to the code which resulted in decoder not getting initialized correctly
when the state was restored from a checkpoint.
This patch adds a startup function to each ISA object. For x86, this function
sets the required state in the decoder. For other ISAs, the function is empty
right now.
The thread context handover code used to break when multiple handovers
were performed during the same quiesce period. Previously, the thread
contexts would assign the TC pointer in the old quiesce event to the
new TC. This obviously broke in cases where multiple switches were
performed within the same quiesce period, in which case the TC pointer
in the quiesce event would point to an old CPU.
The new implementation deschedules pending quiesce events in the old
TC and schedules a new quiesce event in the new TC. The code has been
refactored to remove most of the code duplication.
The O3 CPU used to copy its thread context to a SimpleThread in order
to do serialization. This was a bit of a hack involving two static
SimpleThread instances and a magic constructor that was only used by
the O3 CPU.
This patch moves the ThreadContext serialization code into two global
procedures that, in addition to the normal serialization parameters,
take a ThreadContext reference as a parameter. This allows us to reuse
the serialization code in all ThreadContext implementations.
After making the ISA an independent SimObject, it is serialized
automatically by the Python world. Previously, this just resulted in
an empty ISA section. This patch moves the contents of the ISA to that
section and removes the explicit ISA serialization from the thread
contexts, which makes it behave like a normal SimObject during
serialization.
Note: This patch breaks checkpoint backwards compatibility! Use the
cpt_upgrader.py utility to upgrade old checkpoints to the new format.
The ISA class on stores the contents of ID registers on many
architectures. In order to make reset values of such registers
configurable, we make the class inherit from SimObject, which allows
us to use the normal generated parameter headers.
This patch introduces a Python helper method, BaseCPU.createThreads(),
which creates a set of ISAs for each of the threads in an SMT
system. Although it is currently only needed when creating
multi-threaded CPUs, it should always be called before instantiating
the system as this is an obvious place to configure ID registers
identifying a thread/CPU.
This interface is no longer used, and getting rid of it simplifies the
decoders and code that sets up the decoders. The thread context had been used
to read architectural state which was used to contextualize the instruction
memory as it came in. That was changed so that the state is now sent to the
decoders to keep locally if/when it changes. That's significantly more
efficient.
Committed by: Nilay Vaish <nilay@cs.wisc.edu>
This patch addresses the comments and feedback on the preceding patch
that reworks the clocks and now more clearly shows where cycles
(relative cycle counts) are used to express time.
Instead of bumping the existing patch I chose to make this a separate
patch, merely to try and focus the discussion around a smaller set of
changes. The two patches will be pushed together though.
This changes done as part of this patch are mostly following directly
from the introduction of the wrapper class, and change enough code to
make things compile and run again. There are definitely more places
where int/uint/Tick is still used to represent cycles, and it will
take some time to chase them all down. Similarly, a lot of parameters
should be changed from Param.Tick and Param.Unsigned to
Param.Cycles.
In addition, the use of curTick is questionable as there should not be
an absolute cycle. Potential solutions can be built on top of this
patch. There is a similar situation in the o3 CPU where
lastRunningCycle is currently counting in Cycles, and is still an
absolute time. More discussion to be had in other words.
An additional change that would be appropriate in the future is to
perform a similar wrapping of Tick and probably also introduce a
Ticks class along with suitable operators for all these classes.
These classes are always used together, and merging them will give the ISAs
more flexibility in how they cache things and manage the process.
--HG--
rename : src/arch/x86/predecoder_tables.cc => src/arch/x86/decoder_tables.cc
This pointer was only being stored in code that came from SE mode. The system
pointer is always meaningful and available, so it should always be stored.
This patch is a trivial simplification, removing the cpu pointer from
SimpleThread and relying on the baseCpu pointer in ThreadState. The
patch does not add or change any functionality, it merely cleans up
the code.
Port proxies are used to replace non-structural ports, and thus enable
all ports in the system to correspond to a structural entity. This has
the advantage of accessing memory through the normal memory subsystem
and thus allowing any constellation of distributed memories, address
maps, etc. Most accesses are done through the "system port" that is
used for loading binaries, debugging etc. For the entities that belong
to the CPU, e.g. threads and thread contexts, they wrap the CPU data
port in a port proxy.
The following replacements are made:
FunctionalPort > PortProxy
TranslatingPort > SETranslatingPortProxy
VirtualPort > FSTranslatingPortProxy
--HG--
rename : src/mem/vport.cc => src/mem/fs_translating_port_proxy.cc
rename : src/mem/vport.hh => src/mem/fs_translating_port_proxy.hh
rename : src/mem/translating_port.cc => src/mem/se_translating_port_proxy.cc
rename : src/mem/translating_port.hh => src/mem/se_translating_port_proxy.hh
This change is a low level and pervasive reorganization of how PCs are managed
in M5. Back when Alpha was the only ISA, there were only 2 PCs to worry about,
the PC and the NPC, and the lsb of the PC signaled whether or not you were in
PAL mode. As other ISAs were added, we had to add an NNPC, micro PC and next
micropc, x86 and ARM introduced variable length instruction sets, and ARM
started to keep track of mode bits in the PC. Each CPU model handled PCs in
its own custom way that needed to be updated individually to handle the new
dimensions of variability, or, in the case of ARMs mode-bit-in-the-pc hack,
the complexity could be hidden in the ISA at the ISA implementation's expense.
Areas like the branch predictor hadn't been updated to handle branch delay
slots or micropcs, and it turns out that had introduced a significant (10s of
percent) performance bug in SPARC and to a lesser extend MIPS. Rather than
perpetuate the problem by reworking O3 again to handle the PC features needed
by x86, this change was introduced to rework PC handling in a more modular,
transparent, and hopefully efficient way.
PC type:
Rather than having the superset of all possible elements of PC state declared
in each of the CPU models, each ISA defines its own PCState type which has
exactly the elements it needs. A cross product of canned PCState classes are
defined in the new "generic" ISA directory for ISAs with/without delay slots
and microcode. These are either typedef-ed or subclassed by each ISA. To read
or write this structure through a *Context, you use the new pcState() accessor
which reads or writes depending on whether it has an argument. If you just
want the address of the current or next instruction or the current micro PC,
you can get those through read-only accessors on either the PCState type or
the *Contexts. These are instAddr(), nextInstAddr(), and microPC(). Note the
move away from readPC. That name is ambiguous since it's not clear whether or
not it should be the actual address to fetch from, or if it should have extra
bits in it like the PAL mode bit. Each class is free to define its own
functions to get at whatever values it needs however it needs to to be used in
ISA specific code. Eventually Alpha's PAL mode bit could be moved out of the
PC and into a separate field like ARM.
These types can be reset to a particular pc (where npc = pc +
sizeof(MachInst), nnpc = npc + sizeof(MachInst), upc = 0, nupc = 1 as
appropriate), printed, serialized, and compared. There is a branching()
function which encapsulates code in the CPU models that checked if an
instruction branched or not. Exactly what that means in the context of branch
delay slots which can skip an instruction when not taken is ambiguous, and
ideally this function and its uses can be eliminated. PCStates also generally
know how to advance themselves in various ways depending on if they point at
an instruction, a microop, or the last microop of a macroop. More on that
later.
Ideally, accessing all the PCs at once when setting them will improve
performance of M5 even though more data needs to be moved around. This is
because often all the PCs need to be manipulated together, and by getting them
all at once you avoid multiple function calls. Also, the PCs of a particular
thread will have spatial locality in the cache. Previously they were grouped
by element in arrays which spread out accesses.
Advancing the PC:
The PCs were previously managed entirely by the CPU which had to know about PC
semantics, try to figure out which dimension to increment the PC in, what to
set NPC/NNPC, etc. These decisions are best left to the ISA in conjunction
with the PC type itself. Because most of the information about how to
increment the PC (mainly what type of instruction it refers to) is contained
in the instruction object, a new advancePC virtual function was added to the
StaticInst class. Subclasses provide an implementation that moves around the
right element of the PC with a minimal amount of decision making. In ISAs like
Alpha, the instructions always simply assign NPC to PC without having to worry
about micropcs, nnpcs, etc. The added cost of a virtual function call should
be outweighed by not having to figure out as much about what to do with the
PCs and mucking around with the extra elements.
One drawback of making the StaticInsts advance the PC is that you have to
actually have one to advance the PC. This would, superficially, seem to
require decoding an instruction before fetch could advance. This is, as far as
I can tell, realistic. fetch would advance through memory addresses, not PCs,
perhaps predicting new memory addresses using existing ones. More
sophisticated decisions about control flow would be made later on, after the
instruction was decoded, and handed back to fetch. If branching needs to
happen, some amount of decoding needs to happen to see that it's a branch,
what the target is, etc. This could get a little more complicated if that gets
done by the predecoder, but I'm choosing to ignore that for now.
Variable length instructions:
To handle variable length instructions in x86 and ARM, the predecoder now
takes in the current PC by reference to the getExtMachInst function. It can
modify the PC however it needs to (by setting NPC to be the PC + instruction
length, for instance). This could be improved since the CPU doesn't know if
the PC was modified and always has to write it back.
ISA parser:
To support the new API, all PC related operand types were removed from the
parser and replaced with a PCState type. There are two warts on this
implementation. First, as with all the other operand types, the PCState still
has to have a valid operand type even though it doesn't use it. Second, using
syntax like PCS.npc(target) doesn't work for two reasons, this looks like the
syntax for operand type overriding, and the parser can't figure out if you're
reading or writing. Instructions that use the PCS operand (which I've
consistently called it) need to first read it into a local variable,
manipulate it, and then write it back out.
Return address stack:
The return address stack needed a little extra help because, in the presence
of branch delay slots, it has to merge together elements of the return PC and
the call PC. To handle that, a buildRetPC utility function was added. There
are basically only two versions in all the ISAs, but it didn't seem short
enough to put into the generic ISA directory. Also, the branch predictor code
in O3 and InOrder were adjusted so that they always store the PC of the actual
call instruction in the RAS, not the next PC. If the call instruction is a
microop, the next PC refers to the next microop in the same macroop which is
probably not desirable. The buildRetPC function advances the PC intelligently
to the next macroop (in an ISA specific way) so that that case works.
Change in stats:
There were no change in stats except in MIPS and SPARC in the O3 model. MIPS
runs in about 9% fewer ticks. SPARC runs with 30%-50% fewer ticks, which could
likely be improved further by setting call/return instruction flags and taking
advantage of the RAS.
TODO:
Add != operators to the PCState classes, defined trivially to be !(a==b).
Smooth out places where PCs are split apart, passed around, and put back
together later. I think this might happen in SPARC's fault code. Add ISA
specific constructors that allow setting PC elements without calling a bunch
of accessors. Try to eliminate the need for the branching() function. Factor
out Alpha's PAL mode pc bit into a separate flag field, and eliminate places
where it's blindly masked out or tested in the PC.
This reduces the scope of those includes and makes it less likely for there to
be a dependency loop. This also moves the hashing functions associated with
ExtMachInst objects to be with the ExtMachInst definitions and out of
utility.hh.
