This patch removes the notion of a peer block size and instead sets
the cache line size on the system level.
Previously the size was set per cache, and communicated through the
interconnect. There were plenty checks to ensure that everyone had the
same size specified, and these checks are now removed. Another benefit
that is not yet harnessed is that the cache line size is now known at
construction time, rather than after the port binding. Hence, the
block size can be locally stored and does not have to be queried every
time it is used.
A follow-on patch updates the configuration scripts accordingly.
While FastAlloc provides a small performance increase (~1.5%) over regular malloc it isn't thread safe.
After removing FastAlloc and using tcmalloc I've seen a performance increase of 12% over libc malloc
when running twolf for ARM.
This patch cleans up a number of minor issues aiming to get closer to
compliance with the C++0x standard as interpreted by gcc and clang
(compile with std=c++0x and -pedantic-errors). In particular, the
patch cleans up enums where the last item was succeded by a comma,
namespaces closed by a curcly brace followed by a semi-colon, and the
use of the GNU-extension typeof (replaced by templated functions). It
does not address variable-length arrays, zero-size arrays, anonymous
structs, range expressions in switch statements, and the use of long
long. The generated CPU code also has a large number of issues that
remain to be fixed, mainly related to overflows in implicit constant
conversion (due to shifts).
Enables the CheckerCPU to be selected at runtime with the --checker option
from the configs/example/fs.py and configs/example/se.py configuration
files. Also merges with the SE/FS changes.
This patch continues the unification of how the different CPU models
create and share their instruction and data ports. Most importantly,
it forces every CPU to have an instruction and a data port, and gives
these ports explicit getters in the BaseCPU (getDataPort and
getInstPort). The patch helps in simplifying the code, make
assumptions more explicit, andfurther ease future patches related to
the CPU ports.
The biggest changes are in the in-order model (that was not modified
in the previous unification patch), which now moves the ports from the
CacheUnit to the CPU. It also distinguishes the instruction fetch and
load-store unit from the rest of the resources, and avoids the use of
indices and casting in favour of keeping track of these two units
explicitly (since they are always there anyways). The atomic, timing
and O3 model simply return references to their already existing ports.
This change adds a master id to each request object which can be
used identify every device in the system that is capable of issuing a request.
This is part of the way to removing the numCpus+1 stats in the cache and
replacing them with the master ids. This is one of a series of changes
that make way for the stats output to be changed to python.
Because there are no longer architecture independent but specialized functions
in arch/XXX/faults.hh, code that isn't using the faults from a particular ISA
no longer needs to be able to include them through the switching header file
arch/faults.hh. By removing that header file (arch/faults.hh), the potential
interface between ISA code and non ISA code is narrowed.
Brings the CheckerCPU back to life to allow FS and SE checking of the
O3CPU. These changes have only been tested with the ARM ISA. Other
ISAs potentially require modification.
Only create a memory ordering violation when the value could have changed
between two subsequent loads, instead of just when loads go out-of-order
to the same address. While not very common in the case of Alpha, with
an architecture with a hardware table walker this can happen reasonably
frequently beacuse a translation will miss and start a table walk and
before the CPU re-schedules the faulting instruction another one will
pass it to the same address (or cache block depending on the dendency
checking).
This patch has been tested with a couple of self-checking hand crafted
programs to stress ordering between two cores.
The performance improvement on SPEC benchmarks can be substantial (2-10%).
Having two StaticInst classes, one nominally ISA dependent and the other ISA
dependent, has not been historically useful and makes the StaticInst class
more complicated that it needs to be. This change merges StaticInstBase into
StaticInst.
This constructor assumes that the ExtMachInst can be decoded directly into a
StaticInst that's useful to execute. With the advent of microcoded
instructions that's no longer true.
readBytes and writeBytes had the word "bytes" in their names because they
accessed blobs of bytes. This distinguished them from the read and write
functions which handled higher level data types. Because those functions don't
exist any more, this change renames readBytes and writeBytes to more general
names, readMem and writeMem, which reflect the fact that they are how you read
and write memory. This also makes their names more consistent with the
register reading/writing functions, although those are still read and set for
some reason.
The comment in the code suggests that the checking granularity should be 16
bytes, however in reality the shift by 8 is 256 bytes which seems much
larger than required.
Some ISAs (like ARM) relies on hardware page table walkers. For those ISAs,
when a TLB miss occurs, initiateTranslation() can return with NoFault but with
the translation unfinished.
Instructions experiencing a delayed translation due to a hardware page table
walk are deferred until the translation completes and kept into the IQ. In
order to keep track of them, the IQ has been augmented with a queue of the
outstanding delayed memory instructions. When their translation completes,
instructions are re-executed (only their initiateAccess() was already
executed; their DTB translation is now skipped). The IEW stage has been
modified to support such a 2-pass execution.
For SPARC ASIs are added to the ExtMachInst. If the ASI is changed simply
marking the instruction as Serializing isn't enough beacuse that only
stops rename. This provides a mechanism to squash all the instructions
and refetch them
ARM instructions updating cumulative flags (ARM FP exceptions and saturation
flags) are not serialized.
Added aliases for ARM FP exceptions and saturation flags in FPSCR. Removed
write accesses to the FP condition codes for most ARM VFP instructions: only
VCMP and VCMPE instructions update the FP condition codes. Removed a potential
cause of seg. faults in the O3 model for NEON memory macro-ops (ARM).
This change modifies the way prefetches work. They are now like normal loads
that don't writeback a register. Previously prefetches were supposed to call
prefetch() on the exection context, so they executed with execute() methods
instead of initiateAcc() completeAcc(). The prefetch() methods for all the CPUs
are blank, meaning that they get executed, but don't actually do anything.
On Alpha dead cache copy code was removed and prefetches are now normal ops.
They count as executed operations, but still don't do anything and IsMemRef is
not longer set on them.
On ARM IsDataPrefetch or IsInstructionPreftech is now set on all prefetch
instructions. The timing simple CPU doesn't try to do anything special for
prefetches now and they execute with the normal memory code path.
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.
Also move the "Fault" reference counted pointer type into a separate file,
sim/fault.hh. It would be better to name this less similarly to sim/faults.hh
to reduce confusion, but fault.hh matches the name of the type. We could change
Fault to FaultPtr to match other pointer types, and then changing the name of
the file would make more sense.
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.
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.
When implementing timing address translations instead of atomic, I
forgot to preserve the faults that are returned from the read and
write calls. This patch reinstates them.
When each load or store is sent to the LSQ, we check whether it will cross a
cache line boundary and, if so, split it in two. This creates two TLB
translations and two memory requests. Care has to be taken if the first
packet of a split load is sent but the second blocks the cache. Similarly,
for a store, if the first packet cannot be sent, we must store the second
one somewhere to retry later.
This modifies the LSQSenderState class to record both packets in a split
load or store.
Finally, a new const variable, HasUnalignedMemAcc, is added to each ISA
to indicate whether unaligned memory accesses are allowed. This is used
throughout the changed code so that compiler can optimise away code dealing
with split requests for ISAs that don't need them.
This initiates a timing translation and passes the read or write on to the
processor before waiting for it to finish. Once the translation is finished,
the instruction's state is updated via the 'finish' function. A new
DataTranslation class is created to handle this.
The idea is taken from the implementation of timing translations in
TimingSimpleCPU by Gabe Black. This patch also separates out the timing
translations from this CPU and uses the new DataTranslation class.
the primary identifier for a hardware context should be contextId(). The
concept of threads within a CPU remains, in the form of threadId() because
sometimes you need to know which context within a cpu to manipulate.
across the subclasses. generally make it so that member data is _cpuId and
accessor functions are cpuId(). The ID val comes from the python (default -1 if
none provided), and if it is -1, the index of cpuList will be given. this has
passed util/regress quick and se.py -n4 and fs.py -n4 as well as standard
switch.
