Calculation of offset to copy from storeQueue[idx].data structure for load to
store forwarding fixed to be difference in bytes between store and load virtual
addresses. Previous method would induce bug where a load would index into
buffer at the wrong location.
If a split load fails on a blocked cache wbOutstanding can be decremented
twice if the first part of the split load succeeds and the second part fails.
Condition the decrementing on not having completed the first part of the load.
At the same time, rename the trace flags to debug flags since they
have broader usage than simply tracing. This means that
--trace-flags is now --debug-flags and --trace-help is now --debug-help
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.
The store queue doesn't need to be ISA specific and architectures can
frequently store more than an int registers worth of data. A 128 bits seems
more common, but even 256 bits may be appropriate. Pretty much anything less
than a cache line size is buildable.
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.
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.
A whole bunch of stuff has been converted to use the new params stuff, but
the CPU wasn't one of them. While we're at it, make some things a bit
more stylish. Most of the work was done by Gabe, I just cleaned stuff up
a bit more at the end.
src/cpu/o3/alpha/cpu_impl.hh:
Pass ISA-specific O3 CPU to FullO3CPU as a constructor parameter instead of using setCPU functions.
--HG--
extra : convert_revision : 74f4b1f5fb6f95a56081f367cce7ff44acb5688a
into zamp.eecs.umich.edu:/z/ktlim2/clean/tmp/clean2
src/cpu/base_dyn_inst.hh:
Hand merge. Line is no longer needed because it's handled in the ISA.
--HG--
extra : convert_revision : 0be4067aa38759a5631c6940f0167d48fde2b680
1. Requests are handled more properly now. They assume the memory system takes control of the request upon sending out an access.
2. load-load ordering is maintained.
src/cpu/base_dyn_inst.hh:
Update how requests are handled. The BaseDynInst should not be able to hold a pointer to the request because the request becomes owned by the memory system once it is sent out.
Also include some functions to allow certain status bits to be cleared.
src/cpu/base_dyn_inst_impl.hh:
Update how requests are handled. The BaseDynInst should not be able to hold a pointer to the request because the request becomes owned by the memory system once it is sent out.
src/cpu/o3/fetch_impl.hh:
General correctness fixes. retryPkt is not necessarily always set, so handle it properly. Also consider the cache unblocked only when recvRetry is called.
src/cpu/o3/lsq_unit.hh:
Handle requests a little more correctly. Now that the requests aren't pointed to by the DynInst, be sure to delete the request if it's not being used by the memory system.
Also be sure to not store-load forward from an uncacheable store.
src/cpu/o3/lsq_unit_impl.hh:
Check to make sure load-load ordering was maintained.
Also handle requests a little more correctly.
--HG--
extra : convert_revision : e86bead2886d02443cf77bf7a7a1492845e1690f
Created MemCmd class to wrap enum and provide handy methods to
check attributes, convert to string/int, etc.
--HG--
extra : convert_revision : 57f147ad893443e3a2040c6d5b4cdb1a8033930b
src/cpu/base_dyn_inst.hh:
Extend BaseDynInst a little bit so it can be use as a TC as well (specifically for ll/sc code).
src/cpu/base_dyn_inst_impl.hh:
Add variable to track if the result of the instruction should be recorded.
src/cpu/o3/alpha/cpu_impl.hh:
Clear lock flag upon hwrei.
src/cpu/o3/lsq_unit.hh:
Use ISA specified handling of locked reads.
src/cpu/o3/lsq_unit_impl.hh:
Use ISA specified handling of locked writes.
--HG--
extra : convert_revision : 1f5c789c35deb4b016573c02af4aab60d726c0e5
into zamp.eecs.umich.edu:/z/ktlim2/clean/o3-merge/newmem
src/cpu/memtest/memtest.cc:
src/cpu/memtest/memtest.hh:
src/cpu/simple/timing.hh:
tests/configs/o3-timing-mp.py:
Hand merge.
--HG--
extra : convert_revision : a58cc439eb5e8f900d175ed8b5a85b6c8723e558
src/cpu/o3/commit_impl.hh:
Fixes for compile and sampling.
src/cpu/o3/cpu.cc:
Deallocate and activate threads properly. Also hopefully fix being able to use caches while switching over.
src/cpu/o3/cpu.hh:
Fixes for deallocating and activating threads.
src/cpu/o3/fetch_impl.hh:
src/cpu/o3/lsq_unit.hh:
Handle getting back a BadAddress result from the access.
src/cpu/o3/iew_impl.hh:
More debug output.
src/cpu/o3/lsq_unit_impl.hh:
Fixup store conditional handling (still a bit of a hack, but works now).
Also handle getting back a BadAddress result from the access.
src/cpu/o3/thread_context_impl.hh:
Deallocate context now records if the context should be fully removed.
--HG--
extra : convert_revision : 55f81660602d0e25367ce1f5b0b9cfc62abe7bf9
src/arch/alpha/isa_traits.hh:
This got changed to the wrong version by accident.
src/cpu/base.cc:
Fix up progress event to not schedule itself if the interval is set to 0.
src/cpu/base.hh:
Fix up the CPU Progress Event to not print itself if it's set to 0. Also remove stats_reset_inst (something I added to m5 but isn't necessary here).
src/cpu/base_dyn_inst.hh:
src/cpu/checker/cpu.hh:
Remove float variable of instResult; it's always held within the double part now.
src/cpu/checker/cpu_impl.hh:
Use thread and not cpuXC.
src/cpu/o3/alpha/cpu_builder.cc:
src/cpu/o3/checker_builder.cc:
src/cpu/ozone/checker_builder.cc:
src/cpu/ozone/cpu_builder.cc:
src/python/m5/objects/BaseCPU.py:
Remove stats_reset_inst.
src/cpu/o3/commit_impl.hh:
src/cpu/ozone/lw_back_end_impl.hh:
Get TC, not XCProxy.
src/cpu/o3/cpu.cc:
Switch out updates from the version of m5 I have. Also remove serialize code that got added twice.
src/cpu/o3/iew_impl.hh:
src/cpu/o3/lsq_impl.hh:
src/cpu/thread_state.hh:
Remove code that was added twice.
src/cpu/o3/lsq_unit.hh:
Add back in stats that got lost in the merge.
src/cpu/o3/lsq_unit_impl.hh:
Use proper method to get flags. Also wake CPU if we're coming back from a cache miss.
src/cpu/o3/thread_context_impl.hh:
src/cpu/o3/thread_state.hh:
Support profiling.
src/cpu/ozone/cpu.hh:
Update to use proper typename.
src/cpu/ozone/cpu_impl.hh:
src/cpu/ozone/dyn_inst_impl.hh:
Updates for newmem.
src/cpu/ozone/lw_lsq_impl.hh:
Get flags correctly.
src/cpu/ozone/thread_state.hh:
Reorder constructor initialization, use tc.
src/sim/pseudo_inst.cc:
Allow for loading of symbol file. Be sure to use ThreadContext and not ExecContext.
--HG--
extra : convert_revision : c5657f84155807475ab4a1e20d944bb6f0d79d94
src/cpu/o3/fetch_impl.hh:
Fix ordering so dereference works
src/cpu/o3/lsq_impl.hh:
Check to make sure we didn't squash already
src/cpu/o3/lsq_unit.hh:
Fix for counting squashed retrys in the WB count
src/cpu/o3/lsq_unit_impl.hh:
Make sure to set retryID for stores, and clear it appropriately
--HG--
extra : convert_revision : 689765a1baea7b36f13eb177d65e97b52b6da09f
src/cpu/o3/lsq_unit.hh:
LSQ needs to decrement the WB counter if the load is going to be replayed.
src/cpu/o3/lsq_unit_impl.hh:
LSQ needs to decrement the WB counter if the load is squashed.
--HG--
extra : convert_revision : 20a10baf0d6ab46065e561ddba231251865ebdbd
src/cpu/o3/lsq.hh:
Update to have LSQ work with only one dcache port for all LSQ Units. LSQ has the dcache port, and the LSQ Units must tell the LSQ if the cache has become blocked.
src/cpu/o3/lsq_impl.hh:
Updates to have the LSQ work with only one dcache port for all LSQUnits.
src/cpu/o3/lsq_unit.hh:
src/cpu/o3/lsq_unit_impl.hh:
Update for LSQ to create dcache port instead of LSQUnits. Now LSQUnits are given the dcache port from the LSQ, and also must check the LSQ if the cache is blocked prior to accessing the cache.
--HG--
extra : convert_revision : 2708adbf323f4e7647dc0c1e31ef5bb4596b89f8
