The target of taken conditional direct branches does not
need to be resolved in IEW: the target can be computed at
decode, usually using the decoded instruction word and the PC.
The higher-than-necessary penalty is taken only on conditional
branches that are predicted taken but miss in the BTB. Thus,
this is mostly inconsequential on IPC if the BTB is big/associative
enough (fewer capacity/conflict misses). Nonetheless, what gem5
simulates is not representative of how conditional branch targets
can be handled.
Signed-off-by: Jason Lowe-Power <jason@lowepower.com>
The drain did not wait until stages were ready again. Therefore, as a
result of messages in the TimeBuffer being drain, the state after the
drain was not consistent and asserts fired in some places when the
draining happened after a stage got blocked, but before the notification
arrived to the previous stages.
Change-Id: Ib50b3b40b7f745b62c1eba2931dec76860824c71
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
The o3 pipeline interlock/stall logic is incorrect. o3 unnecessicarily stalled
fetch and decode due to later stages in the pipeline. In general, a stage
should usually only consider if it is stalled by the adjacent, downstream stage.
Forcing stalls due to later stages creates and results in bubbles in the
pipeline. Additionally, o3 stalled the entire frontend (fetch, decode, rename)
on a branch mispredict while the ROB is being serially walked to update the
RAT (robSquashing). Only should have stalled at rename.
O3CPU has a compile-time maximum width set in o3/impl.hh, but checking
the configuration against this limit was not implemented anywhere
except for fetch. Configuring a wider pipe than the limit can silently
cause various issues during the simulation. This patch adds the proper
checking in the constructor of the various pipeline stages.
Fixes the tick used from rename:
- previously this gathered the tick on leaving rename which was always 1 less
than the dispatch. This conflated the decode ticks when back pressure built
in the pipeline.
- now picks up tick on entry.
Added --store_completions flag:
- will additionally display the store completion tail in the viewer.
- this highlights periods when large numbers of stores are outstanding (>16 LSQ
blocking)
Allows selection by tick range (previously this caused an infinite loop)
Fix a case in the O3 CPU where the decode stage blocks and unblocks in a
single cycle sending both signals to fetch which causes an assert or worse.
The previous check could never work before since the status was set to Blocked
before a test for the status being Unblocking was executed.
Previously, the O3 CPU could stop in the middle of a microcode
sequence. This patch makes sure that the pipeline stops when it has
committed a normal instruction or exited from a microcode
sequence. Additionally, it makes sure that the pipeline has no
instructions in flight when it is drained, which should make draining
more robust.
Draining is controlled in the commit stage, which checks if the next
PC after a committed instruction is in microcode. If this isn't the
case, it requests a squash of all instructions after that the
instruction that just committed and immediately signals a drain stall
to the fetch stage. The CPU then continues to execute until the
pipeline and all associated buffers are empty.
1. Updates the Branch Predictor correctly to the state
just after a mispredicted branch, if a squash occurs.
2. If a BTB does not find an entry, the branch is predicted not taken.
The global history is modified to correctly reflect this prediction.
3. Local history is now updated at the fetch stage instead of
execute stage.
4. In the Update stage of the branch predictor the local predictors are
now correctly updated according to the state of local history during
fetch stage.
This patch also improves performance by as much as 17% on some benchmarks
This patch adds the necessary flags to the SConstruct and SConscript
files for compiling using clang 2.9 and later (on Ubuntu et al and OSX
XCode 4.2), and also cleans up a bunch of compiler warnings found by
clang. Most of the warnings are related to hidden virtual functions,
comparisons with unsigneds >= 0, and if-statements with empty
bodies. A number of mismatches between struct and class are also
fixed. clang 2.8 is not working as it has problems with class names
that occur in multiple namespaces (e.g. Statistics in
kernel_stats.hh).
clang has a bug (http://llvm.org/bugs/show_bug.cgi?id=7247) which
causes confusion between the container std::set and the function
Packet::set, and this is currently addressed by not including the
entire namespace std, but rather selecting e.g. "using std::vector" in
the appropriate places.
Implemented a pipeline activity viewer as a python script (util/o3-pipeview.py)
and modified O3 code base to support an extra trace flag (O3PipeView) for
generating traces to be used as inputs by the tool.
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
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.
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
The removed ones were unnecessary. The commented out ones could be useful in the future, should this problem get fixed. See flyspray task #243.
src/cpu/o3/commit_impl.hh:
src/cpu/o3/decode_impl.hh:
src/cpu/o3/fetch_impl.hh:
src/cpu/o3/iew_impl.hh:
src/cpu/o3/inst_queue_impl.hh:
src/cpu/o3/lsq_impl.hh:
src/cpu/o3/lsq_unit_impl.hh:
src/cpu/o3/rename_impl.hh:
src/cpu/o3/rob_impl.hh:
Remove/comment out DPRINTFs that were causing a segfault.
--HG--
extra : convert_revision : b5aeda1c6300dfde5e0a3e9b8c4c5f6fa00b9862
Also don't call (*activeThreads).end() over and over. Just
call activeThreads->end() once and save the result.
Make sure we always check that there are elements in the list
before we grab the first one.
--HG--
extra : convert_revision : d769d8ed52da99532d57a9bbc93e92ddf22b7e58
succesfully but there are some minor quirks to iron out. Who would've known a DELAY SLOT introduces that much complexity?! arrgh!
Anyways, a lot of this stuff had to do with my project at MIPS and me needing to know how I was going to get this working for the MIPS
ISA. So I figured I would try to touch it up and throw it in here (I hate to introduce non-completely working components... )
src/arch/alpha/isa/mem.isa:
spacing
src/arch/mips/faults.cc:
src/arch/mips/faults.hh:
Gabe really authored this
src/arch/mips/isa/decoder.isa:
add StoreConditional Flag to instruction
src/arch/mips/isa/formats/basic.isa:
Steven really did this file
src/arch/mips/isa/formats/branch.isa:
fix bug for uncond/cond control
src/arch/mips/isa/formats/mem.isa:
Adjust O3CPU memory access to use new memory model interface.
src/arch/mips/isa/formats/util.isa:
update LoadStoreBase template
src/arch/mips/isa_traits.cc:
update SERIALIZE partially
src/arch/mips/process.cc:
src/arch/mips/process.hh:
no need for this for NOW. ASID/Virtual addressing handles it
src/arch/mips/regfile/misc_regfile.hh:
add in clear() function and comments for future usage of special misc. regs
src/cpu/base_dyn_inst.hh:
add in nextNPC variable and supporting functions.
add isCondDelaySlot function
Update predTaken and mispredicted functions
src/cpu/base_dyn_inst_impl.hh:
init nextNPC
src/cpu/o3/SConscript:
add MIPS files to compile
src/cpu/o3/alpha/thread_context.hh:
no need for my name on this file
src/cpu/o3/bpred_unit_impl.hh:
Update RAS appropriately for MIPS
src/cpu/o3/comm.hh:
add some extra communication variables to aid in handling the
delay slots
src/cpu/o3/commit.hh:
minor name fix for nextNPC functions.
src/cpu/o3/commit_impl.hh:
src/cpu/o3/decode_impl.hh:
src/cpu/o3/fetch_impl.hh:
src/cpu/o3/iew_impl.hh:
src/cpu/o3/inst_queue_impl.hh:
src/cpu/o3/rename_impl.hh:
Fix necessary variables and functions for squashes with delay slots
src/cpu/o3/cpu.cc:
Update function interface ...
adjust removeInstsNotInROB function to recognize delay slots insts
src/cpu/o3/cpu.hh:
update removeInstsNotInROB
src/cpu/o3/decode.hh:
declare necessary variables for handling delay slot
src/cpu/o3/dyn_inst.hh:
Add in MipsDynInst
src/cpu/o3/fetch.hh:
src/cpu/o3/iew.hh:
src/cpu/o3/rename.hh:
declare necessary variables and adjust functions for handling delay slot
src/cpu/o3/inst_queue.hh:
src/cpu/simple/base.cc:
no need for my name here
src/cpu/o3/isa_specific.hh:
add in MIPS files
src/cpu/o3/scoreboard.hh:
dont include alpha specific isa traits!
src/cpu/o3/thread_context.hh:
no need for my name here, i just rearranged where the file goes
src/cpu/static_inst.hh:
add isCondDelaySlot function
src/cpu/o3/mips/cpu.cc:
src/cpu/o3/mips/cpu.hh:
src/cpu/o3/mips/cpu_builder.cc:
src/cpu/o3/mips/cpu_impl.hh:
src/cpu/o3/mips/dyn_inst.cc:
src/cpu/o3/mips/dyn_inst.hh:
src/cpu/o3/mips/dyn_inst_impl.hh:
src/cpu/o3/mips/impl.hh:
src/cpu/o3/mips/params.hh:
src/cpu/o3/mips/thread_context.cc:
src/cpu/o3/mips/thread_context.hh:
MIPS file for O3CPU...mirrors ALPHA definition
--HG--
extra : convert_revision : 9bb199b4085903e49ffd5a4c8ac44d11460d988c
src/cpu/o3/commit.hh:
src/cpu/o3/commit_impl.hh:
src/cpu/o3/decode.hh:
src/cpu/o3/decode_impl.hh:
src/cpu/o3/fetch.hh:
src/cpu/o3/fetch_impl.hh:
src/cpu/o3/iew.hh:
src/cpu/o3/iew_impl.hh:
src/cpu/o3/rename.hh:
src/cpu/o3/rename_impl.hh:
Return a value so that the CPU can instantly return from draining if the pipeline is already drained.
src/cpu/o3/cpu.cc:
Use values returned from pipeline stages so that the CPU can instantly return from draining if the pipeline is already drained.
--HG--
extra : convert_revision : d8ef6b811644ea67c8b40c4719273fa224105811
First: Rename FullCPU and its variants in the o3 directory to O3CPU to differentiate from the old model, and also to specify it's an out of order model.
Second: Include build options for selecting the Checker to be used. These options make sure if the Checker is being used there is a CPU that supports it also being compiled.
SConstruct:
Add in option USE_CHECKER to allow for not compiling in checker code. The checker is enabled through this option instead of through the CPU_MODELS list. However it's still necessary to treat the Checker like a CPU model, so it is appended onto the CPU_MODELS list if enabled.
configs/test/test.py:
Name change for DetailedCPU to DetailedO3CPU. Also include option for max tick.
src/base/traceflags.py:
Add in O3CPU trace flag.
src/cpu/SConscript:
Rename AlphaFullCPU to AlphaO3CPU.
Only include checker sources if they're necessary. Also add a list of CPUs that support the Checker, and only allow the Checker to be compiled in if one of those CPUs are also being included.
src/cpu/base_dyn_inst.cc:
src/cpu/base_dyn_inst.hh:
Rename typedef to ImplCPU instead of FullCPU, to differentiate from the old FullCPU.
src/cpu/cpu_models.py:
src/cpu/o3/alpha_cpu.cc:
src/cpu/o3/alpha_cpu.hh:
src/cpu/o3/alpha_cpu_builder.cc:
src/cpu/o3/alpha_cpu_impl.hh:
Rename AlphaFullCPU to AlphaO3CPU to differentiate from old FullCPU model.
src/cpu/o3/alpha_dyn_inst.hh:
src/cpu/o3/alpha_dyn_inst_impl.hh:
src/cpu/o3/alpha_impl.hh:
src/cpu/o3/alpha_params.hh:
src/cpu/o3/commit.hh:
src/cpu/o3/cpu.hh:
src/cpu/o3/decode.hh:
src/cpu/o3/decode_impl.hh:
src/cpu/o3/fetch.hh:
src/cpu/o3/iew.hh:
src/cpu/o3/iew_impl.hh:
src/cpu/o3/inst_queue.hh:
src/cpu/o3/lsq.hh:
src/cpu/o3/lsq_impl.hh:
src/cpu/o3/lsq_unit.hh:
src/cpu/o3/regfile.hh:
src/cpu/o3/rename.hh:
src/cpu/o3/rename_impl.hh:
src/cpu/o3/rob.hh:
src/cpu/o3/rob_impl.hh:
src/cpu/o3/thread_state.hh:
src/python/m5/objects/AlphaO3CPU.py:
Rename FullCPU to O3CPU to differentiate from old FullCPU model.
src/cpu/o3/commit_impl.hh:
src/cpu/o3/cpu.cc:
src/cpu/o3/fetch_impl.hh:
src/cpu/o3/lsq_unit_impl.hh:
Rename FullCPU to O3CPU to differentiate from old FullCPU model.
Also #ifdef the checker code so it doesn't need to be included if it's not selected.
--HG--
rename : src/cpu/checker/o3_cpu_builder.cc => src/cpu/checker/o3_builder.cc
rename : src/cpu/checker/cpu_builder.cc => src/cpu/checker/ozone_builder.cc
rename : src/python/m5/objects/AlphaFullCPU.py => src/python/m5/objects/AlphaO3CPU.py
extra : convert_revision : 86619baf257b8b7c8955efd447eba56e0d7acd6a
src/cpu/base_dyn_inst.cc:
src/cpu/base_dyn_inst.hh:
Minor code cleanup by putting several bools into a bitset instead.
src/cpu/o3/commit_impl.hh:
src/cpu/o3/decode_impl.hh:
src/cpu/o3/iew_impl.hh:
src/cpu/o3/inst_queue_impl.hh:
src/cpu/o3/lsq_unit.hh:
src/cpu/o3/lsq_unit_impl.hh:
src/cpu/o3/rename_impl.hh:
src/cpu/o3/rob_impl.hh:
Changed around some things in BaseDynInst.
--HG--
extra : convert_revision : 1db363d69a863cc8744cc9f9ec542ade8472eb42
src/cpu/base_dyn_inst.cc:
Delete the allocated data in destructor.
src/cpu/base_dyn_inst.hh:
Only copy the addresses if the translation succeeded.
src/cpu/o3/alpha_cpu.hh:
Return actual translating port.
Don't panic on setNextNPC() as it's always called, regardless of the architecture, when the process initializes.
src/cpu/o3/alpha_cpu_impl.hh:
Pass in memobject to the thread state in SE mode.
src/cpu/o3/commit_impl.hh:
Initialize all variables.
src/cpu/o3/decode_impl.hh:
Handle early resolution of branches properly.
src/cpu/o3/fetch.hh:
Switch structure back to requests.
src/cpu/o3/fetch_impl.hh:
Initialize all variables, create/delete requests properly.
src/cpu/o3/lsq_unit.hh:
Include sender state along with the packet. Also include a more generic writeback event that's only used for stores forwarding data to loads.
src/cpu/o3/lsq_unit_impl.hh:
Redo writeback code to support the response path of the memory system.
src/cpu/o3/mem_dep_unit.cc:
src/cpu/o3/mem_dep_unit_impl.hh:
Wrap variables in #ifdefs.
src/cpu/o3/store_set.cc:
Include to get panic() function.
src/cpu/o3/thread_state.hh:
Create with MemObject as well.
src/cpu/thread_state.hh:
Have a translating port in the thread state object.
src/python/m5/objects/AlphaFullCPU.py:
Mem parameter no longer needed.
--HG--
extra : convert_revision : a99381fb25cb183322882ce20935a6f3d1f2b64d
