| Age | Commit message (Collapse) | Author |
|---|
|
The o3 cpu could attempt to schedule inactive threads under round-robin SMT
mode.
This is because it maintained an independent priority list of threads from the
active thread list. This priority list could be come stale once threads were
inactive, leading to the cpu trying to fetch/commit from inactive threads.
Additionally the fetch queue is now forcibly flushed of instrctuctions
from the de-scheduled thread.
Relevant output:
24557000: system.cpu: [tid:1]: Calling deactivate thread.
24557000: system.cpu: [tid:1]: Removing from active threads list
24557500: system.cpu:
FullO3CPU: Ticking main, FullO3CPU.
24557500: system.cpu.fetch: Running stage.
24557500: system.cpu.fetch: Attempting to fetch from [tid:1]
|
|
When a branch mispredicted gem5 would squash all history after and including
the mispredicted branch. However, the mispredicted branch is still speculative
and its history is required to rollback state if another, older, branch
mispredicts. This leads to things like RAS corruption.
|
|
This patch adds a fetch queue that sits between fetch and decode to the
o3 cpu. This effectively decouples fetch from decode stalls allowing it
to be more aggressive, running futher ahead in the instruction stream.
|
|
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.
|
|
As highlighed on the mailing list gem5's writeback modeling can impact
performance. This patch removes the limitation on maximum outstanding issued
instructions, however the number that can writeback in a single cycle is still
respected in instToCommit().
|
|
isa_parser.py guesses the OpClass if none were given based upon the StaticInst
flags. The existing code does not take into account optionally set flags.
This code hoists the setting of optional flags so OpClass is properly assigned.
|
|
If a set of LL/SC requests contend on the same cache block we
can get into a situation where CPUs will deadlock if they expect
a failed SC to supply them data. This case happens where 3 or
more cores are contending for a cache block using LL/SC and the system
is configured where 2 cores are connected to a local bus and the
third is connected to a remote bus. If a core on the local bus
sends an SCUpgrade and the core on the remote bus sends and SCUpgrade
they will race to see who will win the SC access. In the meantime
if the other core appends a read to one of the SCUpgrades it will expect
to be supplied data by that SCUpgrade transaction. If it happens that
the SCUpgrade that was picked to supply the data is failed, it will
drop the appended request for data and never respond, leaving the requesting
core to deadlock. This patch makes all SC's behave as normal stores to
prevent this case but still makes sure to check whether it can perform
the update.
|
|
|
|
No change in functionality, just a bit of tidying up.
|
|
The first DPRINTF() in PL390::writeDistributor always read a uint32_t, though a
packet may have only been 1 or 2 bytes. This caused an assertion in
packet->get().
|
|
This patch makes restoring the 'lastStopped' value for Ticked-containing
objects (including MinorCPU) optional so that Ticked-containing objects
can be restored from non-Ticked-containing objects (such as AtomicSimpleCPU).
|
|
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%.
|
|
This patch prunes unused values, and also unifies how the values are
defined (not using an enum for ALPHA), aligning the use of int vs Addr
etc.
The patch also removes the duplication of PageBytes/PageShift and
VMPageSize/LogVMPageSize. For all ISAs the two pairs had identical
values and the latter has been removed.
|
|
When passed from a configuration script with a hexadecimal value (like
"0x80000000"), gem5 would error out. This is because it would call
"toMemorySize" which requires the argument to end with a size specifier (like
1MB, etc).
This modification makes it so raw hex values can be passed through Addr
parameters from the configuration scripts.
|
|
This patch fixes the hash operator used for ARM ExtMachInst, which
incorrectly was still using uint32_t. Instead of changing it to
uint64_t it is not using the underlying data type of the BitUnion.
|
|
The Index type defined as typedef int64 does not really provide any help
since in most places we use primitive types instead of Index. Also, the name
Index is very generic that it does not merit being used as a typename.
|
|
This patch sets op class of two fp instructions: movfp and pop x87 stack
as IntAluOp since these instructions do not make use of the fp alu.
|
|
This patch moves code from the wakeup() function to a operateVnet().
The aim is to improve the readiblity of the code.
|
|
This patch is the final patch in a series of patches. The aim of the series
is to make ruby more configurable than it was. More specifically, the
connections between controllers are not at all possible (unless one is ready
to make significant changes to the coherence protocol). Moreover the buffers
themselves are magically connected to the network inside the slicc code.
These connections are not part of the configuration file.
This patch makes changes so that these connections will now be made in the
python configuration files associated with the protocols. This requires
each state machine to expose the message buffers it uses for input and output.
So, the patch makes these buffers configurable members of the machines.
The patch drops the slicc code that usd to connect these buffers to the
network. Now these buffers are exposed to the python configuration system
as Master and Slave ports. In the configuration files, any master port
can be connected any slave port. The file pyobject.cc has been modified to
take care of allocating the actual message buffer. This is inline with how
other port connections work.
|
|
A later changeset changes the file src/python/swig/pyobject.cc to include
a header file that includes a header file generated at build time depending
on the PROTOCOL in use. Since NULL ISA was not specifying any protocol,
this resulted in compilation problems. Hence, the changeset.
|
|
The namespace Message conflicts with the Message data type used extensively
in Ruby. Since Ruby is being moved to the same Master/Slave ports based
configuration style as the rest of gem5, this conflict needs to be resolved.
Hence, the namespace is being renamed to ProtoMessage.
|
|
There are two changes this patch makes to the way configurable members of a
state machine are specified in SLICC. The first change is that the data
member declarations will need to be separated by a semi-colon instead of a
comma. Secondly, the default value to be assigned would now use SLICC's
assignment operator i.e. ':='.
|
|
This patch changes the grammar for SLICC so as to remove some of the
redundant / duplicate rules. In particular rules for object/variable
declaration and class member declaration have been unified. Similarly, the
rules for a general function and a class method have been unified.
One more change is in the priority of two rules. The first rule is on
declaring a function with all the params typed and named. The second rule is
on declaring a function with all the params only typed. Earlier the second
rule had a higher priority. Now the first rule has a higher priority.
|
|
|
|
This changeset does away with prefixing of member variables of state machines
with the identity of the machine itself.
|
|
|
|
All the implementations were doing the same things.
|
|
There is another type Time in src/base class which results in a conflict.
|
|
The directory ruby/system is crowded and unorganized. Hence, the files the
hold actual physical structures, are being moved to the directory
ruby/structures. This includes Cache Memory, Directory Memory,
Memory Controller, Wire Buffer, TBE Table, Perfect Cache Memory, Timer Table,
Bank Array.
The directory ruby/systems has the glue code that holds these structures
together.
--HG--
rename : src/mem/ruby/system/MachineID.hh => src/mem/ruby/common/MachineID.hh
rename : src/mem/ruby/buffers/MessageBuffer.cc => src/mem/ruby/network/MessageBuffer.cc
rename : src/mem/ruby/buffers/MessageBuffer.hh => src/mem/ruby/network/MessageBuffer.hh
rename : src/mem/ruby/buffers/MessageBufferNode.cc => src/mem/ruby/network/MessageBufferNode.cc
rename : src/mem/ruby/buffers/MessageBufferNode.hh => src/mem/ruby/network/MessageBufferNode.hh
rename : src/mem/ruby/system/AbstractReplacementPolicy.hh => src/mem/ruby/structures/AbstractReplacementPolicy.hh
rename : src/mem/ruby/system/BankedArray.cc => src/mem/ruby/structures/BankedArray.cc
rename : src/mem/ruby/system/BankedArray.hh => src/mem/ruby/structures/BankedArray.hh
rename : src/mem/ruby/system/Cache.py => src/mem/ruby/structures/Cache.py
rename : src/mem/ruby/system/CacheMemory.cc => src/mem/ruby/structures/CacheMemory.cc
rename : src/mem/ruby/system/CacheMemory.hh => src/mem/ruby/structures/CacheMemory.hh
rename : src/mem/ruby/system/DirectoryMemory.cc => src/mem/ruby/structures/DirectoryMemory.cc
rename : src/mem/ruby/system/DirectoryMemory.hh => src/mem/ruby/structures/DirectoryMemory.hh
rename : src/mem/ruby/system/DirectoryMemory.py => src/mem/ruby/structures/DirectoryMemory.py
rename : src/mem/ruby/system/LRUPolicy.hh => src/mem/ruby/structures/LRUPolicy.hh
rename : src/mem/ruby/system/MemoryControl.cc => src/mem/ruby/structures/MemoryControl.cc
rename : src/mem/ruby/system/MemoryControl.hh => src/mem/ruby/structures/MemoryControl.hh
rename : src/mem/ruby/system/MemoryControl.py => src/mem/ruby/structures/MemoryControl.py
rename : src/mem/ruby/system/MemoryNode.cc => src/mem/ruby/structures/MemoryNode.cc
rename : src/mem/ruby/system/MemoryNode.hh => src/mem/ruby/structures/MemoryNode.hh
rename : src/mem/ruby/system/MemoryVector.hh => src/mem/ruby/structures/MemoryVector.hh
rename : src/mem/ruby/system/PerfectCacheMemory.hh => src/mem/ruby/structures/PerfectCacheMemory.hh
rename : src/mem/ruby/system/PersistentTable.cc => src/mem/ruby/structures/PersistentTable.cc
rename : src/mem/ruby/system/PersistentTable.hh => src/mem/ruby/structures/PersistentTable.hh
rename : src/mem/ruby/system/PseudoLRUPolicy.hh => src/mem/ruby/structures/PseudoLRUPolicy.hh
rename : src/mem/ruby/system/RubyMemoryControl.cc => src/mem/ruby/structures/RubyMemoryControl.cc
rename : src/mem/ruby/system/RubyMemoryControl.hh => src/mem/ruby/structures/RubyMemoryControl.hh
rename : src/mem/ruby/system/RubyMemoryControl.py => src/mem/ruby/structures/RubyMemoryControl.py
rename : src/mem/ruby/system/SparseMemory.cc => src/mem/ruby/structures/SparseMemory.cc
rename : src/mem/ruby/system/SparseMemory.hh => src/mem/ruby/structures/SparseMemory.hh
rename : src/mem/ruby/system/TBETable.hh => src/mem/ruby/structures/TBETable.hh
rename : src/mem/ruby/system/TimerTable.cc => src/mem/ruby/structures/TimerTable.cc
rename : src/mem/ruby/system/TimerTable.hh => src/mem/ruby/structures/TimerTable.hh
rename : src/mem/ruby/system/WireBuffer.cc => src/mem/ruby/structures/WireBuffer.cc
rename : src/mem/ruby/system/WireBuffer.hh => src/mem/ruby/structures/WireBuffer.hh
rename : src/mem/ruby/system/WireBuffer.py => src/mem/ruby/structures/WireBuffer.py
rename : src/mem/ruby/recorder/CacheRecorder.cc => src/mem/ruby/system/CacheRecorder.cc
rename : src/mem/ruby/recorder/CacheRecorder.hh => src/mem/ruby/system/CacheRecorder.hh
|
|
This patch enables the use of page tables that are stored in system memory
and respect x86 specification, in SE mode. It defines an architectural
page table for x86 as a MultiLevelPageTable class and puts a placeholder
class for other ISAs page tables, giving the possibility for future
implementation.
|
|
This patch defines a multi-level page table class that stores the page table in
system memory, consistent with ISA specifications. In this way, cpu models that
use the actual hardware to execute (e.g. KvmCPU), are able to traverse the page
table.
|
|
This patch ensures the cycle check is still valid even restoring from
a checkpoint. In this case the DRAMSim2 cycle count is relative to the
startTick rather than 0.
|
|
This patch ensures the functions can be used on const variables.
|
|
We currently use our own home-baked support for type-safe variadic
functions. This is confusing and somewhat limited (e.g., cprintf only
supports a limited number of arguments). This changeset converts all
uses of our internal varargs support to use C++11 variadic macros.
|
|
Add the macros M5_ATTR_FINAL and M5_ATTR_OVERRIDE which are defined to
final and override respectively if supported by the compiler. This is
done to allow a smooth transition to gcc >= 4.7.
|
|
MIPS defined RLIMIT_RSS in a way that could cause a naming conflict with
RLIMIT_RSS from the host system. Broke clang+MacOS build.
|
|
If a bit field in a bit union specified as Bitfield<LSB, MSB> instead
of Bitfield<MSB, LSB> the code silently fails and the field is read as
zero. This changeset introduces a static assert that tests, at compile
time, that the bit order is correct.
|
|
The order of the MSB and LSB bit of the mm field in the PSTATE union
is wrong. Any access to this field will currently be ignored and reads
will always return zero. This patch fixes the ordering so it is <MSB,
LSB> instead of <LSB, MSB>.
|
|
Update comments and add a reference for more information.
|
|
This patch fixes a bug in the DRAM controller address decoding. In
cases where the DRAM burst size (e.g. 32 bytes in a rank with a single
LPDDR3 x32) was smaller than the channel interleaving size
(e.g. systems with a 64-byte cache line) one address bit effectively
got used as a channel bit when it should have been a low-order column
bit.
This patch adds a notion of "columns per stripe", and more clearly
deals with the low-order column bits and high-order column bits. The
patch also relaxes the granularity check such that it is possible to
use interleaving granularities other than the cache line size.
The patch also adds a missing M5_CLASS_VAR_USED to the tCK member as
it is only used in the debug build for now.
|
|
This patch adds a fix for older checkpoints before support for
multiple event queues were added in changeset 2cce74fe359e. The change
in checkpoint version should really hav ebeen part of the
aforementioned changeset.
|
|
Some newer binaries compiled for Versatile Express TC2 contain access
to implementation specific L2MERRSR registers. This causes an infinite
loop of undefined exceptions. This patch changes the behavior to "warn
not fail" to keep the workloads going.
|
|
Baremetal workloads are specified using the "kernel" parameter, but
don't always have the correct address mappings. This patch adds a
boolean flag to the system and bypasses the kernel addr mapping checks
when running in baremetal mode.
|
|
The branch predictor is normally only built when a CPU that uses a
branch predictor is built. The list of CPUs is currently incomplete as
the simple CPUs support branch predictors (for warming, branch stats,
etc). In practice, all CPU models now use branch predictors, so this
changeset removes the CPU model check and replaces it with a check for
the NULL ISA.
|
|
Certain versions of clang complain about unused private members if
they are not used. This changeset removes such members from the
MIPS-specific classes to silence the warning.
|
|
Certain versions of clang complain about unused private members if
they are not used. This changeset removes such members from the
POWER-specific ProcessInfo struct to silence the warning.
|
|
This changeset fixes three types of warnings that occur in clang 3.4
on Ubuntu 12.04:
* Certain versions of libstdc++ (primarily 4.8) use struct and class
interchangeably. This triggers a warning in clang.
* Swig has a tendency to generate code with the register class which
was deprecated in C++11. This triggers a deprecation warning in
clang.
* Swig sometimes generates Python wrapper code which returns
uninitialized values. It's unclear if this is actually a problem
(the cases might be limited to failure paths). We'll silence these
warnings for now since there is little we can do about the
generated code.
|
|
The M5_PRAGMA_NORETURN macro was only used in for
__exit_message. Since the macro only holds a stub definition and all
functions with noreturn semantics use the M5_ATTR_NORETURN, this
macros is completely redundant.
|
|
RefCountingPtr is sometimes forward declared to avoid having to
include refcnt.hh. This does not work since we typically return
instances of RefCountingPtr rather than references to instances. The
only reason this currently works is that we include refcnt.hh in
cprintf.hh, which "leaks" the header to most other source files. This
changeset replaces such forward declarations with an include of
refcnt.hh.
|
|
When a cacheline is written back to a lower-level cache,
tags->insertBlock() sets various status parameters. However these
status bits were cleared immediately after calling. This patch makes
it so that these status fields are not cleared by moving them outside
of the tags->insertBlock() call.
|
