This changeset adds support for aarch64 in kvm. The CPU module
supports both checkpointing and online CPU model switching as long as
no devices are simulated by the host kernel. It currently has the
following limitations:
* The system register based generic timer can only be simulated by
the host kernel. Workaround: Use a memory mapped timer instead to
simulate the timer in gem5.
* Simulating devices (e.g., the generic timer) in the host kernel
requires that the host kernel also simulates the GIC.
* ID registers in the host and in gem5 must match for switching
between simulated CPUs and KVM. This is particularly important
for ID registers describing memory system capabilities (e.g.,
ASID size, physical address size).
* Switching between a virtualized CPU and a simulated CPU is
currently not supported if in-kernel device emulation is
used. This could be worked around by adding support for switching
to the gem5 (e.g., the KvmGic) side of the device models. A
simpler workaround is to avoid in-kernel device models
altogether.
This changeset adds a GIC implementation that uses the kernel's
built-in support for simulating the interrupt controller. Since there
is currently no support for state transfer between gem5 and the
kernel, the device model does not support serialization and CPU
switching (which would require switching to a gem5-simulated GIC).
There are cases (particularly when attaching GDB) when instruction
events are scheduled at the current instruction tick. This used to
trigger an assertion error in kvm. This changeset adds a check for
this condition and forces KVM to do a quick entry that completes any
pending IO operations, but does not execute any new instructions,
before servicing the event. We could check if we need to enter KVM at
all, but forcing a quick entry is makes the code slightly cleaner and
does not hurt correctness (performance is hardly an issue in these
cases).
This changeset moves the ARM-specific KVM CPU implementation to
arch/arm/kvm/. This change is expected to keep the source tree
somewhat cleaner as we start adding support for ARMv8 and KVM
in-kernel interrupt controller simulation.
--HG--
rename : src/cpu/kvm/ArmKvmCPU.py => src/arch/arm/kvm/ArmKvmCPU.py
rename : src/cpu/kvm/arm_cpu.cc => src/arch/arm/kvm/arm_cpu.cc
rename : src/cpu/kvm/arm_cpu.hh => src/arch/arm/kvm/arm_cpu.hh
This patch changes how the address range calculates intersection such
that a system can have a number of non-overlapping interleaved ranges
without complaining. Without this patch we end up with a panic.
The MinorCPU would count bubbles in Execute::issue as part of
the num_insts_issued and so sometimes reach the instruction
issue limit incorrectly.
Fixed by checking for a bubble in one new place.
A step towards removing RubyMemoryControl and shift users to
DRAMCtrl. The latter is faster, more representative, very versatile,
and is integrated with power models.
There are cases when we don't want to use a system register mapped
generic timer, but can't use the SP804. For example, when using KVM on
aarch64, we want to intercept accesses to the generic timer, but can't
do so if it is using the system register interface. In such cases,
we need to use a memory-mapped generic timer.
This changeset adds a device model that implements the memory mapped
generic timer interface. The current implementation only supports a
single frame (i.e., one virtual timer and one physical timer).
The ArmSystem class has a parameter to indicate whether it is
configured to use the generic timer extension or not. This parameter
doesn't affect any feature flags in the current implementation and is
therefore completely unnecessary. In fact, we usually don't set it
even if a system has a generic timer. If we ever need to check if
there is a generic timer present, we should just request a pointer and
check if it is non-null instead.
The generic timer model currently does not support virtual
counters. Virtual and physical counters both tick with the same
frequency. However, virtual timers allow a hypervisor to set an offset
that is subtracted from the counter when it is read. This enables the
hypervisor to present a time base that ticks with virtual time in the
VM (i.e., doesn't tick when the VM isn't running). Modern Linux
kernels generally assume that virtual counters exist and try to use
them by default.
This changeset cleans up the generic timer a bit and moves most of the
register juggling from the ISA code into a separate class in the same
source file as the rest of the generic timer. It also removes the
assumption that there is always 8 or fewer CPUs in the system. Instead
of having a fixed limit, we now instantiate per-core timers as they
are requested. This is all in preparation for other patches that add
support for virtual timers and a memory mapped interface.
The register dumping code in kvm tries to print the bytes in large
registers (128 bits and larger) instead of printing them as hex. This
changeset fixes that.
Some versions of the kernel incorrectly swap the red and blue color
select registers. This changeset adds a workaround for that by
swapping them when instantiating a PixelConverter.
Currently, frame buffer handling in gem5 is quite ad hoc. In practice,
we pass around naked pointers to raw pixel data and expect consumers
to convert frame buffers using the (broken) VideoConverter.
This changeset completely redesigns the way we handle frame buffers
internally. In summary, it fixes several color conversion bugs, adds
support for more color formats (e.g., big endian), and makes the code
base easier to follow.
In the new world, gem5 always represents pixel data using the Pixel
struct when pixels need to be passed between different classes (e.g.,
a display controller and the VNC server). Producers of entire frames
(e.g., display controllers) should use the FrameBuffer class to
represent a frame.
Frame producers are expected to create one instance of the FrameBuffer
class in their constructors and register it with its consumers
once. Consumers are expected to check the dimensions of the frame
buffer when they consume it.
Conversion between the external representation and the internal
representation is supported for all common "true color" RGB formats of
up to 32-bit color depth. The external pixel representation is
expected to be between 1 and 4 bytes in either big endian or little
endian. Color channels are assumed to be contiguous ranges of bits
within each pixel word. The external pixel value is scaled to an 8-bit
internal representation using a floating multiplication to map it to
the entire 8-bit range.
The bitmap generation code is hard to follow and incorrectly uses the
size of an enum member to calculate the size of a pixel. This
changeset cleans up the code and adds some documentation.
The processes of warming up and cooling down Ruby caches are simulation-wide
processes, not just RubySystem instance-specific processes. Thus, the warm-up
and cool-down variables should be globally visible to any Ruby components
participating in either process. Make these variables static members and track
the warm-up and cool-down processes as appropriate.
This patch also has two side benefits:
1) It removes references to the RubySystem g_system_ptr, which are problematic
for allowing multiple RubySystem instances in a single simulation. Warmup and
cooldown variables being static (global) reduces the need for instance-specific
dereferences through the RubySystem.
2) From the AbstractController, it removes local RubySystem pointers, which are
used inconsistently with other uses of the RubySystem: 11 other uses reference
the RubySystem with the g_system_ptr. Only sequencers have local pointers.
Three minor issues are resolved:
1. Apparently gcc 5.1 does not like negation of booleans followed by
bitwise AND.
2. Somehow the compiler also gets confused and warns about
NoopMachInst being unused (removing it causes compilation errors
though). Most likely a compiler bug.
3. There seems to be a number of instances where loop unrolling causes
false positives for the array-bounds check. For now, switch to
std::array. Potentially we could disable the warning for newer gcc
versions, but switching to std::array is probably a good move in
any case.
The system class currently clears the vector of active CPUs in
initState(). CPUs are added to the list by registerThreadContext()
which is called from BaseCPU::init(). This obviously breaks when the
System object is initialized after the CPUs. This changeset removes
the offending clear() call since the list will be empty after it has
been instantiated anyway.
The current ignoreWarnOnceFunc doesn't really work as expected,
since it will only generate one warning total, for whichever
"warn-once" syscall is invoked first. This patch fixes that
behavior by keeping a "warned" flag in the SyscallDesc object,
allowing suitably flagged syscalls to warn exactly once per
syscall.
Sometimes, we need to defer an express snoop in an MSHR, but the original
request might complete and deallocate the original pkt->req. In those cases,
create a copy of the request so that someone who is inspecting the delayed
snoop can also inspect the request still. All of this is rather hacky, but the
allocation / linking and general life-time management of Packet and Request is
rather tricky. Deleting the copy is another tricky area, testing so far has
shown that the right copy is deleted at the right time.
We currently assume that all uncacheable memory accesses are strictly
ordered. Instead of always enforcing strict ordering, we now only
enforce it if the required memory type is device memory or strongly
ordered memory.
The Request::UNCACHEABLE flag currently has two different
functions. The first, and obvious, function is to prevent the memory
system from caching data in the request. The second function is to
prevent reordering and speculation in CPU models.
This changeset gives the order/speculation requirement a separate flag
(Request::STRICT_ORDER). This flag prevents CPU models from doing the
following optimizations:
* Speculation: CPU models are not allowed to issue speculative
loads.
* Write combining: CPU models and caches are not allowed to merge
writes to the same cache line.
Note: The memory system may still reorder accesses unless the
UNCACHEABLE flag is set. It is therefore expected that the
STRICT_ORDER flag is combined with the UNCACHEABLE flag to prevent
this behavior.
With the recent patches addressing how we deal with uncacheable
accesses there is no longer need for the work arounds put in place to
enforce certain sections of memory to be uncacheable during boot.
This patch takes a last step in fixing issues related to uncacheable
accesses. We do not separate uncacheable memory from uncacheable
devices, and in cases where it is really memory, there are valid
scenarios where we need to snoop since we do not support cache
maintenance instructions (yet). On snooping an uncacheable access we
thus provide data if possible. In essence this makes uncacheable
accesses IO coherent.
The snoop filter is also queried to steer the snoops, but not updated
since the uncacheable accesses do not allocate a block.
This patch simplifies the overall CPU by changing the TLB caches such
that they do not forward snoops to the table walker port(s). Note that
only ARM and X86 are affected.
There is no reason for the ports to snoop as they do not actually take
any action, and from a performance point of view we are better of not
snooping more than we have to.
Should it at a later point be required to snoop for a particular TLB
design it is easy enough to add it back.
This patch ensures that we pass on information about a packet being
shared (rather than exclusive), when forwarding a packet downstream.
Without this patch there is a risk that a downstream cache considers
the line exclusive when it really isn't.
This patch adds a missing counter update for the uncacheable
accesses. By updating this counter we also get a meaningful average
latency for uncacheable accesses (previously inf).
This patch changes the cache implementation to rely on virtual methods
rather than using the replacement policy as a template argument.
There is no impact on the simulation performance, and overall the
changes make it easier to modify (and subclass) the cache and/or
replacement policy.
This patch fixes a recent issue with gcc 4.9 (and possibly more) being
convinced that indices outside the array bounds are used when
initialising the FUPool members.
Both open_adaptive and close_adaptive page polices keep the page
open if a row hit is found. If a row hit is not found, close_adaptive
page policy precharges the row, and open_adaptive policy precharges
the row only if there is a bank conflict request waiting in the queue.
This patch makes the checks for above conditions simpler.
Committed by: Nilay Vaish <nilay@cs.wisc.edu>
Currently, each op class has a parameter issueLat that denotes the cycles after
which another op of the same class can be issued. As of now, this latency can
either be one cycle (fully pipelined) or same as execution latency of the op
(not at all pipelined). The fact that issueLat is a parameter of type Cycles
makes one believe that it can be set to any value. To avoid the confusion, the
parameter is being renamed as 'pipelined' with type boolean. If set to true,
the op would execute in a fully pipelined fashion. Otherwise, it would execute
in an unpipelined fashion.
This patch sets the default latency of the division microop to a single cycle
on x86. This is because the division instructions DIV and IDIV have been
implemented as loops of div microops, where each microop computes a single bit
of the quotient.
Same exception is raised whether division with zero is performed or the
quotient is greater than the maximum value that the provided space can hold.
Divide-by-Zero is the AMD terminology, while Divide-Error is Intel's.
This patch introduces a UFS host controller and a UFS device. More
information about the UFS standard can be found at the JEDEC site:
http://www.jedec.org/standards-documents/results/jesd220
Note that the model does not implement the complete standard, and as
such is not an actual implementation of UFS. The following SCSI
commands are implemented: inquiry, read, read capacity, report LUNs,
start/stop, test unit ready, verify, write, format unit, send
diagnostic, synchronize cache, mode select, mode sense, request sense,
unmap, write buffer and read buffer. This is sufficient for usage with
Linux and Android.
To interact with this model a kernel version 3.9 or above is
needed.
This adds a NAND flash timing model. This model takes the number of
planes into account and is ultimately intended to be used as a
high-level performance model for any device using flash. To access the
memory, use either readMemory or writeMemory.
To make use of the model you will need an interface model
such as UFSHostDevice, which is part of a separate patch.
At the moment the flash device is part of the ARM device tree since
the only use if the UFSHostDevice, and that in turn relies on the ARM
GIC.
This patch adds an I2C bus and base device. I2C is used to connect a
variety of sensors, and this patch serves as a starting point to
enable a range of I2C devices.
This patch fixes a few small issues to ensure gem5 compiles when using
gcc 5.1.
First, the GDB_REG_BYTES in the RemoteGDB header are, rather
surprisingly, flagged as unused for both ARM and X86. Removing them,
however, causes compilation errors as they are actually used in the
source file. Moving the constant into the class definition fixes the
issue. Possibly a gcc bug.
Second, we have an unused EthPktData constructor using auto_ptr, and
the latter is deprecated. Since the code is never used it is simply
removed.
The o3 cpu instruction queue model uses the count variable to track the number
of unissued instructions in the queue. Previously, the squash method used
this variable to avoid executing the doSquash method when there were no
unissued instructions in the pipeline. A corner case problem exists when
only issued instructions exist in the pipeline and a squash occurs; the
doSquash code is not invoked and subsequently does not clean up state properly.
This patch takes the final step in removing the InOrderCPU from the
tree. Rest in peace.
The MinorCPU is now used to model an in-order microarchitecture, and
long term the MinorCPU will eventually be renamed InOrderCPU.
This patch ensures that the CPU progress Event is triggered for the new set of
switched_cpus that get scheduled (e.g. during fast-forwarding). it also avoids
printing the interval state if the cpu is currently switched out.
Committed by: Nilay Vaish <nilay@cs.wisc.edu>
Restoring from a checkpoint with ruby + the DRAMCtrl memory model was not
working, because ruby and DRAMCtrl disagreed on the current tick during warmup.
Since there is no reason to do timing requests during warmup, use functional
requests instead.
Committed by: Nilay Vaish <nilay@cs.wisc.edu>
This patch adds an example configuration in ext/sst/tests/ that allows
an SST/gem5 instance to simulate a 4-core AArch64 system with SST's
memHierarchy components providing all the caches and memories.
Restoring from a checkpoint fails if either the RTC or the RTC Timer
Interrrupt event is disabled. The restored machine tried incorrectly
to schedule the next event with negative offset.
Committed by: Nilay Vaish <nilay@cs.wisc.edu>
Add 32-bit access width for PrimaryTiming register and 16bit for UDMAControl
register as FreeBSD required.
Committed by: Nilay Vaish <nilay@cs.wisc.edu>
The totalInstructions counter is only incremented when the whole instruction is
commited and not on every microop. It was incorrectly reset in atomic and
timing cpus.
Committed by: Nilay Vaish <nilay@cs.wisc.edu>"
When running with the Exec flag, the mwait instruction attempted
to print out its source registers, which were never actually
initialized. This led to sporadic assertion failures when the
value stored there was invalid.
Committed by: Nilay Vaish <nilay@cs.wisc.edu>
The stride prefetcher had a hardcoded number of contexts (i.e. master-IDs)
that it could handle. Since master IDs need to be unique per system, and
every core, cache etc. requires a separate master port, a static limit on
these does not make much sense.
Instead, this patch adds a small hash map that will map all master IDs to
the right prefetch state and dynamically allocates new state for new master
IDs.
This patch changes the order of writeback allocation such that any
writebacks resulting from a tag lookup (e.g. for an uncacheable
access), are added to the writebuffer before any new MSHR entries are
allocated. This ensures that the writebacks logically precedes the new
allocations.
The patch also changes the uncacheable flush to use proper timed (or
atomic) writebacks, as opposed to functional writes.
This patch simplifies the code dealing with uncacheable timing
accesses, aiming to align it with the existing miss handling. Similar
to what we do in atomic, a timing request now goes through
Cache::access (where the block is also flushed), and then proceeds to
ignore any existing MSHR for the block in question. This unifies the
flow for cacheable and uncacheable accesses, and for atomic and timing.
This patch changes how we search for matching MSHRs, ignoring any MSHR
that is allocated for an uncacheable access. By doing so, this patch
fixes a corner case in the MSHRs where incorrect data ended up being
copied into a (cacheable) read packet due to a first uncacheable MSHR
target of size 4, followed by a cacheable target to the same MSHR of
size 64. The latter target was filled with nonsense data.
This patch removes the no-longer-needed
allocateUncachedReadBuffer. Besides the checks it is exactly the same
as allocateMissBuffer and thus provides no value.
This patch aligns all MSHR queue entries to block boundaries to
simplify checks for matches. Previously there were corner cases that
could lead to existing entries not being identified as matches.
There are, rather alarmingly, a few regressions that change with this
patch.
This patch subsumes the PREFETCH_SNOOP_SQUASH flag with the more
generic BLOCK_CACHED flag. Future patches implementing cache eviction
messages can use the BLOCK_CACHED flag in almost the same manner as
hardware prefetches use the PREFETCH_SNOOP_SQUASH flag. The
PREFTECH_SNOOP_FLAG is set if the prefetch target is found in the tags
or the MSHRs in any state, so we are simply replacing calls to
setPrefetchSquashed() with setBlockCached(). The case of where the
prefetch target is found in the writeback MSHRs of upper level caches
continues to be covered by the MEM_INHIBIT flag.
Currently if there are shell special characters in a
command-line argument, you can't copy and paste the
echoed command line onto a shell prompt because the
characters aren't quoted properly. This patch fixes
that problem.
This patch accomplishes two things:
1. Makes simulate()'s GlobalSimLoopExitEvent a singleton reused
across calls. This is slightly more efficient than recreating
it every time.
2. Gives callers to simulate() (especially other simulators) a
foolproof way of knowing that the simulation period ended
successfully by hitting the limit event. They can call
getLimitEvent() and compare it to the return
value of simulate().
This change was motivated by an ongoing effort to integrate gem5
and SST, with SST as the master sim and gem5 as the slave sim.
This patch adds a new PIO-accessible GICv2m shim. This shim has a PIO
slave port on one side, and SPI 'wires' on the other. It accepts MSIs
from the system and triggers SPIs on the GIC. It is configurable with
a number of frames, each of which has a number of SPIs and a base SPI
offset.
A Linux driver for GICv2m is available upstream.
This patch removes the code that added this magic register. A
follow-up patch provides a GICv2m MSI shim that gives the same
functionality in a standard ARM system architecture way.
Fix erroneous message format for fatal error.
Previously, code did not have type indicator (% instead of %d).
Also removed redundant fatal check.
Ran modified sweep.py with in range and out of range values to test.
The CommMonitor by default only allows memory traces to be gathered in
timing mode. This patch allows memory traces to be gathered in atomic
mode if all one needs is a functional trace of memory addresses used
and timing information is of a secondary concern.
For some reason we were checking mshr->hasTargets() even though
we had already called mshr->getTarget() unconditionally earlier
in the same function (which asserts if there are no targets).
Get rid of this useless check, and while we're at it get rid
of the redundant call to mshr->getTarget(), since we still have
the value saved in a local var.
Refactor the way that specific MemCmd values are generated for packets.
The new approach is a little more elegant in that we assign the right
value up front, and it's also more amenable to non-heap-allocated
Packet objects.
Also replaced the code in the Minor model that was still doing it the
ad-hoc way.
This is basically a refinement of http://repo.gem5.org/gem5/rev/711eb0e64249.
The 'if (writebacks.size)' check was redundant, because
writeBuffer.findMatches() would return false if the
writebacks list was empty.
Also renamed 'mshr' to 'wb_entry' in this context since
we are pointing at a writebuffer entry and not an MSHR
(even though it's the same C++ class).
The variable is used in only one place and a whole new function setNextStatus()
has been defined just to compute the value of the variable. Instead of calling
the function, the value is now computed in the loop that preceded the function
call.
This patch changes all the DPRINTF messages in the cache to use
'%#llx' every time a packet address is printed. The inclusion of '#'
ensures '0x' is prepended, and since the address type is a uint64_t %x
really should be %llx.
This patch fixes a rather subtle issue in the sending of MSHR requests
in the cache, where the logic previously did not check for conflicts
between the MSRH queue and the write queue when requests were not
ready. The correct thing to do is to always check, since not having a
ready MSHR does not guarantee that there is no conflict.
The underlying problem seems to have slipped past due to the symmetric
timings used for the write queue and MSHR queue. However, with the
recent timing changes the bug caused regressions to fail.
This patch changes the valid-bytes start/end to a proper byte
mask. With the changes in timing introduced in previous patches there
are more packets waiting in queues, and there are regressions using
the checker CPU failing due to non-contigous read data being found in
the various cache queues.
This patch also adds some more comments explaining what is going on,
and adds the fourth and missing case to Packet::checkFunctional.
By default, the packet queue is ordered by the ticks of the to-be-sent
packages. With the recent modifications of packages sinking their header time
when their resposne leaves the caches, there could be cases of MSHR targets
being allocated and ordered A, B, but their responses being sent out in the
order B,A. This led to inconsistencies in bus traffic, in particular the snoop
filter observing first a ReadExResp and later a ReadRespWithInv. Logically,
these were ordered the other way around behind the MSHR, but due to the timing
adjustments when inserting into the PacketQueue, they were sent out in the
wrong order on the bus, confusing the snoop filter.
This patch adds a flag (off by default) such that these special cases can
request in-order insertion into the packet queue, which might offset timing
slighty. This is expected to occur rarely and not affect timing results.
This patch makes the caches and memory controllers consume the delay
that is annotated to a packet by the crossbar. Previously many
components simply threw these delays away. Note that the devices still
do not pay for these delays.
This patch introduces a few subclasses to the CoherentXBar and
NoncoherentXBar to distinguish the different uses in the system. We
use the crossbar in a wide range of places: interfacing cores to the
L2, as a system interconnect, connecting I/O and peripherals,
etc. Needless to say, these crossbars have very different performance,
and the clock frequency alone is not enough to distinguish these
scenarios.
Instead of trying to capture every possible case, this patch
introduces dedicated subclasses for the three primary use-cases:
L2XBar, SystemXBar and IOXbar. More can be added if needed, and the
defaults can be overridden.
This patch introduces latencies in crossbar that were neglected
before. In particular, it adds three parameters in crossbar model:
front_end_latency, forward_latency, and response_latency. Along with
these parameters, three corresponding members are added:
frontEndLatency, forwardLatency, and responseLatency. The coherent
crossbar has an additional snoop_response_latency.
The latency of the request path through the xbar is set as
--> frontEndLatency + forwardLatency
In case the snoop filter is enabled, the request path latency is charged
also by look-up latency of the snoop filter.
--> frontEndLatency + SF(lookupLatency) + forwardLatency.
The latency of the response path through the xbar is set instead as
--> responseLatency.
In case of snoop response, if the response is treated as a normal response
the latency associated is again
--> responseLatency;
If instead it is forwarded as snoop response we add an additional variable
+ snoopResponseLatency
and the latency associated is
--> snoopResponseLatency;
Furthermore, this patch lets the crossbar progress on the next clock
edge after an unused retry, changing the time the crossbar considers
itself busy after sending a retry that was not acted upon.
The ARM PL011 UART model didn't clear and raise interrupts
correctly. This changeset rewrites the whole interrupt handling and
makes it both simpler and fixes several cases where the correct
interrupts weren't raised or cleared. Additionally, it cleans up many
other aspects of the code.
This patch changes how the MMU and table walkers are created such that
a single port is used to connect the MMU and the TLBs to the memory
system. Previously two ports were needed as there are two table walker
objects (stage one and stage two), and they both had a port. Now the
port itself is moved to the Stage2MMU, and each TableWalker is simply
using the port from the parent.
By using the same port we also remove the need for having an
additional crossbar joining the two ports before the walker cache or
the L2. This simplifies the creation of the CPU cache topology in
BaseCPU.py considerably. Moreover, for naming and symmetry reasons,
the TLB walker port is connected through the stage-one table walker
thus making the naming identical to x86. Along the same line, we use
the stage-one table walker to generate the master id that is used by
all TLB-related requests.
Now, prior to the renaming, the instruction requests the exact amount of
registers it will need, and the rename_map decides whether the instruction is
allowed to proceed or not.
This patch fixes a long-standing isue with the port flow
control. Before this patch the retry mechanism was shared between all
different packet classes. As a result, a snoop response could get
stuck behind a request waiting for a retry, even if the send/recv
functions were split. This caused message-dependent deadlocks in
stress-test scenarios.
The patch splits the retry into one per packet (message) class. Thus,
sendTimingReq has a corresponding recvReqRetry, sendTimingResp has
recvRespRetry etc. Most of the changes to the code involve simply
clarifying what type of request a specific object was accepting.
The biggest change in functionality is in the cache downstream packet
queue, facing the memory. This queue was shared by requests and snoop
responses, and it is now split into two queues, each with their own
flow control, but the same physical MasterPort. These changes fixes
the previously seen deadlocks.
This patch resolves a bug with hardware prefetches. Before a hardware prefetch
is sent towards the memory, the system generates a snoop request to check all
caches above the prefetch generating cache for the presence of the prefetth
target. If the prefetch target is found in the tags or the MSHRs of the upper
caches, the cache sets the prefetchSquashed flag in the snoop packet. When the
snoop packet returns with the prefetchSquashed flag set, the prefetch
generating cache deallocates the MSHR reserved for the prefetch. If the
prefetch target is found in the writeback buffer of the upper cache, the cache
sets the memInhibit flag, which signals the prefetch generating cache to
expect the data from the writeback. When the snoop packet returns with the
memInhibitAsserted flag set, it marks the allocated MSHR as inService and
waits for the data from the writeback.
If the prefetch target is found in multiple upper level caches, specifically
in the tags or MSHRs of one upper level cache and the writeback buffer of
another, the snoop packet will return with both prefetchSquashed and
memInhibitAsserted set, while the current code is not written to handle such
an outcome. Current code checks for the prefetchSquashed flag first, if it
finds the flag, it deallocates the reserved MSHR. This leads to assert failure
when the data from the writeback appears at cache. In this fix, we simply
switch the order of checks. We first check for memInhibitAsserted and then for
prefetch squashed.
Have the traffic generator add its masterID as the PC address to the
requests. That way, prefetchers (and other components) that use a PC
for request classification will see per-tester streams of requests.
This enables us to test strided prefetchers with the memchecker, too.
The ISA code sometimes stores 16-bit ASIDs as 8-bit unsigned integers
and has a couple of inverted checks that mask out the high 8 bits of
an ASID if 16-bit ASIDs have been /enabled/. This changeset fixes both
of those issues.
We curently use INTREG_X31 instead of INTREG_SPX when accessing the
stack pointer in GDB. gem5 normally uses INTREG_SPX to access the
stack pointer, which gets mapped to the stack pointer corresponding
(INTREG_SPn) to the current exception level. This changeset updates
the GDB interface to use SPX instead of X31 (which is always zero)
when transfering CPU state to gdb.
The remote GDB interface currently doesn't check if translations are
valid before reading memory. This causes a panic when GDB tries to
access unmapped memory (e.g., when getting a stack trace). There are
two reasons for this: 1) The function used to check for valid
translations (virtvalid()) doesn't work and panics on invalid
translations. 2) The method in the GDB interface used to test if a
translation is valid (RemoteGDB::acc) always returns true regardless
of the return from virtvalid().
This changeset fixes both of these issues.
Previously, the user would have to manually set access_backing_store=True
on all RubyPorts (Sequencers) in the config files.
Now, instead there is one global option that each RubyPort checks on
initialization.
Committed by: Nilay Vaish <nilay@cs.wisc.edu>
To be able to use the TrafficGen in a system with caches we need to
allow it to sink incoming snoop requests. By default the master port
panics, so silently ignore any snoops.
In highly loaded cases, reads might actually overlap with writes to the
initial memory state. The mem checker needs to detect such cases and
permit the read reading either from the writes (what it is doing now) or
read from the initial, unknown value.
This patch adds this logic.
This patch fixes a rather unfortunate oversight where the annotation
pointer was used even though it is null. Somehow the code still works,
but UBSan is rather unhappy. The use is now guarded, and the variable
is initialised in the constructor (as well as init()).
Move the (common) GIC initialization code that notifies the platform
code of the new GIC to the base class (BaseGic) instead of the Pl390
implementation.
This patch ensures we can run simulations with very large simulated
memories (at least 64 TB based on some quick runs on a Linux
workstation). In essence this allows us to efficiently deal with
sparse address maps without having to implement a redirection layer in
the backing store.
This opens up for run-time errors if we eventually exhausts the hosts
memory and swap space, but this should hopefully never happen.
This patch changes the range cache used in the global physical memory
to be an iterator so that we can use it not only as part of isMemAddr,
but also access and functionalAccess. This matches use-cases where a
core is using the atomic non-caching memory mode, and repeatedly calls
isMemAddr and access.
Linux boot on aarch32, with a single atomic CPU, is now more than 30%
faster when using "--fastmem" compared to not using the direct memory
access.
This changeset moves the pseudo instructions used to signal unknown
instructions and unimplemented instructions to the same source files
as the decoder fault.
This patch clarifies the packet timings annotated
when going through a crossbar.
The old 'firstWordDelay' is replaced by 'headerDelay' that represents
the delay associated to the delivery of the header of the packet.
The old 'lastWordDelay' is replaced by 'payloadDelay' that represents
the delay needed to processing the payload of the packet.
For now the uses and values remain identical. However, going forward
the payloadDelay will be additive, and not include the
headerDelay. Follow-on patches will make the headerDelay capture the
pipeline latency incurred in the crossbar, whereas the payloadDelay
will capture the additional serialisation delay.
This patch adds some much-needed clarity in the specification of the
cache timing. For now, hit_latency and response_latency are kept as
top-level parameters, but the cache itself has a number of local
variables to better map the individual timing variables to different
behaviours (and sub-components).
The introduced variables are:
- lookupLatency: latency of tag lookup, occuring on any access
- forwardLatency: latency that occurs in case of outbound miss
- fillLatency: latency to fill a cache block
We keep the existing responseLatency
The forwardLatency is used by allocateInternalBuffer() for:
- MSHR allocateWriteBuffer (unchached write forwarded to WriteBuffer);
- MSHR allocateMissBuffer (cacheable miss in MSHR queue);
- MSHR allocateUncachedReadBuffer (unchached read allocated in MSHR
queue)
It is our assumption that the time for the above three buffers is the
same. Similarly, for snoop responses passing through the cache we use
forwardLatency.
The MemTest class really only tests false sharing, and as such there
was a lot of old cruft that could be removed. This patch cleans up the
tester, and also makes it more clear what the assumptions are. As part
of this simplification the reference functional memory is also
removed.
The regression configs using MemTest are updated to reflect the
changes, and the stats will be bumped in a separate patch. The example
config will be updated in a separate patch due to more extensive
re-work.
In a follow-on patch a new tester will be introduced that uses the
MemChecker to implement true sharing.
The TLB-related code is generally architecture dependent and should
live in the arch directory to signify that.
--HG--
rename : src/sim/BaseTLB.py => src/arch/generic/BaseTLB.py
rename : src/sim/tlb.cc => src/arch/generic/tlb.cc
rename : src/sim/tlb.hh => src/arch/generic/tlb.hh
Gcc and clang both provide an attribute that can be used to flag a
function as deprecated at compile time. This changeset adds a gem5
compiler macro for that compiler feature. The macro can be used to
indicate that a legacy API within gem5 has been deprecated and provide
a graceful migration to the new API.
This patch fixes the CompoundFlag constructor, ensuring that it does
not dereference NULL. Doing so has undefined behaviuor, and both clang
and gcc's undefined-behaviour sanitiser was rather unhappy.
The Platform base class contains a pointer to an instance of the
System which is never initialized. This can lead to subtle bugs since
some architecture-specific platform implementations contain their own
system pointer which is normally used. However, if the platform is
accessed through a pointer to its base class, the dangling pointer
will be used instead.
This patch adds a bit of clarification around the assumptions made in
the cache when packets are sent out, and dirty responses are
pending. As part of the change, the marking of an MSHR as in service
is simplified slightly, and comments are added to explain what
assumptions are made.
This patch extends the current address interleaving with basic hashing
support. Instead of directly comparing a number of address bits with a
matching value, it is now possible to use two independent set of
address bits XOR'ed together. This avoids issues where strided address
patterns are heavily biased to a subset of the interleaved ranges.
This patch changes the DRAM channel interleaving default behaviour to
be more representative. The default address mapping (RoRaBaCoCh) moves
the channel bits towards the least significant bits, and uses 128 byte
as the default channel interleaving granularity.
These defaults can be overridden if desired, but should serve as a
sensible starting point for most use-cases.
The method Event::initialized() tests if this != NULL as a part of the
expression that tests if an event is initialized. The only case when
this check could be false is if the method is called on a null
pointer, which is illegal and leads to undefined behavior (such as
eating your pets) according to the C++ standard. Because of this,
modern compilers (specifically, recent versions of clang) warn about
this which we treat as an error. This changeset removes the redundant
check to fix said warning.
This patch changes how the timing CPU deals with processing responses,
always scheduling an event, even if it is for the current tick. This
helps to avoid situations where a new request shows up before a
response is finished in the crossbar, and also is more in line with
any realistic behaviour.
The Float param was not settable on the command line
due to a typo in the class definition in
python/m5/params.py. This corrects the typo and allows
floats to be set on the command line as intended.
While the IsFirstMicroop flag exists it was only occasionally used in the ARM
instructions that gem5 microOps and therefore couldn't be relied on to be correct.
When gem5 is a slave to another simulator and the Python is only used
to initialize the configuration (and not perform actual simulation), a
"debug start" (--debug-start) event will get freed during or immediately
after the initial Python frame's execution rather than remaining in the
event queue. This tricky patch fixes the GC issue causing this.
This patch takes the final step in removing the src and dest fields in
the packet. These fields were rather confusing in that they only
remember a single multiplexing component, and pushed the
responsibility to the bridge and caches to store the fields in a
senderstate, thus effectively creating a stack. With the recent
changes to the crossbar response routing the crossbar is now
responsible without relying on the packet fields. Thus, these
variables are now unused and can be removed.
This patch removes the source field from the ForwardResponseRecord,
but keeps the class as it is part of how the cache identifies
responses to hardware prefetches that are snooped upwards.
This patch aligns how the response routing is done in the RubyPort,
using the SenderState for both memory and I/O accesses. Before this
patch, only the I/O used the SenderState, whereas the memory accesses
relied on the src field in the packet. With this patch we shift to
using SenderState in both cases, thus not relying on the src field any
longer.
This patch removes the need for a source and destination field in the
packet by shifting the onus of the tracking to the crossbar, much like
a real implementation. This change in behaviour also means we no
longer need a SenderState to remember the source/dest when ever we
have multiple crossbars in the system. Thus, the stack that was
created by the SenderState is not needed, and each crossbar locally
tracks the response routing.
The fields in the packet are still left behind as the RubyPort (which
also acts as a crossbar) does routing based on them. In the succeeding
patches the uses of the src and dest field will be removed. Combined,
these patches improve the simulation performance by roughly 2%.
This patch fixes a minor issue in the X86 page table walker where it
ended up sending new request packets to the crossbar before the
response processing was finished (recvTimingResp is directly calling
sendTimingReq). Under certain conditions this caused the crossbar to
see illegal combinations of request/response overlap, in turn causing
problems with a slightly modified crossbar implementation.
This patch tidies up how we create and set the fields of a Request. In
essence it tries to use the constructor where possible (as opposed to
setPhys and setVirt), thus avoiding spreading the information across a
number of locations. In fact, setPhys is made private as part of this
patch, and a number of places where we callede setVirt instead uses
the appropriate constructor.
The ppCommit should notify the attached listener every time the cpu commits
a microop or non microcoded insturction. The listener can then decide
whether it will process only the last microop (eg. SimPoint probe).
Committed by: Nilay Vaish <nilay@cs.wisc.edu>
This patch fixes a bug where the DRAM controller tried to access the
system cacheline size before the system pointer was initialised. It
also fixes a bug where the granularity is 0 (no interleaving).
This patch corrects the FXSAVE and FXRSTOR Macroops. The actual code used for
saving/restore the FP registers is in the file but it was not used.
The FXSAVE and FXRSTOR instructions are used in the kernel for saving and
loading the state of the mmx,xmm and fpu registers.
This operation is triggered in FS by issuing a Device Not Available Fault. The
cr0 register has a TS flag that is set upon each context change. Every time a
task access any FP related register (SIMD as well) if the TS flag is set to
one, the device not available fault is issued. The kernel saves the current
state of the registers, and restore the previous state of the currently running
task.
Right now Gem5 lacks of this capability. the Device Not Available Fault is
never issued, leading to several problems when different threads share the same
CPU and SMT is not used. The PARSEC Ferret benchmark is an example of this
behavior.
In order to test this a hack in the atomic cpu code was done to detect if a
static instruction has any FP operands and the cr0 reg TS bit is set. This
check must be done in the ISA dependent code. But it seems to be tricky to
access the cr0 register while executing an instruction.
Committed by: Nilay Vaish <nilay@cs.wisc.edu>
This change includes edits to Intel8254Timer to prevent counter events firing
before startup to comply with SimObject initialization call sequence.
Committed by: Nilay Vaish <nilay@cs.wisc.edu>
If two bitfields are of the same type, also implying that they have the same
first and last bit positions, the existing implementation would copy the
entire bitfield. That includes the __data member which is shared among all the
bitfields, effectively overwritting the entire bitunion.
This change also adjusts the write only signed bitfield assignment operator to
be like the unsigned version, using "using" instead of implementing it again
and calling down to the underlying implementation.
That change enables CPUID bits for features that aren't implemented in gem5.
If a simulated system tries to use those features because it was told it
could, bad things can happen.
Minor was reporting the data cache access as ".inst" accesses.
This just switches the MasterPortID to dataMasterPortId.
Committed by: Nilay Vaish <nilay@cs.wisc.edu>
added ARM aarch64 unlinkat syscall support, modeled on other <xxx>at syscalls.
This gets all of the cpu2006 int workloads passing in SE mode on aarch64.
Committed by: Nilay Vaish <nilay@cs.wisc.edu>
This patch implements the simd128 ADDSUBPD instruction for the x86 architecture.
Tested with a simple program in assembly language which executes the
instruction. Checked that different versions of the instruction are executed
by using the execution tracing option.
Committed by: Nilay Vaish <nilay@cs.wisc.edu
This change includes edits to MC146818 timer to prevent RTC events
firing before startup to comply with SimObject initialization call sequence.
Committed by: Nilay Vaish <nilay@cs.wisc.edu>
According to Linux man pages, if writev is successful, it returns the total
number of bytes written. Otherwise, it returns an error code. Instead of
returning 0, return the result from the actual call to writev in the system
call.
The cache's MemSidePacketQueue schedules a sendEvent based upon
nextMSHRReadyTime() which is the time when the next MSHR is ready or whenever
a future prefetch is ready. However, a prefetch being ready does not guarentee
that it can obtain an MSHR. So, when all MSHRs are full,
the simulation ends up unnecessiciarly scheduling a sendEvent every picosecond
until an MSHR is finally freed and the prefetch can happen.
This patch fixes this by not signaling the prefetch ready time if the prefetch
could not be generated. The event is rescheduled as soon as a MSHR becomes
available.
Previously the code commented about an unhandled case where it might be
possible for a writeback to arrive after a prefetch was generated but
before it was sent to the memory system. I hit that case. Luckily
the prefetchSquash() logic already in the code handles dropping prefetch
request in certian circumstances.
Re-organizes the prefetcher class structure. Previously the
BasePrefetcher forced multiple assumptions on the prefetchers that
inherited from it. This patch makes the BasePrefetcher class truly
representative of base functionality. For example, the base class no
longer enforces FIFO order. Instead, prefetchers with FIFO requests
(like the existing stride and tagged prefetchers) now inherit from a
new QueuedPrefetcher base class.
Finally, the stride-based prefetcher now assumes a custimizable lookup table
(sets/ways) rather than the previous fully associative structure.
Adds a new parameter that reserves some number of MSHR entries for demand
accesses. This helps prevent prefetchers from taking all MSHRs, forcing demand
requests from the CPU to stall.
This patch adds table walker stats for:
- Walk events
- Instruction vs Data
- Page size histogram
- Wait time and service time histograms
- Pending requests histogram (per cycle) - measures dist. of L
(p(1..) = how often busy, p(0) = how often idle)
- Squashes, before starting and after completion
This patch gives the user direct influence over the number of DRAM
ranks to make it easier to tune the memory density without affecting
the bandwidth (previously the only means of scaling the device count
was through the number of channels).
The patch also adds some basic sanity checks to ensure that the number
of ranks is a power of two (since we rely on bit slices in the address
decoding).
This patch addresses an issue seen with the KVM CPU where the refresh
events scheduled by the DRAM controller forces the simulator to switch
out of the KVM mode, thus killing performance.
The current patch works around the fact that we currently have no
proper API to inform a SimObject of the mode switches. Instead we rely
on drainResume being called after any switch, and cache the previous
mode locally to be able to decide on appropriate actions.
The switcheroo regression require a minor stats bump as a result.
This patch adds rank-wise refresh to the controller, as opposed to the
channel-wide refresh currently in place. In essence each rank can be
refreshed independently, and for this to be possible the controller
is extended with a state machine per rank.
Without this patch the data bus is always idle during a refresh, as
all the ranks are refreshing at the same time. With the rank-wise
refresh it is possible to use one rank while another one is
refreshing, and thus the data bus can be kept busy.
The patch introduces a Rank class to encapsulate the state per rank,
and also shifts all the relevant banks, activation tracking etc to the
rank. The arbitration is also updated to consider the state of the rank.
This patch adds a stand-alone stack distance calculator. The stack
distance calculator is a passive SimObject that observes the addresses
passed to it. It calculates stack distances (LRU Distances) of
incoming addresses based on the partial sum hierarchy tree algorithm
described by Alamasi et al. http://doi.acm.org/10.1145/773039.773043.
For each transaction a hashtable look-up is performed. At every
non-unique transaction the tree is traversed from the leaf at the
returned index to the root, the old node is deleted from the tree, and
the sums (to the right) are collected and decremented. The collected
sum represets the stack distance of the found node. At every unique
transaction the stack distance is returned as
numeric_limits<uint64>::max().
In addition to the basic stack distance calculation, a feature to mark
an old node in the tree is added. This is useful if it is required to
see the reuse pattern. For example, Writebacks to the lower level
(e.g. membus from L2), can be marked instead of being removed from the
stack (isMarked flag of Node set to True). And then later if this same
address is accessed (by L1), the value of the isMarked flag would be
True. This gives some insight on how the Writeback policy of the
lower level affect the read/write accesses in an application.
Debugging is enabled by setting the verify flag to true. Debugging is
implemented using a dummy stack that behaves in a naive way, using STL
vectors. Note that this has a large impact on run time.
This patch adds the MemChecker and MemCheckerMonitor classes. While
MemChecker can be integrated anywhere in the system and is independent,
the most convenient usage is through the MemCheckerMonitor -- this
however, puts limitations on where the MemChecker is able to observe
read/write transactions.
We currently don't handle unaligned PCs correctly. There is one check
for unaligned PCs in the TLB when running in aarch64 mode, but this
check does not cover cases where the CPU does not do a TLB lookup when
decoding an instruction (e.g., a branch stays within the same cache
line). Additionally, the Decoder class sometimes throws an assertion
for unaligned PCs which breaks speculation.
This changeset introduces a decoder fault bit field in the ExtMachInst
structure. This field can be used to signal a decoder failure. If set,
the decoder generates an internal gem5fault instruction instead of a
normal instruction. This instruction in turns either panics (fault
type PANIC), returns an PCAlignmentFault (fault type UNALIGNED,
aarch64) or PrefetchAbort (fault type UNALIGNED, aarch32).
The patch causes minor changes to the realview64 regressions, and a
stats bump will follow.
This patch adds support for filtering events in the PMU. In order to
do so, it updates the ISADevice base class to forward an ISA pointer
to ISA devices. This enables such devices to access the MiscReg file
to determine the current execution level.
The aarch64 system register decoder is currently not decoding
PMXEVTYPER_EL0 and PMCCFILTR_EL0 correctly. This changeset updates the
decoder so that they are decoded using the values in table C5-6 in ARM
DDI 0478A.c.
Add an assert in the PioPort that checks if a response packet from a
device has the right flags set before passing it to them rest of the
memory system.
The new single stepping implementation for x86 doesn't rely on any ISA
specific properties or functionality. This change pulls out the per ISA
implementation of those functions and promotes the X86 implementation to the
base class.
One drawback of that implementation is that the CPU might stop on an
instruction twice if it's affected by both breakpoints and single stepping.
While that might be a little surprising, it's harmless and would only happen
under somewhat unlikely circumstances.
This stub should allow remote debugging of 32 bit and 64 bit targets. Single
stepping seems to work, as do breakpoints. If both breakpoints and single
stepping affect an instruction, gdb will stop at the instruction twice before
continuing. That's a little surprising, but is generally harmless.
