Andreas Sandberg [Fri, 2 Nov 2012 16:32:00 +0000 (11:32 -0500)]
Partly revert [
4f54b0f229b5] and move draining to m5.changeToTiming
Changeset
4f54b0f229b5 removed the call to doDrain in changeToTiming
based on the assumption that the system does not need draining when
running in atomic mode. This is a false assumption since at least the
System class requires the system to be drained before it allows
switching of memory modes. This patch reverts that part of the
changeset.
Andreas Hansson [Wed, 31 Oct 2012 13:28:23 +0000 (09:28 -0400)]
mem: Fix typo in port comments
This patch merely fixes a few typos in the port comments.
Andreas Hansson [Wed, 31 Oct 2012 12:39:45 +0000 (08:39 -0400)]
stats: Update stats for fixed simple-atomic-mp config
This patch updates the stats for the regressions that were affected by
the typo in the simple-atomic-mp configuration.
Andreas Hansson [Wed, 31 Oct 2012 12:39:43 +0000 (08:39 -0400)]
config: Fix a typo in the simple-atomic-mp configuration
This patch fixes a minor typo that managed to sneak into the
simple-atomic-mp regression configuration.
Andreas Hansson [Tue, 30 Oct 2012 13:35:32 +0000 (09:35 -0400)]
stats: Update stats for unified cache configuration
This patch updates the stats to reflect the changes in the L2 MSHRs,
as the latter are now uniform across the regressions.
Andreas Hansson [Tue, 30 Oct 2012 11:44:08 +0000 (07:44 -0400)]
config: Unify caches used in regressions and adjust L2 MSHRs
This patch unified the L1 and L2 caches used throughout the
regressions instead of declaring different, but very similar,
configurations in the different scripts.
The patch also changes the default L2 configuration to match what it
used to be for the fs and se scripts (until the last patch that
updated the regressions to also make use of the cache config). The
MSHRs and targets per MSHR are now set to a more realistic default of
20 and 12, respectively.
As a result of both the aforementioned changes, many of the regression
stats are changed. A follow-on patch will bump the stats.
Nilay Vaish [Sat, 27 Oct 2012 21:05:06 +0000 (16:05 -0500)]
regressions: update stats for ruby fs test
Malek Musleh [Sat, 27 Oct 2012 21:04:30 +0000 (16:04 -0500)]
ruby: set the is_icache param for caches
This patch sets the is_icache param for the L1 caches used in
the MESI and the MOESI CMP directory protocols.
Ruby: Use block size in configuring directory bits in address
This patch replaces hard coded values used in Ruby's configuration files
for setting directory bits with values based on the block size in use.
Andreas Hansson [Fri, 26 Oct 2012 10:42:45 +0000 (06:42 -0400)]
config: Add a check for fastmem only used with Atomic CPU
This patch adds an additional check to ensure that the fastmem option
is only used if the system is using the Atomic CPU.
Andreas Hansson [Fri, 26 Oct 2012 10:42:43 +0000 (06:42 -0400)]
config: Remove unused mem_size in fs.py
This patch removes a segment of dead code that is never used.
Andreas Hansson [Fri, 26 Oct 2012 10:42:42 +0000 (06:42 -0400)]
config: Fix the cache class naming in regression scripts
This patch unifies the naming of the default L1 and L2 caches in the
regression configs to be in line with what is used in the se and fs
scripts.
Andreas Hansson [Thu, 25 Oct 2012 17:15:59 +0000 (13:15 -0400)]
stats: Update the stats to reflect the 1GHz default system clock
This patch updates the stats to reflect the change in the default
system clock from 1 THz to 1GHz. The changes are due to the DMA
devices now injecting requests at a lower pace.
Andreas Hansson [Thu, 25 Oct 2012 17:14:44 +0000 (13:14 -0400)]
dev: Make default clock more reasonable for system and devices
This patch changes the default system clock from 1THz to 1GHz. This
clock is used by all modules that do not override the default (parent
clock), and primarily affects the IO subsystem. Every DMA device uses
its clock to schedule the next transfer, and the change will thus
cause this inter-transfer delay to be longer.
The default clock of the bus is removed, as the clock inherited from
the system provides exactly the same value.
A follow-on patch will bump the stats.
Andreas Hansson [Thu, 25 Oct 2012 17:14:42 +0000 (13:14 -0400)]
stats: Update stats to reflect use of SimpleDRAM
This patch bumps the stats to match the use of SimpleDRAM instead of
SimpleMemory in all inorder and O3 regressions, and also all
full-system regressions. A number of performance-related stats change,
and a whole bunch of stats are added for the memory controller.
Andreas Hansson [Thu, 25 Oct 2012 17:14:38 +0000 (13:14 -0400)]
config: Use SimpleDRAM in full-system, and with o3 and inorder
This patch favours using SimpleDRAM with the default timing instead of
SimpleMemory for all regressions that involve the o3 or inorder CPU,
or are full system (in other words, where the actual performance of
the memory is important for the overall performance).
Moving forward, the solution for FSConfig and the users of fs.py and
se.py is probably something similar to what we use to choose the CPU
type. I envision a few pre-set configurations SimpleLPDDR2,
SimpleDDR3, etc that can be choosen by a dram_type option. Feedback on
this part is welcome.
This patch changes plenty stats and adds all the DRAM controller
related stats. A follow-on patch updates the relevant statistics. The
total run-time for the entire regression goes up with ~5% with this
patch due to the added complexity of the SimpleDRAM model. This is a
concious trade-off to ensure that the model is properly tested.
Andreas Hansson [Thu, 25 Oct 2012 08:32:44 +0000 (04:32 -0400)]
config: Use shared cache config for regressions
This patch uses the common L1, L2 and IOCache configuration for the
regressions that all share the same cache parameters. There are a few
regressions that use a slightly different configuration (memtest,
o3-timing=mp, simple-atomic-mp and simple-timing-mp), and the latter
are not changed in this patch. They will be updated in a future patch.
The common cache configurations are changed to match the ones used in
the regressions, and are slightly changed with respect to what they
were. Hopefully this means we can converge on a common base
configuration, used both in the normal user configurations and
regressions.
As only regressions that shared the same cache configuration are
updated, no regressions are affected.
Andreas Hansson [Thu, 25 Oct 2012 08:32:42 +0000 (04:32 -0400)]
arm: Use table walker clock that is inherited from CPU
This patch simplifies the scheduling of the next walk for the ARM
table walker. Previously it used the CPU clock, but as the table
walker inherits the clock from the CPU, it is cleaner to simply use
its own clock (which is the same).
Andreas Hansson [Tue, 23 Oct 2012 08:49:48 +0000 (04:49 -0400)]
stats: Update stats for DMA port send
This patch updates the stats after removing the zero-time send used in
the DMA port.
Andreas Hansson [Tue, 23 Oct 2012 08:49:33 +0000 (04:49 -0400)]
dev: Remove zero-time loop in DMA timing send
This patch removes the zero-time loop used to send items from the DMA
port transmit list. Instead of having a loop, the DMA port now uses an
event to schedule sending of a single packet.
Ultimately this patch serves to ease the transition to a blocking
4-phase handshake.
A follow-on patch will update the regression statistics.
Andreas Hansson [Tue, 23 Oct 2012 08:24:32 +0000 (04:24 -0400)]
stats: Update t1000 stats to match recent changes
This patch brings the t1000 stats up to date.
Nilay Vaish [Thu, 18 Oct 2012 23:35:42 +0000 (18:35 -0500)]
ruby: functional access updates to network test protocol
I had forgotten to change the network test protocol while making changes to
ruby for supporting functional accesses. This patch updates the protocol so
that it can compile correctly.
Nilay Vaish [Tue, 16 Oct 2012 19:47:31 +0000 (14:47 -0500)]
regressions: update stats for eio tests
Nilay Vaish [Tue, 16 Oct 2012 00:13:59 +0000 (19:13 -0500)]
regressions: update stats due to change to ruby memory system
Nilay Vaish [Mon, 15 Oct 2012 22:51:57 +0000 (17:51 -0500)]
ruby: improved support for functional accesses
This patch adds support to different entities in the ruby memory system
for more reliable functional read/write accesses. Only the simple network
has been augmented as of now. Later on Garnet will also support functional
accesses.
The patch adds functional access code to all the different types of messages
that protocols can send around. These messages are functionally accessed
by going through the buffers maintained by the network entities.
The patch also rectifies some of the bugs found in coherence protocols while
testing the patch.
With this patch applied, functional writes always succeed. But functional
reads can still fail.
Nilay Vaish [Mon, 15 Oct 2012 22:27:17 +0000 (17:27 -0500)]
memtest: move check on outstanding requests
The Memtest tester allows for only one request to be outstanding for a
particular physical address. The check has been written separately for
reads and writes. This patch moves the check earlier than its current
position so that it need not be written separately for reads and writes.
Nilay Vaish [Mon, 15 Oct 2012 22:27:17 +0000 (17:27 -0500)]
ruby: register multiple memory controllers
Currently the Ruby System maintains pointer to only one of the memory
controllers. But there can be multiple controllers in the system. This
patch adds a vector of memory controllers.
Nilay Vaish [Mon, 15 Oct 2012 22:27:16 +0000 (17:27 -0500)]
ruby: remove AbstractMemOrCache
The only place where this abstract class is in use is the memory controller,
which it self is an abstract class. Does not seem useful at all.
Nilay Vaish [Mon, 15 Oct 2012 22:27:16 +0000 (17:27 -0500)]
ruby: allow function definition in slicc structs
This patch adds support for function definitions to appear in slicc structs.
This is required for supporting functional accesses for different types of
messages. Subsequent patches will use this to development.
Nilay Vaish [Mon, 15 Oct 2012 22:27:15 +0000 (17:27 -0500)]
ruby banked array: do away with event scheduling
It seems unecessary that the BankedArray class needs to schedule an event
to figure out when the access ends. Instead only the time for the end of access
needs to be tracked.
Nilay Vaish [Mon, 15 Oct 2012 22:27:15 +0000 (17:27 -0500)]
ruby: reset timing after cache warm up
Ruby system was recently converted to a clocked object. Such objects maintain
state related to the time that has passed so far. During the cache warmup, Ruby
system changes its own time and the global time. Later on, the global time is
restored. So Ruby system also needs to reset its own time.
Andreas Hansson [Mon, 15 Oct 2012 16:51:21 +0000 (12:51 -0400)]
Mem: Fix incorrect logic in bus blocksize check
This patch fixes the logic in the blocksize check such that the
warning is printed if the size is not 16, 32, 64 or 128.
Andreas Hansson [Mon, 15 Oct 2012 12:12:35 +0000 (08:12 -0400)]
Port: Add protocol-agnostic ports in the port hierarchy
This patch adds an additional level of ports in the inheritance
hierarchy, separating out the protocol-specific and protocl-agnostic
parts. All the functionality related to the binding of ports is now
confined to use BaseMaster/BaseSlavePorts, and all the
protocol-specific parts stay in the Master/SlavePort. In the future it
will be possible to add other protocol-specific implementations.
The functions used in the binding of ports, i.e. getMaster/SlavePort
now use the base classes, and the index parameter is updated to use
the PortID typedef with the symbolic InvalidPortID as the default.
Andreas Hansson [Mon, 15 Oct 2012 12:12:32 +0000 (08:12 -0400)]
Mem: Separate the host and guest views of memory backing store
This patch moves all the memory backing store operations from the
independent memory controllers to the global physical memory. The main
reason for this patch is to allow address striping in a future set of
patches, but at this point it already provides some useful
functionality in that it is now possible to change the number of
memory controllers and their address mapping in combination with
checkpointing. Thus, the host and guest view of the memory backing
store are now completely separate.
With this patch, the individual memory controllers are far simpler as
all responsibility for serializing/unserializing is moved to the
physical memory. Currently, the functionality is more or less moved
from AbstractMemory to PhysicalMemory without any major
changes. However, in a future patch the physical memory will also
resolve any ranges that are interleaved and properly assign the
backing store to the memory controllers, and keep the host memory as a
single contigous chunk per address range.
Functionality for future extensions which involve CPU virtualization
also enable the host to get pointers to the backing store.
Andreas Hansson [Mon, 15 Oct 2012 12:12:29 +0000 (08:12 -0400)]
Checkpoint: Make system serialize call children
This patch changes how the serialization of the system works. The base
class had a non-virtual serialize and unserialize, that was hidden by
a function with the same name for a number of subclasses (most likely
not intentional as the base class should have been virtual). A few of
the derived systems had no specialization at all (e.g. Power and x86
that simply called the System::serialize), but MIPS and Alpha adds
additional symbol table entries to the checkpoint.
Instead of overriding the virtual function, the additional entries are
now printed through a virtual function (un)serializeSymtab. The reason
for not calling System::serialize from the two related systems is that
a follow up patch will require the system to also serialize the
PhysicalMemory, and if this is done in the base class if ends up being
between the general parts and the specialized symbol table.
With this patch, the checkpoint is not modified, as the order of the
segments is unchanged.
Andreas Hansson [Mon, 15 Oct 2012 12:12:25 +0000 (08:12 -0400)]
Mem: Use deque instead of list for bus retries
This patch changes the data structure used to keep track of ports that
should be told to retry. As the bus is doing this in an FCFS way,
there is no point having a list. A deque is a better match (and is at
least in theory a better choice from a performance point of view).
Andreas Hansson [Mon, 15 Oct 2012 12:12:23 +0000 (08:12 -0400)]
Fix: Address a few minor issues identified by cppcheck
This patch addresses a number of smaller issues identified by the code
inspection utility cppcheck. There are a number of identified leaks in
the arm/linux/system.cc (although the function only get's called once
so it is not a major problem), a few deletes in dev/x86/i8042.cc that
were not array deletes, and sprintfs where the character array had one
element less than needed. In the IIC tags there was a function
allocating an array of longs which is in fact never used.
Andreas Hansson [Mon, 15 Oct 2012 12:12:21 +0000 (08:12 -0400)]
Stats: Update stats for cache timings in cycles
This patch updates the stats to reflect the change in how cache
latencies are expressed. In addition, the latencies are now rounded to
multiples of the clock period, thus also affecting other stats.
Andreas Hansson [Mon, 15 Oct 2012 12:10:54 +0000 (08:10 -0400)]
Mem: Use cycles to express cache-related latencies
This patch changes the cache-related latencies from an absolute time
expressed in Ticks, to a number of cycles that can be scaled with the
clock period of the caches. Ultimately this patch serves to enable
future work that involves dynamic frequency scaling. As an immediate
benefit it also makes it more convenient to specify cache performance
without implicitly assuming a specific CPU core operating frequency.
The stat blocked_cycles that actually counter in ticks is now updated
to count in cycles.
As the timing is now rounded to the clock edges of the cache, there
are some regressions that change. Plenty of them have very minor
changes, whereas some regressions with a short run-time are perturbed
quite significantly. A follow-on patch updates all the statistics for
the regressions.
Andreas Hansson [Mon, 15 Oct 2012 12:10:52 +0000 (08:10 -0400)]
Stats: Update memtest stats after setting clock
This patch updates the memtest stats to reflect the addition of a
clock other than the default one.
Andreas Hansson [Mon, 15 Oct 2012 12:09:57 +0000 (08:09 -0400)]
Configs: Set the memtest clock to a reasonable value
This patch changes the memtest clock from 1THz (the default) to 2GHz,
similar to the CPUs in the other regressions. This is useful as the
caches will adopt the same clock as the CPU. The bus clock rate is
scaled accordingly, and the L1-L2 bus is kept at the CPU clock while
the memory bus is at half that frequency.
A separate patch updates the affected stats.
Andreas Hansson [Mon, 15 Oct 2012 12:09:54 +0000 (08:09 -0400)]
Stats: Update stats for new default L1-to-L2 bus clock and width
This patch updates the stats to reflect the changes in the clock speed
and width for the bus connecting the L1 and L2 caches.
Andreas Hansson [Mon, 15 Oct 2012 12:08:08 +0000 (08:08 -0400)]
Regression: Use CPU clock and 32-byte width for L1-L2 bus
This patch changes the CoherentBus between the L1s and L2 to use the
CPU clock and also four times the width compared to the default
bus. The parameters are not intending to fit every single scenario,
but rather serve as a better startingpoint than what we previously
had.
Note that the scripts that do not use the addTwoLevelCacheHiearchy are
not affected by this change.
A separate patch will update the stats.
Andreas Hansson [Mon, 15 Oct 2012 12:08:06 +0000 (08:08 -0400)]
Stats: Update stats for use of two-level builder
This patch updates the name of the l2 stats.
Andreas Hansson [Mon, 15 Oct 2012 12:07:09 +0000 (08:07 -0400)]
Regression: Use addTwoLevelCacheHierarchy in configs
This patch unifies the full-system regression config scripts and uses
the BaseCPU convenience method addTwoLevelCacheHierarchy to connect up
the L1s and L2, and create the bus inbetween.
The patch is a step on the way to use the clock period to express the
cache latencies, as the CPU is now the parent of the L1, L2 and L1-L2
bus, and these modules thus use the CPU clock.
The patch does not change the value of any stats, but plenty names,
and a follow-up patch contains the update to the stats, chaning
system.l2c to system.cpu.l2cache.
Andreas Hansson [Mon, 15 Oct 2012 12:07:07 +0000 (08:07 -0400)]
Clock: Inherit the clock from parent by default
This patch changes the default 1 Tick clock period to a proxy that
resolves the parents clock. As a result of this, the caches and
L1-to-L2 bus, for example, will automatically use the clock period of
the CPU unless explicitly overridden.
To ensure backwards compatibility, the System class overrides the
proxy and specifies a 1 Tick clock. We could change this to something
more reasonable in a follow-on patch, perhaps 1 GHz or something
similar.
With this patch applied, all clocked objects should have a reasonable
clock period set, and could start specifying delays in Cycles instead
of absolute time.
Andreas Hansson [Mon, 15 Oct 2012 12:07:06 +0000 (08:07 -0400)]
Param: Fix proxy traversal to support chained proxies
This patch modifies how proxies are traversed and unproxied to allow
chained proxies. The issue that is solved manifested itself when a
proxy during its evaluation ended up being hitting another proxy, and
the second one got evaluated using the object that was originally used
for the first proxy.
For a more tangible example, see the following patch on making the
default clock being inherited from the parent. In this patch, the CPU
clock is a proxy Parent.clock, which is overridden in the system to be
an actual value. This all works fine, but the AlphaLinuxSystem has a
boot_cpu_frequency parameter that is Self.cpu[0].clock.frequency. When
the latter is evaluated, it all happens relative to the current object
of the proxy, i.e. the system. Thus the cpu.clock is evaluated as
Parent.clock, but using the system rather than the cpu as the object
to enquire.
Andreas Hansson [Mon, 15 Oct 2012 12:07:04 +0000 (08:07 -0400)]
Mem: Use range operations in bus in preparation for striping
This patch transitions the bus to use the AddrRange operations instead
of directly accessing the start and end. The change facilitates the
move to a more elaborate AddrRange class that also supports address
striping in the bus by specifying interleaving bits in the ranges.
Two new functions are added to the AddrRange to determine if two
ranges intersect, and if one is a subset of another. The bus
propagation of address ranges is also tweaked such that an update is
only propagated if the bus received information from all the
downstream slave modules. This avoids the iteration and need for the
cycle-breaking scheme that was previously used.
Andreas Hansson [Thu, 11 Oct 2012 10:38:43 +0000 (06:38 -0400)]
Mem: Determine bus block size during initialisation
This patch moves the block size computation from findBlockSize to
initialisation time, once all the neighbouring ports are connected.
There is no need to dynamically update the block size, and the caching
of the value effectively avoided that anyhow. This is very similar to
what was already in place, just with a slightly leaner implementation.
Andreas Hansson [Thu, 11 Oct 2012 10:38:42 +0000 (06:38 -0400)]
Doxygen: Update the version of the Doxyfile
This patch bumps the Doxyfile to match more recent versions of
Doxygen. The sections that are deprecated have been removed, and the
new ones added. The project name has also been updated.
Nilay Vaish [Tue, 2 Oct 2012 19:35:46 +0000 (14:35 -0500)]
Regression Tests: Update statistics
Nilay Vaish [Tue, 2 Oct 2012 19:35:45 +0000 (14:35 -0500)]
ruby: makes some members non-static
This patch makes some of the members (profiler, network, memory vector)
of ruby system non-static.
Nilay Vaish [Tue, 2 Oct 2012 19:35:45 +0000 (14:35 -0500)]
ruby: changes to simple network
This patch makes the Switch structure inherit from BasicRouter, as is
done in two other networks.
Nilay Vaish [Tue, 2 Oct 2012 19:35:44 +0000 (14:35 -0500)]
ruby: rename template_hack to template
I don't like using the word hack. Hence, the patch.
Nilay Vaish [Tue, 2 Oct 2012 19:35:44 +0000 (14:35 -0500)]
ruby: remove unused code in protocols
Nilay Vaish [Tue, 2 Oct 2012 19:35:43 +0000 (14:35 -0500)]
ruby: remove some unused things in slicc
This patch removes the parts of slicc that were required for multi-chip
protocols. Going ahead, it seems multi-chip protocols would be implemented
by playing with the network itself.
Nilay Vaish [Tue, 2 Oct 2012 19:35:42 +0000 (14:35 -0500)]
ruby: move functional access to ruby system
This patch moves the code for functional accesses to ruby system. This is
because the subsequent patches add support for making functional accesses
to the messages in the interconnect. Making those accesses from the ruby port
would be cumbersome.
Nilay Vaish [Sun, 30 Sep 2012 18:20:53 +0000 (13:20 -0500)]
MI coherence protocol: add copyright notice
Malek Musleh [Fri, 28 Sep 2012 13:35:25 +0000 (09:35 -0400)]
Configs: SE script fix for Alpha and Ruby simulations
PIO interrupt port is only present for x86. Do not attempt to connect
for other ISAs.
Andreas Hansson [Thu, 27 Sep 2012 12:59:25 +0000 (08:59 -0400)]
Configs: Fix memtest cache latency to match new parameters
This patch changes the memtest config to use the new response latency
of the cache model.
Andreas Hansson [Thu, 27 Sep 2012 07:24:21 +0000 (03:24 -0400)]
Configs: Fix memtest.py by moving the system port
The memtest.py script used to connect the system port directly to the
SimpleMemory, but the latter is now single ported. Since the system
port is not used for anything in this particular example, a quick fix
is to attach it to the functional bus instead.
Ali Saidi [Tue, 25 Sep 2012 16:49:41 +0000 (11:49 -0500)]
ARM: update stats for bp and squash fixes.
Djordje Kovacevic [Tue, 25 Sep 2012 16:49:41 +0000 (11:49 -0500)]
MEM: Put memory system document into doxygen
Mrinmoy Ghosh [Tue, 25 Sep 2012 16:49:41 +0000 (11:49 -0500)]
Cache: add a response latency to the caches
In the current caches the hit latency is paid twice on a miss. This patch lets
a configurable response latency be set of the cache for the backward path.
Sascha Bischoff [Tue, 25 Sep 2012 16:49:41 +0000 (11:49 -0500)]
Statistics: Add a function to configure periodic stats dumping
This patch adds a function, periodicStatDump(long long period), which will dump
and reset the statistics every period. This function is designed to be called
from the python configuration scripts. This allows the periodic stats dumping to
be configured more easilly at run time.
The period is currently specified as a long long as there are issues passing
Tick into the C++ from the python as they have conflicting definitions. If the
period is less than curTick, the first occurance occurs at curTick. If the
period is set to 0, then the event is descheduled and the stats are not
periodically dumped.
Due to issues when resumung from a checkpoint, the StatDump event must be moved
forward such that it occues AFTER the current tick. As the function is called
from the python, the event is scheduled before the system resumes from the
checkpoint. Therefore, the event is moved using the updateEvents() function.
This is called from simulate.py once the system has resumed from the checkpoint.
NOTE: It should be noted that this is a fairly temporary patch which re-adds the
capability to extract temporal information from the communication monitors. It
should not be used at the same time as anything that relies on dumping the
statistics based on in simulation events i.e. a context switch.
Dam Sunwoo [Tue, 25 Sep 2012 16:49:41 +0000 (11:49 -0500)]
ARM: added support for flattened device tree blobs
Newer Linux kernels require DTB (device tree blobs) to specify platform
configurations. The input DTB filename can be specified through gem5 parameters
in LinuxArmSystem.
Ali Saidi [Tue, 25 Sep 2012 16:49:40 +0000 (11:49 -0500)]
O3: Pack the comm structures a bit better to reduce their size.
Ali Saidi [Tue, 25 Sep 2012 16:49:40 +0000 (11:49 -0500)]
mem: Add a gasket that allows memory ranges to be re-mapped.
For example if DRAM is at two locations and mirrored this patch allows the
mirroring to occur.
Ali Saidi [Tue, 25 Sep 2012 16:49:40 +0000 (11:49 -0500)]
ARM: Squash outstanding walks when instructions are squashed.
Sascha Bischoff [Tue, 25 Sep 2012 16:49:40 +0000 (11:49 -0500)]
Util: Added script to semantically diff two config.ini files
This script (util/diff_config.pl) takes two config.ini files and compares them.
It highlights value changes, as well as displaying which parts are unique to
a specific config.ini file. This is useful when trying to replicate an earlier
experiment and when trying to make small changes to an existing configuration.
Andreas Sandberg [Tue, 25 Sep 2012 16:49:40 +0000 (11:49 -0500)]
arm: Use a static_assert to test that miscRegName[] is complete
Instead of statically defining miscRegName to contain NUM_MISCREGS
elements, let the compiler determine the length of the array. This
allows us to use a static_assert to test that all registers are listed
in the name vector.
Andreas Sandberg [Tue, 25 Sep 2012 16:49:40 +0000 (11:49 -0500)]
base: Check for static_assert support and provide fallback
C++11 has support for static_asserts to provide compile-time assertion
checking. This is very useful when testing, for example, structure
sizes to make sure that the compiler got the right alignment or vector
sizes.
Andreas Sandberg [Tue, 25 Sep 2012 16:49:40 +0000 (11:49 -0500)]
sim: Move CPU-specific methods from SimObject to the BaseCPU class
Andreas Sandberg [Tue, 25 Sep 2012 16:49:40 +0000 (11:49 -0500)]
sim: Remove SimObject::setMemoryMode
Remove SimObject::setMemoryMode from the main SimObject class since it
is only valid for the System class. In addition to removing the method
from the C++ sources, this patch also removes getMemoryMode and
changeTiming from SimObject.py and updates the simulation code to call
the (get|set)MemoryMode method on the System object instead.
Djordje Kovacevic [Tue, 25 Sep 2012 16:49:40 +0000 (11:49 -0500)]
CPU: Add abandoned instructions to O3 Pipe Viewer
Nathanael Premillieu [Tue, 25 Sep 2012 16:49:40 +0000 (11:49 -0500)]
ARM: Inst writing to cntrlReg registers not set as control inst
Deletion of the fact that instructions that writes to registers of type
"cntrlReg" are not set as control instruction (flag IsControl not set).
Ali Saidi [Tue, 25 Sep 2012 16:49:40 +0000 (11:49 -0500)]
ARM: Predict target of more instructions that modify PC.
Ali Saidi [Tue, 25 Sep 2012 16:49:40 +0000 (11:49 -0500)]
gem5: Update the README file to be a bit less out-of-date.
Andreas Sandberg [Tue, 25 Sep 2012 16:49:40 +0000 (11:49 -0500)]
build: Add missing dependencies when building param SWIG interfaces
This patch adds an explicit dependency between param_%s.i and the
Python source file defining the object. Previously, the build system
didn't rebuild SWIG interfaces correctly when an object's Python
sources were updated.
Andreas Hansson [Mon, 24 Sep 2012 22:03:43 +0000 (18:03 -0400)]
Stats: Update stats for twosys-tsunami after setting CPU clock
This patch updates the stats to reflect the addition of a clock
period other than the default 1 Tick.
Andreas Hansson [Mon, 24 Sep 2012 22:03:41 +0000 (18:03 -0400)]
Regression: Set the clock for twosys-tsunami CPUs
This patch merely adds a clock other than the default 1 Tick for the
CPUs of both the test system and drive system for the twosys-tsunami
regression.
The CPU frequency of the driver system is choosed to be twice that of
the test system to ensure it is not the bottleneck (although in this
case it mostly serves as a demonstration of a two-system setup),
Joel Hestness [Sun, 23 Sep 2012 18:57:08 +0000 (13:57 -0500)]
RubyPort and Sequencer: Fix draining
Fix the drain functionality of the RubyPort to only call drain on child ports
during a system-wide drain process, instead of calling each time that a
ruby_hit_callback is executed.
This fixes the issue of the RubyPort ports being reawakened during the drain
simulation, possibly with work they didn't previously have to complete. If
they have new work, they may call process on the drain event that they had
not registered work for, causing an assertion failure when completing the
drain event.
Also, in RubyPort, set the drainEvent to NULL when there are no events
to be drained. If not set to NULL, the drain loop can result in stale
drainEvents used.
Andreas Hansson [Fri, 21 Sep 2012 15:48:14 +0000 (11:48 -0400)]
SimpleDRAM: A basic SimpleDRAM regression
--HG--
rename : tests/configs/tgen-simple-mem.py => tests/configs/tgen-simple-dram.py
rename : tests/quick/se/70.tgen/tgen-simple-mem.cfg => tests/quick/se/70.tgen/tgen-simple-dram.cfg
rename : tests/quick/se/70.tgen/tgen-simple-mem.trc => tests/quick/se/70.tgen/tgen-simple-dram.trc
Andreas Hansson [Fri, 21 Sep 2012 15:48:13 +0000 (11:48 -0400)]
DRAM: Introduce SimpleDRAM to capture a high-level controller
This patch introduces a high-level model of a DRAM controller, with a
basic read/write buffer structure, a selectable and customisable
arbiter, a few address mapping options, and the basic DRAM timing
constraints. The parameters make it possible to turn this model into
any desired DDRx/LPDDRx/WideIOx memory controller.
The intention is not to be cycle accurate or capture every aspect of a
DDR DRAM interface, but rather to enable exploring of the high-level
knobs with a good simulation speed. Thus, contrary to e.g. DRAMSim
this module emphasizes simulation speed with a good-enough accuracy.
This module is merely a starting point, and there are plenty additions
and improvements to come. A notable addition is the support for
address-striping in the bus to enable a multi-channel DRAM
controller. Also note that there are still a few "todo's" in the code
base that will be addressed as we go along.
A follow-up patch will add basic performance regressions that use the
traffic generator to exercise a few well-defined corner cases.
Andreas Hansson [Fri, 21 Sep 2012 15:48:11 +0000 (11:48 -0400)]
TrafficGen: Add a basic traffic generator regression
This patch adds a basic regression for the traffic generator. The
regression also serves as an example of the file formats used. More
complex regressions that make use of a DRAM controller model will
follow shortly.
Andreas Hansson [Fri, 21 Sep 2012 15:48:08 +0000 (11:48 -0400)]
TrafficGen: Add a basic traffic generator
This patch adds a traffic generator to the code base. The generator is
aimed to be used as a black box model to create appropriate use-cases
and benchmarks for the memory system, and in particular the
interconnect and the memory controller.
The traffic generator is a master module, where the actual behaviour
is captured in a state-transition graph where each state generates
some sort of traffic. By constructing a graph it is possible to create
very elaborate scenarios from basic generators. Currencly the set of
generators include idling, linear address sweeps, random address
sequences and playback of traces (recording will be done by the
Communication Monitor in a follow-up patch). At the moment the graph
and the states are described in an ad-hoc line-based format, and in
the future this should be aligned with our used of e.g. the Google
protobufs. Similarly for the traces, the format is currently a
simplistic ad-hoc line-based format that merely serves as a starting
point.
In addition to being used as a black-box model for system components,
the traffic generator is also useful for creating test cases and
regressions for the interconnect and memory system. In future patches
we will use the traffic generator to create DRAM test cases for the
controller model.
The patch following this one adds a basic regressions which also
contains an example configuration script and trace file for playback.
Andreas Hansson [Fri, 21 Sep 2012 14:11:24 +0000 (10:11 -0400)]
Mem: Tidy up bus member variables types
This patch merely tidies up the types used for the bus member
variables. It also makes the constant ones const.
Andreas Hansson [Fri, 21 Sep 2012 14:11:22 +0000 (10:11 -0400)]
Scons: Verbose messages when dependencies are not installed
This patch adds a few more checks to ensure that a compiler is present
on the system, along with swig. It references the relevant packages on
Ubuntu/RedHat, and also adds a similar line for the Python headers.
Lluc Alvarez [Fri, 21 Sep 2012 08:51:18 +0000 (04:51 -0400)]
SE: Ignore FUTEX_PRIVATE_FLAG of sys_futex
This patch ignores the FUTEX_PRIVATE_FLAG of the sys_futex system call
in SE mode.
With this patch, when sys_futex with the options FUTEX_WAIT_PRIVATE or
FUTEX_WAKE_PRIVATE is emulated, the FUTEX_PRIVATE_FLAG is ignored and
so their behaviours are the regular FUTEX_WAIT and FUTEX_WAKE.
Emulating FUTEX_WAIT_PRIVATE and FUTEX_WAKE_PRIVATE as if they were
non-private is safe from a functional point of view. The
FUTEX_PRIVATE_FLAG does not change the semantics of the futex, it's
just a mechanism to improve performance under certain circunstances
that can be ignored in SE mode.
Anthony Gutierrez [Thu, 20 Sep 2012 21:25:52 +0000 (17:25 -0400)]
bus: removed outdated warn regarding 64 B block sizes
this warn is outdated as 64 B blocks are very common, and even
the default size for some CPU types. E.g., arm_detailed.
Andreas Hansson [Wed, 19 Sep 2012 10:15:46 +0000 (06:15 -0400)]
Mem: Remove the file parameter from AbstractMemory
This patch removes the unused file parameter from the
AbstractMemory. The patch serves to make it easier to transition to a
separation of the actual contigious host memory backing store, and the
gem5 memory controllers.
Without the file parameter it becomes easier to hide the creation of
the mmap in the PhysicalMemory, as there are no longer any reasons to
expose the actual contigious ranges to the user.
To the best of my knowledge there is no use of the parameter, so the
change should not affect anyone.
Andreas Hansson [Wed, 19 Sep 2012 10:15:44 +0000 (06:15 -0400)]
AddrRange: Transition from Range<T> to AddrRange
This patch takes the final plunge and transitions from the templated
Range class to the more specific AddrRange. In doing so it changes the
obvious Range<Addr> to AddrRange, and also bumps the range_map to be
AddrRangeMap.
In addition to the obvious changes, including the removal of redundant
includes, this patch also does some house keeping in preparing for the
introduction of address interleaving support in the ranges. The Range
class is also stripped of all the functionality that is never used.
--HG--
rename : src/base/range.hh => src/base/addr_range.hh
rename : src/base/range_map.hh => src/base/addr_range_map.hh
Andreas Hansson [Wed, 19 Sep 2012 10:15:43 +0000 (06:15 -0400)]
AddrRange: Simplify Range by removing stream input/output
This patch simplifies the Range class in preparation for the
introduction of a more specific AddrRange class that allows
interleaving/striping.
The only place where the parsing was used was in the unit test.
Andreas Hansson [Wed, 19 Sep 2012 10:15:42 +0000 (06:15 -0400)]
AddrRange: Remove unused range_multimap
This patch simply removes the unused range_multimap in preparation for
a more specific AddrRangeMap that also allows interleaving in addition
to pure ranges.
Andreas Hansson [Wed, 19 Sep 2012 10:15:41 +0000 (06:15 -0400)]
AddrRange: Simplify AddrRange params Python hierarchy
This patch simplifies the Range object hierarchy in preparation for an
address range class that also allows striping (e.g. selecting a few
bits as matching in addition to the range).
To extend the AddrRange class to an AddrRegion, the first step is to
simplify the hierarchy such that we can make it as lean as possible
before adding the new functionality. The only class using Range and
MetaRange is AddrRange, and the three classes are now collapsed into
one.
Nilay Vaish [Wed, 19 Sep 2012 03:49:12 +0000 (22:49 -0500)]
ruby: eliminate typedef integer_t
Nilay Vaish [Wed, 19 Sep 2012 03:46:34 +0000 (22:46 -0500)]
ruby: avoid using g_system_ptr for event scheduling
This patch removes the use of g_system_ptr for event scheduling. Each consumer
object now needs to specify upfront an EventManager object it would use for
scheduling events. This makes the ruby memory system more amenable for a
multi-threaded simulation.
Andreas Hansson [Tue, 18 Sep 2012 14:30:04 +0000 (10:30 -0400)]
Stats: Update stats to reflect SimpleMemory bandwidth
This patch simply bumps the stats to reflect the introduction of a
bandwidth limit of 12.8GB/s for SimpleMemory.
Andreas Hansson [Tue, 18 Sep 2012 14:30:02 +0000 (10:30 -0400)]
Mem: Add a maximum bandwidth to SimpleMemory
This patch makes a minor addition to the SimpleMemory by enforcing a
maximum data rate. The bandwidth is configurable, and a reasonable
value (12.8GB/s) has been choosen as the default.
The changes do add some complexity to the SimpleMemory, but they
should definitely be justifiable as this enables a far more realistic
setup using even this simple memory controller.
The rate regulation is done for reads and writes combined to reflect
the bidirectional data busses used by most (if not all) relevant
memories. Moreover, the regulation is done per packet as opposed to
long term, as it is the short term data rate (data bus width times
frequency) that is the limiting factor.
A follow-up patch bumps the stats for the regressions.
Andreas Hansson [Fri, 14 Sep 2012 16:13:22 +0000 (12:13 -0400)]
gcc: Enable Link-Time Optimization for gcc >= 4.6
This patch adds Link-Time Optimization when building the fast target
using gcc >= 4.6, and adds a scons flag to disable it (-no-lto). No
check is performed to guarantee that the linker supports LTO and use
of the linker plugin, so the user has to ensure that binutils GNU ld
>= 2.21 or the gold linker is available. Typically, if gcc >= 4.6 is
available, the latter should not be a problem. Currently the LTO
option is only useful for gcc >= 4.6, due to the limited support on
clang and earlier versions of gcc. The intention is to also add
support for clang once the LTO integration matures.
The same number of jobs is used for the parallel phase of LTO as the
jobs specified on the scons command line, using the -flto=n flag that
was introduced with gcc 4.6. The gold linker also supports concurrent
and incremental linking, but this is not used at this point.
The compilation and linking time is increased by almost 50% on
average, although ARM seems to be particularly demanding with an
increase of almost 100%. Also beware when using this as gcc uses a
tremendous amount of memory and temp space in the process. You have
been warned.
After some careful consideration, and plenty discussions, the flag is
only added to the fast target, and the warning that was issued in an
earlier version of this patch is now removed. Similarly, the flag used
to enable LTO, now the default is to use it, and the flag has been
modified to disable LTO. The rationale behind this decision is that
opt is used for development, whereas fast is only used for long runs,
e.g. regressions or more elaborate experiments where the additional
compile and link time is amortized by a much larger run time.
When it comes to the return on investment, the regression seems to be
roughly 15% faster with LTO. For a bit more detail, I ran twolf on
ARM.fast, with three repeated runs, and they all finish within 42
minutes (+- 25 seconds) without LTO and 31 minutes (+- 25 seconds)
with LTO, i.e. LTO gives an impressive >25% speed-up for this case.
Without LTO (ARM.fast twolf)
real 42m37.632s
user 42m34.448s
sys 0m0.390s
real 41m51.793s
user 41m50.384s
sys 0m0.131s
real 41m45.491s
user 41m39.791s
sys 0m0.139s
With LTO (ARM.fast twolf)
real 30m33.588s
user 30m5.701s
sys 0m0.141s
real 31m27.791s
user 31m24.674s
sys 0m0.111s
real 31m25.500s
user 31m16.731s
sys 0m0.106s