gem5.git
13 years agoCurtick: Fix mysql.cc build needing curTick.
Gabe Black [Mon, 10 Jan 2011 12:53:20 +0000 (04:53 -0800)]
Curtick: Fix mysql.cc build needing curTick.

13 years agoRefCount: Add a unit test for reference counting pointers.
Gabe Black [Mon, 10 Jan 2011 11:56:42 +0000 (03:56 -0800)]
RefCount: Add a unit test for reference counting pointers.

This test exercises each of the functions in the reference counting pointer
implementation individually (except get()) and verifies they have some
minimially expected behavior. It also checks that reference counted objects
are freed when their usage count goes to 0 in some basic situations,
specifically a pointer being set to NULL and a pointer being deleted.

13 years agoReplace curTick global variable with accessor functions.
Steve Reinhardt [Sat, 8 Jan 2011 05:50:29 +0000 (21:50 -0800)]
Replace curTick global variable with accessor functions.
This step makes it easy to replace the accessor functions
(which still access a global variable) with ones that access
per-thread curTick values.

13 years agostats: rename StatEvent() function to schedStatEvent().
Steve Reinhardt [Sat, 8 Jan 2011 05:50:29 +0000 (21:50 -0800)]
stats: rename StatEvent() function to schedStatEvent().
This follows the style rules and is more descriptive.

13 years agosim: clean up CountedDrainEvent slightly.
Steve Reinhardt [Sat, 8 Jan 2011 05:50:29 +0000 (21:50 -0800)]
sim: clean up CountedDrainEvent slightly.
There's no reason for it to derive from SimLoopExitEvent.
This whole drain thing needs to be redone eventually,
but this is a stopgap to make later changes to
SimLoopExitEvent feasible.

13 years agosim: delete unused CheckSwapEvent code.
Steve Reinhardt [Sat, 8 Jan 2011 05:50:29 +0000 (21:50 -0800)]
sim: delete unused CheckSwapEvent code.
There's no way to even create one of these anymore.

13 years agopseudoinst: get rid of mainEventQueue references.
Steve Reinhardt [Sat, 8 Jan 2011 05:50:29 +0000 (21:50 -0800)]
pseudoinst: get rid of mainEventQueue references.
Avoid direct references to mainEventQueue in pseudo-insts
by indirecting through associated CPU object.
Made exitSimLoop() more flexible to enable some of these.

13 years agoinorder: replace schedEvent() code with reschedule().
Steve Reinhardt [Sat, 8 Jan 2011 05:50:29 +0000 (21:50 -0800)]
inorder: replace schedEvent() code with reschedule().
There were several copies of similar functions that looked
like they all replicated reschedule(), so I replaced them
with direct calls.  Keeping this separate from the previous
cset since there may be some subtle functional differences
if the code ever reschedules an event that is scheduled but
not squashed (though none were detected in the regressions).

13 years agoinorder: get rid of references to mainEventQueue.
Steve Reinhardt [Sat, 8 Jan 2011 05:50:29 +0000 (21:50 -0800)]
inorder: get rid of references to mainEventQueue.
Events need to be scheduled on the queue assigned
to the SimObject, not on the global queue (which
should be going away).
Also cleaned up a number of redundant expressions
that made the code unnecessarily verbose.

13 years agoscons: show sources and targets when building, and colorize output.
Steve Reinhardt [Sat, 8 Jan 2011 05:50:13 +0000 (21:50 -0800)]
scons: show sources and targets when building, and colorize output.

I like the brevity of Ali's recent change, but the ambiguity of
sometimes showing the source and sometimes the target is a little
confusing.  This patch makes scons typically list all sources and
all targets for each action, with the common path prefix factored
out for brevity.  It's a little more verbose now but also more
informative.

Somehow Ali talked me into adding colors too, which is a whole
'nother story.

13 years agoRuby: Updates MOESI Hammer protocol
Nilay Vaish [Wed, 5 Jan 2011 03:40:49 +0000 (21:40 -0600)]
Ruby: Updates MOESI Hammer protocol
This patch changes the manner in which data is copied from L1 to L2 cache in
the implementation of the Hammer's cache coherence protocol. Earlier, data was
copied directly from one cache entry to another. This has been broken in to
two parts. First, the data is copied from the source cache entry to a
transaction buffer entry. Then, data is copied from the transaction buffer
entry to the destination cache entry.

This has been done to maintain the invariant - at any given instant, multiple
caches under a controller are exclusive with respect to each other.

13 years agoParams: Print the IP components in the right order.
Gabe Black [Tue, 4 Jan 2011 22:11:49 +0000 (17:11 -0500)]
Params: Print the IP components in the right order.

13 years agoMove sched_list.hh and timebuf.hh from src/base to src/cpu.
Steve Reinhardt [Mon, 3 Jan 2011 22:35:47 +0000 (14:35 -0800)]
Move sched_list.hh and timebuf.hh from src/base to src/cpu.
These files really aren't general enough to belong in src/base.
This patch doesn't reorder include lines, leaving them unsorted
in many cases, but Nate's magic script will fix that up shortly.

--HG--
rename : src/base/sched_list.hh => src/cpu/sched_list.hh
rename : src/base/timebuf.hh => src/cpu/timebuf.hh

13 years agoDelete unused files from src/base directory.
Steve Reinhardt [Mon, 3 Jan 2011 22:35:45 +0000 (14:35 -0800)]
Delete unused files from src/base directory.

13 years agoMake commenting on close namespace brackets consistent.
Steve Reinhardt [Mon, 3 Jan 2011 22:35:43 +0000 (14:35 -0800)]
Make commenting on close namespace brackets consistent.

Ran all the source files through 'perl -pi' with this script:

s|\s*(};?\s*)?/\*\s*(end\s*)?namespace\s*(\S+)\s*\*/(\s*})?|} // namespace $3|;
s|\s*};?\s*//\s*(end\s*)?namespace\s*(\S+)\s*|} // namespace $2\n|;
s|\s*};?\s*//\s*(\S+)\s*namespace\s*|} // namespace $1\n|;

Also did a little manual editing on some of the arch/*/isa_traits.hh files
and src/SConscript.

13 years agoRefCount: Fix reference counting pointer == and != with a T* on the left.
Gabe Black [Mon, 3 Jan 2011 20:31:20 +0000 (15:31 -0500)]
RefCount: Fix reference counting pointer == and != with a T* on the left.

These operators were expecting a const T& instead of a const T*, and were not
being picked up and used by gcc in the right places as a result. Apparently no
one used these operators before. A unit test which exposed these problems,
verified the solution, and checks other basic functionality is on the way.

13 years agoRuby: Add option for random seed to Ruby.py
Nilay Vaish [Mon, 3 Jan 2011 18:40:31 +0000 (12:40 -0600)]
Ruby: Add option for random seed to Ruby.py
This patch adds an option to the script Ruby.py for setting the parameter
m_random_seed used for randomizing delays in the memory system. The option
can be specified as "--random_seed <seed value>".

14 years agoswig: use <> for system %includes instead of ""
Nathan Binkert [Thu, 30 Dec 2010 17:51:04 +0000 (12:51 -0500)]
swig: use <> for system %includes instead of ""

14 years agostyle: make style hook work with pre-qrefresh and update to use new code
Nathan Binkert [Thu, 30 Dec 2010 17:53:56 +0000 (12:53 -0500)]
style: make style hook work with pre-qrefresh and update to use new code
clean up the code a little bit while we're at it.

I recommend that everyone adds the pre-qrefresh hook below since it
will make qref run the style hook and not just commit/qpush

[extensions]
style = <m5 path>/util/style.py

[hooks]
pretxncommit.style = python:style.check_whitespace
pre-qrefresh.style = python:style.check_whitespace

14 years agoPerfectCacheMemory: Add return statements to two functions.
Nilay Vaish [Thu, 23 Dec 2010 19:36:18 +0000 (13:36 -0600)]
PerfectCacheMemory: Add return statements to two functions.
Two functions in src/mem/ruby/system/PerfectCacheMemory.hh, tryCacheAccess()
and cacheProbe(), end with calls to panic(). Both of these functions have
return type other than void. Any file that includes this header file fails
to compile because of the missing return statement. This patch adds dummy
values so as to avoid the compiler warnings.

14 years agoThis patch removes the WARN_* and ERROR_* from src/mem/ruby/common/Debug.hh file...
Nilay Vaish [Thu, 23 Dec 2010 05:15:24 +0000 (23:15 -0600)]
This patch removes the WARN_* and ERROR_* from src/mem/ruby/common/Debug.hh file. These statements have been replaced with warn(), panic() and fatal() defined in src/base/misc.hh

14 years agomemtest: delete some crufty dead code
Steve Reinhardt [Wed, 22 Dec 2010 06:57:29 +0000 (22:57 -0800)]
memtest: delete some crufty dead code

14 years agoGet rid of unused file src/base/dbl_list.hh
Steve Reinhardt [Wed, 22 Dec 2010 06:39:26 +0000 (22:39 -0800)]
Get rid of unused file src/base/dbl_list.hh

14 years agostats: allow stats to be reset even if no objects have been instantiated
Nathan Binkert [Tue, 21 Dec 2010 16:02:41 +0000 (08:02 -0800)]
stats: allow stats to be reset even if no objects have been instantiated

14 years agoimporter: fix error message
Nathan Binkert [Tue, 21 Dec 2010 16:02:40 +0000 (08:02 -0800)]
importer: fix error message

14 years agoscons: remove extra dependencies
Nathan Binkert [Tue, 21 Dec 2010 16:02:39 +0000 (08:02 -0800)]
scons: remove extra dependencies

14 years agoStyle: Replace some tabs with spaces.
Gabe Black [Mon, 20 Dec 2010 21:24:40 +0000 (16:24 -0500)]
Style: Replace some tabs with spaces.

14 years agoParams: Fix a broken error message in verifyIp.
Gabe Black [Mon, 20 Dec 2010 09:20:58 +0000 (04:20 -0500)]
Params: Fix a broken error message in verifyIp.

14 years agoARM: Take advantage of new PCState syntax.
Gabe Black [Thu, 9 Dec 2010 22:45:17 +0000 (14:45 -0800)]
ARM: Take advantage of new PCState syntax.

14 years agoARM: Get rid of some unused FP operands.
Gabe Black [Thu, 9 Dec 2010 22:45:04 +0000 (14:45 -0800)]
ARM: Get rid of some unused FP operands.

14 years agoMerge.
Gabe Black [Thu, 9 Dec 2010 00:52:38 +0000 (16:52 -0800)]
Merge.

14 years agoruby: remove Ruby asserts for m5.fast
Brad Beckmann [Wed, 8 Dec 2010 19:52:02 +0000 (11:52 -0800)]
ruby: remove Ruby asserts for m5.fast

This diff is for changing the way ASSERT is handled in Ruby. m5.fast
compiles out the assert statements by using the macro NDEBUG. Ruby uses the
macro RUBY_NO_ASSERT to do so. This macro has been removed and NDEBUG has
been put in its place.

14 years agoAlpha: Take advantage of new PCState syntax.
Gabe Black [Wed, 8 Dec 2010 18:55:33 +0000 (10:55 -0800)]
Alpha: Take advantage of new PCState syntax.

14 years agoMIPS: Take advantage of new PCState syntax.
Gabe Black [Wed, 8 Dec 2010 18:45:14 +0000 (10:45 -0800)]
MIPS: Take advantage of new PCState syntax.

14 years agoPOWER: Take advantage of new PCState syntax.
Gabe Black [Wed, 8 Dec 2010 18:33:03 +0000 (10:33 -0800)]
POWER: Take advantage of new PCState syntax.

14 years agoSPARC: Take advantage of new PCState syntax.
Gabe Black [Wed, 8 Dec 2010 08:27:43 +0000 (00:27 -0800)]
SPARC: Take advantage of new PCState syntax.

14 years agoX86: Take advantage of new PCState syntax.
Gabe Black [Wed, 8 Dec 2010 08:27:23 +0000 (00:27 -0800)]
X86: Take advantage of new PCState syntax.

14 years agoISA: Get the parser to support pc state components more elegantly.
Gabe Black [Wed, 8 Dec 2010 07:08:05 +0000 (23:08 -0800)]
ISA: Get the parser to support pc state components more elegantly.

14 years agoConfigs: Automatically choose the correct hello world binary.
Ali Saidi [Wed, 8 Dec 2010 00:19:57 +0000 (16:19 -0800)]
Configs: Automatically choose the correct hello world binary.

14 years agoO3: Allow a store entry to store up to 16 bytes (instead of TheISA::IntReg).
Ali Saidi [Wed, 8 Dec 2010 00:19:57 +0000 (16:19 -0800)]
O3: Allow a store entry to store up to 16 bytes (instead of TheISA::IntReg).

The store queue doesn't need to be ISA specific and architectures can
frequently store more than an int registers worth of data. A 128 bits seems
more common, but even 256 bits may be appropriate. Pretty much anything less
than a cache line size is buildable.

14 years agoStats: Fix stats for cumulative flags change.
Ali Saidi [Wed, 8 Dec 2010 00:19:57 +0000 (16:19 -0800)]
Stats: Fix stats for cumulative flags change.

14 years agoO3: Support squashing all state after special instruction
Ali Saidi [Wed, 8 Dec 2010 00:19:57 +0000 (16:19 -0800)]
O3: Support squashing all state after special instruction

For SPARC ASIs are added to the ExtMachInst. If the ASI is changed simply
marking the instruction as Serializing isn't enough beacuse that only
stops rename. This provides a mechanism to squash all the instructions
and refetch them

14 years agoO3: Make all instructions that write a misc. register not perform the write until...
Giacomo Gabrielli [Wed, 8 Dec 2010 00:19:57 +0000 (16:19 -0800)]
O3: Make all instructions that write a misc. register not perform the write until commit.

ARM instructions updating cumulative flags (ARM FP exceptions and saturation
flags) are not serialized.

Added aliases for ARM FP exceptions and saturation flags in FPSCR.  Removed
write accesses to the FP condition codes for most ARM VFP instructions: only
VCMP and VCMPE instructions update the FP condition codes.  Removed a potential
cause of seg. faults in the O3 model for NEON memory macro-ops (ARM).

14 years agoO3: Support SWAP and predicated loads/store in ARM.
Min Kyu Jeong [Wed, 8 Dec 2010 00:19:57 +0000 (16:19 -0800)]
O3: Support SWAP and predicated loads/store in ARM.

14 years agoARM: Support switchover with hardware table walkers
Ali Saidi [Wed, 8 Dec 2010 00:19:57 +0000 (16:19 -0800)]
ARM: Support switchover with hardware table walkers

14 years agoruby: Converted old ruby debug calls to M5 debug calls
Nilay Vaish [Wed, 1 Dec 2010 19:30:04 +0000 (11:30 -0800)]
ruby: Converted old ruby debug calls to M5 debug calls

This patch developed by Nilay Vaish converts all the old GEMS-style ruby
debug calls to the appropriate M5 debug calls.

14 years agoIGbE: return 0 on an invalid descriptor size instead of -1.
Ali Saidi [Sat, 27 Nov 2010 01:47:23 +0000 (20:47 -0500)]
IGbE: return 0 on an invalid descriptor size instead of -1.

Asserts where descSize() get called with assert if we end up returning
0.

14 years agoCopyright: Add AMD copyright to the param changes I just made.
Gabe Black [Tue, 23 Nov 2010 22:08:41 +0000 (17:08 -0500)]
Copyright: Add AMD copyright to the param changes I just made.

14 years agoParams: Add parameter types for IP addresses in various forms.
Gabe Black [Tue, 23 Nov 2010 20:54:43 +0000 (15:54 -0500)]
Params: Add parameter types for IP addresses in various forms.

New parameter forms are:
IP address in the format "a.b.c.d" where a-d are from decimal 0 to 255.
IP address with netmask which is an IP followed by "/n" where n is a netmask
length in bits from decimal 0 to 32 or by "/e.f.g.h" where e-h are from
decimal 0 to 255 and which is all 1 bits followed by all 0 bits when
represented in binary. These can also be specified as an integral IP and
netmask passed in separately.
IP address with port which is an IP followed by ":p" where p is a port index
from decimal 0 to 65535. These can also be specified as an integral IP and
port value passed in separately.

14 years agoX86: Loosen an assert for x86 and connect the APIC ports when caches are used.
Gabe Black [Tue, 23 Nov 2010 11:11:50 +0000 (06:11 -0500)]
X86: Loosen an assert for x86 and connect the APIC ports when caches are used.

14 years agoX86: Obey the PCD (cache disable) bit in the page tables.
Gabe Black [Tue, 23 Nov 2010 11:10:17 +0000 (06:10 -0500)]
X86: Obey the PCD (cache disable) bit in the page tables.

14 years agoX86: Mark IO space accesses as uncachable.
Gabe Black [Mon, 22 Nov 2010 10:49:03 +0000 (05:49 -0500)]
X86: Mark IO space accesses as uncachable.

14 years agoX86: Remove reserved* from the m5 utility program for x86.
Gabe Black [Mon, 22 Nov 2010 10:47:54 +0000 (05:47 -0500)]
X86: Remove reserved* from the m5 utility program for x86.

14 years agoIDE,X86: Fix IDE controller BAR configuration for x86.
Gabe Black [Mon, 22 Nov 2010 07:33:47 +0000 (02:33 -0500)]
IDE,X86: Fix IDE controller BAR configuration for x86.

14 years agorandom: small comment about our random number generator and its origin
Nathan Binkert [Sat, 20 Nov 2010 20:12:27 +0000 (12:12 -0800)]
random: small comment about our random number generator and its origin

14 years agoSE: Fix simulating more than 4GB of RAM in SE mode
Ali Saidi [Sat, 20 Nov 2010 00:01:01 +0000 (18:01 -0600)]
SE: Fix simulating more than 4GB of RAM in SE mode

This change removes some dead code in PhysicalMemory, uses a 64 bit type
for the page pointer in System (instead of 32 bit) and cleans up some style.

14 years agoSCons: Fix compilation on OS X
Ali Saidi [Sat, 20 Nov 2010 00:00:59 +0000 (18:00 -0600)]
SCons: Fix compilation on OS X

14 years agoSCons: Support building without an ISA
Ali Saidi [Sat, 20 Nov 2010 00:00:39 +0000 (18:00 -0600)]
SCons: Support building without an ISA

14 years agoO3: Fix fp destination register flattening, and index offset adjusting.
Gabe Black [Thu, 18 Nov 2010 18:11:36 +0000 (13:11 -0500)]
O3: Fix fp destination register flattening, and index offset adjusting.

This change makes O3 flatten floating point destination registers, and also
fixes misc register flattening so that it's correctly repositioned relative to
the resized regions for integer and floating point indices.

It also fixes some overly long lines.

14 years agoConfig: Change misleading "cycle" message to say "tick".
Gabe Black [Thu, 18 Nov 2010 04:16:19 +0000 (23:16 -0500)]
Config: Change misleading "cycle" message to say "tick".

Most of the messages in the config scripts that report a time value already
print "@ tick" followed by the current tick value, but a few were printing
"@ cycle". Since this is a distinction that's frequently confusing to new
users, this changes those message to the more accurate and consistent "@ tick".

14 years agoStats: Update the O3 fetch stats for SPARC.
Gabe Black [Tue, 16 Nov 2010 03:37:15 +0000 (19:37 -0800)]
Stats: Update the O3 fetch stats for SPARC.

14 years agoO3: Make O3 support variably lengthed instructions.
Gabe Black [Tue, 16 Nov 2010 03:37:03 +0000 (19:37 -0800)]
O3: Make O3 support variably lengthed instructions.

14 years agoO3: reset architetural state by calling clear()
Ali Saidi [Mon, 15 Nov 2010 20:04:05 +0000 (14:04 -0600)]
O3: reset architetural state by calling clear()

14 years agoARM: Add comment about the organization of the IT state register
Ali Saidi [Mon, 15 Nov 2010 20:04:05 +0000 (14:04 -0600)]
ARM: Add comment about the organization of the IT state register

14 years agoRegressions: Update regressions for SIMD opclass changes
Ali Saidi [Mon, 15 Nov 2010 20:04:05 +0000 (14:04 -0600)]
Regressions: Update regressions for SIMD opclass changes

14 years agoCPU/ARM: Add SIMD op classes to CPU models and ARM ISA.
Giacomo Gabrielli [Mon, 15 Nov 2010 20:04:04 +0000 (14:04 -0600)]
CPU/ARM: Add SIMD op classes to CPU models and ARM ISA.

14 years agoARM: Compile O3 CPU by default
Ali Saidi [Mon, 15 Nov 2010 20:04:04 +0000 (14:04 -0600)]
ARM: Compile O3 CPU by default

14 years agoO3: prevent a squash when completeAcc() modifies misc reg through TC.
Min Kyu Jeong [Mon, 15 Nov 2010 20:04:04 +0000 (14:04 -0600)]
O3: prevent a squash when completeAcc() modifies misc reg through TC.

This happens on ARM instructions when they update the IT state bits.
Code and associated comment was copied from execute() and initiateAcc() methods

14 years agoARM: Return an FailUnimp instruction when an unimplemented CP15 register is accessed.
Ali Saidi [Mon, 15 Nov 2010 20:04:04 +0000 (14:04 -0600)]
ARM: Return an FailUnimp instruction when an unimplemented CP15 register is accessed.

Just panicing in readMiscReg() doesn't work because a speculative access
in the o3 model can end the simulation.

14 years agoSCons: Cleanup SCons output during compile
Ali Saidi [Mon, 15 Nov 2010 20:04:04 +0000 (14:04 -0600)]
SCons: Cleanup SCons output during compile

14 years agoARM: Update regressions for CLCD and KMI additions
Ali Saidi [Mon, 15 Nov 2010 20:04:03 +0000 (14:04 -0600)]
ARM: Update regressions for CLCD and KMI additions

14 years agoARM: Add a Keyboard Mouse Interface controller
William Wang [Mon, 15 Nov 2010 20:04:03 +0000 (14:04 -0600)]
ARM: Add a Keyboard Mouse Interface controller

14 years agoARM: Implement a CLCD Frame buffer
William Wang [Mon, 15 Nov 2010 20:04:03 +0000 (14:04 -0600)]
ARM: Implement a CLCD Frame buffer

14 years agoARM: Add support for GDB on ARM
William Wang [Mon, 15 Nov 2010 20:04:03 +0000 (14:04 -0600)]
ARM: Add support for GDB on ARM

--HG--
rename : src/arch/alpha/remote_gdb.cc => src/arch/arm/remote_gdb.cc

14 years agoARM: Make utility.hh meet style guidelines
Ali Saidi [Mon, 15 Nov 2010 20:04:03 +0000 (14:04 -0600)]
ARM: Make utility.hh meet style guidelines

14 years agoARM: Add support for a dumb IDE controller
Ali Saidi [Mon, 15 Nov 2010 20:04:03 +0000 (14:04 -0600)]
ARM: Add support for a dumb IDE controller

14 years agoARM: Cache the misc regs at the TLB to limit readMiscReg() calls.
Ali Saidi [Mon, 15 Nov 2010 20:04:03 +0000 (14:04 -0600)]
ARM: Cache the misc regs at the TLB to limit readMiscReg() calls.

14 years agoARM: Add support for switching CPUs
Ali Saidi [Mon, 15 Nov 2010 20:04:03 +0000 (14:04 -0600)]
ARM: Add support for switching CPUs

14 years agoARM: Use the correct delete operator for RFE
Ali Saidi [Mon, 15 Nov 2010 20:04:03 +0000 (14:04 -0600)]
ARM: Use the correct delete operator for RFE

14 years agoARM: Fix SRS instruction to micro-code memory operation and register update.
Ali Saidi [Mon, 15 Nov 2010 20:04:03 +0000 (14:04 -0600)]
ARM: Fix SRS instruction to micro-code memory operation and register update.

Previously the SRS instruction attempted to writeback in initiateAcc() which
worked until a recent change, but was incorrect.

14 years agoCPU: Fix bug when a split transaction is issued to a faster cache
Ali Saidi [Mon, 15 Nov 2010 20:04:03 +0000 (14:04 -0600)]
CPU: Fix bug when a split transaction is issued to a faster cache

In the case of a split transaction and a cache that is faster than a CPU we
could get two responses before next_tick expires. Add an event that is
scheduled in this case and return false rather than asserting.

14 years agoARM: Do something predictable for an UNPREDICTABLE branch.
Ali Saidi [Mon, 15 Nov 2010 20:04:03 +0000 (14:04 -0600)]
ARM: Do something predictable for an UNPREDICTABLE branch.

14 years agoParams: Fix an off by one error and a misleading comment.
Gabe Black [Thu, 11 Nov 2010 19:58:09 +0000 (11:58 -0800)]
Params: Fix an off by one error and a misleading comment.

14 years agoSimObject: Add a comment near clear_child that it's unlikely to be called.
Gabe Black [Thu, 11 Nov 2010 19:41:13 +0000 (11:41 -0800)]
SimObject: Add a comment near clear_child that it's unlikely to be called.

14 years agoSPARC: Clean up some historical style issues.
Gabe Black [Thu, 11 Nov 2010 10:03:58 +0000 (02:03 -0800)]
SPARC: Clean up some historical style issues.

14 years agoUpdate EIO regressions for last set of patches
Ali Saidi [Wed, 10 Nov 2010 06:48:03 +0000 (00:48 -0600)]
Update EIO regressions for last set of patches

14 years agoscons: Work around for old versions of scons mistaking strings for sequences.
Gabe Black [Tue, 9 Nov 2010 19:03:40 +0000 (11:03 -0800)]
scons: Work around for old versions of scons mistaking strings for sequences.

14 years agoSimObject: Use "self" when calling the clear_child method.
Gabe Black [Tue, 9 Nov 2010 18:45:02 +0000 (10:45 -0800)]
SimObject: Use "self" when calling the clear_child method.

14 years agoX86: Fix X86_FS compilation.
Gabe Black [Mon, 8 Nov 2010 20:43:38 +0000 (12:43 -0800)]
X86: Fix X86_FS compilation.

14 years agoARM: Update SE stats for TLB stats additions
Ali Saidi [Mon, 8 Nov 2010 19:59:35 +0000 (13:59 -0600)]
ARM: Update SE stats for TLB stats additions

14 years agoARM: Add full-system regressions
Ali Saidi [Mon, 8 Nov 2010 19:58:25 +0000 (13:58 -0600)]
ARM: Add full-system regressions

14 years agoARM: Add some TLB statistics for ARM
Ali Saidi [Mon, 8 Nov 2010 19:58:25 +0000 (13:58 -0600)]
ARM: Add some TLB statistics for ARM

14 years agoARM: Add checkpointing support
Ali Saidi [Mon, 8 Nov 2010 19:58:25 +0000 (13:58 -0600)]
ARM: Add checkpointing support

14 years agoARM: Add support for M5 ops in the ARM ISA
Ali Saidi [Mon, 8 Nov 2010 19:58:24 +0000 (13:58 -0600)]
ARM: Add support for M5 ops in the ARM ISA

14 years agoARM: Keep the warnings to a minimum.
Ali Saidi [Mon, 8 Nov 2010 19:58:24 +0000 (13:58 -0600)]
ARM: Keep the warnings to a minimum.

These warnings still need to be addresses, but pages of them is
counterproductive.

14 years agoMem: Finish half-baked support for mmaping file in physmem.
Ali Saidi [Mon, 8 Nov 2010 19:58:24 +0000 (13:58 -0600)]
Mem: Finish half-baked support for mmaping file in physmem.

Physmem has a parameter to be able to mem map a file, however
it isn't actually used. This changeset utilizes the parameter
so a file can be mmapped.

14 years agoBus: Have the I/O devices that return address ranges print them out.
Ali Saidi [Mon, 8 Nov 2010 19:58:24 +0000 (13:58 -0600)]
Bus: Have the I/O devices that return address ranges print them out.

This way we actually get device names associated with the devices.

14 years agoARM: Don't return the result of a table walk the same cycle it's completed.
Ali Saidi [Mon, 8 Nov 2010 19:58:24 +0000 (13:58 -0600)]
ARM: Don't return the result of a table walk the same cycle it's completed.

The L1 cache may have been accessed to provide this data, which confuses
it, if it ends up being accesses twice in one cycle. Instead wait 1 tick
which will force the timing simple CPU to forward to its next clock cycle
when the translation completes.

Also prevent multiple outstanding table walks from occuring at once.

14 years agoscons: add a parameter to configure SCons' build cache
Ali Saidi [Mon, 8 Nov 2010 19:58:24 +0000 (13:58 -0600)]
scons: add a parameter to configure SCons' build cache

14 years agoARM/Alpha/Cpu: Stats change for prefetchs to be more like normal loads.
Ali Saidi [Mon, 8 Nov 2010 19:58:24 +0000 (13:58 -0600)]
ARM/Alpha/Cpu: Stats change for prefetchs to be more like normal loads.