Gabe Black [Wed, 10 Jun 2009 06:41:35 +0000 (23:41 -0700)]
ARM: Add a memory_barrier function to the "comm page".
This function doesn't actually provide a memory barrier (I don't think they're
implemented) and instead just returns.
Gabe Black [Wed, 10 Jun 2009 06:41:03 +0000 (23:41 -0700)]
ARM: Add a cmpxchg implementation to the "comm page".
This implementation does what it's supposed to (I think), but it's not atomic
and doesn't have memory barriers like the kernel's version.
Gabe Black [Wed, 10 Jun 2009 06:39:07 +0000 (23:39 -0700)]
ARM: Implement TLS. This is not tested.
Gabe Black [Wed, 10 Jun 2009 06:38:50 +0000 (23:38 -0700)]
ARM: Make ArmLinuxProcess understand "ARM private" system calls.
Gabe Black [Wed, 10 Jun 2009 06:37:41 +0000 (23:37 -0700)]
ARM: Update the kernel version M5 reports to 2.6.16.19
Nathan Binkert [Sat, 6 Jun 2009 00:01:19 +0000 (17:01 -0700)]
cleanup: Make use of types properly and make the loop a little more clear.
Nathan Binkert [Fri, 5 Jun 2009 22:20:09 +0000 (15:20 -0700)]
scons: Make it so that the processing of trace flags does not depend on order
Nathan Binkert [Fri, 5 Jun 2009 18:40:02 +0000 (11:40 -0700)]
types: need typename keyword to get the type.
Nathan Binkert [Fri, 5 Jun 2009 06:21:12 +0000 (23:21 -0700)]
types: clean up types, especially signed vs unsigned
Nathan Binkert [Fri, 5 Jun 2009 04:50:20 +0000 (21:50 -0700)]
move: put predictor includes and cc files into the same place
--HG--
rename : src/cpu/2bit_local_pred.cc => src/cpu/pred/2bit_local.cc
rename : src/cpu/o3/2bit_local_pred.hh => src/cpu/pred/2bit_local.hh
rename : src/cpu/btb.cc => src/cpu/pred/btb.cc
rename : src/cpu/o3/btb.hh => src/cpu/pred/btb.hh
rename : src/cpu/ras.cc => src/cpu/pred/ras.cc
rename : src/cpu/o3/ras.hh => src/cpu/pred/ras.hh
rename : src/cpu/tournament_pred.cc => src/cpu/pred/tournament.cc
rename : src/cpu/o3/tournament_pred.hh => src/cpu/pred/tournament.hh
Nathan Binkert [Fri, 5 Jun 2009 04:41:46 +0000 (21:41 -0700)]
style: cleanup style
Nathan Binkert [Mon, 1 Jun 2009 23:38:57 +0000 (16:38 -0700)]
swig: %include Event before PythonEvent so python gets the subclass correct.
Before this change, some versions of swig would cause PythonEvent to be
derived from object instead of Event
Nathan Binkert [Fri, 29 May 2009 22:30:16 +0000 (15:30 -0700)]
request: add accessor and constructor for setting time other than curTick
Gabe Black [Fri, 29 May 2009 06:27:56 +0000 (23:27 -0700)]
X86: Keep track of more descriptor state to accomodate KVM.
Nathan Binkert [Tue, 26 May 2009 16:23:13 +0000 (09:23 -0700)]
types: add a type for thread IDs and try to use it everywhere
Gabe Black [Tue, 26 May 2009 09:23:08 +0000 (02:23 -0700)]
X86: Really set up the GDT and various hidden/visible segment registers.
Steve Reinhardt [Sat, 23 May 2009 04:24:09 +0000 (21:24 -0700)]
util: mkblankimage.sh should be executable
Korey Sewell [Thu, 21 May 2009 15:04:24 +0000 (11:04 -0400)]
build_opts: update ALPHA_FS cpu models
Steve Reinhardt [Thu, 21 May 2009 04:52:32 +0000 (21:52 -0700)]
igbe: Fix descriptor cache bug.
Nathan Binkert [Sun, 17 May 2009 21:34:52 +0000 (14:34 -0700)]
includes: sort includes again
Nathan Binkert [Sun, 17 May 2009 21:34:51 +0000 (14:34 -0700)]
includes: use base/types.hh not inttypes.h or stdint.h
Nathan Binkert [Sun, 17 May 2009 21:34:50 +0000 (14:34 -0700)]
types: Move stuff for global types into src/base/types.hh
--HG--
rename : src/sim/host.hh => src/base/types.hh
Nathan Binkert [Wed, 13 May 2009 14:18:03 +0000 (07:18 -0700)]
stats: tidy up the Distribution type a little bit
Nathan Binkert [Wed, 13 May 2009 14:18:02 +0000 (07:18 -0700)]
stats: fancy is a bad name
Nathan Binkert [Wed, 13 May 2009 14:18:01 +0000 (07:18 -0700)]
stats: clean up the code for printing stats
Korey Sewell [Wed, 13 May 2009 06:02:05 +0000 (02:02 -0400)]
mips-merge: merge hello world regress for inorder cpu
w/latest changes
Korey Sewell [Wed, 13 May 2009 05:55:04 +0000 (01:55 -0400)]
inorder-regress: add hello MIPS_SE
Nathan Binkert [Wed, 13 May 2009 05:33:05 +0000 (22:33 -0700)]
ruby: deal with printf warnings and convert some to cprintf
Nathan Binkert [Wed, 13 May 2009 05:33:05 +0000 (22:33 -0700)]
ruby: remove random uint typedef and use unsigned
Nathan Binkert [Wed, 13 May 2009 05:33:05 +0000 (22:33 -0700)]
ruby: Make ruby's Map use hashmap.hh to simplify things.
Nathan Binkert [Wed, 13 May 2009 05:33:05 +0000 (22:33 -0700)]
gcc: work around a bogus gcc error
Nathan Binkert [Wed, 13 May 2009 05:33:05 +0000 (22:33 -0700)]
slicc: work around improper initialization of a global in slicc.
Nathan Binkert [Wed, 13 May 2009 05:33:04 +0000 (22:33 -0700)]
slicc: clean up the slicc environment so things build properly on mac.
Korey Sewell [Wed, 13 May 2009 05:26:47 +0000 (01:26 -0400)]
mips_se: add cpu_models to build options
Korey Sewell [Wed, 13 May 2009 05:26:46 +0000 (01:26 -0400)]
inorder-mips: Remove eaComp & memAcc; use 'visible' eaComp
Inorder expects eaComp to be visible through StaticInst object. This mirrors a similar change
to ALPHA... Needs to be done for SPARC and whatever other ISAs want to use InOrderCPU
Korey Sewell [Wed, 13 May 2009 05:26:38 +0000 (01:26 -0400)]
arch-mips: add regWidth constant to float regfile
Korey Sewell [Wed, 13 May 2009 00:55:21 +0000 (20:55 -0400)]
cpus: add InOrderCPU to default build
regressions need this so they build the model
Korey Sewell [Wed, 13 May 2009 00:30:40 +0000 (20:30 -0400)]
inorder-regress: missing regress config file
regressions need to access this file to setup the InOrderCPU object
Korey Sewell [Wed, 13 May 2009 00:18:34 +0000 (20:18 -0400)]
alpha-isa: add mt.hh so it can compile with inorder
Korey Sewell [Tue, 12 May 2009 19:01:17 +0000 (15:01 -0400)]
inorder-regress: add vortex ALPHA_SE
Korey Sewell [Tue, 12 May 2009 19:01:16 +0000 (15:01 -0400)]
inorder-regress: add twolf ALPHA-SE
Korey Sewell [Tue, 12 May 2009 19:01:16 +0000 (15:01 -0400)]
inorder-regress: add hello world
Korey Sewell [Tue, 12 May 2009 19:01:16 +0000 (15:01 -0400)]
inorder-resources: delete events
make sure unrecognized events in the resource pool are deleted and also delete resource events in destructor
Korey Sewell [Tue, 12 May 2009 19:01:16 +0000 (15:01 -0400)]
inorder-tlb-cunit: merge the TLB as implicit to any memory access
TLBUnit no longer used and we also get rid of memAccSize and memAccFlags functions added to ISA and StaticInst
since TLB is not a separate resource to acquire. Instead, TLB access is done before any read/write to memory
and the result is checked before it's sent out to memory.
* * *
Korey Sewell [Tue, 12 May 2009 19:01:16 +0000 (15:01 -0400)]
inorder-tlb: squash insts in TLB correctly
TLB had a bug where if it was stalled and waiting , it would not squash all instructions older than squashed instruction correctly
* * *
Korey Sewell [Tue, 12 May 2009 19:01:16 +0000 (15:01 -0400)]
inorder-faults: ignore unalign translation faults for prefetches
Korey Sewell [Tue, 12 May 2009 19:01:15 +0000 (15:01 -0400)]
inorder-stc: update interface to handle store conditionals
Korey Sewell [Tue, 12 May 2009 19:01:15 +0000 (15:01 -0400)]
inorder-float: Fix storage of FP results
inorder was incorrectly storing FP values and confusing the integer/fp storage view of floating point operations. A big issue was knowing trying to infer when were doing single or double precision access
because this lets you know the size of value to store (32-64 bits). This isnt exactly straightforward since alpha uses all 64-bit regs while mips/sparc uses a dual-reg view. by getting this value from
the actual floating point register file, the model can figure out what it needs to store
Korey Sewell [Tue, 12 May 2009 19:01:15 +0000 (15:01 -0400)]
inorder-fetch: update model to use predecoder
Korey Sewell [Tue, 12 May 2009 19:01:15 +0000 (15:01 -0400)]
inorder-mem: clean up allocation/deletion of requests/packets
* * *
Korey Sewell [Tue, 12 May 2009 19:01:15 +0000 (15:01 -0400)]
inorder-mem: skeleton support for prefetch/writehints
Korey Sewell [Tue, 12 May 2009 19:01:14 +0000 (15:01 -0400)]
inorder-o3: allow both to compile together
allow InOrder and O3CPU to be compiled at the same time: need to make branch prediction filed shared by both models
Korey Sewell [Tue, 12 May 2009 19:01:14 +0000 (15:01 -0400)]
inorder-unified-tlb: use unified TLB instead of old TLB model
Korey Sewell [Tue, 12 May 2009 19:01:14 +0000 (15:01 -0400)]
inorder-miscregs: Fix indexing for misc. reg operands and update result-types for better tracing of these types of values
Korey Sewell [Tue, 12 May 2009 19:01:14 +0000 (15:01 -0400)]
inorder/alpha-isa: create eaComp object visible to StaticInst through ISA
Remove subinstructions eaComp/memAcc since unused in CPU Models. Instead, create eaComp that is visible from StaticInst object. Gives InOrder model capability of generating address without actually initiating access
* * *
Korey Sewell [Tue, 12 May 2009 19:01:14 +0000 (15:01 -0400)]
inorder-bpred: edits to handle non-delay-slot ISAs
Changes so that InOrder can work for a non-delay-slot ISA like Alpha. Typically, changes have to do with handling misspeculated branches at different points in pipeline
Korey Sewell [Tue, 12 May 2009 19:01:13 +0000 (15:01 -0400)]
inorder-alpha-port: initial inorder support of ALPHA
Edit AlphaISA to support the inorder model. Mostly alternate constructor functions and also a few skeleton multithreaded support functions
* * *
Remove namespace from header file. Causes compiler issues that are hard to find
* * *
Separate the TLB from the CPU and allow it to live in the TLBUnit resource. Give CPU accessor functions for access and also bind at construction time
* * *
Expose memory access size and flags through instruction object
(temporarily memAccSize and memFlags to get TLB stuff working.)
Korey Sewell [Tue, 12 May 2009 19:01:13 +0000 (15:01 -0400)]
isa-parser: made a few changes, but not author-worthy
Korey Sewell [Mon, 11 May 2009 23:44:34 +0000 (19:44 -0400)]
Merge Ruby Stuff
Nathan Binkert [Mon, 11 May 2009 23:32:32 +0000 (16:32 -0700)]
ruby: assert(false) should be panic.
This also fixes some compiler warnings
Nathan Binkert [Mon, 11 May 2009 18:18:09 +0000 (11:18 -0700)]
stats: remove a few compat leftovers
Nathan Binkert [Mon, 11 May 2009 18:18:09 +0000 (11:18 -0700)]
python: pull out common code from main that processes arguments
Nathan Binkert [Mon, 11 May 2009 18:18:09 +0000 (11:18 -0700)]
stats: forgot an include for the mysql stuff
Nathan Binkert [Mon, 11 May 2009 18:18:09 +0000 (11:18 -0700)]
scons: add include guards to info.hh
Nathan Binkert [Mon, 11 May 2009 17:38:46 +0000 (10:38 -0700)]
ruby: add RUBY sticky option that must be set to add ruby to the build
Default is false
Steve Reinhardt [Mon, 11 May 2009 17:38:46 +0000 (10:38 -0700)]
ruby: Initial references for ruby regressions
Steve Reinhardt [Mon, 11 May 2009 17:38:46 +0000 (10:38 -0700)]
ruby: Set up Ruby regression tests.
Daniel Sanchez [Mon, 11 May 2009 17:38:46 +0000 (10:38 -0700)]
ruby: Working M5 interface and updated Ruby interface.
This changeset also includes a lot of work from Derek Hower <drh5@cs.wisc.edu>
RubyMemory is now both a driver for Ruby and a port for M5. Changed
makeRequest/hitCallback interface. Brought packets (superficially)
into the sequencer. Modified tester infrastructure to be packet based.
and Ruby can be used together through the example ruby_se.py
script. SPARC parallel applications work, and the timing *seems* right
from combined M5/Ruby debug traces. To run,
% build/ALPHA_SE/m5.debug configs/example/ruby_se.py -c
tests/test-progs/hello/bin/alpha/linux/hello -n 4 -t
Steve Reinhardt [Mon, 11 May 2009 17:38:46 +0000 (10:38 -0700)]
ruby: Check stderr and not stdin before hanging on an assert.
Polina Dudnik [Mon, 11 May 2009 17:38:46 +0000 (10:38 -0700)]
ruby: decommission code
1. Set.* and BigSet.* are replaced with OptBigSet.* which was renamed Set.*
2. Decomissioned all bloom filters
3. Decomissioned ruby/simics directory
Derek Hower [Mon, 11 May 2009 17:38:46 +0000 (10:38 -0700)]
ruby: removed dead functions from the sequencer
Polina Dudnik [Mon, 11 May 2009 17:38:46 +0000 (10:38 -0700)]
ruby: Removed g_SIMULATING flag
1. removed checks from tester files
2. removed else clause in Sequencer and DirectoryMemory else clause is
needed by the tester, it is up to Derek to revive it elsewhere when he
gets to it
Also:
1. Changed m_entries in DirectoryMemory to a map
2. And replaced SIMICS_read_physical_memory with a call to now-dummy
Derek's-to-be readPhysMem function
Polina Dudnik [Mon, 11 May 2009 17:38:46 +0000 (10:38 -0700)]
ruby: Remove transactional access types (e.g. LD_XACT) from CacheRequestType
1. Modified enumeration
2. Also modified profiler
3. Remove transactions from Tester
4. Edited XACT_MEM out of Synthetic Driver
Polina Dudnik [Mon, 11 May 2009 17:38:46 +0000 (10:38 -0700)]
ruby: reordered Debug and RubyConfig::init to fix segfault
due to uninitialized output file pointer.
Dan Gibson [Mon, 11 May 2009 17:38:46 +0000 (10:38 -0700)]
ruby: Disabled RubyEventQueue's deletion of its home-grown priority heap.
Temporarily to fix unusual memory problem.
Nathan Binkert [Mon, 11 May 2009 17:38:46 +0000 (10:38 -0700)]
ruby: Migrate all of ruby and slicc to SCons.
Add the PROTOCOL sticky option sets the coherence protocol that slicc
will parse and therefore ruby will use. This whole process was made
difficult by the fact that the set of files that are output by slicc
are not easily known ahead of time. The easiest thing wound up being
to write a parser for slicc that would tell me. Incidentally this
means we now have a slicc grammar written in python.
Nathan Binkert [Mon, 11 May 2009 17:38:45 +0000 (10:38 -0700)]
ruby: clean up a few warnings
Dan Gibson [Mon, 11 May 2009 17:38:45 +0000 (10:38 -0700)]
ruby: Fixed some unresolved references.
Nathan Binkert [Mon, 11 May 2009 17:38:45 +0000 (10:38 -0700)]
ruby: Make ruby #includes use full paths to the files they're including.
This basically means changing all #include statements and changing
autogenerated code so that it generates the correct paths. Because
slicc generates #includes, I had to hard code the include paths to
mem/protocol.
Dan Gibson [Mon, 11 May 2009 17:38:45 +0000 (10:38 -0700)]
ruby: remove unnecessary code.
1) Removing files from the ruby build left some unresovled
symbols. Those have been fixed.
2) Most of the dependencies on Simics data types and the simics
interface files have been removed.
3) Almost all mention of opal is gone.
4) Huge chunks of LogTM are now gone.
5) Handling 1-4 left ~hundreds of unresolved references, which were
fixed, yielding a snowball effect (and the massive size of this
delta).
Derek Hower [Mon, 11 May 2009 17:38:45 +0000 (10:38 -0700)]
ruby: Cleaned up sequencer. Removed LogTM specific code.
Derek Hower [Mon, 11 May 2009 17:38:45 +0000 (10:38 -0700)]
ruby: added Packet interface to makeRequest and isReady.
Also pushed Packet usage into the Sequencer
Nathan Binkert [Mon, 11 May 2009 17:38:45 +0000 (10:38 -0700)]
ruby: fold the debugging options into Debug.cc
Derek Hower [Mon, 11 May 2009 17:38:45 +0000 (10:38 -0700)]
ruby: Renamed Ruby's EventQueue to RubyEventQueue
--HG--
rename : src/mem/ruby/eventqueue/EventQueue.cc => src/mem/ruby/eventqueue/RubyEventQueue.cc
rename : src/mem/ruby/eventqueue/EventQueue.hh => src/mem/ruby/eventqueue/RubyEventQueue.hh
rename : src/mem/ruby/eventqueue/EventQueueNode.cc => src/mem/ruby/eventqueue/RubyEventQueueNode.cc
rename : src/mem/ruby/eventqueue/EventQueueNode.hh => src/mem/ruby/eventqueue/RubyEventQueueNode.hh
Daniel Sanchez [Mon, 11 May 2009 17:38:44 +0000 (10:38 -0700)]
ruby: Removed System name clash by renaming ruby's System to RubySystem
Nathan Binkert [Mon, 11 May 2009 17:38:44 +0000 (10:38 -0700)]
ruby: rename config.include to config.hh and clean up the macro stuff.
I did the macro cleanup because I was worried that the SCons scanner
would get confused. This code will hopefully go away soon anyway.
--HG--
rename : src/mem/ruby/config/config.include => src/mem/ruby/config/config.hh
Nathan Binkert [Mon, 11 May 2009 17:38:44 +0000 (10:38 -0700)]
ruby: strip out some unused defines
Nathan Binkert [Mon, 11 May 2009 17:38:43 +0000 (10:38 -0700)]
ruby: Import ruby and slicc from GEMS
We eventually plan to replace the m5 cache hierarchy with the GEMS
hierarchy, but for now we will make both live alongside eachother.
Korey Sewell [Tue, 5 May 2009 07:01:57 +0000 (03:01 -0400)]
branch merge
Korey Sewell [Tue, 5 May 2009 06:51:31 +0000 (02:51 -0400)]
cpus: fix cpu progress event
this was double scheduling itself (once in constructor and once in cpu code). also add support for stopping / starting
progress events through repeatEvent flag and also changing the interval of the progress event as well
Korey Sewell [Tue, 5 May 2009 06:44:21 +0000 (02:44 -0400)]
merge code
Korey Sewell [Tue, 5 May 2009 06:39:05 +0000 (02:39 -0400)]
cpus: fix cpu progress event
this was double scheduling itself (once in constructor and once in cpu code). also add support for stopping / starting
progress events through repeatEvent flag and also changing the interval of the progress event as well
Nathan Binkert [Mon, 4 May 2009 23:58:24 +0000 (16:58 -0700)]
scons: re-work the *Source functions to take more information.
Start by turning all of the *Source functions into classes
so we can do more calculations and more easily collect the data we need.
Add parameters to the new classes for indicating what sorts of flags the
objects should be compiled with so we can allow certain files to be compiled
without Werror for example.
Gabe Black [Sun, 26 Apr 2009 23:49:24 +0000 (16:49 -0700)]
X86: Precompute the default and alternate address and operand size and the stack size.
Gabe Black [Sun, 26 Apr 2009 23:48:44 +0000 (16:48 -0700)]
X86: Split out the internal memory space from the regular translate() and precompute mode.
Gabe Black [Sun, 26 Apr 2009 23:47:48 +0000 (16:47 -0700)]
X86: Centralize updates to the handy M5 reg.
Gabe Black [Sun, 26 Apr 2009 09:09:54 +0000 (02:09 -0700)]
X86: Implement lowest priority interrupts more correctly.
Lowest priority interrupts are now delivered based on a rotating offset into
the list of potential recipients. There could be parasitic cases were a
processor gets picked on and ends up at that rotating offset all the time, but
it's much more likely that the group will stay consistent and the pain will be
distributed evenly.
Gabe Black [Sun, 26 Apr 2009 09:09:27 +0000 (02:09 -0700)]
X86: Tell the function that sends int messages who to send to instead of figuring it out itself.
Gabe Black [Sun, 26 Apr 2009 09:09:13 +0000 (02:09 -0700)]
X86: Make the local APICs register themselves with the IO APIC.
This is a hack so that the IO APIC can figure out information about the local
APICs. The local APICs still have no way to find out about each other.
Ideally, when the local APICs update state that's relevant to somebody else,
they'd send an update to everyone. Without being able to do a broadcast, that
would still require knowing who else there is to notify. Other broadcasts are
implemented using assumptions that may not always be true.
Gabe Black [Sun, 26 Apr 2009 09:06:21 +0000 (02:06 -0700)]
X86: Record the initial APIC ID which identifies an APIC in M5.
The ID as exposed to software can be changed. Tracking those changes in M5
would be cumbersome, especially since there's no guarantee the IDs will remain
unique.