Korey Sewell [Mon, 20 Jun 2011 01:43:42 +0000 (21:43 -0400)]
inorder: se compile fixes
Korey Sewell [Mon, 20 Jun 2011 01:43:41 +0000 (21:43 -0400)]
inorder: add necessary debug flag header files
Korey Sewell [Mon, 20 Jun 2011 01:43:41 +0000 (21:43 -0400)]
mips: mark unaligned access flag as true
Korey Sewell [Mon, 20 Jun 2011 01:43:41 +0000 (21:43 -0400)]
inorder: clear fetchbuffer on traps
implement clearfetchbufferfunction
extend predecoder to use multiple threads and clear those on trap
Korey Sewell [Mon, 20 Jun 2011 01:43:41 +0000 (21:43 -0400)]
inorder: use separate float-reg bits function in dyninst
this will make sure we get the correct view of a FP register
Korey Sewell [Mon, 20 Jun 2011 01:43:41 +0000 (21:43 -0400)]
inorder: use trapPending flag to manage traps
Korey Sewell [Mon, 20 Jun 2011 01:43:41 +0000 (21:43 -0400)]
inorder/dtb: make sure DTB translate correct address
The DTB expects the correct PC in the ThreadContext
but how if the memory accesses are speculative? Shouldn't
we send along the requestor's PC to the translate functions?
Korey Sewell [Mon, 20 Jun 2011 01:43:41 +0000 (21:43 -0400)]
inorder: handle serializing instructions
including IPR accesses and store-conditionals. These class of instructions will not
execute correctly in a superscalar machine
Korey Sewell [Mon, 20 Jun 2011 01:43:40 +0000 (21:43 -0400)]
alpha: fix warn_once for prefetches
Korey Sewell [Mon, 20 Jun 2011 01:43:40 +0000 (21:43 -0400)]
alpha: naming for dtb faults
Just "dfault" gets confusing while debugging. Why not
differentiate whether it's an access violation or page
fault
Korey Sewell [Mon, 20 Jun 2011 01:43:40 +0000 (21:43 -0400)]
inorder: dont handle multiple faults on same cycle
if a faulting instruction reaches an execution unit,
then ignore it and pass it through the pipeline.
Once we recognize the fault in the graduation unit,
dont allow a second fault to creep in on the same cycle.
Korey Sewell [Mon, 20 Jun 2011 01:43:40 +0000 (21:43 -0400)]
inorder: register ports for FS mode
handle "snoop" port registration as well as functional
port setup for FS mode
Korey Sewell [Mon, 20 Jun 2011 01:43:40 +0000 (21:43 -0400)]
inorder: check for interrupts each tick
use a dummy instruction to facilitate the squash after
the interrupts trap
Korey Sewell [Mon, 20 Jun 2011 01:43:40 +0000 (21:43 -0400)]
inorder: explicit fault check
Before graduating an instruction, explicitly check fault
by making the fault check it's own separate command
that can be put on an instruction schedule.
Korey Sewell [Mon, 20 Jun 2011 01:43:39 +0000 (21:43 -0400)]
inorder: squash and trap behind a tlb fault
Korey Sewell [Mon, 20 Jun 2011 01:43:39 +0000 (21:43 -0400)]
inorder: stall stores on store conditionals & compare/swaps
Korey Sewell [Mon, 20 Jun 2011 01:43:39 +0000 (21:43 -0400)]
alpha: make hwrei a control inst
this always changes the PC and is basically an impromptu branch instruction. why
not speculate on this instead of always be forced to mispredict/squash after the
hwrei gets resolved?
The InOrder model needs this marked as "isControl" so it knows to update the PC
after the ALU executes it. If this isnt marked as control, then it's going to
force the model to check the PC of every instruction at commit (what O3 does?),
and that would be a wasteful check for a very high percentage of instructions.
Korey Sewell [Mon, 20 Jun 2011 01:43:39 +0000 (21:43 -0400)]
inorder: make InOrder CPU FS compilable/visible
make syscall a SE mode only functionality
copy over basic FS functions (hwrei) to make FS compile
Korey Sewell [Mon, 20 Jun 2011 01:43:39 +0000 (21:43 -0400)]
inorder: remove memdep tracking for default pipeline
speculative load/store pipelines can reenable this
Korey Sewell [Mon, 20 Jun 2011 01:43:39 +0000 (21:43 -0400)]
inorder: fetchBuffer tracking
calculate blocks in use for the fetch buffer to figure out how many total blocks
are pending
Korey Sewell [Mon, 20 Jun 2011 01:43:38 +0000 (21:43 -0400)]
inorder: redefine DynInst FP result type
Sharing the FP value w/the integer values was giving inconsistent results esp. when
their is a 32-bit integer register matched w/a 64-bit float value
Korey Sewell [Mon, 20 Jun 2011 01:43:38 +0000 (21:43 -0400)]
inorder: treat SE mode syscalls as a trapping instruction
define a syscallContext to schedule the syscall and then use syscall() to actually perform the action
Korey Sewell [Mon, 20 Jun 2011 01:43:38 +0000 (21:43 -0400)]
inorder: bug in mdu
segfault was caused by squashed multiply thats in the process of an event.
use isProcessing flag to handle this and cleanup the MDU code
Korey Sewell [Mon, 20 Jun 2011 01:43:38 +0000 (21:43 -0400)]
inorder: optionally track faulting instructions
Korey Sewell [Mon, 20 Jun 2011 01:43:38 +0000 (21:43 -0400)]
inorder: cleanup events in resource pool
remove events in the resource pool that can be called from the CPU event, since the CPU
event is scheduled at the same time at the resource pool event.
----
Also, match the resPool event function names to the cpu event function names
----
Korey Sewell [Mon, 20 Jun 2011 01:43:38 +0000 (21:43 -0400)]
inorder: don't stall after stores
once a ST is sent off, it's OK to keep processing, however it's a little more
complicated to handle the packet acknowledging the store is completed
Korey Sewell [Mon, 20 Jun 2011 01:43:37 +0000 (21:43 -0400)]
inorder: don't stall after stores
once a ST is sent off, it's OK to keep processing, however it's a little more
complicated to handle the packet acknowledging the store is completed
Korey Sewell [Mon, 20 Jun 2011 01:43:37 +0000 (21:43 -0400)]
inorder: remove decode squash
also, cleanup comments for gem5.fast compilation
Korey Sewell [Mon, 20 Jun 2011 01:43:37 +0000 (21:43 -0400)]
inorder: support for compare and swap insts
dont treat read() and write() fields as mut. exclusive
Korey Sewell [Mon, 20 Jun 2011 01:43:37 +0000 (21:43 -0400)]
inorder: branch predictor update
only update BTB on a taken branch and update branch predictor w/pcstate from instruction
---
only pay attention to branch predictor updates if the the inst. is in fact a branch
Korey Sewell [Mon, 20 Jun 2011 01:43:37 +0000 (21:43 -0400)]
inorder: priority for grad/squash events
define separate priority resource pool squash and graduate events
Korey Sewell [Mon, 20 Jun 2011 01:43:37 +0000 (21:43 -0400)]
inorder: remove stalls on trap squash
Korey Sewell [Mon, 20 Jun 2011 01:43:37 +0000 (21:43 -0400)]
inorder: no dep. tracking for zero reg
this causes forwarding a bad value register value
Korey Sewell [Mon, 20 Jun 2011 01:43:37 +0000 (21:43 -0400)]
imported patch recoverPCfromTrap
Korey Sewell [Mon, 20 Jun 2011 01:43:36 +0000 (21:43 -0400)]
imported patch squash_from_next_stage
Korey Sewell [Mon, 20 Jun 2011 01:43:36 +0000 (21:43 -0400)]
inorder: add flatDestReg member to dyninst
use it in reg. dep. tracking
Korey Sewell [Mon, 20 Jun 2011 01:43:36 +0000 (21:43 -0400)]
inorder: update event priorities
dont use offset to calculate this but rather an enum
that can be updated
Korey Sewell [Mon, 20 Jun 2011 01:43:36 +0000 (21:43 -0400)]
inorder: implement trap handling
Korey Sewell [Mon, 20 Jun 2011 01:43:35 +0000 (21:43 -0400)]
inorder: cleanup intercomm. structs/squash info
Korey Sewell [Mon, 20 Jun 2011 01:43:35 +0000 (21:43 -0400)]
inorder: use setupSquash for misspeculation
implement a clean interface to handle branch misprediction and eventually all pipeline
flushing
Korey Sewell [Mon, 20 Jun 2011 01:43:35 +0000 (21:43 -0400)]
sparc: init. cache state in TLB
valgrind complains and its a potential source of instability, so go ahead
and set it to 0 to start
Korey Sewell [Mon, 20 Jun 2011 01:43:35 +0000 (21:43 -0400)]
inorder: DynInst handling of stores for big-endian ISAs
The DynInst was not performing the host-to-guest translation
which ended up breaking stores for SPARC
Korey Sewell [Mon, 20 Jun 2011 01:43:35 +0000 (21:43 -0400)]
inorder: make marking of dest. regs an explicit request
formerly, this was implicit when you accessed the execution unit
or the use-def unit but it's better that this just be something
that a user can specify.
Korey Sewell [Mon, 20 Jun 2011 01:43:35 +0000 (21:43 -0400)]
inorder: simplify handling of split accesses
Korey Sewell [Mon, 20 Jun 2011 01:43:35 +0000 (21:43 -0400)]
inorder: addtl functionaly for inst. skeds
add find and end functions for inst. schedules
that can search by stage number
Korey Sewell [Mon, 20 Jun 2011 01:43:34 +0000 (21:43 -0400)]
inorder: register file stats
keep stats for int/float reg file usage instead
of aggregating across reg file types
Korey Sewell [Mon, 20 Jun 2011 01:43:34 +0000 (21:43 -0400)]
inorder: scheduling for nonspec insts
make handling of speculative and nonspeculative insts
more explicit
Korey Sewell [Mon, 20 Jun 2011 01:43:34 +0000 (21:43 -0400)]
inorder: find register dependencies "lazily"
Architectures like SPARC need to read the window pointer
in order to figure out it's register dependence. However,
this may not get updated until after an instruction gets
executed, so now we lazily detect the register dependence
in the EXE stage (execution unit or use_def). This
makes sure we get the mapping after the most current change.
Korey Sewell [Mon, 20 Jun 2011 01:43:34 +0000 (21:43 -0400)]
inorder: assert on macro-ops
provide a sanity check for someone coding
a new architecture
Korey Sewell [Mon, 20 Jun 2011 01:43:34 +0000 (21:43 -0400)]
inorder: handle faults at writeback stage
call trap function when a fault is received
Korey Sewell [Mon, 20 Jun 2011 01:43:34 +0000 (21:43 -0400)]
inorder: ISA-zero reg handling
ignore writes to the ISA zero register
Korey Sewell [Mon, 20 Jun 2011 01:43:34 +0000 (21:43 -0400)]
inorder: update support for branch delay slots
Korey Sewell [Mon, 20 Jun 2011 01:43:34 +0000 (21:43 -0400)]
inorder: inst. iterator cleanup
get rid of accessing iterators (for instructions) by reference
Korey Sewell [Mon, 20 Jun 2011 01:43:33 +0000 (21:43 -0400)]
cpus/isa: add a != operator for pcstate
Korey Sewell [Mon, 20 Jun 2011 01:43:33 +0000 (21:43 -0400)]
inorder: update bpred code
clean up control flow to make it easier to understand
Korey Sewell [Mon, 20 Jun 2011 01:43:33 +0000 (21:43 -0400)]
inorder: add types for dependency checks
Korey Sewell [Mon, 20 Jun 2011 01:43:33 +0000 (21:43 -0400)]
inorder: use flattenIdx for reg indexing
- also use "threadId()" instead of readTid() everywhere
- this will help support more complex ISA indexing
Korey Sewell [Mon, 20 Jun 2011 01:43:33 +0000 (21:43 -0400)]
simple-thread: give a name() function for debugging w/the SimpleThread object
Korey Sewell [Mon, 20 Jun 2011 01:43:33 +0000 (21:43 -0400)]
inorder: use m5_hash_map for skedCache
since we dont care about if the cache of instruction schedules is sorted or not,
then the hash map should be faster
Ali Saidi [Fri, 17 Jun 2011 17:20:11 +0000 (12:20 -0500)]
ARM: Cleanup m5ops usage of r0 and r1 a bit.
Gedare Bloom [Fri, 17 Jun 2011 17:20:10 +0000 (12:20 -0500)]
ARM: Add m5ops and related support for workbegin() and workend() to ARM ISA.
Ali Saidi [Thu, 16 Jun 2011 20:08:12 +0000 (15:08 -0500)]
ARM: Handle case where new TLB size is different from previous TLB size.
After a checkpoint we need to make sure that we restore the right
number of entries.
Chander Sudanthi [Thu, 16 Jun 2011 20:08:11 +0000 (15:08 -0500)]
ARM: Fix memset on TLB flush and initialization
Instead of clearing the entire TLB on initialization and flush, the code was
clearing only one element. This patch corrects the memsets in the init and
flush routines.
Nilay Vaish [Wed, 15 Jun 2011 00:51:44 +0000 (19:51 -0500)]
Ruby: Correct set LONG_BITS and INDEX_SHIFT in class Set.
The code for Set class was written under the assumption that
std::numeric_limits<long>::digits returns the number of bits used for
data type long, which was presumed to be either 32 or 64. But return value
is actually one less, that is, it is either 31 or 63. The value is now
being incremented by 1 so as to correctly set it.
Gabe Black [Mon, 13 Jun 2011 06:52:21 +0000 (23:52 -0700)]
Loader: Handle bad section names when loading an ELF file.
If there's a problem when reading the section names from a supposed ELF file,
this change makes gem5 print an error message as returned by libelf and die.
Previously these sorts of errors would make gem5 segfault when it tried to
access the section name through a NULL pointer.
Gabe Black [Mon, 13 Jun 2011 06:51:59 +0000 (23:51 -0700)]
LibElf: Build the error management code in libelf.
This change makes some minor changes to get the error management code in
libelf to build on Linux and to build it into the library.
Korey Sewell [Mon, 13 Jun 2011 01:35:03 +0000 (21:35 -0400)]
sparc: update long regressions
Korey Sewell [Sat, 11 Jun 2011 02:15:34 +0000 (22:15 -0400)]
sparc: update o3 regressions
Korey Sewell [Sat, 11 Jun 2011 02:15:32 +0000 (22:15 -0400)]
o3: missing newlines on some dprintfs
Korey Sewell [Sat, 11 Jun 2011 02:15:32 +0000 (22:15 -0400)]
sparc: don't use directcntrl branch flag
this flag is only used for early branch resolution in the O3 model (of pc-relative branches)
but this isnt cleanly working even when the branch target code is added for sparc. For now,
we'll ignore this optimization and add a todo in the SPARC ISA for future developers
Korey Sewell [Fri, 10 Jun 2011 07:49:23 +0000 (03:49 -0400)]
sparc: merge regr. updates w/last update
Korey Sewell [Fri, 10 Jun 2011 07:45:24 +0000 (03:45 -0400)]
sparc: update simple cpu regressions
use stats file generated by zizzer
Korey Sewell [Thu, 9 Jun 2011 05:34:06 +0000 (01:34 -0400)]
sparc: compilation fixes for inorder
Add a few constants and functions that the InOrder model wants for SPARC.
* * *
sparc: add eaComp function
InOrder separates the address generation from the actual access so give
Sparc that functionality
* * *
sparc: add control flags for branches
branch predictors and other cpu model functions need to know specific information
about branches, so add the necessary flags here
Nilay Vaish [Wed, 8 Jun 2011 16:58:09 +0000 (11:58 -0500)]
Ruby: Correctly set access permissions for directory entries
The access permissions for the directory entries are not being set correctly.
This is because pointers are not used for handling directory entries.
function. get and set functions for access permissions have been added to the
Controller state machine. The changePermission() function provided by the
AbstractEntry and AbstractCacheEntry classes has been exposed to SLICC
code once again. The set_permission() functionality has been removed.
NOTE: Each protocol will have to define these get and set functions in order
to compile successfully.
Gabe Black [Wed, 8 Jun 2011 07:57:50 +0000 (00:57 -0700)]
Mem: Use sysconf to get the page size instead of the PAGE_SIZE macro.
Gabe Black [Tue, 7 Jun 2011 07:46:54 +0000 (00:46 -0700)]
ISA parser: Loosen the regular expressions matching filenames.
The regular expressions matching filenames in the ##include directives and the
internally generated ##newfile directives where only looking for filenames
composed of alpha numeric characters, periods, and dashes. In Unix/Linux, the
rules for what characters can be in a filename are much looser than that. This
change replaces those expressions with ones that look for anything other than
a quote character. Technically quote characters are allowed as well so we
should allow escaping them somehow, but the additional complexity probably
isn't worth it.
Gabe Black [Tue, 7 Jun 2011 07:24:49 +0000 (00:24 -0700)]
gcc 4.0: Add some virtual destructors to make gcc 4.0 happy.
Nilay Vaish [Fri, 3 Jun 2011 18:52:18 +0000 (13:52 -0500)]
SLICC: Remove machine name as prefix to functions
Currently, the machine name is appended before any of the functions
defined with in the sm files. This is not necessary and it also
means that these functions cannot be used outside the sm files.
This patch does away with the prefixes. Note that the generated
C++ files in which the code for these functions is present are
still named such that the machine name is the prefix.
Steve Reinhardt [Fri, 3 Jun 2011 04:23:02 +0000 (21:23 -0700)]
SConstruct: automatically update .hg/hgrc with style hooks.
Seems easier than pestering people about it.
Note also that path is now absolute, so you don't get errors
when invoking hg from subdirectories.
Also whacked unused mercurial_bin_not_found message (the
code that used this was deleted a couple months ago in
rev
5138d1e453f1).
Nathan Binkert [Fri, 3 Jun 2011 00:36:21 +0000 (17:36 -0700)]
scons: rename TraceFlags to DebugFlags
Nathan Binkert [Fri, 3 Jun 2011 00:36:18 +0000 (17:36 -0700)]
scons: rename some things from m5 to gem5
The default generated binary is now gem5.<type> instead of m5.<type>.
The latter does still work but gem5.<type> will be generated first and
then m5.<type> will be hard linked to it.
Nathan Binkert [Fri, 3 Jun 2011 00:36:07 +0000 (17:36 -0700)]
copyright: Add code for finding all copyright blocks and create a COPYING file
The end of the COPYING file was generated with:
% python ./util/find_copyrights.py configs src system tests util
Update -C command line option to spit out COPYING file
Nathan Binkert [Thu, 2 Jun 2011 21:36:35 +0000 (14:36 -0700)]
copyright: clean up copyright blocks
Steve Reinhardt [Thu, 2 Jun 2011 04:43:13 +0000 (21:43 -0700)]
SimObject: allow modules in subclass definitions
In particular, this avoids crashing when you do
an import (like "import pdb") inside a SimObject
subclass definition.
Tushar Krishna [Tue, 31 May 2011 06:56:22 +0000 (02:56 -0400)]
orion: bug fix in link power, and some reorg
Tushar Krishna [Tue, 31 May 2011 06:55:14 +0000 (02:55 -0400)]
garnet: added network ptr to links to be used by orion
Gabe Black [Mon, 30 May 2011 04:48:58 +0000 (21:48 -0700)]
Misc: Remove the URL from warnings, fatals, panics, etc.
Gabe Black [Wed, 25 May 2011 08:32:07 +0000 (01:32 -0700)]
Name: Replace M5 with gem5 in a few places it's printed on startup.
Nathan Binkert [Wed, 25 May 2011 04:19:31 +0000 (21:19 -0700)]
style: Make the style hook work in directories other than the root.
Steve Reinhardt [Mon, 23 May 2011 21:29:23 +0000 (14:29 -0700)]
sim: style fixes in sim/process.hh
Steve Reinhardt [Mon, 23 May 2011 21:29:23 +0000 (14:29 -0700)]
syscall emul: fix Power Linux mmap constant, plus other cleanup
We were getting a spurious warning in the regressions that turned
out to be due to having the wrong value for TGT_MAP_ANONYMOUS for
Power Linux, but in the process of tracking it down I ended up
doing some cleanup of the mmap handling in general.
Steve Reinhardt [Mon, 23 May 2011 21:29:23 +0000 (14:29 -0700)]
config: revamp x86 config to avoid appending to SimObjectVectors
A significant contributor to the need for adoptOrphanParams()
is the practice of appending to SimObjectVectors which have
already been assigned as children. This practice sidesteps the
assignment operation for those appended SimObjects, which is
where parent/child relationships are typically established.
This patch reworks the config scripts that use append() on
SimObjectVectors, which all happen to be in the x86 system
configuration. At some point in the future, I hope to make
SimObjectVectors immutable (by deriving from tuple rather than
list), at which time this patch will be necessary for correct
operation. For now, it just avoids some of the warning
messages that get printed in adoptOrphanParams().
Steve Reinhardt [Mon, 23 May 2011 21:29:23 +0000 (14:29 -0700)]
config: tweak ruby configs to clean up hierarchy
Re-enabling implicit parenting (see previous patch) causes current
Ruby config scripts to create some strange hierarchies and generate
several warnings. This patch makes three general changes to address
these issues.
1. The order of object creation in the ruby config files makes the L1
caches children of the sequencer rather than the controller; these
config ciles are rewritten to assign the L1 caches to the
controller first.
2. The assignment of the sequencer list to system.ruby.cpu_ruby_ports
causes the sequencers to be children of system.ruby, generating
warnings because they are already parented to their respective
controllers. Changing this attribute to _cpu_ruby_ports fixes this
because the leading underscore means this is now treated as a plain
Python attribute rather than a child assignment. As a result, the
configuration hierarchy changes such that, e.g.,
system.ruby.cpu_ruby_ports0 becomes system.l1_cntrl0.sequencer.
3. In the topology classes, the routers become children of some random
internal link node rather than direct children of the topology.
The topology classes are rewritten to assign the routers to the
topology object first.
Steve Reinhardt [Mon, 23 May 2011 21:29:08 +0000 (14:29 -0700)]
config: reinstate implicit parenting on parameter assignment
Last summer's big rewrite of the initialization code (in
particular cset
6efc3672733b) got rid of the implicit parenting
that used to occur when an unparented SimObject was assigned as
a parameter value to another SimObject. The idea was that the
new adoptOrphanParams() step would catch these anyway so it was
unnecessary.
Unfortunately it turns out that adoptOrphanParams() has some
inherent instability in that the parent that does the adoption
depends on the config tree traversal order. Even making this
order deterministic (e.g., by traversing children in
alphabetical order) can introduce unwanted and unexpected
hierarchy changes between similar configs (e.g., when adding a
switch_cpu in place of a cpu), causing problems when trying to
restore checkpoints across similar configs. The hierarchy
created by implicit parenting is more stable and more
controllable, so this patch turns that behavior back on.
This patch also cleans up some long-standing holes regarding
parenting of SimObjects that are created in class definitions
(either in the body of the class, or as default parameters).
To avoid breaking some existing config files, this necessitated
changing the error on reparenting children to a warning. This
change fixes another bug where attempting to print the prior
error message would fail on reparenting SimObjectVectors
because they lack a _parent attribute. Some further issues
with SimObjectVectors were cleaned up by getting rid of the
get_parent() call (which could cause errors with some
SimObjectVectors where there was no single parent to return)
with has_parent() (since all the uses of get_parent() were just
boolean tests anyway).
Finally, since the adoptOrphanParam() step turned out to be so
problematic, we now issue a warning when it actually has to do
an adoption. Future cleanup of config files will get rid of
current warnings.
Steve Reinhardt [Mon, 23 May 2011 21:27:20 +0000 (14:27 -0700)]
sim: add some DPRINTFs for debugging unserialization
Also got rid of unused C++ unserializeAll() method
(this is now handled in Python)
Steve Reinhardt [Mon, 23 May 2011 21:27:20 +0000 (14:27 -0700)]
util/regress: make default action a more thorough regression
Changed the --variants option to --test-variants and added a new
--compile-variants option for variants that are only compiled
(not tested). The former still defaults to 'opt' and the latter
defaults to 'debug,fast'.
Also changed the behavior when no tests are specified from just
compiling to running the 'quick' tests.
As a result, a plain 'util/regress' invocation will now compile
(but not test) the debug and fast builds, and compile and run the
quick regressions on the opt build. This should be the default
set of tests that are run before committing. Since the nightly
regressions use this same script, this will also be the new
nightly regression behavior.
Test-only regressions can still be done by setting --compile=''.
Compile-only regressions can be done by setting --test=''.
Korey Sewell [Mon, 23 May 2011 18:36:22 +0000 (14:36 -0400)]
configs: missed spot progress-interval change
Ali Saidi [Mon, 23 May 2011 15:59:13 +0000 (10:59 -0500)]
Stats: Update stats for minor O3 changes below.
Geoffrey Blake [Mon, 23 May 2011 15:40:21 +0000 (10:40 -0500)]
O3: Fix offset calculation into storeQueue buffer for store->load forwarding
Calculation of offset to copy from storeQueue[idx].data structure for load to
store forwarding fixed to be difference in bytes between store and load virtual
addresses. Previous method would induce bug where a load would index into
buffer at the wrong location.
Geoffrey Blake [Mon, 23 May 2011 15:40:19 +0000 (10:40 -0500)]
O3: Fix issue w/wbOutstading being decremented multiple times on blocked cache.
If a split load fails on a blocked cache wbOutstanding can be decremented
twice if the first part of the split load succeeds and the second part fails.
Condition the decrementing on not having completed the first part of the load.