cpu: Added the Multiperspective Perceptron Predictor with TAGE (8KB and 64KB)

Described by the following article:
Jiménez, D. "Multiperspective perceptron predictor with TAGE."
Championship Branch Prediction (CBP-5) (2016).

Change-Id: Ica3c121a4c94657d9015573085040e8a1984b069
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/19188
Tested-by: kokoro <noreply+kokoro@google.com>
Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
Reviewed-by: Ilias Vougioukas <ilias.vougioukas@arm.com>

cpu: Added the Multiperspective Perceptron Predictor (8KB and 64KB)

Described by the following article:
Jiménez, D. "Multiperspective perceptron predictor."
Championship Branch Prediction (CBP-5) (2016).

Change-Id: Iaa68ead7696e0b6ba05b4417d0322e8053e10d30
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/15495
Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
Reviewed-by: Ilias Vougioukas <ilias.vougioukas@arm.com>
Tested-by: kokoro <noreply+kokoro@google.com>

arch, base, cpu, dev, mem, sim: Remove #if 0-ed out code.

This code will be preserved through version control, but otherwise
creates clutter and will rot in place since it's never compiled.

Change-Id: Id265f6deac445116843956ea5cf1210d8127274e
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/18608
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Reviewed-by: Brandon Potter <Brandon.Potter@amd.com>
Maintainer: Jason Lowe-Power <jason@lowepower.com>
Tested-by: kokoro <noreply+kokoro@google.com>

base: Move SatCounter to base directory

Saturating counters are used by many objects, not only
the cpu predictors. Therefore, move the class to the
base folder so that it can be more easily used.

Change-Id: I26f799324bdd8720ab8834c72a2002149cee777c
Signed-off-by: Daniel <odanrc@yahoo.com.br>
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/17993
Maintainer: Jason Lowe-Power <jason@lowepower.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>

cpu: Revamp saturating counters

Revamp the SatCounter class, improving comments, implementing
increment, decrement and read operators to solve an old todo,
and adding missing error checking.

Change-Id: Ia057c423c90652ebd966b6b91a3471b17800f933
Signed-off-by: Daniel <odanrc@yahoo.com.br>
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/17992
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Maintainer: Jason Lowe-Power <jason@lowepower.com>
Tested-by: kokoro <noreply+kokoro@google.com>

cpu: Make the indirect predictor into a SimObject

Change-Id: Ice6549773def7d3e944fae450d4a079bc351e2ba
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/15319
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
Tested-by: kokoro <noreply+kokoro@google.com>

misc: Removed inconsistency in O3* debug msgs

Added consistency in the DEBUG message form, to allow a better parsing.
Fixed sn/tid type parameter.
Removed some annoying newlines

Change-Id: I4761c49fc12b874a7d8b46779475b606865cad4b
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/17248
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Maintainer: Jason Lowe-Power <jason@lowepower.com>

cpu: Fixed the indirect branch predictor GHR handling

The internal indirect predictor global history was not being updated
properly, resulting in higher than expected miss rates

Also added a parameter to set the size of the indirect predictor GHR

Change-Id: Ibc797816974cba6719da65122801e8919559a003
Signed-off-by: Pau Cabre <pau.cabre@metempsy.com>
Reported-by: Daniel Carvalho <odanrc@yahoo.com.br>
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/16928
Reviewed-by: Daniel Carvalho <odanrc@yahoo.com.br>
Reviewed-by: Andrea Mondelli <Andrea.Mondelli@ucf.edu>
Reviewed-by: Sudhanshu Jha <sudhanshu.jha@arm.com>
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>

cpu: Fix indirect branch history updates

Recent changes to indirect branch predictor interface accesses
non-existent buffers even when indirect predictor is not in use.

Change-Id: I0df9ac4d5f6f3cb63e4d1bd36949c27f7611eef6
Reviewed-on: https://gem5-review.googlesource.com/c/16668
Reviewed-by: Anthony Gutierrez <anthony.gutierrez@amd.com>
Maintainer: Anthony Gutierrez <anthony.gutierrez@amd.com>

cpu: Added 8KB and 64KB TAGE-SC-L branch predictor

The original paper of the branch predictor can be found here:
http://www.jilp.org/cbp2016/paper/AndreSeznecLimited.pdf

Change-Id: I684863752407685adaacedebb699205c3559c528
Reviewed-on: https://gem5-review.googlesource.com/c/14855
Reviewed-by: Sudhanshu Jha <sudhanshu.jha@arm.com>
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>

cpu: Proposal for changing the indirect branch predictor interface

Now the indirect branch predictor handles its own GHR instead of getting
the one from the direction predictor.

Also, now the commit method of the indirect predictor is called for every
pending branch on an update, as the indirect predictors may want to update
their interal structures/histories with the information of each branch.

Change-Id: I7053fbea42a53960a3bc1ba32912cc99c160511e
Reviewed-on: https://gem5-review.googlesource.com/c/15318
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>

cpu: Made the Loop Predictor a SimObject

The Loop Predictor implementation is now a SimObject so that other branch
predictors can easily use it (including LTAGE, which is now using it).
It has also been updated with the latest
available loop predictor implementation from Andre Seznec:

http://www.irisa.fr/alf/downloads/seznec/TAGE-GSC-IMLI.tar

Change-Id: I60ad079a2c49b00a1f84d5cfd3611631883a4b57
Reviewed-on: https://gem5-review.googlesource.com/c/15775
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>

cpu: Made TAGE a SimObject that can be used by other predictors

The TAGE implementation is now a SimObject so that other branch predictors
can easily use it. It has also been updated with the latest available TAGE
implementation from Andre Seznec:

http://www.irisa.fr/alf/downloads/seznec/TAGE-GSC-IMLI.tar

Change-Id: I2251b8b2d7f94124f9955f52b917dc3b064f090e
Reviewed-on: https://gem5-review.googlesource.com/c/15317
Reviewed-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>

cpu: Fixed typos in parameter/stats descriptions

Change-Id: I7b3274a3e37128da35f497da150af08343e97ee6
Signed-off-by: Pau Cabre <pau.cabre@metempsy.com>
Reviewed-on: https://gem5-review.googlesource.com/c/14795
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Reviewed-by: Ilias Vougioukas <ilias.vougioukas@arm.com>
Maintainer: Jason Lowe-Power <jason@lowepower.com>

cpu: Added parameters to enable/disable features in LTAGE

They are for the following features in the LTAGE loop predictor:
- Hashing for calculating the loop table entry
- Add direction information
- Add speculative iteration number information

Change-Id: I395f4526163ee0d0229d1e87cde2bb046f1dd43a
Signed-off-by: Pau Cabre <pau.cabre@metempsy.com>
Reviewed-on: https://gem5-review.googlesource.com/c/14597
Reviewed-by: Ilias Vougioukas <ilias.vougioukas@arm.com>
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Reviewed-by: Louis Delhez <ldelhez@ucla.edu>
Maintainer: Jason Lowe-Power <jason@lowepower.com>

cpu: Added new stats to TAGE and LTAGE branch predictors

They are basically used to tell wich component of the predictor is
providing the prediction and whether it is correct or wrong

Change-Id: I7b3db66535f159091f1b37d70c2d942d50b20fb2
Signed-off-by: Pau Cabre <pau.cabre@metempsy.com>
Reviewed-on: https://gem5-review.googlesource.com/c/14535
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Maintainer: Jason Lowe-Power <jason@lowepower.com>

cpu: split LTAGE implementation into a base TAGE and a derived LTAGE

The new derived LTAGE is equivalent to the original LTAGE implementation
The default values of the TAGE branch predictor match the settings of the
8C-TAGE configuration described in https://www.jilp.org/vol8/v8paper1.pdf

Change-Id: I8323adbfd5c9a45db23cfff234218280e639f9ed
Signed-off-by: Pau Cabre <pau.cabre@metempsy.com>
Reviewed-on: https://gem5-review.googlesource.com/c/14435
Reviewed-by: Sudhanshu Jha <sudhanshu.jha@arm.com>
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Maintainer: Jason Lowe-Power <jason@lowepower.com>

cpu: Made LTAGE parameters configurable

This includes TAGE tag sizes, TAGE table sizes, U counters reset period,
loop predictor associativity, path history size, the USE_ALT_ON_NA size
and the WITHLOOP size

Change-Id: I935823f0a5794f5d55b744263798897a813dc1bd
Signed-off-by: Pau Cabre <pau.cabre@metempsy.com>
Reviewed-on: https://gem5-review.googlesource.com/c/14417
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Maintainer: Jason Lowe-Power <jason@lowepower.com>

cpu: Fixed useful counter handling in LTAGE

Increased to 2 bits of useful counter per TAGE entry as described in the
LTAGE paper (and made the size configurable)

Changed how the useful counters are incremented/decremented as described
in the LTAGE paper

Change-Id: I8c692cc7c180d29897cb77781681ff498a1d16c8
Signed-off-by: Pau Cabre <pau.cabre@metempsy.com>
Reviewed-on: https://gem5-review.googlesource.com/c/14215
Reviewed-by: Ilias Vougioukas <ilias.vougioukas@arm.com>
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Maintainer: Jason Lowe-Power <jason@lowepower.com>

cpu: Fixes on the loop predictor part of LTAGE

Fixed the following fields of the loop predictor entries as described on
the LTAGE paper:
- Age counter (it was 3 bits and it should be 8 bits)
- Tag (it was 16 bits and it should be 14 bits). Also some times it used
int variables and some times uint16_t, leading to wrong behaviour
- Confidence counter (it was 2 bits ins some parts of the code and 3 bits
in some other parts. It should be 2 bits)
- Iteration counters (they were 16 bits and they should be 14 bits)
All the new sizes are now configurable

Change-Id: I8884c7454c1e510b65160eb4d5749d3259d34096
Signed-off-by: Pau Cabre <pau.cabre@metempsy.com>
Reviewed-on: https://gem5-review.googlesource.com/c/14216
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Maintainer: Jason Lowe-Power <jason@lowepower.com>

cpu: Fix LTAGE max number of allocations on update

The LTAGE paper states that only one TAGE entry can be
allocated when updating

Change-Id: I6cfb4d80ce835e93d4bf5099ef88a7d425abaddd
Signed-off-by: Pau Cabre <pau.cabre@metempsy.com>
Reviewed-on: https://gem5-review.googlesource.com/c/14195
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
Reviewed-by: Ilias Vougioukas <ilias.vougioukas@arm.com>
Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>

configs: Added an option for choosing branch predictor type

Added the parameter "--bp-type" to set the branch predictor type
Added the parameter "--list-bp-types" to list all the available branch
predictor types

Change-Id: Ia6aae90c784aef359b6d8233c8383cd7a871aca1
Signed-off-by: Pau Cabre <pau.cabre@metempsy.com>
Reviewed-on: https://gem5-review.googlesource.com/c/14015
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>

cpu: Fixed ratio of pred to hyst bits for LTAGE Bimodal

The LTAGE paper states 1 hyst bit shared for 4 pred bits.
Made this ratio configurable use 4 by default.
Also changed the Bimodal structure to use two std::vector<bool> (one for
pred and one for hyst bits)

Change-Id: I6793e8e358be01b75b8fd181ddad50f259862d79
Signed-off-by: Pau Cabre <pau.cabre@metempsy.com>
Reviewed-on: https://gem5-review.googlesource.com/c/14120
Reviewed-by: Ilias Vougioukas <ilias.vougioukas@arm.com>
Reviewed-by: Sudhanshu Jha <sudhanshu.jha@arm.com>
Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>

cpu: Fixed PC shifting on LTAGE branch predictor

The PC needs to be shifted according to the instShiftAmt parameter

Change-Id: I272619c093695b56cf7f8ff7163e3b5d23205d16
Signed-off-by: Pau Cabre <pau.cabre@metempsy.com>
Reviewed-on: https://gem5-review.googlesource.com/c/14035
Reviewed-by: Sudhanshu Jha <sudhanshu.jha@arm.com>
Reviewed-by: Ilias Vougioukas <ilias.vougioukas@arm.com>
Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com>
Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>

misc: Rename misc.(hh|cc) to logging.(hh|cc)

These files aren't a collection of miscellaneous stuff, they're the
definition of the Logger interface, and a few utility macros for
calling into that interface (panic, warn, etc.).

Change-Id: I84267ac3f45896a83c0ef027f8f19c5e9a5667d1
Reviewed-on: https://gem5-review.googlesource.com/6226
Reviewed-by: Brandon Potter <Brandon.Potter@amd.com>
Maintainer: Gabe Black <gabeblack@google.com>

cpu: Fix bi-mode branch predictor thresholds

When different sizes were set for the choice and global saturation
counter (e.g. ex5_big), the threshold calculation used the wrong
size. Thus the branch predictor always predicted "not taken" for
choice > global.

Change-Id: I076549ff1482e2280cef24a0d16b7bb2122d4110
Reviewed-on: https://gem5-review.googlesource.com/4560
Reviewed-by: Curtis Dunham <curtis.dunham@arm.com>
Maintainer: Jason Lowe-Power <jason@lowepower.com>

style: [patch 1/22] use /r/3648/ to reorganize includes

cpu: implement an L-TAGE branch predictor

This patch implements an L-TAGE predictor, based on André Seznec's code
available from CBP-2
(http://hpca23.cse.tamu.edu/taco/camino/cbp2/cbp-src/realistic-seznec.h).

Signed-off-by Jason Lowe-Power <jason@lowepower.com>

cpu: disallow speculative update of branch predictor tables (o3)

The Minor and o3 cpu models share the branch prediction
code. Minor relies on the BPredUnit::squash() function
to update the branch predictor tables on a branch mispre-
diction. This is fine because Minor executes in-order, so
the update is on the correct path. However, this causes the
branch predictor to be updated on out-of-order branch
mispredictions when using the o3 model, which should not
be the case.

This patch guards against speculative update of the branch
prediction tables. On a branch misprediction, BPredUnit::squash()
calls BpredUnit::update(..., squashed = true). The underlying
branch predictor tests against the value of squashed. If it is
true, it restores any speculatively updated internal state
it might have (e.g., global/local branch history), then returns.
If false, it updates its prediction tables. Previously, exist-
ing predictors did not test against the "squashed" parameter.

To accomodate for this change, the Minor model must now call
BPredUnit::squash() then BPredUnit::update(..., squashed = false)
on branch mispredictions. Before, calling BpredUnit::squash()
performed the prediction tables update.

The effect is a slight MPKI improvement when using the o3
model. A further patch should perform the same modifications
for the indirect target predictor and BTB (less critical).

Signed-off-by: Jason Lowe-Power <jason@lowepower.com>

cpu: correct comments in tournament branch predictor

The tournament predictor is presented as doing speculative
update of the global history and non-speculative update
of the local history used to generate the branch prediction.
However, the code does speculative update of both histories.

Signed-off-by: Jason Lowe-Power <jason@lowepower.com>

cpu: Remove branch predictor function predictInOrder

This function was used by the now-defunct InOrderCPU model. Since this
model is no longer in gem5, this function was not called from anywhere in
the code.

stats: Fixing regStats function for some SimObjects

Fixing an issue with regStats not calling the parent class method
for most SimObjects in Gem5. This causes issues if one adds new
stats in the base class (since they are never initialized properly!).

Change-Id: Iebc5aa66f58816ef4295dc8e48a357558d76a77c
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>

cpu: Implement per-thread GHRs

Branch predictors that use GHRs should index them on a
per-thread basis. This makes that so.

This is a re-spin of fb51231 after the revert (bd1c6789).

cpu: Add an indirect branch target predictor

This patch adds a configurable indirect branch predictor that can be indexed
by a combination of GHR and path history hashes. Implements the functionality
described in:

"Target prediction for indirect jumps" by Chang, Hao, and Patt
http://dl.acm.org/citation.cfm?id=264209

This is a re-spin of fb9d142 after the revert (bd1c6789).

cpu: Fix BTB threading oversight

The extant BTB code doesn't hash on the thread id but does check the
thread id for 'btb hits'. This results in 1-thread of a multi-threaded
workload taking a BTB entry, and all other threads missing for the same branch
missing.

Revert power patch sets with unexpected interactions

The following patches had unexpected interactions with the current
upstream code and have been reverted for now:

e07fd01651f3: power: Add support for power models
831c7f2f9e39: power: Low-power idle power state for idle CPUs
4f749e00b667: power: Add power states to ClockedObject

Signed-off-by: Andreas Sandberg <andreas.sandberg@arm.com>

cpu: Implement per-thread GHRs

Branch predictors that use GHRs should index them on a
per-thread basis. This makes that so.

cpu: Add an indirect branch target predictor

This patch adds a configurable indirect branch predictor that can be indexed
by a combination of GHR and path history hashes. Implements the functionality
described in:

"Target prediction for indirect jumps" by Chang, Hao, and Patt
http://dl.acm.org/citation.cfm?id=264209

cpu: Fix BTB threading oversight

The extant BTB code doesn't hash on the thread id but does check the
thread id for 'btb hits'. This results in 1-thread of a multi-threaded
workload taking a BTB entry, and all other threads missing for the same branch
missing.

style: fix missing spaces in control statements

Result of running 'hg m5style --skip-all --fix-control -a'.

scons: Enable -Wextra by default

Make best use of the compiler, and enable -Wextra as well as
-Wall. There are a few issues that had to be resolved, but they are
all trivial.

misc: Add explicit overrides and fix other clang >= 3.5 issues

This patch adds explicit overrides as this is now required when using
"-Wall" with clang >= 3.5, the latter now part of the most recent
XCode. The patch consequently removes "virtual" for those methods
where "override" is added. The latter should be enough of an
indication.

As part of this patch, a few minor issues that clang >= 3.5 complains
about are also resolved (unused methods and variables).

misc: Remove redundant compiler-specific defines

This patch moves away from using M5_ATTR_OVERRIDE and the m5::hashmap
(and similar) abstractions, as these are no longer needed with gcc 4.7
and clang 3.1 as minimum compiler versions.

cpu: pred: Local Predictor Reset in Tournament Predictor

When a branch gets squashed, it's speculative branch predictor state should get
rolled back in squash(). However, only the globalHistory state was being
rolled back. This patch adds (at least some) support for rolling back the
local predictor state also.

Committed by: Nilay Vaish <nilay@cs.wisc.edu>

cpu: re-organizes the branch predictor structure.

Committed by: Nilay Vaish <nilay@cs.wisc.edu>

cpu: Add branch predictor PMU probe points

This changeset adds probe points that can be used to implement PMU
counters for branch predictor stats. The following probes are
supported:

* BPRedUnit::ppBranches / Branches
* BPRedUnit::ppMisses / Misses

arch: Use const StaticInstPtr references where possible

This patch optimises the passing of StaticInstPtr by avoiding copying
the reference-counting pointer. This avoids first incrementing and
then decrementing the reference-counting pointer.

cpu: fix bimodal predictor to use correct global history reg

A small bug in the bimodal predictor caused significant degradation in
performance on some benchmarks. This was caused by using the wrong
globalHistoryReg during the update phase. This patches fixes the bug
and brings the performance to normal level.

cpu: Fix incorrect speculative branch predictor behavior

When a branch mispredicted gem5 would squash all history after and including
the mispredicted branch. However, the mispredicted branch is still speculative
and its history is required to rollback state if another, older, branch
mispredicts. This leads to things like RAS corruption.

scons: Build the branch predictor for all CPUs

The branch predictor is normally only built when a CPU that uses a
branch predictor is built. The list of CPUs is currently incomplete as
the simple CPUs support branch predictors (for warming, branch stats,
etc). In practice, all CPU models now use branch predictors, so this
changeset removes the CPU model check and replaces it with a check for
the NULL ISA.

cpu: Modernise the branch predictor (STL and C++11)

This patch does some minor house keeping of the branch predictor by
adopting STL containers, and shifting some iterator to use range-based
for loops.

The predictor history is also changed from a list to a deque as we
never to insertion/deletion other than at the front and back.

cpu: `Minor' in-order CPU model

This patch contains a new CPU model named `Minor'. Minor models a four
stage in-order execution pipeline (fetch lines, decompose into
macroops, decompose macroops into microops, execute).

The model was developed to support the ARM ISA but should be fixable
to support all the remaining gem5 ISAs. It currently also works for
Alpha, and regressions are included for ARM and Alpha (including Linux
boot).

Documentation for the model can be found in src/doc/inside-minor.doxygen and
its internal operations can be visualised using the Minorview tool
utils/minorview.py.

Minor was designed to be fairly simple and not to engage in a lot of
instruction annotation. As such, it currently has very few gathered
stats and may lack other gem5 features.

Minor is faster than the o3 model. Sample results:

Benchmark | Stat host_seconds (s)
---------------+--------v--------v--------
(on ARM, opt) | simple | o3 | minor
| timing | timing | timing
---------------+--------+--------+--------
10.linux-boot | 169 | 1883 | 1075
10.mcf | 117 | 967 | 491
20.parser | 668 | 6315 | 3146
30.eon | 542 | 3413 | 2414
40.perlbmk | 2339 | 20905 | 11532
50.vortex | 122 | 1094 | 588
60.bzip2 | 2045 | 18061 | 9662
70.twolf | 207 | 2736 | 1036

cpu: implement a bi-mode branch predictor

cpu: add consistent guarding to *_impl.hh files.

cpu: remove local/globalHistoryBits params from branch pred

having separate params for the local/globalHistoryBits and the
local/globalPredictorSize can lead to inconsistencies if they
are not carefully set. this patch dervies the number of bits
necessary to index into the local/global predictors based on
their size.

the value of the localHistoryTableSize for the ARM O3 CPU has been
increased to 1024 from 64, which is more accurate for an A15 based
on some correlation against A15 hardware.

branch predictor: move out of o3 and inorder cpus
This patch moves the branch predictor files in the o3 and inorder directories
to src/cpu/pred. This allows sharing the branch predictor across different
cpu models.

This patch was originally posted by Timothy Jones in July 2010
but never made it to the repository.

TournamentBP: Fix some bugs with table sizes and counters
globalHistoryBits, globalPredictorSize, and choicePredictorSize are decoupled.
globalHistoryBits controls how much history is kept, global and choice
predictor sizes control how much of that history is used when accessing
predictor tables. This way, global and choice predictors can actually be
different sizes, and it is no longer possible to walk off the predictor arrays
and cause a seg fault.

There are now individual thresholds for choice, global, and local saturating
counters, so that taken/not taken decisions are correct even when the
predictors' counters' sizes are different.

The interface for localPredictorSize has been removed from TournamentBP because
the value can be calculated from localHistoryBits.

Committed by: Nilay Vaish <nilay@cs.wisc.edu>

o3: Fix a couple of issues with the local predictor.

Fix some issues with the local predictor and the way it's indexed.

BPred: Fix RAS to handle predicated call/return instructions.

Change RAS to fix issues with predicated call/return instructions.
Handled all cases in the life of a predicated call and return instruction.

BP: Fix several Branch Predictor issues.
1. Updates the Branch Predictor correctly to the state
just after a mispredicted branch, if a squash occurs.
2. If a BTB does not find an entry, the branch is predicted not taken.
The global history is modified to correctly reflect this prediction.
3. Local history is now updated at the fetch stage instead of
execute stage.
4. In the Update stage of the branch predictor the local predictors are
now correctly updated according to the state of local history during
fetch stage.

This patch also improves performance by as much as 17% on some benchmarks

O3: Fix uninitialized variable in the tournament branch predictor.

Branch predictor: Fixes the tournament branch predictor.

Branch predictor could not predict a branch in a nested loop because:
1. The global history was not updated after a mispredict squash.
2. The global history was updated in the fetch stage. The choice predictors
that were updated used the changed global history. This is incorrect, as
it incorporates the state of global history after the branch in
encountered. Fixed update to choice predictor using the global history
state before the branch happened.
3. The global predictor table was also updated using the global history state
before the branch happened as above.

Additionally, parameters to initialize ctr and history size were reversed.

scons: rename TraceFlags to DebugFlags

trace: reimplement the DTRACE function so it doesn't use a vector
At the same time, rename the trace flags to debug flags since they
have broader usage than simply tracing. This means that
--trace-flags is now --debug-flags and --trace-help is now --debug-help

ISA,CPU,etc: Create an ISA defined PC type that abstracts out ISA behaviors.



This change is a low level and pervasive reorganization of how PCs are managed
in M5. Back when Alpha was the only ISA, there were only 2 PCs to worry about,
the PC and the NPC, and the lsb of the PC signaled whether or not you were in
PAL mode. As other ISAs were added, we had to add an NNPC, micro PC and next
micropc, x86 and ARM introduced variable length instruction sets, and ARM
started to keep track of mode bits in the PC. Each CPU model handled PCs in
its own custom way that needed to be updated individually to handle the new
dimensions of variability, or, in the case of ARMs mode-bit-in-the-pc hack,
the complexity could be hidden in the ISA at the ISA implementation's expense.
Areas like the branch predictor hadn't been updated to handle branch delay
slots or micropcs, and it turns out that had introduced a significant (10s of
percent) performance bug in SPARC and to a lesser extend MIPS. Rather than
perpetuate the problem by reworking O3 again to handle the PC features needed
by x86, this change was introduced to rework PC handling in a more modular,
transparent, and hopefully efficient way.


PC type:

Rather than having the superset of all possible elements of PC state declared
in each of the CPU models, each ISA defines its own PCState type which has
exactly the elements it needs. A cross product of canned PCState classes are
defined in the new "generic" ISA directory for ISAs with/without delay slots
and microcode. These are either typedef-ed or subclassed by each ISA. To read
or write this structure through a *Context, you use the new pcState() accessor
which reads or writes depending on whether it has an argument. If you just
want the address of the current or next instruction or the current micro PC,
you can get those through read-only accessors on either the PCState type or
the *Contexts. These are instAddr(), nextInstAddr(), and microPC(). Note the
move away from readPC. That name is ambiguous since it's not clear whether or
not it should be the actual address to fetch from, or if it should have extra
bits in it like the PAL mode bit. Each class is free to define its own
functions to get at whatever values it needs however it needs to to be used in
ISA specific code. Eventually Alpha's PAL mode bit could be moved out of the
PC and into a separate field like ARM.

These types can be reset to a particular pc (where npc = pc +
sizeof(MachInst), nnpc = npc + sizeof(MachInst), upc = 0, nupc = 1 as
appropriate), printed, serialized, and compared. There is a branching()
function which encapsulates code in the CPU models that checked if an
instruction branched or not. Exactly what that means in the context of branch
delay slots which can skip an instruction when not taken is ambiguous, and
ideally this function and its uses can be eliminated. PCStates also generally
know how to advance themselves in various ways depending on if they point at
an instruction, a microop, or the last microop of a macroop. More on that
later.

Ideally, accessing all the PCs at once when setting them will improve
performance of M5 even though more data needs to be moved around. This is
because often all the PCs need to be manipulated together, and by getting them
all at once you avoid multiple function calls. Also, the PCs of a particular
thread will have spatial locality in the cache. Previously they were grouped
by element in arrays which spread out accesses.


Advancing the PC:

The PCs were previously managed entirely by the CPU which had to know about PC
semantics, try to figure out which dimension to increment the PC in, what to
set NPC/NNPC, etc. These decisions are best left to the ISA in conjunction
with the PC type itself. Because most of the information about how to
increment the PC (mainly what type of instruction it refers to) is contained
in the instruction object, a new advancePC virtual function was added to the
StaticInst class. Subclasses provide an implementation that moves around the
right element of the PC with a minimal amount of decision making. In ISAs like
Alpha, the instructions always simply assign NPC to PC without having to worry
about micropcs, nnpcs, etc. The added cost of a virtual function call should
be outweighed by not having to figure out as much about what to do with the
PCs and mucking around with the extra elements.

One drawback of making the StaticInsts advance the PC is that you have to
actually have one to advance the PC. This would, superficially, seem to
require decoding an instruction before fetch could advance. This is, as far as
I can tell, realistic. fetch would advance through memory addresses, not PCs,
perhaps predicting new memory addresses using existing ones. More
sophisticated decisions about control flow would be made later on, after the
instruction was decoded, and handed back to fetch. If branching needs to
happen, some amount of decoding needs to happen to see that it's a branch,
what the target is, etc. This could get a little more complicated if that gets
done by the predecoder, but I'm choosing to ignore that for now.


Variable length instructions:

To handle variable length instructions in x86 and ARM, the predecoder now
takes in the current PC by reference to the getExtMachInst function. It can
modify the PC however it needs to (by setting NPC to be the PC + instruction
length, for instance). This could be improved since the CPU doesn't know if
the PC was modified and always has to write it back.


ISA parser:

To support the new API, all PC related operand types were removed from the
parser and replaced with a PCState type. There are two warts on this
implementation. First, as with all the other operand types, the PCState still
has to have a valid operand type even though it doesn't use it. Second, using
syntax like PCS.npc(target) doesn't work for two reasons, this looks like the
syntax for operand type overriding, and the parser can't figure out if you're
reading or writing. Instructions that use the PCS operand (which I've
consistently called it) need to first read it into a local variable,
manipulate it, and then write it back out.


Return address stack:

The return address stack needed a little extra help because, in the presence
of branch delay slots, it has to merge together elements of the return PC and
the call PC. To handle that, a buildRetPC utility function was added. There
are basically only two versions in all the ISAs, but it didn't seem short
enough to put into the generic ISA directory. Also, the branch predictor code
in O3 and InOrder were adjusted so that they always store the PC of the actual
call instruction in the RAS, not the next PC. If the call instruction is a
microop, the next PC refers to the next microop in the same macroop which is
probably not desirable. The buildRetPC function advances the PC intelligently
to the next macroop (in an ISA specific way) so that that case works.


Change in stats:

There were no change in stats except in MIPS and SPARC in the O3 model. MIPS
runs in about 9% fewer ticks. SPARC runs with 30%-50% fewer ticks, which could
likely be improved further by setting call/return instruction flags and taking
advantage of the RAS.


TODO:

Add != operators to the PCState classes, defined trivially to be !(a==b).
Smooth out places where PCs are split apart, passed around, and put back
together later. I think this might happen in SPARC's fault code. Add ISA
specific constructors that allow setting PC elements without calling a bunch
of accessors. Try to eliminate the need for the branching() function. Factor
out Alpha's PAL mode pc bit into a separate flag field, and eliminate places
where it's blindly masked out or tested in the PC.

BPRED: Fixed the treshold-bug in the tournament predictor.

Suppose the saturating counters of a branch predictor contain n bits. When the
counter is between 0 and (2^(n-1) - 1), boundaries included, the branch is
predicted as not taken. When the counter is between 2^(n-1) and (2^n - 1),
boundaries included, the branch is predicted as taken.

types: clean up types, especially signed vs unsigned

move: put predictor includes and cc files into the same place