struct EnergyModelOptimizer<'a> {
eq_pds: Vec<Vec<&'a PerfDomain>>,
nr_cpus_combinations: Vec<Vec<usize>>,
cpus_topological_order: Vec<usize>,
pd_cpu_order: BTreeMap<usize, RefCell<Vec<usize>>>,
tot_perf: usize,
pdss_infos: RefCell<BTreeMap<usize, RefCell<HashSet<PDSetInfo<'a>>>>>,
perf_pdsi: RefCell<BTreeMap<usize, PDSetInfo<'a>>>,
perf_cpu_order: RefCell<BTreeMap<usize, PerfCpuOrder>>,
}Fields§
§eq_pds: Vec<Vec<&'a PerfDomain>>§nr_cpus_combinations: Vec<Vec<usize>>§cpus_topological_order: Vec<usize>§pd_cpu_order: BTreeMap<usize, RefCell<Vec<usize>>>§tot_perf: usize§pdss_infos: RefCell<BTreeMap<usize, RefCell<HashSet<PDSetInfo<'a>>>>>§perf_pdsi: RefCell<BTreeMap<usize, PDSetInfo<'a>>>§perf_cpu_order: RefCell<BTreeMap<usize, PerfCpuOrder>>Implementations§
Source§impl<'a> EnergyModelOptimizer<'a>
impl<'a> EnergyModelOptimizer<'a>
fn new(em: &'a EnergyModel, cpus_pf: &'a Vec<CpuId>) -> EnergyModelOptimizer<'a>
fn get_perf_cpu_order_table( em: &'a EnergyModel, cpus_pf: &'a Vec<CpuId>, ) -> BTreeMap<usize, PerfCpuOrder>
fn get_fake_perf_cpu_order_table( cpus_pf: &'a Vec<CpuId>, cpus_ps: &'a Vec<CpuId>, ) -> BTreeMap<usize, PerfCpuOrder>
fn fake_pco( tot_perf: usize, cpuids: &'a Vec<CpuId>, powersave: bool, ) -> PerfCpuOrder
Sourcefn gen_perf_cpu_order_table(&'a self)
fn gen_perf_cpu_order_table(&'a self)
Generate the performance versus CPU preference order table based on the system’s CPU topology and energy model. The table consists of the following information (PerfCpuOrder):
-
PerfCpuOrder::perf_cap: The upper bound of the performance capacity covered by this tuple.
-
PerfCpuOrder::cpus_perf: Primary CPUs to be used is ordered by preference.
-
PerfCpuOrder::cpus_ovrflw: When the system load goes beyond @perf_cap, the list of CPUs to be used is ordered by preference.
Sourcefn assign_cpu_vids(&'a self)
fn assign_cpu_vids(&'a self)
Generate a CPU order table for each performance range.
Sourcefn sort_cpus_by_topological_order(&'a self, cpus: &Vec<usize>) -> Vec<usize>
fn sort_cpus_by_topological_order(&'a self, cpus: &Vec<usize>) -> Vec<usize>
Sort the CPU IDs by topological order (@self.cpus_topological_order).
Sourcefn gen_perf_pds_table(&'a self)
fn gen_perf_pds_table(&'a self)
Generate a table of performance vs. performance domain sets (@self.perf_pdss) from all the possible performance domain & state combinations (@self.pdss_infos).
An example result is as follows: PERF: [, 300] pd:id: 0 – cpu_vid: 0 pd:id: 0 – cpu_vid: 1 PERF: [, 1138] pd:id: 0 – cpu_vid: 0 pd:id: 0 – cpu_vid: 1 pd:id: 1 – cpu_vid: 0 pd:id: 1 – cpu_vid: 1 PERF: [, 3386] pd:id: 1 – cpu_vid: 0 pd:id: 1 – cpu_vid: 1 pd:id: 1 – cpu_vid: 2 pd:id: 2 – cpu_vid: 0 pd:id: 2 – cpu_vid: 1 PERF: [, 3977] pd:id: 0 – cpu_vid: 0 pd:id: 1 – cpu_vid: 0 pd:id: 1 – cpu_vid: 1 pd:id: 1 – cpu_vid: 2 pd:id: 2 – cpu_vid: 0 pd:id: 2 – cpu_vid: 1 PERF: [, 4508] pd:id: 0 – cpu_vid: 0 pd:id: 0 – cpu_vid: 1 pd:id: 1 – cpu_vid: 0 pd:id: 1 – cpu_vid: 1 pd:id: 1 – cpu_vid: 2 pd:id: 2 – cpu_vid: 0 pd:id: 2 – cpu_vid: 1 PERF: [, 5627] pd:id: 0 – cpu_vid: 0 pd:id: 0 – cpu_vid: 1 pd:id: 1 – cpu_vid: 0 pd:id: 1 – cpu_vid: 1 pd:id: 1 – cpu_vid: 2 pd:id: 2 – cpu_vid: 0 pd:id: 2 – cpu_vid: 1 pd:id: 3 – cpu_vid: 0
fn find_perf_pds_for( &'a self, util: f32, base: Option<&PDSetInfo<'a>>, ) -> Option<PDSetInfo<'a>>
Sourcefn gen_all_pds_combinations(&'a self)
fn gen_all_pds_combinations(&'a self)
Generate all possible performance domain & state combinations, @self.pdss_infos. Each combination represents a set of performance domains (and their corresponding performance states) that achieve the requested performance with minimal power consumption.
We assume a ‘reasonable load balancer,’ so the CPU utilization of all the involved CPUs is similar.
An example result is as follows:
PERF: [_, 5135]
perf: 5135 -- power: 5475348
pd:id: 0 -- cpu_vid: 0
pd:id: 1 -- cpu_vid: 0
pd:id: 1 -- cpu_vid: 1
pd:id: 1 -- cpu_vid: 2
pd:id: 2 -- cpu_vid: 0
pd:id: 2 -- cpu_vid: 1
pd:id: 3 -- cpu_vid: 0
PERF: [_, 5187]
perf: 5187 -- power: 4844969
pd:id: 0 -- cpu_vid: 0
pd:id: 0 -- cpu_vid: 1
pd:id: 1 -- cpu_vid: 0
pd:id: 1 -- cpu_vid: 1
pd:id: 1 -- cpu_vid: 2
pd:id: 2 -- cpu_vid: 0
pd:id: 2 -- cpu_vid: 1
pd:id: 3 -- cpu_vid: 0
PERF: [_, 5195]
perf: 5195 -- power: 5924606
pd:id: 1 -- cpu_vid: 0
pd:id: 1 -- cpu_vid: 1
pd:id: 1 -- cpu_vid: 2
pd:id: 2 -- cpu_vid: 0
pd:id: 2 -- cpu_vid: 1
pd:id: 3 -- cpu_vid: 0
PERF: [_, 5217]
perf: 5217 -- power: 4894911
pd:id: 0 -- cpu_vid: 0
pd:id: 0 -- cpu_vid: 1
pd:id: 1 -- cpu_vid: 0
pd:id: 1 -- cpu_vid: 1
pd:id: 1 -- cpu_vid: 2
pd:id: 2 -- cpu_vid: 0
pd:id: 2 -- cpu_vid: 1
pd:id: 3 -- cpu_vid: 0
PERF: [_, 5225]
perf: 5225 -- power: 5665770
pd:id: 0 -- cpu_vid: 0
pd:id: 1 -- cpu_vid: 0
pd:id: 1 -- cpu_vid: 1
pd:id: 1 -- cpu_vid: 2
pd:id: 2 -- cpu_vid: 0
pd:id: 2 -- cpu_vid: 1
pd:id: 3 -- cpu_vid: 0
PERF: [_, 5316]
perf: 5316 -- power: 5860568
pd:id: 0 -- cpu_vid: 0
pd:id: 1 -- cpu_vid: 0
pd:id: 1 -- cpu_vid: 1
pd:id: 1 -- cpu_vid: 2
pd:id: 2 -- cpu_vid: 0
pd:id: 2 -- cpu_vid: 1
pd:id: 3 -- cpu_vid: 0fn gen_perf_cpuset_table_range(&'a self, low: isize, high: isize)
Sourcefn rank_perf_doms(cpus_pf: &[CpuId]) -> BTreeMap<usize, usize>
fn rank_perf_doms(cpus_pf: &[CpuId]) -> BTreeMap<usize, usize>
Rank the performance domains by CPU preference: a performance domain is as preferred as its most preferred CPU, which is the first one appearing in @cpus_pf. A performance domain with no CPU in @cpus_pf is unranked and comes last.
Sourcefn sort_eq_pds(
em: &'a EnergyModel,
cpus_pf: &'a [CpuId],
) -> Vec<Vec<&'a PerfDomain>>
fn sort_eq_pds( em: &'a EnergyModel, cpus_pf: &'a [CpuId], ) -> Vec<Vec<&'a PerfDomain>>
Collect the member performance domains of each equivalence performance domain of @em, ordering both the members and the equivalence performance domains by CPU preference. See @EnergyModelOptimizer::eq_pds.
Sourcefn gen_nr_cpus_combinations(max_nr_cpus: &[usize]) -> Vec<Vec<usize>>
fn gen_nr_cpus_combinations(max_nr_cpus: &[usize]) -> Vec<Vec<usize>>
Enumerate how many CPUs to take from each equivalence performance domain, taking at most @max_nr_cpus[i] CPUs from the i-th one. See @EnergyModelOptimizer::nr_cpus_combinations.
Sourcefn gen_all_nr_cpus(max_nr_cpus: &[usize]) -> Vec<Vec<usize>>
fn gen_all_nr_cpus(max_nr_cpus: &[usize]) -> Vec<Vec<usize>>
Enumerate how many CPUs to take from each equivalence performance domain
in every possible way. An equivalence performance domain independently
takes 0, 1, … up to all of its CPUs, so a combination picks one count
from the range 0..=max_nr_cpus[i] of every equivalence performance
domain. Picking one element from each of several ranges, in all the
possible ways, is the cartesian product of those ranges, which
multi_cartesian_product enumerates one combination at a time. See
@EnergyModelOptimizer::nr_cpus_combinations for an example.
Sourcefn gen_run_nr_cpus(max_nr_cpus: &[usize]) -> Vec<Vec<usize>>
fn gen_run_nr_cpus(max_nr_cpus: &[usize]) -> Vec<Vec<usize>>
Enumerate how many CPUs to take from each equivalence performance domain when there are too many combinations to consider them all (@MAX_EQPD_COMBINATIONS). Only the runs of equivalence performance domains are considered, where a run takes all the CPUs of consecutive equivalence performance domains and some of the CPUs of the last one:
- A forward run grows from the first equivalence performance domain, adding one more equivalence performance domain at a time.
- A backward run grows from the last equivalence performance domain in the opposite direction.
- A single run takes CPUs from one equivalence performance domain and none from the others.
For example, with three equivalence performance domains of 1, 2, and 1 CPUs, the runs are:
forward: [1, 0, 0]
[1, 1, 0], [1, 2, 0]
[1, 2, 1]
backward: [0, 0, 1]
[0, 1, 1], [0, 2, 1]
[1, 2, 1]
single: [1, 0, 0]
[0, 1, 0], [0, 2, 0]
[0, 0, 1]which is 3 * nr_cpus = 12 combinations, or 9 once the duplicates are removed. Since there are at most 3 * nr_cpus of them, the runs always fit in @MAX_EQPD_COMBINATIONS.
Sourcefn gen_pds_combinations(&'a self, util: f32) -> Vec<PDSetInfo<'a>>
fn gen_pds_combinations(&'a self, util: f32) -> Vec<PDSetInfo<'a>>
Generate the combinations of performance domains and states to consider for a given CPU utilization (@util), one for each combination of per-equivalence performance domain CPU counts.
Sourcefn gen_pdsi(&'a self, nr_cpus: &[usize], util: f32) -> PDSetInfo<'a>
fn gen_pdsi(&'a self, nr_cpus: &[usize], util: f32) -> PDSetInfo<'a>
Build the performance domains and states taking @nr_cpus[i] CPUs from the i-th equivalence performance domain at the performance state for @util. The CPUs are taken from the member performance domains in order, so the CPUs for a count of N are always a subset of the ones for N + 1.
fn insert_pds_combinations(&self, new_pdsi_vec: &Vec<PDSetInfo<'a>>) -> bool
Trait Implementations§
Auto Trait Implementations§
impl<'a> !Freeze for EnergyModelOptimizer<'a>
impl<'a> !RefUnwindSafe for EnergyModelOptimizer<'a>
impl<'a> !Sync for EnergyModelOptimizer<'a>
impl<'a> !UnwindSafe for EnergyModelOptimizer<'a>
impl<'a> Send for EnergyModelOptimizer<'a>
impl<'a> Unpin for EnergyModelOptimizer<'a>
impl<'a> UnsafeUnpin for EnergyModelOptimizer<'a>
Blanket Implementations§
Source§impl<T> BorrowMut<T> for Twhere
T: ?Sized,
impl<T> BorrowMut<T> for Twhere
T: ?Sized,
Source§fn borrow_mut(&mut self) -> &mut T
fn borrow_mut(&mut self) -> &mut T
§impl<T> Conv for T
impl<T> Conv for T
§impl<T> FmtForward for T
impl<T> FmtForward for T
§fn fmt_binary(self) -> FmtBinary<Self>where
Self: Binary,
fn fmt_binary(self) -> FmtBinary<Self>where
Self: Binary,
self to use its Binary implementation when Debug-formatted.§fn fmt_display(self) -> FmtDisplay<Self>where
Self: Display,
fn fmt_display(self) -> FmtDisplay<Self>where
Self: Display,
self to use its Display implementation when
Debug-formatted.§fn fmt_lower_exp(self) -> FmtLowerExp<Self>where
Self: LowerExp,
fn fmt_lower_exp(self) -> FmtLowerExp<Self>where
Self: LowerExp,
self to use its LowerExp implementation when
Debug-formatted.§fn fmt_lower_hex(self) -> FmtLowerHex<Self>where
Self: LowerHex,
fn fmt_lower_hex(self) -> FmtLowerHex<Self>where
Self: LowerHex,
self to use its LowerHex implementation when
Debug-formatted.§fn fmt_octal(self) -> FmtOctal<Self>where
Self: Octal,
fn fmt_octal(self) -> FmtOctal<Self>where
Self: Octal,
self to use its Octal implementation when Debug-formatted.§fn fmt_pointer(self) -> FmtPointer<Self>where
Self: Pointer,
fn fmt_pointer(self) -> FmtPointer<Self>where
Self: Pointer,
self to use its Pointer implementation when
Debug-formatted.§fn fmt_upper_exp(self) -> FmtUpperExp<Self>where
Self: UpperExp,
fn fmt_upper_exp(self) -> FmtUpperExp<Self>where
Self: UpperExp,
self to use its UpperExp implementation when
Debug-formatted.§fn fmt_upper_hex(self) -> FmtUpperHex<Self>where
Self: UpperHex,
fn fmt_upper_hex(self) -> FmtUpperHex<Self>where
Self: UpperHex,
self to use its UpperHex implementation when
Debug-formatted.§fn fmt_list(self) -> FmtList<Self>where
&'a Self: for<'a> IntoIterator,
fn fmt_list(self) -> FmtList<Self>where
&'a Self: for<'a> IntoIterator,
§impl<T> Instrument for T
impl<T> Instrument for T
§fn instrument(self, span: Span) -> Instrumented<Self> ⓘ
fn instrument(self, span: Span) -> Instrumented<Self> ⓘ
Source§impl<T> IntoEither for T
impl<T> IntoEither for T
Source§fn into_either(self, into_left: bool) -> Either<Self, Self> ⓘ
fn into_either(self, into_left: bool) -> Either<Self, Self> ⓘ
self into a Left variant of Either<Self, Self>
if into_left is true.
Converts self into a Right variant of Either<Self, Self>
otherwise. Read moreSource§fn into_either_with<F>(self, into_left: F) -> Either<Self, Self> ⓘ
fn into_either_with<F>(self, into_left: F) -> Either<Self, Self> ⓘ
self into a Left variant of Either<Self, Self>
if into_left(&self) returns true.
Converts self into a Right variant of Either<Self, Self>
otherwise. Read more§impl<T> Pipe for Twhere
T: ?Sized,
impl<T> Pipe for Twhere
T: ?Sized,
§fn pipe<R>(self, func: impl FnOnce(Self) -> R) -> Rwhere
Self: Sized,
fn pipe<R>(self, func: impl FnOnce(Self) -> R) -> Rwhere
Self: Sized,
§fn pipe_ref<'a, R>(&'a self, func: impl FnOnce(&'a Self) -> R) -> Rwhere
R: 'a,
fn pipe_ref<'a, R>(&'a self, func: impl FnOnce(&'a Self) -> R) -> Rwhere
R: 'a,
self and passes that borrow into the pipe function. Read more§fn pipe_ref_mut<'a, R>(&'a mut self, func: impl FnOnce(&'a mut Self) -> R) -> Rwhere
R: 'a,
fn pipe_ref_mut<'a, R>(&'a mut self, func: impl FnOnce(&'a mut Self) -> R) -> Rwhere
R: 'a,
self and passes that borrow into the pipe function. Read more§fn pipe_borrow<'a, B, R>(&'a self, func: impl FnOnce(&'a B) -> R) -> R
fn pipe_borrow<'a, B, R>(&'a self, func: impl FnOnce(&'a B) -> R) -> R
§fn pipe_borrow_mut<'a, B, R>(
&'a mut self,
func: impl FnOnce(&'a mut B) -> R,
) -> R
fn pipe_borrow_mut<'a, B, R>( &'a mut self, func: impl FnOnce(&'a mut B) -> R, ) -> R
§fn pipe_as_ref<'a, U, R>(&'a self, func: impl FnOnce(&'a U) -> R) -> R
fn pipe_as_ref<'a, U, R>(&'a self, func: impl FnOnce(&'a U) -> R) -> R
self, then passes self.as_ref() into the pipe function.§fn pipe_as_mut<'a, U, R>(&'a mut self, func: impl FnOnce(&'a mut U) -> R) -> R
fn pipe_as_mut<'a, U, R>(&'a mut self, func: impl FnOnce(&'a mut U) -> R) -> R
self, then passes self.as_mut() into the pipe
function.§fn pipe_deref<'a, T, R>(&'a self, func: impl FnOnce(&'a T) -> R) -> R
fn pipe_deref<'a, T, R>(&'a self, func: impl FnOnce(&'a T) -> R) -> R
self, then passes self.deref() into the pipe function.§impl<T> Pointable for T
impl<T> Pointable for T
§impl<T> Tap for T
impl<T> Tap for T
§fn tap_borrow<B>(self, func: impl FnOnce(&B)) -> Self
fn tap_borrow<B>(self, func: impl FnOnce(&B)) -> Self
Borrow<B> of a value. Read more§fn tap_borrow_mut<B>(self, func: impl FnOnce(&mut B)) -> Self
fn tap_borrow_mut<B>(self, func: impl FnOnce(&mut B)) -> Self
BorrowMut<B> of a value. Read more§fn tap_ref<R>(self, func: impl FnOnce(&R)) -> Self
fn tap_ref<R>(self, func: impl FnOnce(&R)) -> Self
AsRef<R> view of a value. Read more§fn tap_ref_mut<R>(self, func: impl FnOnce(&mut R)) -> Self
fn tap_ref_mut<R>(self, func: impl FnOnce(&mut R)) -> Self
AsMut<R> view of a value. Read more§fn tap_deref<T>(self, func: impl FnOnce(&T)) -> Self
fn tap_deref<T>(self, func: impl FnOnce(&T)) -> Self
Deref::Target of a value. Read more§fn tap_deref_mut<T>(self, func: impl FnOnce(&mut T)) -> Self
fn tap_deref_mut<T>(self, func: impl FnOnce(&mut T)) -> Self
Deref::Target of a value. Read more§fn tap_dbg(self, func: impl FnOnce(&Self)) -> Self
fn tap_dbg(self, func: impl FnOnce(&Self)) -> Self
.tap() only in debug builds, and is erased in release builds.§fn tap_mut_dbg(self, func: impl FnOnce(&mut Self)) -> Self
fn tap_mut_dbg(self, func: impl FnOnce(&mut Self)) -> Self
.tap_mut() only in debug builds, and is erased in release
builds.§fn tap_borrow_dbg<B>(self, func: impl FnOnce(&B)) -> Self
fn tap_borrow_dbg<B>(self, func: impl FnOnce(&B)) -> Self
.tap_borrow() only in debug builds, and is erased in release
builds.§fn tap_borrow_mut_dbg<B>(self, func: impl FnOnce(&mut B)) -> Self
fn tap_borrow_mut_dbg<B>(self, func: impl FnOnce(&mut B)) -> Self
.tap_borrow_mut() only in debug builds, and is erased in release
builds.§fn tap_ref_dbg<R>(self, func: impl FnOnce(&R)) -> Self
fn tap_ref_dbg<R>(self, func: impl FnOnce(&R)) -> Self
.tap_ref() only in debug builds, and is erased in release
builds.§fn tap_ref_mut_dbg<R>(self, func: impl FnOnce(&mut R)) -> Self
fn tap_ref_mut_dbg<R>(self, func: impl FnOnce(&mut R)) -> Self
.tap_ref_mut() only in debug builds, and is erased in release
builds.§fn tap_deref_dbg<T>(self, func: impl FnOnce(&T)) -> Self
fn tap_deref_dbg<T>(self, func: impl FnOnce(&T)) -> Self
.tap_deref() only in debug builds, and is erased in release
builds.