diff options
| author | Sebastian Graf <sebastian.graf@kit.edu> | 2019-01-28 16:49:04 +0100 |
|---|---|---|
| committer | Sebastian Graf <sebastian.graf@kit.edu> | 2020-02-12 11:00:58 +0100 |
| commit | 059c3c9d7c84fc37c69e9f414ff736d47081e72c (patch) | |
| tree | da3c17ac002b9c6d31542af78553769fd40d5d65 /compiler/coreSyn/CoreSeq.hs | |
| parent | f0c0ee7d9a942a19361e72553cd08f42cc12b04a (diff) | |
| download | haskell-wip/sep-cpr.tar.gz | |
Separate CPR analysis from the Demand analyserwip/sep-cpr
The reasons for that can be found in the wiki:
https://gitlab.haskell.org/ghc/ghc/wikis/nested-cpr/split-off-cpr
We now run CPR after demand analysis (except for after the final demand
analysis run just before code gen). CPR got its own dump flags
(`-ddump-cpr-anal`, `-ddump-cpr-signatures`), but not its own flag to
activate/deactivate. It will run with `-fstrictness`/`-fworker-wrapper`.
As explained on the wiki page, this step is necessary for a sane Nested
CPR analysis. And it has quite positive impact on compiler performance:
Metric Decrease:
T9233
T9675
T9961
T15263
Diffstat (limited to 'compiler/coreSyn/CoreSeq.hs')
| -rw-r--r-- | compiler/coreSyn/CoreSeq.hs | 2 |
1 files changed, 2 insertions, 0 deletions
diff --git a/compiler/coreSyn/CoreSeq.hs b/compiler/coreSyn/CoreSeq.hs index 7de8923a71..aa94a24215 100644 --- a/compiler/coreSyn/CoreSeq.hs +++ b/compiler/coreSyn/CoreSeq.hs @@ -15,6 +15,7 @@ import GhcPrelude import CoreSyn import IdInfo import Demand( seqDemand, seqStrictSig ) +import Cpr( seqCprSig ) import BasicTypes( seqOccInfo ) import VarSet( seqDVarSet ) import Var( varType, tyVarKind ) @@ -34,6 +35,7 @@ megaSeqIdInfo info seqDemand (demandInfo info) `seq` seqStrictSig (strictnessInfo info) `seq` + seqCprSig (cprInfo info) `seq` seqCaf (cafInfo info) `seq` seqOneShot (oneShotInfo info) `seq` seqOccInfo (occInfo info) |