#4342: Review containers changes
----------------------------------+-----------------------------------------
Reporter: igloo | Owner:
Type: bug | Status: new
Priority: high | Milestone: 7.2.1
Component: libraries (other) | Version: 7.1
Keywords: | Testcase:
Blockedby: | Difficulty:
Os: Unknown/Multiple | Blocking:
Architecture: Unknown/Multiple | Failure: None/Unknown
----------------------------------+-----------------------------------------
Changes (by milan):
* cc: f...@… (added)
Comment:
Most of the allocation overhead is causes by the INLINABLE and INLINE
pragmas.
Here are my numbers:
* T1969:
- ghc-head, containers-from-ghc-7.0-branch: 248471644
- ghc-head, containers-dons/tibbe/milan: 282720916
- ghc-head, containers-dons/tibbe/milan + all INLINE converted to
INLINABLE: 282689256
- ghc-head, containers-dons/tibbe/milan + all INLINE and INLINABLE
removed: 256416620
* T3294:
- ghc-head, containers-from-ghc-7.0-branch: 826836708
- ghc-head, containers-dons/tibbe/milan: 943910592
- ghc-head, containers-dons/tibbe/milan + all INLINE converted to
INLINABLE: 943819492
- ghc-head, containers-dons/tibbe/milan + all INLINE and INLINABLE
removed: 859844404
By removing INLINE and INLINABLE annotations, the allocation is higher by
3.2% and 3.9%, instead of 15% and 14%.
What is also worth mentioning is that with all methods INLINABLE the
allocation is approximately the same as with some explicit INLINE.
Maybe we can remove the INLINE / INLINABLE annotations for know?
I will investigate further after I return from ICFP.
--
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/4342#comment:2>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler
_______________________________________________
Glasgow-haskell-bugs mailing list
[email protected]
http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
