Re: [GHC] #7162: RULES that never fire (automatically)
#7162: RULES that never fire (automatically)
---+
Reporter: andygill | Owner:
Type: feature request | Status: closed
Priority: normal| Milestone: 7.8.1
Component: Compiler |Version: 7.7
Resolution: fixed | Keywords:
Os: Unknown/Multiple | Architecture: Unknown/Multiple
Failure: None/Unknown | Difficulty: Unknown
Testcase:| Blockedby:
Blocking:|Related:
---+
Changes (by simonpj):
* status: patch = closed
* resolution: = fixed
Comment:
OK I've pushed this
{{{
commit c56c355b0f5504f8afd062ca4c78a8bb40905299
Author: Andrew Farmer afar...@ittc.ku.edu
Date: Thu Dec 6 15:47:33 2012 -0600
Only allow special NeverActive syntax for RULES.
compiler/parser/Parser.y.pp | 17 +
docs/users_guide/glasgow_exts.xml |3 +++
2 files changed, 16 insertions(+), 4 deletions(-)
}}}
plus an extra sentence in the docs
{{{
commit d3e2912ac2048346828539e0dfef6c0cefef0d38
Author: Simon Peyton Jones simo...@microsoft.com
Date: Fri Dec 21 13:17:26 2012 +
Clarify documentation of [~] syntax on RULE activations
docs/users_guide/glasgow_exts.xml |3 +++
1 file changed, 3 insertions(+)
}}}
--
Ticket URL: http://hackage.haskell.org/trac/ghc/ticket/7162#comment:13
GHC http://www.haskell.org/ghc/
The Glasgow Haskell Compiler
___
Glasgow-haskell-bugs mailing list
Glasgow-haskell-bugs@haskell.org
http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
Re: [GHC] #7162: RULES that never fire (automatically)
#7162: RULES that never fire (automatically) ---+ Reporter: andygill | Owner: Type: feature request | Status: patch Priority: normal| Milestone: 7.8.1 Component: Compiler |Version: 7.7 Resolution:| Keywords: Os: Unknown/Multiple | Architecture: Unknown/Multiple Failure: None/Unknown | Difficulty: Unknown Testcase:| Blockedby: Blocking:|Related: ---+ Changes (by afarmer): * status: new = patch Comment: Here are two patches that implement option 1. This includes a note in the user's guide about [~] on RULES. -- Ticket URL: http://hackage.haskell.org/trac/ghc/ticket/7162#comment:12 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler ___ Glasgow-haskell-bugs mailing list Glasgow-haskell-bugs@haskell.org http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
Re: [GHC] #7162: RULES that never fire (automatically)
#7162: RULES that never fire (automatically) ---+ Reporter: andygill | Owner: Type: feature request | Status: new Priority: normal| Milestone: 7.8.1 Component: Compiler |Version: 7.7 Resolution:| Keywords: Os: Unknown/Multiple | Architecture: Unknown/Multiple Failure: None/Unknown | Difficulty: Unknown Testcase:| Blockedby: Blocking:|Related: ---+ Comment(by simonpj): Yes (1) looks best to me. Alternatively, you could make `[~]` parse as an activation but check in INLINE and SPECIALISE productions that it didn't come back as `NeverActive`. Regardless, please add comments in the parser to explain what is going on. -- Ticket URL: http://hackage.haskell.org/trac/ghc/ticket/7162#comment:11 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler ___ Glasgow-haskell-bugs mailing list Glasgow-haskell-bugs@haskell.org http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
Re: [GHC] #7162: RULES that never fire (automatically)
#7162: RULES that never fire (automatically)
---+
Reporter: andygill | Owner:
Type: feature request | Status: new
Priority: normal| Milestone: 7.8.1
Component: Compiler |Version: 7.7
Resolution:| Keywords:
Os: Unknown/Multiple | Architecture: Unknown/Multiple
Failure: None/Unknown | Difficulty: Unknown
Testcase:| Blockedby:
Blocking:|Related:
---+
Comment(by afarmer):
So while documenting this, I realized this patch makes an unintended
change. Namely, the INLINE pragma uses the same parser rules for
activation as the RULES pragma. So it is now possible to write:
{-# INLINE [~] ... #-}
and
{-# NOINLINE [~] ... #-}
Looking at RdrHsSyn.lhs, the first case (INLINE [~]) is equivalent to
NOINLINE. Worryingly, I have yet to figure out how the second case
(NOINLINE [~]) behaves.
So as I see it, there are two possible solutions.
1. I submit a new patch to the parser that separates the activation
parsing for RULES and INLINE so RULES [~] is possible but INLINE [~] is
not.
2. I figure out what NOINLINE [~] means and document that.
Opinions? I'm inclined towards option 1, as I think it'll be less
confusing for the user.
--
Ticket URL: http://hackage.haskell.org/trac/ghc/ticket/7162#comment:10
GHC http://www.haskell.org/ghc/
The Glasgow Haskell Compiler
___
Glasgow-haskell-bugs mailing list
Glasgow-haskell-bugs@haskell.org
http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
Re: [GHC] #7162: RULES that never fire (automatically)
#7162: RULES that never fire (automatically)
-+--
Reporter: andygill | Owner:
Type: feature request | Status: patch
Priority: normal| Milestone: 7.8.1
Component: Compiler | Version: 7.7
Keywords:| Os: Unknown/Multiple
Architecture: Unknown/Multiple | Failure: None/Unknown
Difficulty: Unknown |Testcase:
Blockedby:|Blocking:
Related:|
-+--
Comment(by afarmer@…):
commit 779f10fdedbeeb31891f3f3772cbc2dfdd32e1ee
{{{
Author: Andrew Farmer afar...@ittc.ku.edu
Date: Thu Oct 4 16:51:28 2012 -0500
Extend parser to allow specification of RULES that never fire. #7162
compiler/parser/Parser.y.pp |1 +
1 files changed, 1 insertions(+), 0 deletions(-)
}}}
--
Ticket URL: http://hackage.haskell.org/trac/ghc/ticket/7162#comment:7
GHC http://www.haskell.org/ghc/
The Glasgow Haskell Compiler
___
Glasgow-haskell-bugs mailing list
Glasgow-haskell-bugs@haskell.org
http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
Re: [GHC] #7162: RULES that never fire (automatically)
#7162: RULES that never fire (automatically) ---+ Reporter: andygill | Owner: Type: feature request | Status: closed Priority: normal| Milestone: 7.8.1 Component: Compiler |Version: 7.7 Resolution: fixed | Keywords: Os: Unknown/Multiple | Architecture: Unknown/Multiple Failure: None/Unknown | Difficulty: Unknown Testcase:| Blockedby: Blocking:|Related: ---+ Changes (by igloo): * status: patch = closed * resolution: = fixed Comment: Applied thanks - and thanks for the test too! -- Ticket URL: http://hackage.haskell.org/trac/ghc/ticket/7162#comment:8 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler ___ Glasgow-haskell-bugs mailing list Glasgow-haskell-bugs@haskell.org http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
Re: [GHC] #7162: RULES that never fire (automatically)
#7162: RULES that never fire (automatically) ---+ Reporter: andygill | Owner: Type: feature request | Status: new Priority: normal| Milestone: 7.8.1 Component: Compiler |Version: 7.7 Resolution:| Keywords: Os: Unknown/Multiple | Architecture: Unknown/Multiple Failure: None/Unknown | Difficulty: Unknown Testcase:| Blockedby: Blocking:|Related: ---+ Changes (by simonpj): * status: closed = new * resolution: fixed = Comment: Whoa! Where's the documentation? Andrew, can you send a patch to the user manual, please? Simon -- Ticket URL: http://hackage.haskell.org/trac/ghc/ticket/7162#comment:9 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler ___ Glasgow-haskell-bugs mailing list Glasgow-haskell-bugs@haskell.org http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
Re: [GHC] #7162: RULES that never fire (automatically)
#7162: RULES that never fire (automatically) -+-- Reporter: andygill | Owner: Type: feature request | Status: patch Priority: normal| Milestone: 7.8.1 Component: Compiler | Version: 7.7 Keywords:| Os: Unknown/Multiple Architecture: Unknown/Multiple | Failure: None/Unknown Difficulty: Unknown |Testcase: Blockedby:|Blocking: Related:| -+-- Changes (by igloo): * milestone: = 7.8.1 Comment: Thanks for the patch. We'll take a look. -- Ticket URL: http://hackage.haskell.org/trac/ghc/ticket/7162#comment:4 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler ___ Glasgow-haskell-bugs mailing list Glasgow-haskell-bugs@haskell.org http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
Re: [GHC] #7162: RULES that never fire (automatically)
#7162: RULES that never fire (automatically) -+-- Reporter: andygill | Owner: Type: feature request | Status: patch Priority: normal| Milestone: 7.8.1 Component: Compiler | Version: 7.7 Keywords:| Os: Unknown/Multiple Architecture: Unknown/Multiple | Failure: None/Unknown Difficulty: Unknown |Testcase: Blockedby:|Blocking: Related:| -+-- Comment(by afarmer): I think Andy had originally hoped that this would make it into 7.6.2. Is that still possible? It's a very small change. -- Ticket URL: http://hackage.haskell.org/trac/ghc/ticket/7162#comment:5 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler ___ Glasgow-haskell-bugs mailing list Glasgow-haskell-bugs@haskell.org http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
Re: [GHC] #7162: RULES that never fire (automatically)
#7162: RULES that never fire (automatically) -+-- Reporter: andygill | Owner: Type: feature request | Status: patch Priority: normal| Milestone: 7.8.1 Component: Compiler | Version: 7.7 Keywords:| Os: Unknown/Multiple Architecture: Unknown/Multiple | Failure: None/Unknown Difficulty: Unknown |Testcase: Blockedby:|Blocking: Related:| -+-- Comment(by igloo): We have a policy of not making interface changes in stable branches, sorry -- Ticket URL: http://hackage.haskell.org/trac/ghc/ticket/7162#comment:6 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler ___ Glasgow-haskell-bugs mailing list Glasgow-haskell-bugs@haskell.org http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
Re: [GHC] #7162: RULES that never fire (automatically)
#7162: RULES that never fire (automatically) -+-- Reporter: andygill | Owner: Type: feature request | Status: patch Priority: normal| Milestone: Component: Compiler | Version: 7.4.2 Keywords:| Os: Unknown/Multiple Architecture: Unknown/Multiple | Failure: None/Unknown Difficulty: Unknown |Testcase: Blockedby:|Blocking: Related:| -+-- Changes (by afarmer): * status: new = patch Comment: Here is a patch that implements this functionality (with [~] meaning never automatically fire), as well as a patch to the testsuite to verify the parser works. I've built HERMIT against head (with this patch applied) and verified that we can still use RULES that are marked to never fire (at both ends of the optimization pipeline). As an aside, all the packages required by HERMIT compiled fine, so it doesn't seem to break any existing code. I've also run the ghc validate script, with no extra unexpected errors. -- Ticket URL: http://hackage.haskell.org/trac/ghc/ticket/7162#comment:2 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler ___ Glasgow-haskell-bugs mailing list Glasgow-haskell-bugs@haskell.org http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
Re: [GHC] #7162: RULES that never fire (automatically)
#7162: RULES that never fire (automatically) -+-- Reporter: andygill | Owner: Type: feature request | Status: patch Priority: normal| Milestone: Component: Compiler | Version: 7.7 Keywords:| Os: Unknown/Multiple Architecture: Unknown/Multiple | Failure: None/Unknown Difficulty: Unknown |Testcase: Blockedby:|Blocking: Related:| -+-- Changes (by afarmer): * version: 7.4.2 = 7.7 -- Ticket URL: http://hackage.haskell.org/trac/ghc/ticket/7162#comment:3 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler ___ Glasgow-haskell-bugs mailing list Glasgow-haskell-bugs@haskell.org http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
Re: [GHC] #7162: RULES that never fire (automatically)
#7162: RULES that never fire (automatically)
-+--
Reporter: andygill | Owner:
Type: feature request | Status: new
Priority: normal| Milestone:
Component: Compiler | Version: 7.4.2
Keywords:| Os: Unknown/Multiple
Architecture: Unknown/Multiple | Failure: None/Unknown
Difficulty: Unknown |Testcase:
Blockedby:|Blocking:
Related:|
-+--
Changes (by simonpj):
* difficulty: = Unknown
Old description:
We want a way of having GHC RULES known by GHC, but not used by the
optimizer.
HERMIT, a interactive plugin for GHC that applies rules - and as well as
built in rules (like alpha conversion, beta-reduction, etc) - also
provide access to the named GHC RULES. Here is the rub: We want to use
GHC RULES that are parsed and typed checked like normal rules, are
visible to the HERMIT system, but never run by the simplifier. Currently
we can say
- attempt this *before* this (opt) pass, or
- attempt *after* this pass, there is no way of saying
- *never* attempt.
We were thinking
{-# RULES [~] map/map forall f g . map f (map g xs) = map (f.g) xs
#-}
Where the [~] says *never* execute this without be explicitly asked,
following on from the [~0] which does not run in first pass.
We happy making the required changes.
New description:
We want a way of having GHC RULES known by GHC, but not used by the
optimizer.
HERMIT, a interactive plugin for GHC that applies rules - and as well as
built in rules (like alpha conversion, beta-reduction, etc) - also provide
access to the named GHC RULES. Here is the rub: We want to use GHC RULES
that are parsed and typed checked like normal rules, are visible to the
HERMIT system, but never run by the simplifier. Currently we can say
- attempt this '''before''' this (opt) pass, or
- attempt '''after''' this pass,
but there is no way of saying
- '''never''' attempt.
We were thinking
{{{
{-# RULES [~] map/map forall f g . map f (map g xs) = map (f.g) xs
#-}
}}}
Where the `[~]` says *never* execute this without be explicitly asked,
following on from the `[~0]` which does not run in first pass.
We are happy making the required changes.
--
Comment:
I think I'm ok with this. All the machinery is there already except for
the concrete syntax to say never execute.
The only question in my mind is whether to be a bit more explicit by
saying
{{{
{-# RULES [NEVER] map/map ... #-}
}}}
but that means more work in the lexer, so it probably isn't worth it. I
don't feel strongly.
Simon
--
Ticket URL: http://hackage.haskell.org/trac/ghc/ticket/7162#comment:1
GHC http://www.haskell.org/ghc/
The Glasgow Haskell Compiler
___
Glasgow-haskell-bugs mailing list
Glasgow-haskell-bugs@haskell.org
http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
[GHC] #7162: RULES that never fire (automatically)
#7162: RULES that never fire (automatically)
--+-
Reporter: andygill | Owner:
Type: feature request | Status: new
Priority: normal| Component: Compiler
Version: 7.4.2 | Keywords:
Os: Unknown/Multiple | Architecture: Unknown/Multiple
Failure: None/Unknown | Testcase:
Blockedby:| Blocking:
Related:|
--+-
We want a way of having GHC RULES known by GHC, but not used by the
optimizer.
HERMIT, a interactive plugin for GHC that applies rules - and as well as
built in rules (like alpha conversion, beta-reduction, etc) - also provide
access to the named GHC RULES. Here is the rub: We want to use GHC RULES
that are parsed and typed checked like normal rules, are visible to the
HERMIT system, but never run by the simplifier. Currently we can say
- attempt this *before* this (opt) pass, or
- attempt *after* this pass, there is no way of saying
- *never* attempt.
We were thinking
{-# RULES [~] map/map forall f g . map f (map g xs) = map (f.g) xs
#-}
Where the [~] says *never* execute this without be explicitly asked,
following on from the [~0] which does not run in first pass.
We happy making the required changes.
--
Ticket URL: http://hackage.haskell.org/trac/ghc/ticket/7162
GHC http://www.haskell.org/ghc/
The Glasgow Haskell Compiler
___
Glasgow-haskell-bugs mailing list
Glasgow-haskell-bugs@haskell.org
http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
