[Git][ghc/ghc][wip/nested-cpr-2019] 55 commits: rts/CNF: Fix fixup comparison function

Sebastian Graf gitlab at gitlab.haskell.org
Mon May 18 11:54:19 UTC 2020 


Sebastian Graf pushed to branch wip/nested-cpr-2019 at Glasgow Haskell Compiler / GHC


Commits:
cf4f1e2f by Ben Gamari at 2020-05-13T02:02:33-04:00
rts/CNF: Fix fixup comparison function

Previously we would implicitly convert the difference between two words
to an int, resulting in an integer overflow on 64-bit machines.

Fixes #16992

- - - - -
a03da9bf by Ömer Sinan Ağacan at 2020-05-13T02:03:16-04:00
Pack some of IdInfo fields into a bit field

This reduces residency of compiler quite a bit on some programs.
Example stats when building T10370:

Before:

   2,871,242,832 bytes allocated in the heap
   4,693,328,008 bytes copied during GC
      33,941,448 bytes maximum residency (276 sample(s))
         375,976 bytes maximum slop
              83 MiB total memory in use (0 MB lost due to fragmentation)

After:

   2,858,897,344 bytes allocated in the heap
   4,629,255,440 bytes copied during GC
      32,616,624 bytes maximum residency (278 sample(s))
         314,400 bytes maximum slop
              80 MiB total memory in use (0 MB lost due to fragmentation)

So -3.9% residency, -1.3% bytes copied and -0.4% allocations.

Fixes #17497

Metric Decrease:
    T9233
    T9675

- - - - -
670c3e5c by Ben Gamari at 2020-05-13T02:03:54-04:00
get-win32-tarballs: Fix base URL

Revert a change previously made for testing purposes.

- - - - -
8ad8dc41 by Ben Gamari at 2020-05-13T02:03:54-04:00
get-win32-tarballs: Improve diagnostics output

- - - - -
8c0740b7 by Simon Jakobi at 2020-05-13T02:04:33-04:00
docs: Add examples for Data.Semigroup.Arg{Min,Max}

Context: #17153

- - - - -
cb22348f by Ben Gamari at 2020-05-13T02:05:11-04:00
Add few cleanups of the CAF logic

Give the NameSet of non-CAFfy names a proper newtype to distinguish it
from all of the other NameSets floating about.

- - - - -
90e38b81 by Emeka Nkurumeh at 2020-05-13T02:05:51-04:00
fix printf warning when using with ghc with clang on mingw
- - - - -
86d8ac22 by Sebastian Graf at 2020-05-13T02:06:29-04:00
CprAnal: Don't attach CPR sigs to expandable bindings (#18154)

Instead, look through expandable unfoldings in `cprTransform`.
See the new Note [CPR for expandable unfoldings]:

```
Long static data structures (whether top-level or not) like

  xs = x1 : xs1
  xs1 = x2 : xs2
  xs2 = x3 : xs3

should not get CPR signatures, because they

  * Never get WW'd, so their CPR signature should be irrelevant after analysis
    (in fact the signature might even be harmful for that reason)
  * Would need to be inlined/expanded to see their constructed product
  * Recording CPR on them blows up interface file sizes and is redundant with
    their unfolding. In case of Nested CPR, this blow-up can be quadratic!

But we can't just stop giving DataCon application bindings the CPR property,
for example

  fac 0 = 1
  fac n = n * fac (n-1)

fac certainly has the CPR property and should be WW'd! But FloatOut will
transform the first clause to

  lvl = 1
  fac 0 = lvl

If lvl doesn't have the CPR property, fac won't either. But lvl doesn't have a
CPR signature to extrapolate into a CPR transformer ('cprTransform'). So
instead we keep on cprAnal'ing through *expandable* unfoldings for these arity
0 bindings via 'cprExpandUnfolding_maybe'.

In practice, GHC generates a lot of (nested) TyCon and KindRep bindings, one
for each data declaration. It's wasteful to attach CPR signatures to each of
them (and intractable in case of Nested CPR).
```

Fixes #18154.

- - - - -
e34bf656 by Ben Gamari at 2020-05-13T02:07:08-04:00
users-guide: Add discussion of shared object naming

Fixes #18074.

- - - - -
5d0f2445 by Ben Gamari at 2020-05-13T02:07:47-04:00
testsuite: Print sign of performance changes

Executes the minor formatting change in the tabulated performance
changes suggested in #18135.

- - - - -
9e4b981f by Ben Gamari at 2020-05-13T02:08:24-04:00
testsuite: Add testcase for #18129

- - - - -
266310c3 by Ivan-Yudin at 2020-05-13T02:09:03-04:00
doc: Reformulate the opening paragraph of Ch. 4 in User's guide

Removes mentioning of Hugs
(it is not helpful for new users anymore).

Changes the wording for the rest of the paragraph.

Fixes #18132.

- - - - -
55e35c0b by Baldur Blöndal at 2020-05-13T20:02:48-04:00
Predicate, Equivalence derive via `.. -> a -> All'

- - - - -
d7e0b57f by Alp Mestanogullari at 2020-05-13T20:03:30-04:00
hadrian: add a --freeze2 option to freeze stage 1 and 2

- - - - -
d880d6b2 by Artem Pelenitsyn at 2020-05-13T20:04:11-04:00
Don't reload environment files on every setSessionDynFlags

Makes `interpretPackageEnv` (which loads envirinment files) a part of
`parseDynamicFlags` (parsing command-line arguments, which is typically
done once) instead of `setSessionDynFlags` (which is typically called
several times). Making several (transitive) calls to `interpretPackageEnv`,
as before, caused #18125 #16318, which should be fixed now.

- - - - -
102cfd67 by Ryan Scott at 2020-05-13T20:04:46-04:00
Factor out HsPatSigType for pat sigs/RULE term sigs (#16762)

This implements chunks (2) and (3) of
https://gitlab.haskell.org/ghc/ghc/issues/16762#note_270170. Namely,
it introduces a dedicated `HsPatSigType` AST type, which represents
the types that can appear in pattern signatures and term-level `RULE`
binders. Previously, these were represented with `LHsSigWcType`.
Although `LHsSigWcType` is isomorphic to `HsPatSigType`, the intended
semantics of the two types are slightly different, as evidenced by
the fact that they have different code paths in the renamer and
typechecker.

See also the new `Note [Pattern signature binders and scoping]` in
`GHC.Hs.Types`.

- - - - -
b17574f7 by Hécate at 2020-05-13T20:05:28-04:00
fix(documentation): Fix the RST links to GHC.Prim

- - - - -
df021fb1 by Baldur Blöndal at 2020-05-13T20:06:06-04:00
Document (->) using inferred quantification for its runtime representations.

Fixes #18142.

- - - - -
1a93ea57 by Takenobu Tani at 2020-05-13T20:06:54-04:00
Tweak man page for ghc command

This commit updates the ghc command's man page as followings:

* Enable `man_show_urls` to show URL addresses in the `DESCRIPTION`
section of ghc.rst, because sphinx currently removes hyperlinks
for man pages.

* Add a `SEE ALSO` section to point to the GHC homepage

- - - - -
a951e1ba by Takenobu Tani at 2020-05-13T20:07:37-04:00
GHCi: Add link to the user's guide in help message

This commit adds a link to the user's guide in ghci's
`:help` message.

Newcomers could easily reach to details of ghci.

- - - - -
404581ea by Jeff Happily at 2020-05-13T20:08:15-04:00
Handle single unused import

- - - - -
1c999e5d by Ben Gamari at 2020-05-13T20:09:07-04:00
Ensure that printMinimalImports closes handle

Fixes #18166.

- - - - -
c9f5a8f4 by Ben Gamari at 2020-05-13T20:09:51-04:00
hadrian: Tell testsuite driver about LLVM availability

This reflects the logic present in the Make build system into Hadrian.

Fixes #18167.

- - - - -
c05c0659 by Simon Jakobi at 2020-05-14T03:31:21-04:00
Improve some folds over Uniq[D]FM

* Replace some non-deterministic lazy folds with
  strict folds.
* Replace some O(n log n) folds in deterministic order
  with O(n) non-deterministic folds.
* Replace some folds with set-operations on the underlying
  IntMaps.

This reduces max residency when compiling
`nofib/spectral/simple/Main.hs` with -O0 by about 1%.

Maximum residency when compiling Cabal also seems reduced on the
order of 3-9%.

- - - - -
477f13bb by Simon Jakobi at 2020-05-14T03:31:58-04:00
Use Data.IntMap.disjoint

Data.IntMap gained a dedicated `disjoint` function in containers-0.6.2.1.

This patch applies this function where appropriate in hopes of modest
compiler performance improvements.

Closes #16806.

- - - - -
e9c0110c by Ben Gamari at 2020-05-14T12:25:53-04:00
IdInfo: Add reference to bitfield-packing ticket

- - - - -
9bd20e83 by Sebastian Graf at 2020-05-15T10:42:09-04:00
DmdAnal: Improve handling of precise exceptions

This patch does two things: Fix possible unsoundness in what was called
the "IO hack" and implement part 2.1 of the "fixing precise exceptions"
plan in
https://gitlab.haskell.org/ghc/ghc/wikis/fixing-precise-exceptions,
which, in combination with !2956, supersedes !3014 and !2525.

**IO hack**

The "IO hack" (which is a fallback to preserve precise exceptions
semantics and thus soundness, rather than some smart thing that
increases precision) is called `exprMayThrowPreciseException` now.
I came up with two testcases exemplifying possible unsoundness (if
twisted enough) in the old approach:

- `T13380d`: Demonstrating unsoundness of the "IO hack" when resorting
             to manual state token threading and direct use of primops.
             More details below.
- `T13380e`: Demonstrating unsoundness of the "IO hack" when we have
             Nested CPR. Not currently relevant, as we don't have Nested
             CPR yet.
- `T13380f`: Demonstrating unsoundness of the "IO hack" for safe FFI
             calls.

Basically, the IO hack assumed that precise exceptions can only be
thrown from a case scrutinee of type `(# State# RealWorld, _ #)`. I
couldn't come up with a program using the `IO` abstraction that violates
this assumption. But it's easy to do so via manual state token threading
and direct use of primops, see `T13380d`. Also similar code might be
generated by Nested CPR in the (hopefully not too) distant future, see
`T13380e`. Hence, we now have a more careful test in `forcesRealWorld`
that passes `T13380{d,e}` (and will hopefully be robust to Nested CPR).

**Precise exceptions**

In #13380 and #17676 we saw that we didn't preserve precise exception
semantics in demand analysis. We fixed that with minimal changes in
!2956, but that was terribly unprincipled.

That unprincipledness resulted in a loss of precision, which is tracked
by these new test cases:

- `T13380b`: Regression in dead code elimination, because !2956 was too
             syntactic about `raiseIO#`
- `T13380c`: No need to apply the "IO hack" when the IO action may not
             throw a precise exception (and the existing IO hack doesn't
             detect that)

Fixing both issues in !3014 turned out to be too complicated and had
the potential to regress in the future. Hence we decided to only fix
`T13380b` and augment the `Divergence` lattice with a new middle-layer
element, `ExnOrDiv`, which means either `Diverges` (, throws an
imprecise exception) or throws a *precise* exception.

See the wiki page on Step 2.1 for more implementational details:
https://gitlab.haskell.org/ghc/ghc/wikis/fixing-precise-exceptions#dead-code-elimination-for-raiseio-with-isdeadenddiv-introducing-exnordiv-step-21

- - - - -
568d7279 by Ben Gamari at 2020-05-15T10:42:46-04:00
GHC.Cmm.Opt: Handle MO_XX_Conv

This MachOp was introduced by 2c959a1894311e59cd2fd469c1967491c1e488f3
but a wildcard match in cmmMachOpFoldM hid the fact that it wasn't
handled. Ideally we would eliminate the match but this appears to be a
larger task.

Fixes #18141.

- - - - -
6f2dd7d1 by Sebastian Graf at 2020-05-17T18:05:28+02:00
Nested CPR

- - - - -
d8ef66c6 by Sebastian Graf at 2020-05-17T18:05:30+02:00
Move tests from stranal to cpranal

- - - - -
5f4a67bc by Sebastian Graf at 2020-05-17T18:05:30+02:00
Accept FacState

- - - - -
84aaab65 by Sebastian Graf at 2020-05-17T18:09:20+02:00
Factor Cpr and Termination into a joint lattice

As a result, we don't even have to export Termination from Cpr. Neat!

Also I realised there is a simpler and more sound way to generate and
unleash CPR signatures.

- - - - -
c2168883 by Sebastian Graf at 2020-05-17T18:09:21+02:00
Consider unboxing effects of WW better and get rid of hack

- - - - -
64e8973c by Sebastian Graf at 2020-05-17T18:09:21+02:00
stuff

- - - - -
58c5e7a2 by Sebastian Graf at 2020-05-17T18:09:21+02:00
Debug output

- - - - -
28fd1541 by Sebastian Graf at 2020-05-17T18:10:15+02:00
A slew of testsuite changes

- - - - -
8db32a83 by Sebastian Graf at 2020-05-17T18:10:17+02:00
Fix T1600

- - - - -
0122dab8 by Sebastian Graf at 2020-05-17T18:10:17+02:00
Inline `decodeDoubleInteger` and constant-fold `decodeDouble_Int64#` instead

Currently, `decodeDoubleInteger` is known-key so that it can be
recognised in constant folding. But that is very brittle and doesn't
survive worker/wrapper, which we even do for
`NOINLINE`/`CONSTANT_FOLDED` things since #13143.
Also it is a trade-off: The implementation of `decodeDoubleInteger`
allocates an `Integer` box that never cancels aways if we don't inline
it.

Hence we recognise the `decodeDouble_Int64#` primop instead in constant
folding, so that we can inline `decodeDoubleInteger`.

You may wonder how this affects performance of code using
`integer-simple`; Apparently, according to @hsyl20 this is not a concern
since we will hopefully land !2231 soon.

Fixes #18092.

- - - - -
a68c809a by Sebastian Graf at 2020-05-17T18:10:17+02:00
Fix primop termination

- - - - -
0df4b24f by Sebastian Graf at 2020-05-17T18:10:17+02:00
Test for DataCon wrapper CPR

- - - - -
d671da8a by Sebastian Graf at 2020-05-17T18:11:16+02:00
Fix CPR of bottoming functions/primops

- - - - -
d6c0898e by Sebastian Graf at 2020-05-17T18:11:18+02:00
Fix DataConWrapperCpr and accept other test outputs

- - - - -
a9a6d4df by Sebastian Graf at 2020-05-17T18:11:18+02:00
Accept two more changed test outputs

- - - - -
e445765f by Sebastian Graf at 2020-05-17T18:11:23+02:00
Update CaseBinderCPR with a new function

- - - - -
20fab53c by Sebastian Graf at 2020-05-17T18:11:25+02:00
Don't give the case binder the CPR property

- - - - -
a1c73a85 by Sebastian Graf at 2020-05-17T18:11:25+02:00
Prune CPR sigs to constant depth on all bindings

- - - - -
ea1c7ed1 by Sebastian Graf at 2020-05-17T18:11:25+02:00
Use variable length coding for ConTags

- - - - -
37796cdb by Sebastian Graf at 2020-05-17T18:11:28+02:00
Accept testuite output

- - - - -
edd8801e by Sebastian Graf at 2020-05-17T18:13:00+02:00
Don't attach CPR sigs to expandable bindings; transform their unfoldings instead

- - - - -
88f8e954 by Sebastian Graf at 2020-05-17T18:13:02+02:00
Revert "Don't give the case binder the CPR property"

This reverts commit 910edd76d5fe68b58c74f3805112f9faef4f2788.

It seems we broke too much with this change. We lost our big win in
`fish`.

- - - - -
654f6727 by Sebastian Graf at 2020-05-17T18:13:02+02:00
A more modular and configurable approach to optimistic case binder CPR

- - - - -
b017a272 by Sebastian Graf at 2020-05-17T18:13:02+02:00
Fix T9291

- - - - -
5324d98d by Sebastian Graf at 2020-05-17T18:13:02+02:00
Document -fcase-binder-cpr-depth in the user's guide

- - - - -
a572aee9 by Sebastian Graf at 2020-05-18T13:14:25+02:00
Rebase woes

- - - - -
0c2eedad by Sebastian Graf at 2020-05-18T13:36:40+02:00
Testsuite changes

- - - - -


30 changed files:

- compiler/GHC.hs
- compiler/GHC/Builtin/Types/Prim.hs
- compiler/GHC/Builtin/primops.txt.pp
- compiler/GHC/Cmm/Info/Build.hs
- compiler/GHC/Cmm/Opt.hs
- compiler/GHC/CmmToAsm/Reg/Graph/SpillClean.hs
- compiler/GHC/Core/Arity.hs
- compiler/GHC/Core/Coercion.hs
- compiler/GHC/Core/FamInstEnv.hs
- compiler/GHC/Core/Lint.hs
- compiler/GHC/Core/Make.hs
- compiler/GHC/Core/Opt/CallArity.hs
- compiler/GHC/Core/Opt/ConstantFold.hs
- compiler/GHC/Core/Opt/CprAnal.hs
- compiler/GHC/Core/Opt/DmdAnal.hs
- compiler/GHC/Core/Opt/FloatIn.hs
- compiler/GHC/Core/Opt/FloatOut.hs
- compiler/GHC/Core/Opt/LiberateCase.hs
- compiler/GHC/Core/Opt/OccurAnal.hs
- compiler/GHC/Core/Opt/SetLevels.hs
- compiler/GHC/Core/Opt/Simplify.hs
- compiler/GHC/Core/Opt/Simplify/Utils.hs
- compiler/GHC/Core/Opt/SpecConstr.hs
- compiler/GHC/Core/Opt/Specialise.hs
- compiler/GHC/Core/Opt/WorkWrap.hs
- compiler/GHC/Core/Opt/WorkWrap/Utils.hs
- compiler/GHC/Core/SimpleOpt.hs
- compiler/GHC/Core/TyCo/FVs.hs
- compiler/GHC/Core/TyCo/Rep.hs
- compiler/GHC/Core/TyCo/Subst.hs


The diff was not included because it is too large.


View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/5397fe394009f07b0c42977af1bc1030c2592b8a...0c2eedad466fabeb8367a3d7617f13e87df9778f

-- 
View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/5397fe394009f07b0c42977af1bc1030c2592b8a...0c2eedad466fabeb8367a3d7617f13e87df9778f
You're receiving this email because of your account on gitlab.haskell.org.


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.haskell.org/pipermail/ghc-commits/attachments/20200518/bac771b3/attachment-0001.html>

More information about the ghc-commits mailing list