[Git][ghc/ghc][wip/gc/aligned-block-allocation] 40 commits: removing x87 register support from native code gen

Ömer Sinan Ağacan gitlab at gitlab.haskell.org
Wed Apr 17 13:45:14 UTC 2019 


Ömer Sinan Ağacan pushed to branch wip/gc/aligned-block-allocation at Glasgow Haskell Compiler / GHC


Commits:
42504f4a by Carter Schonwald at 2019-04-11T00:28:41Z
removing x87 register support from native code gen

* simplifies registers to have GPR, Float and Double, by removing the SSE2 and X87 Constructors
* makes -msse2 assumed/default for x86 platforms, fixing a long standing nondeterminism in rounding
behavior in 32bit haskell code
* removes the 80bit floating point representation from the supported float sizes
* theres still 1 tiny bit of x87 support needed,
for handling float and double return values in FFI calls  wrt the C ABI on x86_32,
but this one piece does not leak into the rest of NCG.
* Lots of code thats not been touched in a long time got deleted as a
consequence of all of this

all in all, this change paves the way towards a lot of future further
improvements in how GHC handles floating point computations, along with
making the native code gen more accessible to a larger pool of contributors.

- - - - -
c401f8a4 by Sylvain Henry at 2019-04-11T23:51:24Z
Hadrian: fix binary-dir with --docs=none

Hadrian's "binary-dist" target must check that the "docs" directory
exists (it may not since we can disable docs generation).

- - - - -
091195a4 by Ömer Sinan Ağacan at 2019-04-11T23:57:38Z
Remove unused remilestoning script

- - - - -
fa0ccbb8 by Ömer Sinan Ağacan at 2019-04-11T23:57:38Z
Update a panic message

Point users to the right URL

- - - - -
beaa07d2 by Sylvain Henry at 2019-04-12T17:17:21Z
Hadrian: fix ghci wrapper script generation (#16508)

- - - - -
e05df3e1 by Ben Gamari at 2019-04-12T17:23:30Z
gitlab-ci: Ensure that version number has three components

- - - - -
885d2e04 by klebinger.andreas at gmx.at at 2019-04-12T18:40:04Z
Add -ddump-stg-final to dump stg as it is used for codegen.

Intermediate STG does not contain free variables which can be useful
sometimes. So adding a flag to dump that info.

- - - - -
3c759ced by Alp Mestanogullari at 2019-04-12T18:46:54Z
Hadrian: add a --test-accept/-a flag, to mimic 'make accept'

When -a or --test-accept is passed, and if one runs the 'test' target, then
any test failing because of mismatching output and which is not expected to
fail will have its expected output adjusted by the test driver, effectively
considering the new output correct from now on.

When this flag is passed, hadrian's 'test' target becomes sensitive to the
PLATFORM and OS environment variable, just like the Make build system:
- when the PLATFORM env var is set to "YES", when accepting a result, accept it
  for the current platform;
- when the OS env var is set to "YES", when accepting a result, accept it
  for all wordsizes of the current operating system.

This can all be combined with `--only="..."` and `TEST="..." to only accept
the new output of a subset of tests.

- - - - -
f4b5a6c0 by Alp Mestanogullari at 2019-04-12T18:46:54Z
Hadrian: document -a/--test-accept

- - - - -
30a0988d by Ben Gamari at 2019-04-12T19:41:07Z
gitlab: Disable windows-hadrian job

Not only is it reliably failing due to #16574 but all of the quickly
failing builds also causes the Windows runners to run out of disk space.

- - - - -
8870a51b by Ben Gamari at 2019-04-12T19:41:07Z
gitlab: Don't run lint-submods job on Marge branches

This broke Marge by creating a second pipeline (consisting of only the
`lint-submods` job). Marge then looked at this pipeline and concluded
that CI for her merge branch passed. However, this is ignores the fact
that the majority of the CI jobs are triggered on `merge_request` and
are therefore in another pipeline.

- - - - -
7876d088 by Ben Gamari at 2019-04-13T13:51:59Z
linters: Fix check-version-number

This should have used `grep -E`, not `grep -e`
- - - - -
2e7b2e55 by Ara Adkins at 2019-04-13T14:00:02Z
[skip ci] Update CI badge in readme

This trivial MR updates the CI badge in the readme to point to the
new CI on gitlab, rather than the very out-of-date badge from
Travis.

- - - - -
40848a43 by Ben Gamari at 2019-04-13T14:02:36Z
base: Better document implementation implications of Data.Timeout

As noted in #16546 timeout uses asynchronous exceptions internally, an
implementation detail which can leak out in surprising ways.  Note this
fact.

Also expose the `Timeout` tycon.

[skip ci]

- - - - -
5f183081 by David Eichmann at 2019-04-14T05:08:15Z
Hadrian: add rts shared library symlinks for backwards compatability

Fixes test T3807 when building with Hadrian.

Trac #16370

- - - - -
9b142c53 by Sylvain Henry at 2019-04-14T05:14:23Z
Hadrian: add binary-dist-dir target

This patch adds an Hadrian target "binary-dist-dir". Compared to
"binary-dist", it only builds a binary distribution directory without
creating the Tar archive. It makes the use/test of the bindist
installation script easier.

- - - - -
6febc444 by Krzysztof Gogolewski at 2019-04-14T05:20:29Z
Fix assertion failures reported in #16533

- - - - -
edcef7b3 by Artem Pyanykh at 2019-04-14T05:26:35Z
codegen: unroll memcpy calls for small bytearrays

- - - - -
6094d43f by Artem Pyanykh at 2019-04-14T05:26:35Z
docs: mention memcpy optimization for ByteArrays in 8.10.1-notes

- - - - -
d2271fe4 by Simon Jakobi at 2019-04-14T12:43:17Z
Ord docs: Add explanation on 'min' and 'max' operator interactions

[ci skip]

- - - - -
e7cad16c by Krzysztof Gogolewski at 2019-04-14T12:49:23Z
Add a safeguard to Core Lint

Lint returns a pair (Maybe a, WarnsAndErrs). The Maybe monad
allows to handle an unrecoverable failure.
In case of such a failure, the error should be added to the second
component of the pair. If this is not done, Lint will silently
accept bad programs. This situation actually happened during
development of linear types. This adds a safeguard.

- - - - -
c54a093f by Ben Gamari at 2019-04-14T12:55:29Z
CODEOWNERS: Add simonmar as owner of rts/linker

I suspect this is why @simonmar wasn't notified of !706.

[skip ci]

- - - - -
1825f50d by Alp Mestanogullari at 2019-04-14T13:01:38Z
Hadrian: don't accept p_dyn for executables, to fix --flavour=prof

- - - - -
b024e289 by Giles Anderson at 2019-04-15T10:20:29Z
Document how -O3 is handled by GHC

    -O2 is the highest value of optimization.
    -O3 will be reverted to -O2.

- - - - -
4b1ef06d by Giles Anderson at 2019-04-15T10:20:29Z
Apply suggestion to docs/users_guide/using-optimisation.rst
- - - - -
71cf94db by Fraser Tweedale at 2019-04-15T10:26:37Z
GHCi: fix load order of .ghci files

Directives in .ghci files in the current directory ("local .ghci")
can be overridden by global files.  Change the order in which the
configs are loaded: global and $HOME/.ghci first, then local.

Also introduce a new field to GHCiState to control whether local
.ghci gets sourced or ignored.  This commit does not add a way to
set this value (a subsequent commit will add this), but the .ghci
sourcing routine respects its value.

Fixes: https://gitlab.haskell.org/ghc/ghc/issues/14689
Related: https://gitlab.haskell.org/ghc/ghc/issues/6017
Related: https://gitlab.haskell.org/ghc/ghc/issues/14250

- - - - -
5c06b60d by Fraser Tweedale at 2019-04-15T10:26:38Z
users-guide: update startup script order

Update users guide to match the new startup script order.  Also
clarify that -ignore-dot-ghci does not apply to scripts specified
via the -ghci-script option.

Part of: https://gitlab.haskell.org/ghc/ghc/issues/14689

- - - - -
aa490b35 by Fraser Tweedale at 2019-04-15T10:26:38Z
GHCi: add 'local-config' setting

Add the ':set local-config { source | ignore }' setting to control
whether .ghci file in current directory will be sourced or not.  The
directive can be set in global config or $HOME/.ghci, which are
processed before local .ghci files.

The default is "source", preserving current behaviour.

Related: https://gitlab.haskell.org/ghc/ghc/issues/6017
Related: https://gitlab.haskell.org/ghc/ghc/issues/14250

- - - - -
ed94d345 by Fraser Tweedale at 2019-04-15T10:26:38Z
users-guide: document :set local-config

Document the ':set local-config' command and add a warning about
sourcing untrusted local .ghci scripts.

Related: https://gitlab.haskell.org/ghc/ghc/issues/6017
Related: https://gitlab.haskell.org/ghc/ghc/issues/14250

- - - - -
be05bd81 by Gabor Greif at 2019-04-15T21:19:03Z
asm-emit-time IND_STATIC elimination

When a new closure identifier is being established to a
local or exported closure already emitted into the same
module, refrain from adding an IND_STATIC closure, and
instead emit an assembly-language alias.

Inter-module IND_STATIC objects still remain, and need to be
addressed by other measures.

Binary-size savings on nofib are around 0.1%.

- - - - -
57eb5bc6 by erthalion at 2019-04-16T19:40:36Z
Show dynamic object files (#16062)

Closes #16062. When -dynamic-too is specified, reflect that in the
progress message, like:

$ ghc Main.hs -dynamic-too
[1 of 1] Compiling Lib              ( Main.hs, Main.o, Main.dyn_o )

instead of:

$ ghc Main.hs -dynamic-too
[1 of 1] Compiling Lib              ( Main.hs, Main.o )

- - - - -
894ec447 by Andrey Mokhov at 2019-04-16T19:46:44Z
Hadrian: Generate GHC wrapper scripts

This is a temporary workaround for #16534. We generate wrapper scripts
<build-root>/ghc-stage1 and <build-root>/ghc-stage2 that can be used to
run Stage1 and Stage2 GHCs with the right arguments.

See https://gitlab.haskell.org/ghc/ghc/issues/16534.

- - - - -
742a694f by Ömer Sinan Ağacan at 2019-04-17T10:50:00Z
rts/BlockAlloc: Allow aligned allocation requests

This implements support for block group allocations which are aligned to
an integral number of blocks.

This will be used by the nonmoving garbage collector, which uses the
block allocator to allocate the segments which back its heap. These
segments are a fixed number of blocks in size, with each segment being
aligned to the segment size boundary. This allows us to easily find the
segment metadata stored at the beginning of the segment.

- - - - -
02c3eb12 by Ben Gamari at 2019-04-17T10:50:00Z
testsuite/testblockalloc: A bit of refactoring

- - - - -
f6685cb5 by Ben Gamari at 2019-04-17T10:50:00Z
testsuite/testblockalloc: Test aligned block group allocation

- - - - -
0131dfbb by Ben Gamari at 2019-04-17T10:50:00Z
rts/BlockAlloc: Wibbles

- - - - -
90f035ab by Ben Gamari at 2019-04-17T10:50:00Z
rts/BlockAlloc: Use allocLargeChunk in aligned block allocation

- - - - -
827574fd by Ömer Sinan Ağacan at 2019-04-17T10:50:00Z
Disallow allocating megablocks, update tests

- - - - -
2a22aa8c by Ömer Sinan Ağacan at 2019-04-17T10:50:00Z
Fix lint errors

- - - - -
81f7df93 by Ömer Sinan Ağacan at 2019-04-17T13:44:56Z
Use allocLargeChunkOnNode to reduce splitting

- - - - -


30 changed files:

- .gitlab-ci.yml
- + .gitlab/linters/check-version-number.sh
- CODEOWNERS
- README.md
- compiler/cmm/CLabel.hs
- compiler/cmm/CmmCallConv.hs
- compiler/cmm/CmmExpr.hs
- compiler/cmm/CmmType.hs
- compiler/codeGen/StgCmmBind.hs
- compiler/codeGen/StgCmmPrim.hs
- compiler/coreSyn/CoreLint.hs
- compiler/ghci/ByteCodeLink.hs
- compiler/llvmGen/Llvm/Types.hs
- compiler/llvmGen/LlvmCodeGen/Base.hs
- compiler/llvmGen/LlvmCodeGen/Data.hs
- compiler/llvmGen/LlvmCodeGen/Ppr.hs
- compiler/main/DynFlags.hs
- compiler/main/HscMain.hs
- compiler/main/HscTypes.hs
- compiler/nativeGen/AsmCodeGen.hs
- compiler/nativeGen/Format.hs
- compiler/nativeGen/NCGMonad.hs
- compiler/nativeGen/PPC/CodeGen.hs
- compiler/nativeGen/PPC/Ppr.hs
- compiler/nativeGen/PPC/Regs.hs
- compiler/nativeGen/Reg.hs
- compiler/nativeGen/RegAlloc/Graph/TrivColorable.hs
- compiler/nativeGen/RegClass.hs
- compiler/nativeGen/SPARC/Instr.hs
- compiler/nativeGen/SPARC/Ppr.hs


The diff was not included because it is too large.


View it on GitLab: https://gitlab.haskell.org/ghc/ghc/compare/9ba77dd0ffd5724c1302e8550750a413589257d2...81f7df9374ace2158cec2a940baf12668ab7d77c

-- 
View it on GitLab: https://gitlab.haskell.org/ghc/ghc/compare/9ba77dd0ffd5724c1302e8550750a413589257d2...81f7df9374ace2158cec2a940baf12668ab7d77c
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/20190417/20553956/attachment-0001.html>

More information about the ghc-commits mailing list