[Git][ghc/ghc][wip/t21766] 192 commits: JS: fix thread-related primops
Finley McIlwaine (@FinleyMcIlwaine)
gitlab at gitlab.haskell.org
Thu May 25 19:44:17 UTC 2023
Finley McIlwaine pushed to branch wip/t21766 at Glasgow Haskell Compiler / GHC
Commits:
d442ac05 by Sylvain Henry at 2023-04-19T20:04:35-04:00
JS: fix thread-related primops
- - - - -
7a96f90b by Bryan Richter at 2023-04-19T20:05:11-04:00
CI: Disable abi-test-nightly
See #23269
- - - - -
ab6c1d29 by Sylvain Henry at 2023-04-19T20:05:50-04:00
Testsuite: don't use obsolescent egrep (#22351)
Recent egrep displays the following message, breaking golden tests:
egrep: warning: egrep is obsolescent; using grep -E
Switch to using "grep -E" instead
- - - - -
f15b0ce5 by Matthew Pickering at 2023-04-20T11:01:06-04:00
hadrian: Pass haddock file arguments in a response file
In !10119 CI was failing on windows because the command line was too
long. We can mitigate this by passing the file arguments to haddock in a
response file.
We can't easily pass all the arguments in a response file because the
`+RTS` arguments can't be placed in the response file.
Fixes #23273
- - - - -
7012ec2f by tocic at 2023-04-20T11:01:42-04:00
Fix doc typo in GHC.Read.readList
- - - - -
5c873124 by sheaf at 2023-04-20T18:33:34-04:00
Implement -jsem: parallelism controlled by semaphores
See https://github.com/ghc-proposals/ghc-proposals/pull/540/ for a
complete description for the motivation for this feature.
The `-jsem` option allows a build tool to pass a semaphore to GHC which
GHC can use in order to control how much parallelism it requests.
GHC itself acts as a client in the GHC jobserver protocol.
```
GHC Jobserver Protocol
~~~~~~~~~~~~~~~~~~~~~~
This proposal introduces the GHC Jobserver Protocol. This protocol allows
a server to dynamically invoke many instances of a client process,
while restricting all of those instances to use no more than <n> capabilities.
This is achieved by coordination over a system semaphore (either a POSIX
semaphore [6]_ in the case of Linux and Darwin, or a Win32 semaphore [7]_
in the case of Windows platforms).
There are two kinds of participants in the GHC Jobserver protocol:
- The *jobserver* creates a system semaphore with a certain number of
available tokens.
Each time the jobserver wants to spawn a new jobclient subprocess, it **must**
first acquire a single token from the semaphore, before spawning
the subprocess. This token **must** be released once the subprocess terminates.
Once work is finished, the jobserver **must** destroy the semaphore it created.
- A *jobclient* is a subprocess spawned by the jobserver or another jobclient.
Each jobclient starts with one available token (its *implicit token*,
which was acquired by the parent which spawned it), and can request more
tokens through the Jobserver Protocol by waiting on the semaphore.
Each time a jobclient wants to spawn a new jobclient subprocess, it **must**
pass on a single token to the child jobclient. This token can either be the
jobclient's implicit token, or another token which the jobclient acquired
from the semaphore.
Each jobclient **must** release exactly as many tokens as it has acquired from
the semaphore (this does not include the implicit tokens).
```
Build tools such as cabal act as jobservers in the protocol and are
responsibile for correctly creating, cleaning up and managing the
semaphore.
Adds a new submodule (semaphore-compat) for managing and interacting
with semaphores in a cross-platform way.
Fixes #19349
- - - - -
52d3e9b4 by Ben Gamari at 2023-04-20T18:34:11-04:00
rts: Initialize Array# header in listThreads#
Previously the implementation of listThreads# failed to initialize the
header of the created array, leading to various nastiness.
Fixes #23071
- - - - -
1db30fe1 by Ben Gamari at 2023-04-20T18:34:11-04:00
testsuite: Add test for #23071
- - - - -
dae514f9 by tocic at 2023-04-21T13:31:21-04:00
Fix doc typos in libraries/base/GHC
- - - - -
113e21d7 by Sylvain Henry at 2023-04-21T13:32:01-04:00
Testsuite: replace some js_broken/js_skip predicates with req_c
Using req_c is more precise.
- - - - -
038bb031 by Krzysztof Gogolewski at 2023-04-21T18:03:04-04:00
Minor doc fixes
- Add docs/index.html to .gitignore.
It is created by ./hadrian/build docs, and it was the only file
in Hadrian's templateRules not present in .gitignore.
- Mention that MultiWayIf supports non-boolean guards
- Remove documentation of optdll - removed in 2007, 763daed95
- Fix markdown syntax
- - - - -
e826cdb2 by amesgen at 2023-04-21T18:03:44-04:00
User's guide: DeepSubsumption is implied by Haskell{98,2010}
- - - - -
499a1c20 by PHO at 2023-04-23T13:39:32-04:00
Implement executablePath for Solaris and make getBaseDir less platform-dependent
Use base-4.17 executablePath when possible, and fall back on
getExecutablePath when it's not available. The sole reason why getBaseDir
had #ifdef's was apparently that getExecutablePath wasn't reliable, and we
could reduce the number of CPP conditionals by making use of
executablePath instead.
Also export executablePath on js_HOST_ARCH.
- - - - -
97a6f7bc by tocic at 2023-04-23T13:40:08-04:00
Fix doc typos in libraries/base
- - - - -
787c6e8c by Ben Gamari at 2023-04-24T12:19:06-04:00
testsuite/T20137: Avoid impl.-defined behavior
Previously we would cast pointers to uint64_t. However, implementations
are allowed to either zero- or sign-extend such casts. Instead cast to
uintptr_t to avoid this.
Fixes #23247.
- - - - -
87095f6a by Cheng Shao at 2023-04-24T12:19:44-04:00
rts: always build 64-bit atomic ops
This patch does a few things:
- Always build 64-bit atomic ops in rts/ghc-prim, even on 32-bit
platforms
- Remove legacy "64bit" cabal flag of rts package
- Fix hs_xchg64 function prototype for 32-bit platforms
- Fix AtomicFetch test for wasm32
- - - - -
2685a12d by Cheng Shao at 2023-04-24T12:20:21-04:00
compiler: don't install signal handlers when the host platform doesn't have signals
Previously, large parts of GHC API will transitively invoke
withSignalHandlers, which doesn't work on host platforms without
signal functionality at all (e.g. wasm32-wasi). By making
withSignalHandlers a no-op on those platforms, we can make more parts
of GHC API work out of the box when signals aren't supported.
- - - - -
1338b7a3 by Cheng Shao at 2023-04-24T16:21:30-04:00
hadrian: fix non-ghc program paths passed to testsuite driver when testing cross GHC
- - - - -
1a10f556 by Bodigrim at 2023-04-24T16:22:09-04:00
Add since pragma to Data.Functor.unzip
- - - - -
0da9e882 by Soham Chowdhury at 2023-04-25T00:15:22-04:00
More informative errors for bad imports (#21826)
- - - - -
ebd5b078 by Josh Meredith at 2023-04-25T00:15:58-04:00
JS/base: provide implementation for mkdir (issue 22374)
- - - - -
8f656188 by Josh Meredith at 2023-04-25T18:12:38-04:00
JS: Fix h$base_access implementation (issue 22576)
- - - - -
74c55712 by Andrei Borzenkov at 2023-04-25T18:13:19-04:00
Give more guarntees about ImplicitParams (#23289)
- Added new section in the GHC user's guide that legends behavior of
nested implicit parameter bindings in these two cases:
let ?f = 1 in let ?f = 2 in ?f
and
data T where MkT :: (?f :: Int) => T
f :: T -> T -> Int
f MkT MkT = ?f
- Added new test case to examine this behavior.
- - - - -
c30ac25f by Sebastian Graf at 2023-04-26T14:50:51-04:00
DmdAnal: Unleash demand signatures of free RULE and unfolding binders (#23208)
In #23208 we observed that the demand signature of a binder occuring in a RULE
wasn't unleashed, leading to a transitively used binder being discarded as
absent. The solution was to use the same code path that we already use for
handling exported bindings.
See the changes to `Note [Absence analysis for stable unfoldings and RULES]`
for more details.
I took the chance to factor out the old notion of a `PlusDmdArg` (a pair of a
`VarEnv Demand` and a `Divergence`) into `DmdEnv`, which fits nicely into our
existing framework. As a result, I had to touch quite a few places in the code.
This refactoring exposed a few small bugs around correct handling of bottoming
demand environments. As a result, some strictness signatures now mention uniques
that weren't there before which caused test output changes to T13143, T19969 and
T22112. But these tests compared whole -ddump-simpl listings which is a very
fragile thing to begin with. I changed what exactly they test for based on the
symptoms in the corresponding issues.
There is a single regression in T18894 because we are more conservative around
stable unfoldings now. Unfortunately it is not easily fixed; let's wait until
there is a concrete motivation before invest more time.
Fixes #23208.
- - - - -
77f506b8 by Josh Meredith at 2023-04-26T14:51:28-04:00
Refactor GenStgRhs to include the Type in both constructors (#23280, #22576, #22364)
Carry the actual type of an expression through the PreStgRhs and into GenStgRhs
for use in later stages. Currently this is used in the JavaScript backend to fix
some tests from the above mentioned issues: EtaExpandLevPoly, RepPolyWrappedVar2,
T13822, T14749.
- - - - -
052e2bb6 by Alan Zimmerman at 2023-04-26T14:52:05-04:00
EPA: Use ExplicitBraces only in HsModule
!9018 brought in exact print annotations in LayoutInfo for open and
close braces at the top level.
But it retained them in the HsModule annotations too.
Remove the originals, so exact printing uses LayoutInfo
- - - - -
d5c4629b by Cheng Shao at 2023-04-27T16:00:35-04:00
ci: update ci.sh to actually run the entire testsuite for wasm backend
For the time being, we still need to use in-tree mode and can't test
the bindist yet.
- - - - -
533d075e by Cheng Shao at 2023-04-27T16:00:35-04:00
ci: additional wasm32 manual jobs in validate pipelines
This patch enables bignum native & unregisterised wasm32 jobs as
manual jobs in validate pipelines, which can be useful to prevent
breakage when working on wasm32 related patches.
- - - - -
b5f00811 by Cheng Shao at 2023-04-27T16:00:35-04:00
testsuite: fix cross prefix stripping
This patch fixes cross prefix stripping in the testsuite driver. The
normalization logic used to only handle prefixes of the triple form
<arch>-<vendor>-<os>, now it's relaxed to allow any number of tokens
in the prefix tuple, so the cross prefix stripping logic would work
when ghc is configured with something like --target=wasm32-wasi.
- - - - -
6f511c36 by Cheng Shao at 2023-04-27T16:00:35-04:00
testsuite: include target exe extension in heap profile filenames
This patch fixes hp2ps related framework failures when testing the
wasm backend by including target exe extension in heap profile
filenames.
- - - - -
e6416b10 by Cheng Shao at 2023-04-27T16:00:35-04:00
testsuite: exclude ghci ways if no rts linker is present
This patch implements logic to automatically exclude ghci ways when
there is no rts linker. It's way better than having to annotate
individual test cases.
- - - - -
791cce64 by Cheng Shao at 2023-04-27T16:00:35-04:00
testsuite: fix permission bits in copy_files
When the testsuite driver copy files instead of symlinking them, it
should also copy the permission bits, otherwise there'll be permission
denied errors. Also, enforce file copying when testing wasm32, since
wasmtime doesn't handle host symlinks quite well
(https://github.com/bytecodealliance/wasmtime/issues/6227).
- - - - -
aa6afe8a by Cheng Shao at 2023-04-27T16:00:35-04:00
testsuite: add the req_ghc_with_threaded_rts predicate
This patch adds the req_ghc_with_threaded_rts predicate to the
testsuite to assert the platform has threaded RTS, and mark some tests
as req_ghc_with_threaded_rts. Also makes ghc_with_threaded_rts a
config field instead of a global variable.
- - - - -
ce580426 by Cheng Shao at 2023-04-27T16:00:35-04:00
testsuite: add the req_process predicate
This patch adds the req_process predicate to the testsuite to assert
the platform has a process model, also marking tests that involve
spawning processes as req_process. Also bumps hpc & process submodule.
- - - - -
cb933665 by Cheng Shao at 2023-04-27T16:00:35-04:00
testsuite: add the req_host_target_ghc predicate
This patch adds the req_host_target_ghc predicate to the testsuite to
assert the ghc compiler being tested can compile both host/target
code. When testing cross GHCs this is not supported yet, but it may
change in the future.
- - - - -
b174a110 by Cheng Shao at 2023-04-27T16:00:35-04:00
testsuite: add missing annotations for some tests
This patch adds missing annotations (req_th, req_dynamic_lib_support,
req_rts_linker) to some tests. They were discovered when testing
wasm32, though it's better to be explicit about what features they
require, rather than simply adding when(arch('wasm32'), skip).
- - - - -
bd2bfdec by Cheng Shao at 2023-04-27T16:00:35-04:00
testsuite: wasm32-specific fixes
This patch includes all wasm32-specific testsuite fixes.
- - - - -
4eaf2c2a by Josh Meredith at 2023-04-27T16:01:11-04:00
JS: change GHC.JS.Transform.identsS/E/V to take a saturated IR (#23304)
- - - - -
57277662 by sheaf at 2023-04-29T20:23:06+02:00
Add the Unsatisfiable class
This commit implements GHC proposal #433, adding the Unsatisfiable
class to the GHC.TypeError module. This provides an alternative to
TypeError for which error reporting is more predictable: we report it
when we are reporting unsolved Wanted constraints.
Fixes #14983 #16249 #16906 #18310 #20835
- - - - -
00a8a5ff by Torsten Schmits at 2023-04-30T03:45:09-04:00
Add structured error messages for GHC.Rename.Names
Tracking ticket: #20115
MR: !10336
This converts uses of `mkTcRnUnknownMessage` to newly added constructors
of `TcRnMessage`.
- - - - -
931c8d82 by Ben Orchard at 2023-05-03T20:16:18-04:00
Add sized primitive literal syntax
Adds a new LANGUAGE pragma ExtendedLiterals, which enables defining
unboxed numeric literals such as `0xFF#Word8 :: Word8#`.
Implements GHC proposal 0451:
https://github.com/ghc-proposals/ghc-proposals/blob/b384a538b34f79d18a0201455b7b3c473bc8c936/proposals/0451-sized-literals.rst
Fixes #21422.
Bumps haddock submodule.
Co-authored-by: Krzysztof Gogolewski <krzysztof.gogolewski at tweag.io>
- - - - -
f3460845 by Bodigrim at 2023-05-03T20:16:57-04:00
Document instances of Double
- - - - -
1e9caa1a by Sylvain Henry at 2023-05-03T20:17:37-04:00
Bump Cabal submodule (#22356)
- - - - -
4eafb52a by sheaf at 2023-05-03T20:18:16-04:00
Don't forget to check the parent in an export list
Commit 3f374399 introduced a bug which caused us to forget to include
the parent of an export item of the form T(..) (that is, IEThingAll)
when checking for duplicate exports.
Fixes #23318
- - - - -
8fde4ac8 by amesgen at 2023-05-03T20:18:57-04:00
Fix unlit path in cross bindists
- - - - -
8cc9a534 by Matthew Pickering at 2023-05-04T14:58:14-04:00
hadrian: Flavour: Change args -> extraArgs
Previously in a flavour definition you could override all the flags
which were passed to GHC. This causes issues when needed to compute a
package hash because we need to know what these extra arguments are
going to be before computing the hash. The solution is to modify flavour
so that the arguments you pass here are just extra ones rather than all
the arguments that you need to compile something.
This makes things work more like how cabal.project files work when you
give extra arguments to a package and also means that flavour
transformers correctly affect the hash.
- - - - -
3fdb18f8 by romes at 2023-05-04T14:58:14-04:00
Hardwire a better unit-id for ghc
Previously, the unit-id of ghc-the-library was fixed as `ghc`.
This was done primarily because the compiler must know the unit-id of
some packages (including ghc) a-priori to define wired-in names.
However, as seen in #20742, a reinstallable `ghc` whose unit-id is fixed
to `ghc` might result in subtle bugs when different ghc's interact.
A good example of this is having GHC_A load a plugin compiled by GHC_B,
where GHC_A and GHC_B are linked to ghc-libraries that are ABI
incompatible. Without a distinction between the unit-id of the ghc library
GHC_A is linked against and the ghc library the plugin it is loading was
compiled against, we can't check compatibility.
This patch gives a slightly better unit-id to ghc (ghc-version) by
(1) Not setting -this-unit-id to ghc, but rather to the new unit-id (modulo stage0)
(2) Adding a definition to `GHC.Settings.Config` whose value is the new unit-id.
(2.1) `GHC.Settings.Config` is generated by Hadrian
(2.2) and also by cabal through `compiler/Setup.hs`
This unit-id definition is imported by `GHC.Unit.Types` and used to
set the wired-in unit-id of "ghc", which was previously fixed to "ghc"
The commits following this one will improve the unit-id with a
cabal-style package hash and check compatibility when loading plugins.
Note that we also ensure that ghc's unit key matches unit id both when
hadrian or cabal builds ghc, and in this way we no longer need to add
`ghc` to the WiringMap.
- - - - -
6689c9c6 by romes at 2023-05-04T14:58:14-04:00
Validate compatibility of ghcs when loading plugins
Ensure, when loading plugins, that the ghc the plugin depends on is the
ghc loading the plugin -- otherwise fail to load the plugin.
Progress towards #20742.
- - - - -
db4be339 by romes at 2023-05-04T14:58:14-04:00
Add hashes to unit-ids created by hadrian
This commit adds support for computing an inputs hash for packages
compiled by hadrian. The result is that ABI incompatible packages should
be given different hashes and therefore be distinct in a cabal store.
Hashing is enabled by the `--flag`, and is off by default as the hash
contains a hash of the source files. We enable it when we produce
release builds so that the artifacts we distribute have the right unit
ids.
- - - - -
944a9b94 by Matthew Pickering at 2023-05-04T14:58:14-04:00
Use hash-unit-ids in release jobs
Includes fix upload_ghc_libs glob
- - - - -
116d7312 by Josh Meredith at 2023-05-04T14:58:51-04:00
JS: fix bounds checking (Issue 23123)
* For ByteArray-based bounds-checking, the JavaScript backend must use the
`len` field, instead of the inbuild JavaScript `length` field.
* Range-based operations must also check both the start and end of the range
for bounds
* All indicies are valid for ranges of size zero, since they are essentially no-ops
* For cases of ByteArray accesses (e.g. read as Int), the end index is
(i * sizeof(type) + sizeof(type) - 1), while the previous implementation
uses (i + sizeof(type) - 1). In the Int32 example, this is (i * 4 + 3)
* IndexByteArrayOp_Word8As* primitives use byte array indicies (unlike
the previous point), but now check both start and end indicies
* Byte array copies now check if the arrays are the same by identity and
then if the ranges overlap.
- - - - -
2d5c1dde by Sylvain Henry at 2023-05-04T14:58:51-04:00
Fix remaining issues with bound checking (#23123)
While fixing these I've also changed the way we store addresses into
ByteArray#. Addr# are composed of two parts: a JavaScript array and an
offset (32-bit number).
Suppose we want to store an Addr# in a ByteArray# foo at offset i.
Before this patch, we were storing both fields as a tuple in the "arr"
array field:
foo.arr[i] = [addr_arr, addr_offset];
Now we only store the array part in the "arr" field and the offset
directly in the array:
foo.dv.setInt32(i, addr_offset):
foo.arr[i] = addr_arr;
It avoids wasting space for the tuple.
- - - - -
98c5ee45 by Luite Stegeman at 2023-05-04T14:59:31-04:00
JavaScript: Correct arguments to h$appendToHsStringA
fixes #23278
- - - - -
ca611447 by Josh Meredith at 2023-05-04T15:00:07-04:00
base/encoding: add an allocations performance test (#22946)
- - - - -
e3ddf58d by Krzysztof Gogolewski at 2023-05-04T15:00:44-04:00
linear types: Don't add external names to the usage env
This has no observable effect, but avoids storing useless data.
- - - - -
b3226616 by Andrei Borzenkov at 2023-05-04T15:01:25-04:00
Improved documentation for the Data.OldList.nub function
There was recomentation to use map head . group . sort instead of nub
function, but containers library has more suitable and efficient
analogue
- - - - -
e8b72ff6 by Ryan Scott at 2023-05-04T15:02:02-04:00
Fix type variable substitution in gen_Newtype_fam_insts
Previously, `gen_Newtype_fam_insts` was substituting the type variable binders
of a type family instance using `substTyVars`, which failed to take type
variable dependencies into account. There is similar code in
`GHC.Tc.TyCl.Class.tcATDefault` that _does_ perform this substitution properly,
so this patch:
1. Factors out this code into a top-level `substATBndrs` function, and
2. Uses `substATBndrs` in `gen_Newtype_fam_insts`.
Fixes #23329.
- - - - -
275836d2 by Torsten Schmits at 2023-05-05T08:43:02+00:00
Add structured error messages for GHC.Rename.Utils
Tracking ticket: #20115
MR: !10350
This converts uses of `mkTcRnUnknownMessage` to newly added constructors
of `TcRnMessage`.
- - - - -
983ce558 by Oleg Grenrus at 2023-05-05T13:11:29-04:00
Use TemplateHaskellQuotes in TH.Syntax to construct Names
- - - - -
a5174a59 by Matthew Pickering at 2023-05-05T18:42:31-04:00
driver: Use hooks from plugin_hsc_env
This fixes a bug in oneshot mode where hooks modified in a plugin
wouldn't be used in oneshot mode because we neglected to use the right
hsc_env. This was observed by @csabahruska.
- - - - -
18a7d03d by Aaron Allen at 2023-05-05T18:42:31-04:00
Rework plugin initialisation points
In general this patch pushes plugin initialisation points to earlier in
the pipeline. As plugins can modify the `HscEnv`, it's imperative that
the plugins are initialised as soon as possible and used thereafter.
For example, there are some new tests which modify hsc_logger and other
hooks which failed to fire before (and now do)
One consequence of this change is that the error for specifying the
usage of a HPT plugin from the command line has changed, because it's
now attempted to be loaded at initialisation rather than causing a
cyclic module import.
Closes #21279
Co-authored-by: Matthew Pickering <matthewtpickering at gmail.com>
- - - - -
6e776ed3 by Matthew Pickering at 2023-05-05T18:42:31-04:00
docs: Add Note [Timing of plugin initialization]
- - - - -
e1df8511 by Matthew Pickering at 2023-05-05T18:43:07-04:00
Incrementally update ghcup metadata in ghc/ghcup-metadata
This job paves the way for distributing nightly builds
* A new repo https://gitlab.haskell.org/ghc/ghcup-metadata stores the
metadata on the "updates" branch.
* Each night this metadata is downloaded and the nightly builds are
appended to the end of the metadata.
* The update job only runs on the scheduled nightly pipeline, not just
when NIGHTLY=1.
Things which are not done yet
* Modify the retention policy for nightly jobs
* Think about building release flavour compilers to distribute nightly.
Fixes #23334
- - - - -
8f303d27 by Rodrigo Mesquita at 2023-05-05T22:04:31-04:00
docs: Remove mentions of ArrayArray# from unlifted FFI section
Fixes #23277
- - - - -
994bda56 by Torsten Schmits at 2023-05-05T22:05:12-04:00
Add structured error messages for GHC.Rename.Module
Tracking ticket: #20115
MR: !10361
This converts uses of `mkTcRnUnknownMessage` to newly added constructors
of `TcRnMessage`.
Only addresses the single warning missing from the previous MR.
- - - - -
3e3a6be4 by Ben Gamari at 2023-05-08T12:15:19+00:00
rts: Fix data-race in hs_init_ghc
As noticed by @Terrorjack, `hs_init_ghc` previously used non-atomic
increment/decrement on the RTS's initialization count. This may go wrong
in a multithreaded program which initializes the runtime multiple times.
Closes #22756.
- - - - -
78c8dc50 by Torsten Schmits at 2023-05-08T21:41:51-04:00
Add structured error messages for GHC.IfaceToCore
Tracking ticket: #20114
MR: !10390
This converts uses of `mkTcRnUnknownMessage` to newly added constructors
of `TcRnMessage`.
- - - - -
0e2df4c9 by Bryan Richter at 2023-05-09T12:03:35+03:00
Fix up rules for ghcup-metadata-nightly-push
- - - - -
b970e64f by Ben Gamari at 2023-05-09T08:41:33-04:00
testsuite: Add test for atomicSwapIORef
- - - - -
81cfefd2 by Ben Gamari at 2023-05-09T08:41:53-04:00
compiler: Implement atomicSwapIORef with xchg
As requested by @treeowl in CLC#139.
- - - - -
6b29154d by Ben Gamari at 2023-05-09T08:41:53-04:00
Make atomicSwapMutVar# an inline primop
- - - - -
64064cfe by doyougnu at 2023-05-09T18:40:01-04:00
JS: add GHC.JS.Optimizer, remove RTS.Printer, add Linker.Opt
This MR changes some simple optimizations and is a first step in re-architecting
the JS backend pipeline to add the optimizer. In particular it:
- removes simple peep hole optimizations from `GHC.StgToJS.Printer` and removes that module
- adds module `GHC.JS.Optimizer`
- defines the same peep hole opts that were removed only now they are `Syntax -> Syntax` transformations rather than `Syntax -> JS code` optimizations
- hooks the optimizer into code gen
- adds FuncStat and ForStat constructors to the backend.
Working Ticket:
- #22736
Related MRs:
- MR !10142
- MR !10000
-------------------------
Metric Decrease:
CoOpt_Read
ManyAlternatives
PmSeriesS
PmSeriesT
PmSeriesV
T10421
T12707
T13253
T13253-spj
T15164
T17516
T18140
T18282
T18698a
T18698b
T18923
T1969
T19695
T20049
T3064
T5321FD
T5321Fun
T783
T9198
T9233
T9630
-------------------------
- - - - -
6738c01d by Krzysztof Gogolewski at 2023-05-09T18:40:38-04:00
Add a regression test for #21050
- - - - -
b2cdb7da by Ben Gamari at 2023-05-09T18:41:14-04:00
nonmoving: Account for mutator allocations in bytes_allocated
Previously we failed to account direct mutator allocations into the
nonmoving heap against the mutator's allocation limit and
`cap->total_allocated`. This only manifests during CAF evaluation (since
we allocate the CAF's blackhole directly into the nonmoving heap).
Fixes #23312.
- - - - -
0657b482 by Sven Tennie at 2023-05-09T22:22:42-04:00
Adjust AArch64 stackFrameHeaderSize
The prologue of each stack frame are the saved LR and FP registers, 8
byte each. I.e. the size of the stack frame header is 2 * 8 byte.
- - - - -
7788c09c by konsumlamm at 2023-05-09T22:23:23-04:00
Make `(&)` representation polymorphic in the return type
- - - - -
b3195922 by Ben Gamari at 2023-05-10T05:06:45-04:00
ghc-prim: Generalize keepAlive#/touch# in state token type
Closes #23163.
- - - - -
1e6861dd by Cheng Shao at 2023-05-10T05:07:25-04:00
Bump hsc2hs submodule
Fixes #22981.
- - - - -
0a513952 by Ben Gamari at 2023-05-11T04:10:17-04:00
base: Export GHC.Conc.Sync.fromThreadId
Closes #22706.
- - - - -
29be39ba by Matthew Pickering at 2023-05-11T04:10:54-04:00
Build vanilla alpine bindists
We currently attempt to build and distribute fully static alpine
bindists (ones which could be used on any linux platform) but most
people who use the alpine bindists want to use alpine to build their own
static applications (for which a fully static bindist is not necessary).
We should build and distribute these bindists for these users whilst the
fully-static bindist is still unusable.
Fixes #23349
- - - - -
40c7daed by Simon Peyton Jones at 2023-05-11T04:11:30-04:00
Look both ways when looking for quantified equalities
When looking up (t1 ~# t2) in the quantified constraints,
check both orientations. Forgetting this led to #23333.
- - - - -
c17bb82f by Rodrigo Mesquita at 2023-05-11T04:12:07-04:00
Move "target has RTS linker" out of settings
We move the "target has RTS linker" information out of configure into a
predicate in GHC, and remove this option from the settings file where it
is unnecessary -- it's information statically known from the platform.
Note that previously we would consider `powerpc`s and `s390x`s other
than `powerpc-ibm-aix*` and `s390x-ibm-linux` to have an RTS linker,
but the RTS linker supports neither platform.
Closes #23361
- - - - -
bd0b056e by Krzysztof Gogolewski at 2023-05-11T04:12:44-04:00
Add a test for #17284
Since !10123 we now reject this program.
- - - - -
630b1fea by Bodigrim at 2023-05-11T04:13:24-04:00
Document unlawfulness of instance Num Fixed
Fixes #22712
- - - - -
87eebf98 by sheaf at 2023-05-11T11:55:22-04:00
Add fused multiply-add instructions
This patch adds eight new primops that fuse a multiplication and an
addition or subtraction:
- `{fmadd,fmsub,fnmadd,fnmsub}{Float,Double}#`
fmadd x y z is x * y + z, computed with a single rounding step.
This patch implements code generation for these primops in the following
backends:
- X86, AArch64 and PowerPC NCG,
- LLVM
- C
WASM uses the C implementation. The primops are unsupported in the
JavaScript backend.
The following constant folding rules are also provided:
- compute a * b + c when a, b, c are all literals,
- x * y + 0 ==> x * y,
- ±1 * y + z ==> z ± y and x * ±1 + z ==> z ± x.
NB: the constant folding rules incorrectly handle signed zero.
This is a known limitation with GHC's floating-point constant folding
rules (#21227), which we hope to resolve in the future.
- - - - -
ad16a066 by Krzysztof Gogolewski at 2023-05-11T11:55:59-04:00
Add a test for #21278
- - - - -
05cea68c by Matthew Pickering at 2023-05-11T11:56:36-04:00
rts: Refine memory retention behaviour to account for pinned/compacted objects
When using the copying collector there is still a lot of data which
isn't copied (such as pinned, compacted, large objects etc). The logic
to decide how much memory to retain didn't take into account that these
wouldn't be copied. Therefore we pessimistically retained 2* the amount
of memory for these blocks even though they wouldn't be copied by the
collector.
The solution is to split up the heap into two parts, the parts which
will be copied and the parts which won't be copied. Then the appropiate
factor is applied to each part individually (2 * for copying and 1.2 *
for not copying).
The T23221 test demonstrates this improvement with a program which first
allocates many unpinned ByteArray# followed by many pinned ByteArray#
and observes the difference in the ultimate memory baseline between the
two.
There are some charts on #23221.
Fixes #23221
- - - - -
1bb24432 by Cheng Shao at 2023-05-11T11:57:15-04:00
hadrian: fix no_dynamic_libs flavour transformer
This patch fixes the no_dynamic_libs flavour transformer and make
fully_static reuse it. Previously building with no_dynamic_libs fails
since ghc program is still dynamic and transitively brings in dyn ways
of rts which are produced by no rules.
- - - - -
0ed493a3 by Josh Meredith at 2023-05-11T23:08:27-04:00
JS: refactor jsSaturate to return a saturated JStat (#23328)
- - - - -
a856d98e by Pierre Le Marre at 2023-05-11T23:09:08-04:00
Doc: Fix out-of-sync using-optimisation page
- Make explicit that default flag values correspond to their -O0 value.
- Fix -fignore-interface-pragmas, -fstg-cse, -fdo-eta-reduction,
-fcross-module-specialise, -fsolve-constant-dicts, -fworker-wrapper.
- - - - -
c176ad18 by sheaf at 2023-05-12T06:10:57-04:00
Don't panic in mkNewTyConRhs
This function could come across invalid newtype constructors, as we
only perform validity checking of newtypes once we are outside the
knot-tied typechecking loop.
This patch changes this function to fake up a stub type in the case of
an invalid newtype, instead of panicking.
This patch also changes "checkNewDataCon" so that it reports as many
errors as possible at once.
Fixes #23308
- - - - -
ab63daac by Krzysztof Gogolewski at 2023-05-12T06:11:38-04:00
Allow Core optimizations when interpreting bytecode
Tracking ticket: #23056
MR: !10399
This adds the flag `-funoptimized-core-for-interpreter`, permitting use
of the `-O` flag to enable optimizations when compiling with the
interpreter backend, like in ghci.
- - - - -
c6cf9433 by Ben Gamari at 2023-05-12T06:12:14-04:00
hadrian: Fix mention of non-existent removeFiles function
Previously Hadrian's bindist Makefile referred to a `removeFiles`
function that was previously defined by the `make` build system. Since
the `make` build system is no longer around, this function is now
undefined. Naturally, make being make, this appears to be silently
ignored instead of producing an error.
Fix this by rewriting it to `rm -f`.
Closes #23373.
- - - - -
eb60ec18 by Bodigrim at 2023-05-12T06:12:54-04:00
Mention new implementation of GHC.IORef.atomicSwapIORef in the changelog
- - - - -
aa84cff4 by Teo Camarasu at 2023-05-12T19:27:23-04:00
rts: Ensure non-moving gc is not running when pausing
- - - - -
5ad776ab by Teo Camarasu at 2023-05-12T19:27:23-04:00
rts: Teach listAllBlocks about nonmoving heap
List all blocks on the non-moving heap.
Resolves #22627
- - - - -
d683b2e5 by Krzysztof Gogolewski at 2023-05-12T19:28:00-04:00
Fix coercion optimisation for SelCo (#23362)
setNominalRole_maybe is supposed to output a nominal coercion.
In the SelCo case, it was not updating the stored role to Nominal,
causing #23362.
- - - - -
59aa4676 by Alexis King at 2023-05-12T19:28:47-04:00
hadrian: Fix linker script flag for MergeObjects builder
This fixes what appears to have been a typo in !9530. The `-t` flag just
enables tracing on all versions of `ld` I’ve looked at, while `-T` is
used to specify a linker script. It seems that this worked anyway for
some reason on some `ld` implementations (perhaps because they
automatically detect linker scripts), but the missing `-T` argument
causes `gold` to complain.
- - - - -
4bf9fa0f by Adam Gundry at 2023-05-12T23:49:49-04:00
Less coercion optimization for non-newtype axioms
See Note [Push transitivity inside newtype axioms only] for an explanation
of the change here. This change substantially improves the performance of
coercion optimization for programs involving transitive type family reductions.
-------------------------
Metric Decrease:
CoOpt_Singletons
LargeRecord
T12227
T12545
T13386
T15703
T5030
T8095
-------------------------
- - - - -
dc0c9574 by Adam Gundry at 2023-05-12T23:49:49-04:00
Move checkAxInstCo to GHC.Core.Lint
A consequence of the previous change is that checkAxInstCo is no longer
called during coercion optimization, so it can be moved back where it belongs.
Also includes some edits to Note [Conflict checking with AxiomInstCo] as
suggested by @simonpj.
- - - - -
8b9b7dbc by Simon Peyton Jones at 2023-05-12T23:50:25-04:00
Use the eager unifier in the constraint solver
This patch continues the refactoring of the constraint solver
described in #23070.
The Big Deal in this patch is to call the regular, eager unifier from the
constraint solver, when we want to create new equalities. This
replaces the existing, unifyWanted which amounted to
yet-another-unifier, so it reduces duplication of a rather subtle
piece of technology. See
* Note [The eager unifier] in GHC.Tc.Utils.Unify
* GHC.Tc.Solver.Monad.wrapUnifierTcS
I did lots of other refactoring along the way
* I simplified the treatment of right hand sides that contain CoercionHoles.
Now, a constraint that contains a hetero-kind CoercionHole is non-canonical,
and cannot be used for rewriting or unification alike. This required me
to add the ch_hertero_kind flag to CoercionHole, with consequent knock-on
effects. See wrinkle (2) of `Note [Equalities with incompatible kinds]` in
GHC.Tc.Solver.Equality.
* I refactored the StopOrContinue type to add StartAgain, so that after a
fundep improvement (for example) we can simply start the pipeline again.
* I got rid of the unpleasant (and inefficient) rewriterSetFromType/Co functions.
With Richard I concluded that they are never needed.
* I discovered Wrinkle (W1) in Note [Wanteds rewrite Wanteds] in
GHC.Tc.Types.Constraint, and therefore now prioritise non-rewritten equalities.
Quite a few error messages change, I think always for the better.
Compiler runtime stays about the same, with one outlier: a 17% improvement in T17836
Metric Decrease:
T17836
T18223
- - - - -
5cad28e7 by Bartłomiej Cieślar at 2023-05-12T23:51:06-04:00
Cleanup of dynflags override in export renaming
The deprecation warnings are normally emitted whenever the name's GRE is being looked up, which calls the GHC.Rename.Env.addUsedGRE function. We do not want those warnings to be emitted when renaming export lists, so they are artificially turned off by removing all warning categories from DynFlags at the beginning of GHC.Tc.Gen.Export.rnExports. This commit removes that dependency by unifying the function used for GRE lookup in lookup_ie to lookupGreAvailRn and disabling the call to addUsedGRE in said function (the warnings are also disabled in a call to lookupSubBndrOcc_helper in lookupChildrenExport), as per #17957. This commit also changes the setting for whether to warn about deprecated names in addUsedGREs to be an explicit enum instead of a boolean.
- - - - -
d85ed900 by Alexis King at 2023-05-13T08:45:18-04:00
Use a uniform return convention in bytecode for unary results
fixes #22958
- - - - -
8a0d45f7 by Bodigrim at 2023-05-13T08:45:58-04:00
Add more instances for Compose: Enum, Bounded, Num, Real, Integral
See https://github.com/haskell/core-libraries-committee/issues/160 for discussion
- - - - -
902f0730 by Simon Peyton Jones at 2023-05-13T14:58:34-04:00
Make GHC.Types.Id.Make.shouldUnpackTy a bit more clever
As #23307, GHC.Types.Id.Make.shouldUnpackTy was leaving money on the
table, failing to unpack arguments that are perfectly unpackable.
The fix is pretty easy; see Note [Recursive unboxing]
- - - - -
a5451438 by sheaf at 2023-05-13T14:59:13-04:00
Fix bad multiplicity role in tyConAppFunCo_maybe
The function tyConAppFunCo_maybe produces a multiplicity coercion
for the multiplicity argument of the function arrow, except that
it could be at the wrong role if asked to produce a representational
coercion. We fix this by using the 'funRole' function, which computes
the right roles for arguments to the function arrow TyCon.
Fixes #23386
- - - - -
5b9e9300 by sheaf at 2023-05-15T11:26:59-04:00
Turn "ambiguous import" error into a panic
This error should never occur, as a lookup of a type or data constructor
should never be ambiguous. This is because a single module cannot export
multiple Names with the same OccName, as per item (1) of
Note [Exporting duplicate declarations] in GHC.Tc.Gen.Export.
This code path was intended to handle duplicate record fields, but the
rest of the code had since been refactored to handle those in a
different way.
We also remove the AmbiguousImport constructor of IELookupError, as
it is no longer used.
Fixes #23302
- - - - -
e305e60c by M Farkas-Dyck at 2023-05-15T11:27:41-04:00
Unbreak some tests with latest GNU grep, which now warns about stray '\'.
Confusingly, the testsuite mangled the error to say "stray /".
We also migrate some tests from grep to grep -E, as it seems the author actually wanted an "POSIX extended" (a.k.a. sane) regex.
Background: POSIX specifies 2 "regex" syntaxen: "basic" and "extended". Of these, only "extended" syntax is actually a regular expression. Furthermore, "basic" syntax is inconsistent in its use of the '\' character — sometimes it escapes a regex metacharacter, but sometimes it unescapes it, i.e. it makes an otherwise normal character become a metacharacter. This baffles me and it seems also the authors of these tests. Also, the regex(7) man page (at least on Linux) says "basic" syntax is obsolete. Nearly all modern tools and libraries are consistent in this use of the '\' character (of which many use "extended" syntax by default).
- - - - -
5ae81842 by sheaf at 2023-05-15T14:49:17-04:00
Improve "ambiguous occurrence" error messages
This error was sometimes a bit confusing, especially when data families
were involved. This commit improves the general presentation of the
"ambiguous occurrence" error, and adds a bit of extra context in the
case of data families.
Fixes #23301
- - - - -
2f571afe by Sylvain Henry at 2023-05-15T14:50:07-04:00
Fix GHCJS OS platform (fix #23346)
- - - - -
86aae570 by Oleg Grenrus at 2023-05-15T14:50:43-04:00
Split DynFlags structure into own module
This will allow to make command line parsing to depend on
diagnostic system (which depends on dynflags)
- - - - -
fbe3fe00 by Josh Meredith at 2023-05-15T18:01:43-04:00
Replace the implementation of CodeBuffers with unboxed types
- - - - -
21f3aae7 by Josh Meredith at 2023-05-15T18:01:43-04:00
Use unboxed codebuffers in base
Metric Decrease:
encodingAllocations
- - - - -
18ea2295 by Ben Gamari at 2023-05-15T18:02:20-04:00
rts: Weak pointer cleanups
Various stylistic cleanups. No functional changes.
- - - - -
c343112f by Ben Gamari at 2023-05-15T18:02:20-04:00
rts: Don't force debug output to stderr
Previously `+RTS -Dw -l` would emit debug output to the eventlog while
`+RTS -l -Dw` would emit it to stderr. This was because the parser for
`-D` would unconditionally override the debug output target. Now we
instead only do so if no it is currently `TRACE_NONE`.
- - - - -
a5f5f067 by Ben Gamari at 2023-05-15T18:02:20-04:00
rts: Forcibly flush eventlog on barf
Previously we would attempt to flush via `endEventLogging` which can
easily deadlock, e.g., if `barf` fails during GC.
Using `flushEventLog` directly may result in slightly less consistent
eventlog output (since we don't take all capabilities before flushing)
but avoids deadlocking.
- - - - -
73b1e87c by Ben Gamari at 2023-05-15T18:02:20-04:00
rts: Assert that pointers aren't cleared by -DZ
This turns many segmentation faults into much easier-to-debug assertion
failures by ensuring that LOOKS_LIKE_*_PTR checks recognize bit-patterns
produced by `+RTS -DZ` clearing as invalid pointers.
This is a bit ad-hoc but this is the debug runtime.
- - - - -
37fb61d8 by Ben Gamari at 2023-05-15T18:02:20-04:00
rts: Introduce printGlobalThreads
- - - - -
451d65a6 by Ben Gamari at 2023-05-15T18:02:20-04:00
rts: Don't sanity-check StgTSO.global_link
See Note [Avoid dangling global_link pointers].
Fixes #19146.
- - - - -
d69cbd78 by sheaf at 2023-05-15T18:03:00-04:00
Split up tyThingToIfaceDecl from GHC.Iface.Make
This commit moves tyThingToIfaceDecl and coAxiomToIfaceDecl
from GHC.Iface.Make into GHC.Iface.Decl.
This avoids GHC.Types.TyThing.Ppr, which needs tyThingToIfaceDecl,
transitively depending on e.g. GHC.Iface.Load and GHC.Tc.Utils.Monad.
- - - - -
4d29ecdf by sheaf at 2023-05-15T18:03:00-04:00
Migrate errors to diagnostics in GHC.Tc.Module
This commit migrates the errors in GHC.Tc.Module to use the new
diagnostic infrastructure.
It required a significant overhaul of the compatibility checks between
an hs-boot or signature module and its implementation; we now use
a Writer monad to accumulate errors; see the BootMismatch datatype
in GHC.Tc.Errors.Types, with its panoply of subtypes.
For the sake of readability, several local functions inside the
'checkBootTyCon' function were split off into top-level functions.
We split off GHC.Types.HscSource into a "boot or sig" vs "normal hs file"
datatype, as this mirrors the logic in several other places where we
want to treat hs-boot and hsig files in a similar fashion.
This commit also refactors the Backpack checks for type synonyms
implementing abstract data, to correctly reject implementations that
contain qualified or quantified types (this fixes #23342 and #23344).
- - - - -
d986c98e by Rodrigo Mesquita at 2023-05-16T00:14:04-04:00
configure: Drop unused AC_PROG_CPP
In configure, we were calling `AC_PROG_CPP` but never making use of the
$CPP variable it sets or reads.
The issue is $CPP will show up in the --help output of configure,
falsely advertising a configuration option that does nothing.
The reason we don't use the $CPP variable is because HS_CPP_CMD is
expected to be a single command (without flags), but AC_PROG_CPP, when
CPP is unset, will set said variable to something like `/usr/bin/gcc -E`.
Instead, we configure HS_CPP_CMD through $CC.
- - - - -
a8f0435f by Cheng Shao at 2023-05-16T00:14:42-04:00
rts: fix --disable-large-address-space
This patch moves
ACQUIRE_ALLOC_BLOCK_SPIN_LOCK/RELEASE_ALLOC_BLOCK_SPIN_LOCK from
Storage.h to HeapAlloc.h. When --disable-large-address-space is passed
to configure, the code in HeapAlloc.h makes use of these two macros.
Fixes #23385.
- - - - -
bdb93cd2 by Oleg Grenrus at 2023-05-16T07:59:21+03:00
Add -Wmissing-role-annotations
Implements #22702
- - - - -
41ecfc34 by Ben Gamari at 2023-05-16T07:28:15-04:00
base: Export {get,set}ExceptionFinalizer from System.Mem.Weak
As proposed in CLC Proposal #126 [1].
[1]: https://github.com/haskell/core-libraries-committee/issues/126
- - - - -
67330303 by Ben Gamari at 2023-05-16T07:28:16-04:00
base: Introduce printToHandleFinalizerExceptionHandler
- - - - -
5e3f9bb5 by Josh Meredith at 2023-05-16T13:59:22-04:00
JS: Implement h$clock_gettime in the JavaScript RTS (#23360)
- - - - -
90e69d5d by Zubin Duggal at 2023-05-16T14:00:00-04:00
compiler: Use compact representation for SourceText
SourceText is serialized along with INLINE pragmas into interface files. Many of
these SourceTexts are identical, for example "{-# INLINE#". When deserialized,
each such SourceText was previously expanded out into a [Char], which is highly
wasteful of memory, and each such instance of the text would allocate an
independent list with its contents as deserializing breaks any sharing that might
have existed.
Instead, we use a `FastString` to represent these, so that each instance unique
text will be interned and stored in a memory efficient manner.
- - - - -
b70bc690 by Zubin Duggal at 2023-05-16T14:00:00-04:00
compiler: Use compact representation/FastStrings for `SourceNote`s
`SourceNote`s should not be stored as [Char] as this is highly wasteful
and in certain scenarios can be highly duplicated.
Metric Decrease:
hard_hole_fits
- - - - -
6231a126 by Zubin Duggal at 2023-05-16T14:00:00-04:00
compiler: Use compact representation for UsageFile (#22744)
Use FastString to store filepaths in interface files, as this data is
highly redundant so we want to share all instances of filepaths in the
compiler session.
- - - - -
47a58150 by Zubin Duggal at 2023-05-16T14:00:00-04:00
testsuite: add test for T22744
This test checks for #22744 by compiling 100 modules which each have
a dependency on 1000 distinct external files.
Previously, when loading these interfaces from disk, each individual instance
of a filepath in the interface will would be allocated as an individual object
on the heap, meaning we have heap objects for 100*1000 files, when there are
only 1000 distinct files we care about.
This test checks this by first compiling the module normally, then measuring
the peak memory usage in a no-op recompile, as the recompilation checking will
force the allocation of all these filepaths.
- - - - -
0451bdc9 by Ben Gamari at 2023-05-16T21:31:40-04:00
users guide: Add glossary
Currently this merely explains the meaning of "technology preview" in
the context of released features.
- - - - -
0ba52e4e by Ben Gamari at 2023-05-16T21:31:40-04:00
Update glossary.rst
- - - - -
3d23060c by Ben Gamari at 2023-05-16T21:31:40-04:00
Use glossary directive
- - - - -
2972fd66 by Sylvain Henry at 2023-05-16T21:32:20-04:00
JS: fix getpid (fix #23399)
- - - - -
5fe1d3e6 by Matthew Pickering at 2023-05-17T21:42:00-04:00
Use setSrcSpan rather than setLclEnv in solveForAll
In subsequent MRs (#23409) we want to remove the TcLclEnv argument from
a CtLoc. This MR prepares us for that by removing the one place where
the entire TcLclEnv is used, by using it more precisely to just set the
contexts source location.
Fixes #23390
- - - - -
385edb65 by Torsten Schmits at 2023-05-17T21:42:40-04:00
Update the users guide paragraph on -O in GHCi
In relation to #23056
- - - - -
87626ef0 by Ben Gamari at 2023-05-18T15:18:53-04:00
base: Add test for #13660
- - - - -
9eef53b1 by Ben Gamari at 2023-05-18T15:18:53-04:00
base: Move implementation of GHC.Foreign to GHC.Internal
- - - - -
174ea2fa by Ben Gamari at 2023-05-18T15:18:53-04:00
base: Introduce {new,with}CStringLen0
These are useful helpers for implementing the internal-NUL code unit
check needed to fix #13660.
- - - - -
a46ced16 by Ben Gamari at 2023-05-18T15:18:53-04:00
base: Clean up documentation
- - - - -
b98d99cc by Ben Gamari at 2023-05-18T15:18:53-04:00
base: Ensure that FilePaths don't contain NULs
POSIX filepaths may not contain the NUL octet but previously we did not
reject such paths. This could be exploited by untrusted input to cause
discrepancies between various `FilePath` queries and the opened
filename. For instance, `readFile "hello.so\x00.txt"` would open the
file `"hello.so"` yet `takeFileExtension` would return `".txt"`.
The same argument applies to Windows FilePaths
Fixes #13660.
- - - - -
7ae45459 by Simon Peyton Jones at 2023-05-18T15:19:29-04:00
Allow the demand analyser to unpack tuple and equality dictionaries
Addresses #23398. The demand analyser usually does not unpack class
dictionaries: see Note [Do not unbox class dictionaries] in
GHC.Core.Opt.DmdAnal.
This patch makes an exception for tuple dictionaries and equality
dictionaries, for reasons explained in wrinkles (DNB1) and (DNB2) of
the above Note.
Compile times fall by 0.1% for some reason (max 0.7% on T18698b).
- - - - -
b53a9086 by Greg Steuck at 2023-05-18T15:20:08-04:00
Use a simpler and more portable construct in ld.ldd check
printf '%q\n' is a bash extension which led to incorrectly
failing an ld.lld test on OpenBSD which uses pdksh as /bin/sh
- - - - -
dd5710af by Torsten Schmits at 2023-05-18T15:20:50-04:00
Update the warning about interpreter optimizations
to reflect that they're not incompatible anymore, but guarded by a flag
- - - - -
4f6dd999 by Matthew Pickering at 2023-05-18T15:21:26-04:00
Remove stray dump flags in GHC.Rename.Names
- - - - -
4bca0486 by Oleg Grenrus at 2023-05-19T11:51:33+03:00
Make Warn = Located DriverMessage
This change makes command line argument parsing use diagnostic
framework for producing warnings.
- - - - -
525ed554 by Simon Peyton Jones at 2023-05-19T10:09:15-04:00
Type inference for data family newtype instances
This patch addresses #23408, a tricky case with data family
newtype instances. Consider
type family TF a where TF Char = Bool
data family DF a
newtype instance DF Bool = MkDF Int
and [W] Int ~R# DF (TF a), with a Given (a ~# Char). We must fully
rewrite the Wanted so the tpye family can fire; that wasn't happening.
- - - - -
c6fb6690 by Peter Trommler at 2023-05-20T03:16:08-04:00
testsuite: fix predicate on rdynamic test
Test rdynamic requires dynamic linking support, which is
orthogonal to RTS linker support. Change the predicate accordingly.
Fixes #23316
- - - - -
735d504e by Matthew Pickering at 2023-05-20T03:16:44-04:00
docs: Use ghc-ticket directive where appropiate in users guide
Using the directive automatically formats and links the ticket
appropiately.
- - - - -
b56d7379 by Sylvain Henry at 2023-05-22T14:21:22-04:00
NCG: remove useless .align directive (#20758)
- - - - -
15b93d2f by Simon Peyton Jones at 2023-05-22T14:21:58-04:00
Add test for #23156
This program had exponential typechecking time in GHC 9.4 and 9.6
- - - - -
2b53f206 by Greg Steuck at 2023-05-22T20:23:11-04:00
Revert "Change hostSupportsRPaths to report False on OpenBSD"
This reverts commit 1e0d8fdb55a38ece34fa6cf214e1d2d46f5f5bf2.
- - - - -
882e43b7 by Greg Steuck at 2023-05-22T20:23:11-04:00
Disable T17414 on OpenBSD
Like on other systems it's not guaranteed that there's sufficient
space in /tmp to write 2G out.
- - - - -
9d531f9a by Greg Steuck at 2023-05-22T20:23:11-04:00
Bring back getExecutablePath to getBaseDir on OpenBSD
Fix #18173
- - - - -
9db0eadd by Krzysztof Gogolewski at 2023-05-22T20:23:47-04:00
Add an error origin for impedance matching (#23427)
- - - - -
33cf4659 by Ben Gamari at 2023-05-23T03:46:20-04:00
testsuite: Add tests for #23146
Both lifted and unlifted variants.
- - - - -
76727617 by Ben Gamari at 2023-05-23T03:46:21-04:00
codeGen: Fix some Haddocks
- - - - -
33a8c348 by Ben Gamari at 2023-05-23T03:46:21-04:00
codeGen: Give proper LFInfo to datacon wrappers
As noted in `Note [Conveying CAF-info and LFInfo between modules]`,
when importing a binding from another module we must ensure that it gets
the appropriate `LambdaFormInfo` if it is in WHNF to ensure that
references to it are tagged correctly.
However, the implementation responsible for doing this,
`GHC.StgToCmm.Closure.mkLFImported`, only dealt with datacon workers and
not wrappers. This lead to the crash of this program in #23146:
module B where
type NP :: [UnliftedType] -> UnliftedType
data NP xs where
UNil :: NP '[]
module A where
import B
fieldsSam :: NP xs -> NP xs -> Bool
fieldsSam UNil UNil = True
x = fieldsSam UNil UNil
Due to its GADT nature, `UNil` produces a trivial wrapper
$WUNil :: NP '[]
$WUNil = UNil @'[] @~(<co:1>)
which is referenced in the RHS of `A.x`. Due to the above-mentioned bug
in `mkLFImported`, the references to `$WUNil` passed to `fieldsSam` were
not tagged. This is problematic as `fieldsSam` expected its arguments to
be tagged as they are unlifted.
The fix is straightforward: extend the logic in `mkLFImported` to cover
(nullary) datacon wrappers as well as workers. This is safe because we
know that the wrapper of a nullary datacon will be in WHNF, even if it
includes equalities evidence (since such equalities are not runtime
relevant).
Thanks to @MangoIV for the great ticket and @alt-romes for his
minimization and help debugging.
Fixes #23146.
- - - - -
2fc18e9e by Rodrigo Mesquita at 2023-05-23T03:46:21-04:00
codeGen: Fix LFInfo of imported datacon wrappers
As noted in #23231 and in the previous commit, we were failing to give a
an LFInfo of LFCon to a nullary datacon wrapper from another module,
failing to properly tag pointers which ultimately led to the
segmentation fault in #23146.
On top of the previous commit which now considers wrappers where we
previously only considered workers, we change the order of the guards so
that we check for the arity of the binding before we check whether it is
a constructor. This allows us to
(1) Correctly assign `LFReEntrant` to imported wrappers whose worker was
nullary, which we previously would fail to do
(2) Remove the `isNullaryRepDataCon` predicate:
(a) which was previously wrong, since it considered wrappers whose
workers had zero-width arguments to be non-nullary and would fail to
give `LFCon` to them
(b) is now unnecessary, since arity == 0 guarantees
- that the worker takes no arguments at all
- and the wrapper takes no arguments and its RHS must be an
application of the worker to zero-width-args only.
- we lint these two items with an assertion that the datacon
`hasNoNonZeroWidthArgs`
We also update `isTagged` to use the new logic in determining the
LFInfos of imported Ids.
The creation of LFInfos for imported Ids and this detail are explained
in Note [The LFInfo of Imported Ids].
Note that before the patch to those issues we would already consider these
nullary wrappers to have `LFCon` lambda form info; but failed to re-construct
that information in `mkLFImported`
Closes #23231, #23146
(I've additionally batched some fixes to documentation I found while
investigating this issue)
- - - - -
0598f7f0 by Rodrigo Mesquita at 2023-05-23T03:46:21-04:00
Make LFInfos for DataCons on construction
As a result of the discussion in !10165, we decided to amend the
previous commit which fixed the logic of `mkLFImported` with regard to
datacon workers and wrappers.
Instead of having the logic for the LFInfo of datacons be in
`mkLFImported`, we now construct an LFInfo for all data constructors on
GHC.Types.Id.Make and store it in the `lfInfo` field.
See the new Note [LFInfo of DataCon workers and wrappers] and
ammendments to Note [The LFInfo of Imported Ids]
- - - - -
12294b22 by Rodrigo Mesquita at 2023-05-23T03:46:21-04:00
Update Note [Core letrec invariant]
Authored by @simonpj
- - - - -
e93ab972 by Rodrigo Mesquita at 2023-05-23T03:46:21-04:00
Rename mkLFImported to importedIdLFInfo
The `mkLFImported` sounded too much like a constructor of sorts, when
really it got the `LFInfo` of an imported Id from its `lf_info` field
when this existed, and otherwise returned a conservative estimate of
that imported Id's LFInfo. This in contrast to functions such as
`mkLFReEntrant` which really are about constructing an `LFInfo`.
- - - - -
e54d9259 by Rodrigo Mesquita at 2023-05-23T03:46:21-04:00
Enforce invariant on typePrimRepArgs in the types
As part of the documentation effort in !10165 I came across this
invariant on 'typePrimRepArgs' which is easily expressed at the
type-level through a NonEmpty list.
It allowed us to remove one panic.
- - - - -
b8fe6a0c by Rodrigo Mesquita at 2023-05-23T03:46:21-04:00
Merge outdated Note [Data con representation] into Note [Data constructor representation]
Introduce new Note [Constructor applications in STG] to better support
the merge, and reference it from the relevant bits in the STG syntax.
- - - - -
e1590ddc by Simon Peyton Jones at 2023-05-23T03:46:57-04:00
Add the SolverStage monad
This refactoring makes a substantial improvement in the
structure of the type-checker's constraint solver: #23070.
Specifically:
* Introduced the SolverStage monad. See GHC.Tc.Solver.Monad
Note [The SolverStage monad]
* Make each solver pipeline (equalities, dictionaries, irreds etc)
deal with updating the inert set, as a separate SolverStage. There
is sometimes special stuff to do, and it means that each full
pipeline can have type SolverStage Void, indicating that they never
return anything.
* Made GHC.Tc.Solver.Equality.zonkEqTypes into a SolverStage. Much nicer.
* Combined the remnants of GHC.Tc.Solver.Canonical and
GHC.Tc.Solver.Interact into a new module GHC.Tc.Solver.Solve.
(Interact and Canonical are removed.)
* Gave the same treatment to dictionary and irred constraints
as I have already done for equality constraints:
* New types (akin to EqCt): IrredCt and DictCt
* Ct is now just a simple sum type
data Ct
= CDictCan DictCt
| CIrredCan IrredCt
| CEqCan EqCt
| CQuantCan QCInst
| CNonCanonical CtEvidence
* inert_dicts can now have the better type DictMap DictCt, instead of
DictMap Ct; and similarly inert_irreds.
* Significantly simplified the treatment of implicit parameters.
Previously we had a number of special cases
* interactGivenIP, an entire function
* special case in maybeKickOut
* special case in findDict, when looking up dictionaries
But actually it's simpler than that. When adding a new Given, implicit
parameter constraint to the InertSet, we just need to kick out any
existing inert constraints that mention that implicit parameter.
The main work is done in GHC.Tc.Solver.InertSet.delIPDict, along with
its auxiliary GHC.Core.Predicate.mentionsIP.
See Note [Shadowing of implicit parameters] in GHC.Tc.Solver.Dict.
* Add a new fast-path in GHC.Tc.Errors.Hole.tcCheckHoleFit.
See Note [Fast path for tcCheckHoleFit]. This is a big win in some cases:
test hard_hole_fits gets nearly 40% faster (at compile time).
* Add a new fast-path for solving /boxed/ equality constraints
(t1 ~ t2). See Note [Solving equality classes] in GHC.Tc.Solver.Dict.
This makes a big difference too: test T17836 compiles 40% faster.
* Implement the PermissivePlan of #23413, which concerns what happens with
insoluble Givens. Our previous treatment was wildly inconsistent as that
ticket pointed out.
A part of this, I simplified GHC.Tc.Validity.checkAmbiguity: now we simply
don't run the ambiguity check at all if -XAllowAmbiguousTypes is on.
Smaller points:
* In `GHC.Tc.Errors.misMatchOrCND` instead of having a special case for
insoluble /occurs/ checks, broaden in to all insouluble constraints.
Just generally better. See Note [Insoluble mis-match] in that module.
As noted above, compile time perf gets better. Here are the changes
over 0.5% on Fedora. (The figures are slightly larger on Windows for
some reason.)
Metrics: compile_time/bytes allocated
-------------------------------------
LargeRecord(normal) -0.9%
MultiLayerModulesTH_OneShot(normal) +0.5%
T11822(normal) -0.6%
T12227(normal) -1.8% GOOD
T12545(normal) -0.5%
T13035(normal) -0.6%
T15703(normal) -1.4% GOOD
T16875(normal) -0.5%
T17836(normal) -40.7% GOOD
T17836b(normal) -12.3% GOOD
T17977b(normal) -0.5%
T5837(normal) -1.1%
T8095(normal) -2.7% GOOD
T9020(optasm) -1.1%
hard_hole_fits(normal) -37.0% GOOD
geo. mean -1.3%
minimum -40.7%
maximum +0.5%
Metric Decrease:
T12227
T15703
T17836
T17836b
T8095
hard_hole_fits
LargeRecord
T9198
T13035
- - - - -
6abf3648 by Simon Peyton Jones at 2023-05-23T03:46:57-04:00
Avoid an assertion failure in abstractFloats
The function GHC.Core.Opt.Simplify.Utils.abstractFloats
was carelessly calling lookupIdSubst_maybe on a CoVar;
but a precondition of the latter is being given an Id.
In fact it's harmless to call it on a CoVar, but still, the
precondition on lookupIdSubst_maybe makes sense, so I added
a test for CoVars.
This avoids a crash in a DEBUG compiler, but otherwise has
no effect. Fixes #23426.
- - - - -
838aaf4b by hainq at 2023-05-24T12:41:19-04:00
Migrate errors in GHC.Tc.Validity
This patch migrates the error messages in GHC.Tc.Validity to use
the new diagnostic infrastructure.
It adds the constructors:
- TcRnSimplifiableConstraint
- TcRnArityMismatch
- TcRnIllegalInstanceDecl, with sub-datatypes for HasField errors
and fundep coverage condition errors.
- - - - -
8539764b by Krzysztof Gogolewski at 2023-05-24T12:41:56-04:00
linear lint: Add missing processing of DEFAULT
In this correct program
f :: a %1 -> a
f x = case x of x { _DEFAULT -> x }
after checking the alternative we weren't popping the case binder 'x'
from the usage environment, which meant that the lambda-bound 'x'
was counted twice: in the scrutinee and (incorrectly) in the alternative.
In fact, we weren't checking the usage of 'x' at all.
Now the code for handling _DEFAULT is similar to the one handling
data constructors.
Fixes #23025.
- - - - -
ae683454 by Matthew Pickering at 2023-05-24T12:42:32-04:00
Remove outdated "Don't check hs-boot type family instances too early" note
This note was introduced in 25b70a29f623 which delayed performing some
consistency checks for type families. However, the change was reverted
later in 6998772043a7f0b0360116eb5ffcbaa5630b21fb but the note was not
removed.
I found it confusing when reading to code to try and work out what
special behaviour there was for hs-boot files (when in-fact there isn't
any).
- - - - -
44af57de by Matthew Pickering at 2023-05-24T12:43:08-04:00
rts: Define ticky macro stubs
These macros have long been undefined which has meant we were missing
reporting these allocations in ticky profiles.
The most critical missing definition was TICK_ALLOC_HEAP_NOCTR which was
missing all the RTS calls to allocate, this leads to a the overall
ALLOC_RTS_tot number to be severaly underreported.
Of particular interest though is the ALLOC_STACK_ctr and ALLOC_STACK_tot
counters which are useful to tracking stack allocations.
Fixes #23421
- - - - -
b2dabe3a by Matthew Pickering at 2023-05-24T12:43:08-04:00
rts: ticky: Rename TICK_ALLOC_HEAP_NOCTR to TICK_ALLOC_RTS
This macro increments the ALLOC_HEAP_tot and ALLOC_HEAP_ctr so it makes
more sense to name it after that rather than the suffix NOCTR, whose
meaning has been lost to the mists of time.
- - - - -
eac4420a by Ben Gamari at 2023-05-24T12:43:45-04:00
users guide: A few small mark-up fixes
- - - - -
a320ca76 by Rodrigo Mesquita at 2023-05-24T12:44:20-04:00
configure: Fix support check for response files.
In failing to escape the '-o' in '-o\nconftest\nconftest.o\n' argument
to printf, the writing of the arguments response file always failed.
The fix is to pass the arguments after `--` so that they are treated
positional arguments rather than flags to printf.
Closes #23435
- - - - -
f21ce0e4 by mangoiv at 2023-05-24T12:45:00-04:00
[feat] add .direnv to the .gitignore file
- - - - -
36d5944d by Bodigrim at 2023-05-24T20:58:34-04:00
Add Data.List.unsnoc
See https://github.com/haskell/core-libraries-committee/issues/165 for discussion
- - - - -
c0f2f9e3 by Bartłomiej Cieślar at 2023-05-24T20:59:14-04:00
Fix crash in backpack signature merging with -ddump-rn-trace
In some cases, backpack signature merging could crash in addUsedGRE
when -ddump-rn-trace was enabled, as pretty-printing the GREInfo would cause
unavailable interfaces to be loaded.
This commit fixes that issue by not pretty-printing the GREInfo in addUsedGRE
when -ddump-rn-trace is enabled.
Fixes #23424
Signed-off-by: Bartłomiej Cieślar <bcieslar2001 at gmail.com>
- - - - -
5a07d94a by Krzysztof Gogolewski at 2023-05-25T03:30:20-04:00
Add a regression test for #13981
The panic was fixed by 6998772043a7f0b. Fixes #13981.
- - - - -
182df90e by Krzysztof Gogolewski at 2023-05-25T03:30:57-04:00
Add a test for #23355
It was fixed by !10061, so I'm adding it in the same group.
- - - - -
1b31b039 by uhbif19 at 2023-05-25T12:08:28+02:00
Migrate errors in GHC.Rename.Splice GHC.Rename.Pat
This commit migrates the errors in GHC.Rename.Splice and GHC.Rename.Pat
to use the new diagnostic infrastructure.
- - - - -
56abe494 by sheaf at 2023-05-25T12:09:55+02:00
Common up Template Haskell errors in TcRnMessage
This commit commons up the various Template Haskell errors into a
single constructor, TcRnTHError, of TcRnMessage.
- - - - -
a487ba9e by Krzysztof Gogolewski at 2023-05-25T14:35:56-04:00
Enable ghci tests for unboxed tuples
The tests were originally skipped because ghci used not to support
unboxed tuples/sums.
- - - - -
60317bcb by Finley McIlwaine at 2023-05-25T13:40:36-06:00
Restructure IPE buffer layout
Reference ticket #21766
This commit restructures IPE buffer list entries to not contain
references to their corresponding info tables. IPE buffer list nodes now
point to two lists of equal length, one holding the list of info table
pointers and one holding the corresponding entries for each info table.
This will allow the entry data to be compressed without losing the
references to the info tables.
- - - - -
2bd8ab36 by Finley McIlwaine at 2023-05-25T13:42:22-06:00
Add IPE compression to configure
Reference ticket #21766
Adds an `--enable-ipe-data-compreesion` flag to the configure script
which will check for libzstd and set the appropriate flags to allow for
IPE data compression in the compiler
- - - - -
30904e96 by Finley McIlwaine at 2023-05-25T13:43:25-06:00
IPE data compression
Reference ticket #21766
When IPE data compression is enabled, compress the emitted IPE buffer
entries and decompress them in the RTS.
- - - - -
edf2ef30 by Finley McIlwaine at 2023-05-25T13:43:32-06:00
Fix libzstd detection in configure and RTS
Ensure that `HAVE_LIBZSTD` gets defined to either 0 or 1 in all cases
and properly check that before IPE data decompression in the RTS. See
ticket #21766.
- - - - -
083f1a1b by Finley McIlwaine at 2023-05-25T13:43:32-06:00
Add note describing IPE data compression
See ticket #21766
- - - - -
7616cf8f by Finley McIlwaine at 2023-05-25T13:43:32-06:00
Fix byte order of IPE data, fix IPE tests
Make sure byte order of written IPE buffer entries matches target.
Make sure the IPE-related tests properly access the fields of IPE buffer
entry nodes with the new IPE layout.
This commit also introduces checks to avoid importing modules if IPE
compression is not enabled.
See ticket #21766.
- - - - -
6f588fb8 by Finley McIlwaine at 2023-05-25T13:43:32-06:00
Fix IPE data decompression buffer allocation
Capacity of buffers allocated for decompressed IPE data was
incorrect due to a misuse of the `ZSTD_findFrameCompressedSize`
function. Fix by always storing decompressed size of IPE data in IPE
buffer list nodes and using `ZSTD_findFrameCompressedSize` to determine
the size of the compressed data.
See ticket #21766
- - - - -
7d34d351 by Finley McIlwaine at 2023-05-25T13:43:32-06:00
Add optional dependencies to ./configure output
Changes the configure script to indicate whether libnuma, libzstd, or
libdw are being used as dependencies due to their optional features
being enabled.
- - - - -
0793beea by Finley McIlwaine at 2023-05-25T13:43:32-06:00
Add IPE-enabled builds to CI
- Adds an IPE job to the CI pipeline which is triggered by the ~IPE label
- Introduces CI logic to enable IPE data compression
- Enables uncompressed IPE data on debug CI job
- Regenerates jobs.yaml
MR https://gitlab.haskell.org/ghc/ci-images/-/merge_requests/112 on the
images repository is meant to ensure that the proper images have
libzstd-dev installed.
- - - - -
f2b76b9c by Finley McIlwaine at 2023-05-25T13:43:59-06:00
Update user's guide and release notes
Add mention of IPE data compression to user's guide and the release
notes for 9.8.1. Also note the impact compression has on binary size in
both places.
See ticket #21766
- - - - -
30 changed files:
- .gitignore
- .gitlab-ci.yml
- .gitlab/ci.sh
- .gitlab/gen_ci.hs
- .gitlab/jobs.yaml
- .gitlab/rel_eng/fetch-gitlab-artifacts/fetch_gitlab.py
- .gitlab/rel_eng/upload_ghc_libs.py
- .gitmodules
- cabal.project-reinstall
- compiler/GHC.hs
- compiler/GHC/Builtin/Names.hs
- compiler/GHC/Builtin/Types.hs
- compiler/GHC/Builtin/Types/Prim.hs
- compiler/GHC/Builtin/primops.txt.pp
- compiler/GHC/ByteCode/Asm.hs
- compiler/GHC/ByteCode/Instr.hs
- compiler/GHC/Cmm/CLabel.hs
- compiler/GHC/Cmm/MachOp.hs
- compiler/GHC/Cmm/Parser.y
- compiler/GHC/CmmToAsm/AArch64/CodeGen.hs
- compiler/GHC/CmmToAsm/AArch64/Instr.hs
- compiler/GHC/CmmToAsm/AArch64/Ppr.hs
- compiler/GHC/CmmToAsm/Dwarf.hs
- compiler/GHC/CmmToAsm/PPC/CodeGen.hs
- compiler/GHC/CmmToAsm/PPC/Instr.hs
- compiler/GHC/CmmToAsm/PPC/Ppr.hs
- compiler/GHC/CmmToAsm/Wasm/FromCmm.hs
- compiler/GHC/CmmToAsm/X86/CodeGen.hs
- compiler/GHC/CmmToAsm/X86/Instr.hs
- compiler/GHC/CmmToAsm/X86/Ppr.hs
The diff was not included because it is too large.
View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/ab0fe323b235236798e6c2fd37b5ab409e0029ff...f2b76b9c6faba11f99197c0908d7f557a1136485
--
View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/ab0fe323b235236798e6c2fd37b5ab409e0029ff...f2b76b9c6faba11f99197c0908d7f557a1136485
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/20230525/d16d3077/attachment-0001.html>
More information about the ghc-commits
mailing list