| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
There were two problems with LLVM version checking:
- The parser would only parse x and x.y formatted versions. E.g. 1.2.3
would be rejected.
- The version check was too strict and would reject x.y formatted
versions. E.g. when we support version 7 it'd reject 7.0 ("LLVM
version 7.0") and only accept 7 ("LLVM version 7").
We now parse versions with arbitrarily deep minor numbering (x.y.z.t...)
and accept versions as long as the major version matches the supported
version (e.g. 7.1, 7.1.2, 7.1.2.3 ...).
(cherry picked from commit bf9dfe1ca32270f5e946e0f8ac1bb97184de6e4c)
|
| |
|
|
|
|
|
| |
This was inadvertently referring to the cabal-install-latest/ directory
which is volatile.
(cherry picked from commit fc746e98d8ee7ac22224ba7d7fd1c38e16dfad30)
|
| | |
|
| |
|
|
| |
This was inadvertently reverted in 0bce423c016a02a0e384378eb729ffeb79df005b
|
| | |
|
| |
|
|
| |
Ensures that #15934 doesn't break the 32-bit Windows release build.
|
| | |
|
| |
|
|
|
|
|
| |
For reasons I don't understand sphinx seems to fail to produce a .idx
file for makeindex.
(cherry picked from commit 9721b40db9b87c0135613649f9f1e129eb49795b)
|
| | |
|
| |
|
|
|
|
| |
When floating a single-alternative case we previously would set the
context level to the level where we were floating the case. However,
this is not what FloatOut did: it rather left
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
It is important that `heapCensus` and `LdvCensusForDead` traverse the
same areas.
`heapCensus` increases the `not_used` counter which tracks how many
closures are live but haven't been used yet.
`LdvCensusForDead` increases the `void_total` counter which tracks how
many dead closures there are.
The `LAG` is then calculated by substracting the `void_total` from
`not_used` and so it is essential that `not_used >= void_total`. This
fact is checked by quite a few assertions.
However, if a program has low maximum residency but allocates a lot in
the nursery then these assertions were failing (see #16753 and #15903)
because `LdvCensusForDead` was observing dead closures from the nursery
which totalled more than the `not_used`. The same closures were not
counted by `heapCensus`.
Therefore, it seems that the correct fix is to make `LdvCensusForDead`
agree with `heapCensus` and not traverse the nursery for dead closures.
Fixes #16100 #16753 #15903 #8982
(cherry picked from commit 1dc61c1ae7b80822612844d760eaebe6eb1f0ed4)
|
| |
|
|
|
|
|
| |
It is possible that void_total is exactly equal to not_used and the
other assertions for this check for <= rather than <.
(cherry picked from commit a196d9c391aa488d58fb3d787637c6c0402958cb)
|
| |
|
|
|
|
|
|
|
|
|
| |
This implements the correct fix for #11627 by skipping over the slop
(which is zeroed) rather than adding special case logic for LARGE
ARR_WORDS which runs the risk of not performing a correct census by
ignoring any subsequent blocks.
This approach implements similar logic to that in Sanity.c
(cherry picked from commit 808a87ff9be4d92d3124c7d70104f6a339880021)
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
The simple optimiser was making an invalid transformation
to join points -- yikes. The fix is easy.
I also added some documentation about the fact that GHC uses
a slightly more restrictive version of join points than does
the paper.
Fix #16918
(cherry picked from commit d2e290d3280841647354ddf5ca9abdd974bce0d5)
|
| |
|
|
|
|
|
|
| |
When `join_ids` is empty `extendVarSetList existing_joins join_ids` is
already no-op, so no need to check whether `join_ids` is empty or not
before extending the joins set.
(cherry picked from commit 897a59a5c0a08985ddccf00a4961cc2d080324e8)
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
| |
For some reason configure seems unable to find it on its own. Let's try
giving it a hint.
Addresses #16398.
(cherry picked from commit 7e404afd7c923bf275093c7fa1b278d484cfdd82)
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Ticket #16978 showed that we were floating a recursive,
unlifted join point to top level. It's very much a corner
case:
joinrec j :: Int#
j = jump j
in ...
But somehow it showed up in a real program.
For non-recursive bindings in SetLevels.lvlBind we were already
(correctly) checking for unlifted bindings, but when I wrote
that code I didn't think that a /recursive/ binding could be
unlifted but /join-points/ can be!
Actually I don't think that SetLevels should be floating
join points at all. SetLevels really floats things to move
stuff out of loops and save allocation; but none of that applies
to join points. The only reason to float join points is in
cases like
join j1 x = join j2 y = ...
in ...
which we might want to swizzle to
join j2 x y = ... in
join j1 x = ...
in ...
because now j1 looks small and might be inlined away altogether.
But this is a very local float perhaps better done in the simplifier.
Still: this patch fixes the crash, and does so in a way that is
harmless if/when we change our strategy for floating join points.
(cherry picked from commit 7d8d0012acd8701c0bb562376fd8321009342dcd)
|
| |
|
|
| |
(cherry picked from commit 3b31a94df4cf7e55d93dfcad9b96d0f49f4d3f8d)
|
| | |
|
| |
|
|
|
|
| |
Issue #16893 reveals that the unsafeCoercion# in eqTypeRep can float in
a way that compromises type safety. A real solution for this sadly won't
be ready for 8.10 so this is a stop-gap measure for 8.8.
|
| |
|
|
|
|
|
|
|
|
|
| |
The `mkEtaWW` case for newtypes forgot to apply the substitution to
the newtype coercion, resulting in the Core Lint errors observed
in #16979. Easily fixed.
Fixes #16979.
Co-authored-by: Ryan Scott <ryan.gl.scott@gmail.com>
(cherry picked from commit 2829f6dab5e860e61dba970a536709380c9d993d)
|
| |
|
|
|
|
|
|
| |
This reverts commit 4e1dfc3767167dddd0e151a2df8305b12aa0f49c.
Due to #16943.
(cherry picked from commit bbdd2c9677ff608873e7eaba984c43fafb8d7273)
|
| |
|
|
|
|
|
|
|
| |
This adds lookup logic for _GLOBAL_OFFSET_TABLE_ as well as
relocation logic for R_ARM_BASE_PREL and R_ARM_GOT_BREL which
the gnu toolchain (gas, gcc, ...) prefers to produce. Apparently
recent llvm toolchains will produce those as well.
(cherry picked from commit 826939387be69b100461dd553937dfea729b15ef)
|
| |
|
|
|
|
|
|
|
|
|
|
| |
To display the free variables for a single breakpoint, GHCi pulls out the
information from the fields `modBreaks_breakInfo` and `modBreaks_vars`
of the `ModBreaks` data structure. For a specific breakpoint this gives 2
lists of types 'Id` (`Var`) and `OccName`. They are used to create the Id's
for the free variables and must be kept in sync:
If we remove an element from the Names list, then we also must remove the
corresponding element from the OccNames list.
(cherry picked from commit 32be44613fed3fa7bff7190381acbdaa8ea15cfc)
|
| | |
|
| |
|
|
| |
It looks like this was an incomplete cherry-pick.
|
| |
|
|
|
| |
Backport of the master changes to the comment to prevent merge
conflicts.
|
| |
|
|
|
|
|
|
|
| |
1. Slightly increased the waiting time for the tested effect to be more
profound.
2. Introduced measuring of the actual time spent waiting and adjusing
CPU time by it to compensate for threadDelay waiting time
inconsistencies.
|
| |
|
|
|
|
|
|
|
|
|
| |
Kqueue/kevent implementation used to ignore events to be unsubscribed
from when events to be subscribed to were provided. This resulted in a
lost notification subscription, when GHC runtime didn't listen for any
events, yet the kernel considered otherwise and kept waking up the IO
manager thread.
This commit fixes this issue by always adding and removing all of the
provided subscriptions.
|
| |
|
|
|
|
|
|
|
|
| |
* Use show! in source tarball job. Since we aren't actually building
anything in this job `show` won't
work.
* Fix Docker image name
* Make `version` file contain only version string
|
| | |
|
| |
|
|
|
| |
As fromFlag is partial. The only case where we used fromFlag is when
determining whether to strip libraries; we now assume that we shouldn't.
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
|
| |
This will hopefully become release 0.10.8.3.
|
| |\ |
|
| | | |
|
| | |
| |
| |
| |
| |
| | |
See Trac #15072, Trac #16349, Trac #16350
(cherry picked from commit 14586f5d737ec5dc828633267b50dcf0d47e1696)
|
| | |
| |
| |
| |
| | |
This is the initial commit of a utility that I have been using to
compare the user-facing interfaces exposed by GHC across releases
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Here the following changes are introduced:
- A read barrier machine op is added to Cmm.
- The order in which a closure's fields are read and written is changed.
- Memory barriers are added to RTS code to ensure correctness on
out-or-order machines with weak memory ordering.
Cmm has a new CallishMachOp called MO_ReadBarrier. On weak memory machines, this
is lowered to an instruction that ensures memory reads that occur after said
instruction in program order are not performed before reads coming before said
instruction in program order. On machines with strong memory ordering properties
(e.g. X86, SPARC in TSO mode) no such instruction is necessary, so
MO_ReadBarrier is simply erased. However, such an instruction is necessary on
weakly ordered machines, e.g. ARM and PowerPC.
Weam memory ordering has consequences for how closures are observed and mutated.
For example, consider a closure that needs to be updated to an indirection. In
order for the indirection to be safe for concurrent observers to enter, said
observers must read the indirection's info table before they read the
indirectee. Furthermore, the entering observer makes assumptions about the
closure based on its info table contents, e.g. an INFO_TYPE of IND imples the
closure has an indirectee pointer that is safe to follow.
When a closure is updated with an indirection, both its info table and its
indirectee must be written. With weak memory ordering, these two writes can be
arbitrarily reordered, and perhaps even interleaved with other threads' reads
and writes (in the absence of memory barrier instructions). Consider this
example of a bad reordering:
- An updater writes to a closure's info table (INFO_TYPE is now IND).
- A concurrent observer branches upon reading the closure's INFO_TYPE as IND.
- A concurrent observer reads the closure's indirectee and enters it. (!!!)
- An updater writes the closure's indirectee.
Here the update to the indirectee comes too late and the concurrent observer has
jumped off into the abyss. Speculative execution can also cause us issues,
consider:
- An observer is about to case on a value in closure's info table.
- The observer speculatively reads one or more of closure's fields.
- An updater writes to closure's info table.
- The observer takes a branch based on the new info table value, but with the
old closure fields!
- The updater writes to the closure's other fields, but its too late.
Because of these effects, reads and writes to a closure's info table must be
ordered carefully with respect to reads and writes to the closure's other
fields, and memory barriers must be placed to ensure that reads and writes occur
in program order. Specifically, updates to a closure must follow the following
pattern:
- Update the closure's (non-info table) fields.
- Write barrier.
- Update the closure's info table.
Observing a closure's fields must follow the following pattern:
- Read the closure's info pointer.
- Read barrier.
- Read the closure's (non-info table) fields.
This patch updates RTS code to obey this pattern. This should fix long-standing
SMP bugs on ARM (specifically newer aarch64 microarchitectures supporting
out-of-order execution) and PowerPC. This fixes issue #15449.
Co-Authored-By: Ben Gamari <ben@well-typed.com>
(cherry picked from commit 11bac11545b19a63f5cec3c5bbd5c3f9a7dae0b2)
|
| | |
| |
| |
| | |
(cherry picked from commit 4ec233ecfc7f061c19d0c5ef98ad05719b1161e7)
|
| | |
| |
| |
| | |
(cherry picked from commit 90e0ab7d80d88463df97bc3514fc89d2ab9fcfca)
|
| | |
| |
| |
| | |
(cherry picked from commit d35cec7a9c07a0fc3b40e45d64e5794c3567a523)
|
| | |
| |
| |
| | |
(cherry picked from commit ef6d9a50db115e296d2d9bec3e94c7369f1d504c)
|
