491 Commits

Author SHA1 Message Date
Johannes Doerfert
55a970fbd4 [Attributor][FIX] Make sure to not ignore non-load users of stores
When we look through memory for a store we used to allow any other use
of the memory that is reachable. This is generally OK but we need to
make sure to actually let the user look at these properly. For now,
we simply require loads (via exact reloads).
2022-03-11 18:41:13 -06:00
Johannes Doerfert
f3ad8cf00e [Attributor] Cleanup manifest and liveness for CGSCC passes
There was some ad-hoc handling of liveness and manifest to avoid
breaking CGSCC guarantees. Things always slipped through though.
This cleanup will:

1) Prevent us from manifesting any "information" outside the CGSCC.
   This might be too conservative but we need to opt-in to annotation
   not try to avoid some problematic ones.
2) Avoid running any liveness analysis outside the CGSCC. We did have
   some AAIsDeadFunction handling to this end but we need this for all
   AAIsDead classes. The reason is that AAIsDead information is only
   correct if we actually manifest it, since we don't (see point 1) we
   cannot actually derive/use it at all. We are currently trying to
   avoid running any AA updates outside the CGSCC but that seems to
   impact things quite a bit.
3) Assert, don't check, that our modifications (during cleanup) modifies
   only CGSCC functions.
2022-03-11 16:46:02 -06:00
Johannes Doerfert
7211dbd01d [Attributor][NFCI] Remove non-deterministic behavior and debug output 2022-03-10 23:27:47 -06:00
Johannes Doerfert
5af11ec34b [Attributor] Determine potentially loaded values through memory
We already look through memory to determine where a value that is stored
might pop up again (potential copies). This patch introduces the other
direction with similar logic. If a value is loaded, we can follow all
the accesses to the pointer (or better object) and try to determine what
value might have been stored.
2022-03-06 23:26:37 -06:00
Johannes Doerfert
ad26e199ff [Attributor] Use CFG reasoning also for read accesses
With D106397 we used CFG reasoning to filter out writes that will not
interfere with a given load instruction. With this patch we use the
same logic (modulo the reversal in reachability check order) for store
instructions. As an example, we can now proof stores to shared memory
are dead if all the loads of the shared memory are not reachable from
them.
2022-03-06 23:26:22 -06:00
Johannes Doerfert
acb3773491 [Attributor] Improve isValidAtPosition (mostly for old PM)
To minimize the test difference between old and new PM we perform some
local dominance check if no dominator tree is available.
2022-03-06 23:26:21 -06:00
Johannes Doerfert
efedf70aa5 [Attributor][NFC] Expose helper with more generic interface
This simply makes the function argument of the
`Attributor::checkForAllInstructions` helper explicit so one can iterate
over instructions in other functions.
2022-03-06 19:59:23 -06:00
Nikita Popov
6b5b367858 [Attributor] Remove function pointer type check (NFCI)
This check is not relevant for correctness, it can only avoid
walking some recursive uses if the cast is to a non-function
pointer type. As this distinction will no longer be possible
with opaque pointers and all users will have to be walked
anyway, I'm dropping the check in advance.
2022-03-04 12:09:51 +01:00
serge-sans-paille
a494ae43be Cleanup includes: TransformsUtils
Estimation on the impact on preprocessor output:
before: 1065307662
after:  1064800684

Discourse thread: https://discourse.llvm.org/t/include-what-you-use-include-cleanup
Differential Revision: https://reviews.llvm.org/D120741
2022-03-01 21:00:07 +01:00
Johannes Doerfert
6ed1ef0643 [Attributor][FIX] Pipe UsedAssumedInformation through more interfaces
`UsedAssumedInformation` is a return argument utilized to determine what
information is known. Most APIs used it already but
`genericValueTraversal` did not. This adds it to `genericValueTraversal`
and replaces `AllCallSitesKnown` of `checkForAllCallSites` with the
commonly used `UsedAssumedInformation`.

This was supposed to be a NFC commit, then the test change appeared.
Turns out, we had one user of `AllCallSitesKnown` (AANoReturn) and the
way we set `AllCallSitesKnown` was wrong as we ignored the fact some
call sites were optimistically assumed dead. Included a dedicated test
for this as well now.

Fixes https://github.com/llvm/llvm-project/issues/53884
2022-02-16 14:44:20 -06:00
Sylvestre Ledru
f2c2e924e7 Fix a typo (occured => occurred)
Reported:
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1005195
2022-02-08 21:35:26 +01:00
Joseph Huber
caf7f05c1c [Attributor] Emit fixed-point remark on function list
This patch replaces the function we emit the remark on when we run into
the fix-point limit. Previously we got a function to emit a remark on
from the worklist's associated function. However, the worklist may not
always have an associated function in the case of global variables.
Replace this with the function set, and if there are no functions don't
emit the remark.

Reviewed By: jdoerfert

Differential Revision: https://reviews.llvm.org/D119248
2022-02-08 12:10:21 -05:00
Johannes Doerfert
3b8ffe668d [Attributor][FIX] Relax assertion in IRPosition::verify
A call base can be a floating value if we talk about the instruction and
not the return value. This distinction was not made before but is
important for liveness, e.g., a call site return value might be unused
(=dead) but the call site is not.
2022-02-01 02:25:44 -06:00
Johannes Doerfert
a265cf22af [Attributor] Introduce the AA::isPotentiallyReachable helper APIs
To make usage easier (compared to the many reachability related AAs),
this patch introduces a helper API, `AA::isPotentiallyReachable`, which
performs all the necessary steps. It also does the "backwards"
reachability (see D106720) as that simplifies the AA a lot (backwards
queries were somewhat different from the other query resolvers), and
ensures we use cached values in every stage.

To test inter-procedural reachability in a reasonable way this patch
includes an extension to `AAPointerInfo::forallInterferingWrites`.
Basically, we can exclude writes if they cannot reach a load "during the
lifetime" of the allocation. That is, we need to go up the call graph to
determine reachability until we can determine the allocation would be
dead in the caller. This leads to new constant propagations (through
memory) in `value-simplify-pointer-info-gpu.ll`.

Note: The new code contains plenty debug output to determine how
reachability queries are resolved.

Parts extracted from D110078.

Differential Revision: https://reviews.llvm.org/D118673
2022-02-01 01:40:45 -06:00
Johannes Doerfert
b51b83f68e [Attributor] Introduce the concept of query AAs
D106720 introduced features that did not work properly as we could add
new queries after a fixpoint was reached and which could not be answered
by the information gathered up to the fixpoint alone.

As an alternative to D110078, which forced eager computation where we
want to continue to be lazy, this patch fixes the problem.

QueryAAs are AAs that allow lazy queries during their lifetime. They are
never fixed if they have no outstanding dependences and always run as
part of the updates in an iteration. To determine if we are done, all
query AAs are asked if they received new queries, if not, we only need
to consider updated AAs, as before. If new queries are present we go for
another iteration.

Differential Revision: https://reviews.llvm.org/D118669
2022-02-01 01:40:44 -06:00
Johannes Doerfert
a1db0e523d [Attributor][FIX] Liveness handling in the isAssumedDead helpers
This fixes a conceptual problem with our AAIsDead usage which conflated
call site liveness with call site return value liveness. Without the
fix tests would obviously miscompile as we make genericValueTraversal
more powerful (in a follow up). The effects on the tests are mixed but
mostly marginal. The most prominent one is the lack of `noreturn` for
functions. The reason is that we make entire blocks live at the same
time (for time reasons). Now that we actually look at the block
liveness, which we need to do, the return instructions are live and
will survive. As an example,  `noreturn_async.ll` has been modified
to retain the `noreturn` even with block granularity. We could address
this easily but there is little need in practice.
2022-02-01 01:18:52 -06:00
Johannes Doerfert
adf0d57f15 [Attributor] Provide convenient helpers for isAssumedRead{None,Only}
We have two attributes that can answer readnone queries. While there is
a dependence between them, it seems best to not force the users to know
what AA to ask. The helpers also allow to check for readonly nicely.

Test changes show where we now deduce readnone but haven't before,
mostly because we only asked AAMemoryBehavior and not AAMemoryLocation.
AANoAlias has not been ported to the new API yet.
2022-02-01 01:18:51 -06:00
Johannes Doerfert
191fa419a6 [Attributor][NFC] Make debug output more useful and concise 2022-02-01 01:18:51 -06:00
Johannes Doerfert
3f0e670498 [Attributor][NFCI] Expose some nosync reasoning to outside users.
No-sync is a property that we need in more places as complex
transformations emerge. To simplify the query we provide an
`AA::isNoSyncInst` helper now and expose two existing helpers through
the `AANoSync` class.
2022-02-01 01:07:50 -06:00
Nikita Popov
cf0357a545 [BasicBlockUtils] Fix typo in API name (NFC)
detatch -> detach. As this requires touching all uses, also
lower-case it in accordance with the style guide.
2022-01-28 16:32:13 +01:00
Nikita Popov
92d55e7336 [MemoryBuiltins] Remove isNoAliasFn() in favor of isNoAliasCall()
We currently have two similar implementations of this concept:
isNoAliasCall() only checks for the noalias return attribute.
isNoAliasFn() also checks for allocation functions.

We should switch to only checking the attribute. SLC is responsible
for inferring the noalias return attribute for non-new allocation
functions (with a missing case fixed in
348bc76e35).
For new, clang is responsible for setting the attribute,
if -fno-assume-sane-operator-new is not passed.

Differential Revision: https://reviews.llvm.org/D116800
2022-01-10 09:18:15 +01:00
Philip Reames
6b0ff0969d Extract utility function for checking initial value of allocation [NFC, try 2]
This is a reoccuring pattern, we can consolidate three copies into one.  The main motivation is to reduce usages of isMallocLike.

The original commit (which was quickly reverted) didn't account for the allocation function could be an invoke, test coverage for that case added in this commit.
2022-01-07 08:44:08 -08:00
Philip Reames
c6a0c1585a Revert "Extract utility function for checking initial value of allocation [NFC]"
This reverts commit 9ce30fe86f58df45b2c5daa601802593601c471d.  Appears to be causing a problem on a buildbot, revert while investigating.

https://green.lab.llvm.org/green//job/clang-stage1-RA/26818/consoleFull#-1502953973d489585b-5106-414a-ac11-3ff90657619c
2022-01-06 19:05:51 -08:00
Philip Reames
9ce30fe86f Extract utility function for checking initial value of allocation [NFC]
This is a reoccuring pattern, we can consolidate three copies into one.  The main motivation is to reduce usages of isMallocLike.
2022-01-06 18:02:14 -08:00
Sjoerd Meijer
e550dfa4a6 Silence a few unused variable warnings. NFC. 2022-01-05 09:15:07 +00:00
Johannes Doerfert
5602c866c0 [Attributor] Look through allocated heap memory
AAPointerInfo, and thereby other places, can look already through
internal global and stack memory. This patch enables them to look
through heap memory returned by functions with a `noalias` return.

In the future we can look through `noalias` arguments as well but that
will require AAIsDead to learn that such memory can be inspected by the
caller later on. We also need teach AAPointerInfo about dominance to
actually deal with memory that might not be `null` or `undef`
initialized. D106397 is a first step in that direction already.

Reviewed By: kuter

Differential Revision: https://reviews.llvm.org/D109170
2021-12-29 00:21:36 -06:00
Johannes Doerfert
6e2fcf8513 [Attributor][FIX] Ensure store uses are correlated with reloads
While we skipped uses in stores if we can find all copies of the value
when the memory is loaded, we did not correlate the use in the store
with the use in the load. So far this lead to less precise results in the
offset calculations which prevented deductions. With the new
EquivalentUseCB callback argument the user of checkForAllUses can be
informed of the correlation and act on it appropriately.

Differential Revision: https://reviews.llvm.org/D109662
2021-12-28 23:53:29 -06:00
Kazu Hirata
36b8a4f9f3 [llvm] Use llvm::is_contained (NFC) 2021-12-11 11:42:09 -08:00
Joseph Huber
b8a825b483 [Attributor] Introduce AAAssumptionInfo to propagate assumptions
This patch introduces a new abstract attributor instance that propagates
assumption information from functions. Conceptually, if a function is
only called by functions that have certain assumptions, then we can
apply the same assumptions to that function. This problem is similar to
calculating the dominator set, but the assumptions are merged instead of
nodes.

Reviewed By: jdoerfert

Differential Revision: https://reviews.llvm.org/D111054
2021-11-09 17:39:18 -05:00
Johannes Doerfert
734f91441d [Attributor][NFC] Improve debug messages 2021-10-27 21:04:31 -05:00
Joseph Huber
f074a6a041 [OpenMP] Add options to change Attributor max iterations in OpenMPOpt
This patch adds a new command line option `openmp-opt-max-iterations`
that controls the maximum number of iterations the attributor will run
for when compiling OpenMP target device code. This patch also adds a
remark to indicate when the attributor failed because it did not run
for enough iterations.

Reviewed By: jdoerfert

Differential Revision: https://reviews.llvm.org/D110749
2021-10-04 09:39:04 -04:00
Kazu Hirata
4f0225f6d2 [Transforms] Migrate from getNumArgOperands to arg_size (NFC)
Note that getNumArgOperands is considered a legacy name.  See
llvm/include/llvm/IR/InstrTypes.h for details.
2021-10-01 09:57:40 -07:00
Arthur Eubanks
52e6d70c40 [NFC] Use newly introduced *AtIndex methods
Introduced in D108788. These are clearer.
2021-09-01 11:18:41 -07:00
Johannes Doerfert
e05940de2a [Attributor][FIX] Recursion via memory needs to be tracked explicitly
Recursion can happen when we see a PHI use the second time or when we
look at a store value operand use again. We already visited the
potential copies and doing so again will just cause endless looping.

Reviewed By: kuter

Differential Revision: https://reviews.llvm.org/D108190
2021-08-27 13:12:13 -05:00
Arthur Eubanks
3f4d00bc3b [NFC] More get/removeAttribute() cleanup 2021-08-17 21:05:41 -07:00
Arthur Eubanks
80ea2bb574 [NFC] Rename AttributeList::getParam/Ret/FnAttributes() -> get*Attributes()
This is more consistent with similar methods.
2021-08-13 11:16:52 -07:00
Johannes Doerfert
4e7d7cae67 [Attributor][FIX] Do not try to rewrite functions with casted call sites
If we cast a function at the call site it is hard(er) to get the rewrite
correct, let's not attempt it for now.

Fixes PR51448.
2021-08-12 10:39:53 -05:00
Johannes Doerfert
5f543919b2 [Attributor][FIX] Guard constant casts with type size checks 2021-08-12 10:39:53 -05:00
Johannes Doerfert
96da6dd6ba [Attributor][FIX] Only avoid visiting PHI uses multiple times (PR51249)
AAPointerInfoFloating needs to visit all uses and some multiple times if
we go through PHI nodes. Attributor::checkForAllUses keeps a visited set
so we don't recurs endlessly. We now allow recursion for non-phi uses so
we track all pointer offsets via PHI nodes properly without endless
recursion.

This replaces the first attempt D107579.

Differential Revision: https://reviews.llvm.org/D107798
2021-08-11 00:49:54 -05:00
Joseph Huber
adbaa39dfc [Attributor] Change function internalization to not replace uses in internalized callers
The current implementation of function internalization creats a copy of each
function and replaces every use. This has the downside that the external
versions of the functions will call into the internalized versions of the
functions. This prevents them from being fully independent of eachother. This
patch replaces the current internalization scheme with a method that creates
all the copies of the functions intended to be internalized first and then
replaces the uses as long as their caller is not already internalized.

Reviewed By: jdoerfert

Differential Revision: https://reviews.llvm.org/D106931
2021-07-28 18:57:28 -04:00
Johannes Doerfert
3dca83961c Reapply "[Attributor] Disable simplification AAs if a callback is present""
This reapplies commit cbb709e25124dc38ee593882051fc88c987fe591 and
includes the use of the lookup method instead of operator[] to avoid
accidentally setting (empty) simplification callbacks.

This reverts commit aa27430a625b2fd059707a87f8ba2df8f480ff11.
2021-07-27 19:14:50 -05:00
Johannes Doerfert
c55e18824d [Attributor][FIX] Copy all members in the assignment operator
Also improve debug output slightly.
2021-07-27 01:44:13 -05:00
Johannes Doerfert
25a3130d89 [Local] Do not introduce a new llvm.trap before unreachable
This is the second attempt to remove the `llvm.trap` insertion after
https://reviews.llvm.org/rGe14e7bc4b889dfaffb7180d176a03311df2d4ae6
reverted the first one. It is not clear what the exact issue was back
then and it might already be gone by now, it has been >5 years after
all.

Replaces D106299.

Differential Revision: https://reviews.llvm.org/D106308
2021-07-26 23:33:36 -05:00
Johannes Doerfert
adddd3dbda [Attributor] Introduce getPotentialCopiesOfStoredValue and use it
This patch introduces `getPotentialCopiesOfStoredValue` which uses
AAPointerInfo to determine all "aliases" or "potential copies" of a
value that is stored into memory. This operation can fail but if it
succeeds it means we can visit all "uses" of a value even if it is
temporarily stored in memory.

There are two users for the function:
  1) `Attributor::checkForAllUses` which will now ignore the value use
     in a store if all "potential copies" can be identified and instead
     be visited. This allows various AAs, including AAPointerInfo
     itself, to look through memory.
  2) `AANoCapture` which uses a custom use tracking through the
     CaptureTracker interface and therefore needs to be thought
     explicitly.

Differential Revision: https://reviews.llvm.org/D106185
2021-07-26 23:33:36 -05:00
Kuter Dinel
96709823ec [AMDGPU] Deduce attributes with the Attributor
This patch introduces a pass that uses the Attributor to deduce AMDGPU specific attributes.

Reviewed By: jdoerfert, arsenm

Differential Revision: https://reviews.llvm.org/D104997
2021-07-24 06:07:15 +03:00
Kuter Dinel
0cd964ff25 [Attributor][FIX] checkForAllInstructions, correctly handle declarations
checkForAllInstructions was not handling declarations correctly.
It should have been returning false when it gets called on a declaration

The patch also fixes a test case for AAFunctionReachability for it to be able
to pass after the changes to the checkForAllinstructions.

Differential Revision: https://reviews.llvm.org/D106625
2021-07-24 02:21:29 +03:00
Johannes Doerfert
6ca969353c [Attributor] If provided, only look at simplification callbacks not IR
A simplification callback can mean that the IR value is modified beyond
the apparent IR semantics. That is, a `i1 true` could be replaced by an
`i1 false` based on high-level domain-specific information. If a user
provides a simplification callback we will not look at the IR but
instead give up if the callback returns a nullptr.
2021-07-22 23:57:37 -05:00
Johannes Doerfert
0c0eb76782 [Attributor][FIX] Improve call graph updating
If we remove a non-intrinsic instruction we need to tell the (old) call
graph about it. This caused problems with some features down the line as
they allowed to removed calls more aggressively.
2021-07-22 00:07:56 -05:00
Johannes Doerfert
94d3b59c56 [Attributor][FIX] Do not introduce multiple instances of SSA values
If we have a recursive function we could create multiple instantiations
of an SSA value, one per recursive invocation of the function. This is a
problem as we use SSA value equality in various places. The basic idea
follows from this test:

```
static int r(int c, int *a) {
  int X;
  return c ? r(false, &X) : a == &X;
}

int test(int c) {
  return r(c, undef);
}
```

If we look through the argument `a` we will end up with `X`. Using SSA
value equality we will fold `a == &X` to true and return true even
though it should have been false because `a` and `&X` are from different
instantiations of the function.

Various tests for this  have been placed in value-simplify-instances.ll
and this commit fixes them all by avoiding to produce simplified values
that could be non-unique at runtime. Thus, the result of a simplify
value call will always be unique at runtime or the original value, both
do not allow to accidentally compare two instances of a value with each
other and conclude they are equal statically (pointer equivalence) while
they are unequal at runtime.
2021-07-22 00:07:55 -05:00
Johannes Doerfert
c819266ecc [Attributor] Improve the Attributor::getAssumedConstant interface
Similar to Attributor::getAssumedSimplified we need to allow IRPs
directly to get the right simplification callback (and context).
2021-07-22 00:07:55 -05:00