Commit Graph

10 Commits

Author SHA1 Message Date
Peter Collingbourne 6f85225ef3 StackLifetime: Remove asserts for multiple lifetime intrinsics.
According to the langref, it is valid to have multiple consecutive
lifetime start or end intrinsics on the same object.

For llvm.lifetime.start:
"If ptr [...] is a stack object that is already alive, it simply
fills all bytes of the object with poison."

For llvm.lifetime.end:
"Calling llvm.lifetime.end on an already dead alloca is no-op."

However, we currently fail an assertion in such cases. I've observed
the assertion failure when the loop vectorization pass duplicates
the intrinsic.

We can conservatively handle these intrinsics by ignoring all but
the first one, which can be implemented by removing the assertions.

Differential Revision: https://reviews.llvm.org/D108337
2021-08-18 18:45:28 -07:00
Vitaly Buka a40660551e [StackSafety] Ignore allocas with partial lifetime markers
Reviewed By: eugenis

Differential Revision: https://reviews.llvm.org/D86672
2020-08-27 13:54:41 -07:00
Chuanqi Xu f6de5306ec [NFC][StackSafety] Test that StackLifetime looks through stripPointerCasts
StackLifetime class collects lifetime marker of an `alloca` by collect
the user of `BitCast` who is the user of the `alloca`. However, either
the `alloca` itself could be used with the lifetime marker or the `BitCast`
of the `alloca` could be transformed to other instructions. (e.g.,
it may be transformed to all zero reps in `InstCombine` pass).
This patch tries to fix this process in `collectMarkers` functions.

Reviewed By: vitalybuka

Differential Revision: https://reviews.llvm.org/D85399
2020-08-18 16:21:00 -07:00
Vitaly Buka e10e7829bf [StackSafety] Skip ambiguous lifetime analysis
If we can't identify alloca used in lifetime marker we
need to assume to worst case scenario.

Reviewed By: eugenis

Differential Revision: https://reviews.llvm.org/D84630
2020-08-16 18:05:52 -07:00
Vitaly Buka 7547508b7a Revert "[StackSafety] Skip ambiguous lifetime analysis"
This reverts commit 0b2616a804.

Crashes with safe-stack.
2020-08-07 14:02:50 -07:00
Vitaly Buka 0b2616a804 [StackSafety] Skip ambiguous lifetime analysis
If we can't identify alloca used in lifetime marker we
need to assume to worst case scenario.

Reviewed By: eugenis

Differential Revision: https://reviews.llvm.org/D84630
2020-08-06 19:10:33 -07:00
Vitaly Buka 306c257b00 [SafeStack,NFC] Print liveness for all instrunctions 2020-06-19 02:32:17 -07:00
Vitaly Buka 7b27c09f63 [StackSafety,NFC] Don't test terminators
Code does not track terminators and do not expose them through interface.
State there is just a state of the last instruction or entry.
So this information is just redundant and doesn't need to be tested.
2020-06-19 02:32:17 -07:00
Vitaly Buka fcd67665a8 [StackSafety] Add "Must Live" logic
Summary:
Extend StackLifetime with option to calculate liveliness
where alloca is only considered alive on basic block entry
if all non-dead predecessors had it alive at terminators.

Depends on D82043.

Reviewers: eugenis

Reviewed By: eugenis

Subscribers: hiraditya, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D82124
2020-06-18 16:53:37 -07:00
Vitaly Buka f672791e08 [StackSafety] Add pass for StackLifetime testing
Summary: lifetime.ll is a copy of SafeStack/X86/coloring2.ll

Reviewers: eugenis

Reviewed By: eugenis

Subscribers: hiraditya, mgrang, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D82043
2020-06-18 16:34:18 -07:00