6 Commits

Author SHA1 Message Date
Aaron Ballman
617cf8a72d
Reapply "Finish deleting the le32/le64 targets" (#99079) (#101983)
This reverts commit d3f8105c65046173e20c4c59394b4a7f1bbe7627.

Halide no longer relies on this target:
https://github.com/llvm/llvm-project/pull/98497#issuecomment-2253358685
2024-08-06 08:35:56 -04:00
Aaron Ballman
d3f8105c65
Revert "Finish deleting the le32/le64 targets" (#99079)
Reverts llvm/llvm-project#98497

We're reverting this for approx 30 days so that the Halide project has
time to transition off the target.
2024-07-16 14:47:09 -04:00
Michael Buch
4497ec293a
[clang][CGRecordLayout] Remove dependency on isZeroSize (#96422)
This is a follow-up from the conversation starting at
https://github.com/llvm/llvm-project/pull/93809#issuecomment-2173729801

The root problem that motivated the change are external AST sources that
compute `ASTRecordLayout`s themselves instead of letting Clang compute
them from the AST. One such example is LLDB using DWARF to get the
definitive offsets and sizes of C++ structures. Such layouts should be
considered correct (modulo buggy DWARF), but various assertions and
lowering logic around the `CGRecordLayoutBuilder` relies on the AST
having `[[no_unique_address]]` attached to them. This is a
layout-altering attribute which is not encoded in DWARF. This causes us
LLDB to trip over the various LLVM<->Clang layout consistency checks.
There has been precedent for avoiding such layout-altering attributes
from affecting lowering with externally-provided layouts (e.g., packed
structs).

This patch proposes to replace the `isZeroSize` checks in
`CGRecordLayoutBuilder` (which roughly means "empty field with
[[no_unique_address]]") with checks for
`CodeGen::isEmptyField`/`CodeGen::isEmptyRecord`.

**Details**
The main strategy here was to change the `isZeroSize` check in
`CGRecordLowering::accumulateFields` and
`CGRecordLowering::accumulateBases` to use the `isEmptyXXX` APIs
instead, preventing empty fields from being added to the `Members` and
`Bases` structures. The rest of the changes fall out from here, to
prevent lookups into these structures (for field numbers or base
indices) from failing.

Added `isEmptyRecordForLayout` and `isEmptyFieldForLayout` (open to
better naming suggestions). The main difference to the existing
`isEmptyRecord`/`isEmptyField` APIs, is that the `isEmptyXXXForLayout`
counterparts don't have special treatment for `unnamed bitfields`/arrays
and also treat fields of empty types as if they had
`[[no_unique_address]]` (i.e., just like the `AsIfNoUniqueAddr` in
`isEmptyField` does).
2024-07-16 04:59:51 +01:00
Aaron Ballman
2369a54fbe
Finish deleting the le32/le64 targets (#98497)
This is a revert of ef5e7f90ea4d5063ce68b952c5de473e610afc02 which was a
temporary partial revert of 77ac823fd285973cfb3517932c09d82e6a32f46d.
The le32 and le64 targets are no longer necessary to retain, so this
removes them entirely.
2024-07-12 06:55:49 -04:00
Nathan Sidwell
49839f97d2 [clang] Better SysV bitfield access units (#65742)
Reimplement bitfield access unit computation for SysV ABIs. Considers
expense of unaligned and non-power-of-2 accesses.
2024-03-29 09:35:31 -04:00
Nathan Sidwell
a8ca4abfcf [clang][NFC] Bitfield access unit tests (#65742)
Verify bitfield access units.
2024-03-29 09:35:31 -04:00