* [PATCH] aarch64: Don't trust TYPE_ALIGN for pointers [PR108910]
@ 2023-04-13 16:00 Richard Sandiford
0 siblings, 0 replies; only message in thread
From: Richard Sandiford @ 2023-04-13 16:00 UTC (permalink / raw)
To: gcc-patches; +Cc: jakub
The aarch64 PCS rules ignore user alignment for scalars and
vectors and use the "natural" alignment of the type. GCC tried
to calculate that natural alignment using:
TYPE_ALIGN (TYPE_MAIN_VARIANT (type))
But as discussed in the PR, it's possible that the main variant
of a pointer type is an overaligned type (although that's usually
accidental).
This isn't known to be a problem for other types, so this patch
changes the bare minimum. It might be that we need to ignore
TYPE_ALIGN in other cases too.
Tested on aarch64-linux-gnu & pushed to trunk so far. Will backport
to GCC 12 soon.
Richard
gcc/
PR target/108910
* config/aarch64/aarch64.cc (aarch64_function_arg_alignment): Do
not trust TYPE_ALIGN for pointer types; use POINTER_SIZE instead.
gcc/testsuite/
PR target/108910
* gcc.dg/torture/pr108910.c: New test.
---
gcc/config/aarch64/aarch64.cc | 15 ++++++++++++++-
gcc/testsuite/gcc.dg/torture/pr108910.c | 8 ++++++++
2 files changed, 22 insertions(+), 1 deletion(-)
create mode 100644 gcc/testsuite/gcc.dg/torture/pr108910.c
diff --git a/gcc/config/aarch64/aarch64.cc b/gcc/config/aarch64/aarch64.cc
index 42617ced73a..f4ef22ce02f 100644
--- a/gcc/config/aarch64/aarch64.cc
+++ b/gcc/config/aarch64/aarch64.cc
@@ -7484,7 +7484,20 @@ aarch64_function_arg_alignment (machine_mode mode, const_tree type,
gcc_assert (TYPE_MODE (type) == mode);
if (!AGGREGATE_TYPE_P (type))
- return TYPE_ALIGN (TYPE_MAIN_VARIANT (type));
+ {
+ /* The ABI alignment is the natural alignment of the type, without
+ any attributes applied. Normally this is the alignment of the
+ TYPE_MAIN_VARIANT, but not always; see PR108910 for a counterexample.
+ For now we just handle the known exceptions explicitly. */
+ type = TYPE_MAIN_VARIANT (type);
+ if (POINTER_TYPE_P (type))
+ {
+ gcc_assert (known_eq (POINTER_SIZE, GET_MODE_BITSIZE (mode)));
+ return POINTER_SIZE;
+ }
+ gcc_assert (!TYPE_USER_ALIGN (type));
+ return TYPE_ALIGN (type);
+ }
if (TREE_CODE (type) == ARRAY_TYPE)
return TYPE_ALIGN (TREE_TYPE (type));
diff --git a/gcc/testsuite/gcc.dg/torture/pr108910.c b/gcc/testsuite/gcc.dg/torture/pr108910.c
new file mode 100644
index 00000000000..59735488c2e
--- /dev/null
+++ b/gcc/testsuite/gcc.dg/torture/pr108910.c
@@ -0,0 +1,8 @@
+extern void foo (float, float *, float *);
+
+void
+bar (void *p)
+{
+ float *__attribute__((aligned (64))) q = __builtin_assume_aligned (p, 64);
+ foo (0.0f, q, q);
+}
--
2.25.1
^ permalink raw reply [flat|nested] only message in thread
only message in thread, other threads:[~2023-04-13 16:00 UTC | newest]
Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-04-13 16:00 [PATCH] aarch64: Don't trust TYPE_ALIGN for pointers [PR108910] Richard Sandiford
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).