https://bugs.llvm.org/show_bug.cgi?id=44395
Bug ID: 44395
Summary: Invalid code generation for win32/32bit windows target
Product: clang
Version: trunk
Hardware: PC
OS: Linux
Status: NEW
Severity: enhancement
Priority: P
Component: LLVM Codegen
Assignee: unassignedclangb...@nondot.org
Reporter: zamazan...@tut.by
CC: llvm-bugs@lists.llvm.org, neeil...@live.com,
richard-l...@metafoo.co.uk
clang(trunk) with -m32 on windows (win32 target) generates wrong code for the
following code ( https://godbolt.org/z/RKzU5M )
#include <vector>
#include <functional>
#include <emmintrin.h>
struct Id { unsigned handle; };
struct QueryView { __m128 val; unsigned i; };
extern void perform_query(std::function<void(QueryView &)>);
template<typename BT> void foo ( QueryView &qv, const BT & block)
{
block( Id{qv.i}, _mm_setzero_ps() );
}
template<typename BT> void bar ( const BT & block)
{
perform_query([&](QueryView& qv) { block(qv.val); });
}
void external_call ( QueryView & qv ) {
{
foo(qv, [&](Id id, const __m128 two)->void{ bar([&](__m128 &
arg)->void{arg = two;});});
}}
the body of function std::function capture will (reasonably) make aligned read
on argument of __m128 type
While unaligned_stack_store will store __m128 register on stack unaligned,
tightly with Id structure.
This results in crash during runtime (unaligned load).
LLVM code (relevant part)
%3 = alloca inalloca <{ %struct.Id, <4 x float> }>, align 4
Assembly disasm (relevant part):
push eax
sub esp, 16
mov eax, esp
xorps xmm0, xmm0
mov dword ptr [eax], ecx
movups xmmword ptr [eax + 4], xmm0 #unaligned!
Possible fix: https://reviews.llvm.org/D71915
--
You are receiving this mail because:
You are on the CC list for the bug._______________________________________________
llvm-bugs mailing list
llvm-bugs@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
