[llvm-bugs] [Bug 30503] New: SSE instructions are generated with SSE disabled
via llvm-bugs
llvm-bugs at lists.llvm.org
Fri Sep 23 06:46:25 PDT 2016
https://llvm.org/bugs/show_bug.cgi?id=30503
Bug ID: 30503
Summary: SSE instructions are generated with SSE disabled
Product: libraries
Version: 3.9
Hardware: PC
OS: Linux
Status: NEW
Severity: normal
Priority: P
Component: Backend: X86
Assignee: unassignedbugs at nondot.org
Reporter: simonas+llvm.org at kazlauskas.me
CC: llvm-bugs at lists.llvm.org
Classification: Unclassified
Compiling
target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
target triple = "x86_64-unknown-linux-gnu"
; Function Attrs: uwtable
define internal i64 @banana() unnamed_addr #0 {
entry-block:
ret i64 0
}
; Function Attrs: nounwind uwtable
define x86_64_win64cc i64 @peach() unnamed_addr #1 {
entry-block:
%0 = call i64 @banana()
ret i64 %0
}
attributes #0 = { uwtable }
attributes #1 = { nounwind uwtable }
with `llc test.ll -mattr="-sse,-mmx,+soft_float" -filetype=asm -o -` produces
assembly which contains SSE (movaps) instructions, despite SSE being disabled:
peach: # @peach
.cfi_startproc
# BB#0: # %entry-block
pushq %rsi
.Ltmp0:
.cfi_def_cfa_offset 16
pushq %rdi
.Ltmp1:
.cfi_def_cfa_offset 24
subq $168, %rsp
movaps %xmm15, 144(%rsp) # 16-byte Spill
movaps %xmm14, 128(%rsp) # 16-byte Spill
movaps %xmm13, 112(%rsp) # 16-byte Spill
movaps %xmm12, 96(%rsp) # 16-byte Spill
# and so forth
Issue is present in 3.8 as well.
--
You are receiving this mail because:
You are on the CC list for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-bugs/attachments/20160923/09cc3862/attachment.html>
More information about the llvm-bugs
mailing list