<div dir="ltr"><div class="gmail_quote"><div>Sorry to dredge up an old commit, but I just saw this test spuriously fail:</div><div><br></div><div><a href="http://lab.llvm.org:8011/builders/llvm-clang-lld-x86_64-scei-ps4-ubuntu-fast/builds/14073">http://lab.llvm.org:8011/builders/llvm-clang-lld-x86_64-scei-ps4-ubuntu-fast/builds/14073</a><br></div><div><br></div><div>My commit seems unlikely to have triggered this, and the next build "fixed" the test. I'm worried there is something non-deterministic going on here. When reading this patch I noticed one thing below:</div><div dir="ltr"><br></div><div dir="ltr">On Wed, Mar 9, 2016 at 2:34 PM Sean Silva via llvm-commits <<a href="mailto:llvm-commits@lists.llvm.org">llvm-commits@lists.llvm.org</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">==============================================================================<br>
--- lld/trunk/test/ELF/lto/save-temps.ll (original)<br>
+++ lld/trunk/test/ELF/lto/save-temps.ll Wed Mar 9 16:30:05 2016<br>
@@ -1,10 +1,11 @@<br>
; REQUIRES: x86<br>
-; RUN: rm -f %t.so %t.so.lto.bc<br>
+; RUN: rm -f %t.so %t.so.lto.bc %t.so.lto.o<br>
; RUN: llvm-as %s -o %t.o<br>
; RUN: llvm-as %p/Inputs/save-temps.ll -o %t2.o<br>
; RUN: ld.lld -shared -m elf_x86_64 %t.o %t2.o -o %t.so -save-temps<br>
; RUN: llvm-nm %t.so | FileCheck %s<br>
; RUN: llvm-nm %t.so.lto.bc | FileCheck %s<br>
+; RUN: llvm-nm %t.so.lto.o | FileCheck %s<br>
<br>
target triple = "x86_64-unknown-linux-gnu"<br>
target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"<br>
@@ -13,5 +14,5 @@ define void @foo() {<br>
ret void<br>
}<br>
<br>
-; CHECK-DAG: T bar<br>
-; CHECK-DAG: T foo<br>
+; CHECK: T bar<br>
+; CHECK: T foo<br></blockquote><div><br></div><div>This change seems odd and unrelated to me.</div><div><br></div><div>If the CHECK-DAG was required, that shows some lack of deterministic output here. <span style="line-height:1.5">This change seems unlikely to have fixed any such non-determinism.</span></div><div><span style="line-height:1.5"><br></span></div><div>Sadly, this seems unlikely to explain the spurious failure I just saw. =/</div></div></div>