1fd9670b71
ROOT now uses llvm/clang version 9 (updated from version 5)
No longer exclude arch s390x (better supported in llvm/clang 9)
Drop patches accepted upstream or previously backported
Backport some fixes that make more tests work
New subpackages: python{2,3}-distrdf, root-roofit-batchcompute
Require js-jsroot >= 6
36 lines
1.5 KiB
Diff
36 lines
1.5 KiB
Diff
From 774ab9358d852e2c004564183de4a60eaaa5ac98 Mon Sep 17 00:00:00 2001
|
|
From: Mattias Ellert <mattias.ellert@physics.uu.se>
|
|
Date: Fri, 23 Apr 2021 21:40:14 +0200
|
|
Subject: [PATCH 2/2] Actually request the use of the large code model for
|
|
ppc64/ppc64le
|
|
|
|
Instead of erroring out with an assert.
|
|
---
|
|
interpreter/cling/lib/Interpreter/IncrementalExecutor.cpp | 6 +++---
|
|
1 file changed, 3 insertions(+), 3 deletions(-)
|
|
|
|
diff --git a/interpreter/cling/lib/Interpreter/IncrementalExecutor.cpp b/interpreter/cling/lib/Interpreter/IncrementalExecutor.cpp
|
|
index 46771aa8c0..93a48f31f3 100644
|
|
--- a/interpreter/cling/lib/Interpreter/IncrementalExecutor.cpp
|
|
+++ b/interpreter/cling/lib/Interpreter/IncrementalExecutor.cpp
|
|
@@ -73,14 +73,14 @@ CreateHostTargetMachine(const clang::CompilerInstance& CI) {
|
|
JTMB->getOptions().EmulatedTLS = false;
|
|
#endif // _WIN32
|
|
|
|
- std::unique_ptr<TargetMachine> TM = cantFail(JTMB->createTargetMachine());
|
|
-
|
|
#if defined(__powerpc64__) || defined(__PPC64__)
|
|
// We have to use large code model for PowerPC64 because TOC and text sections
|
|
// can be more than 2GB apart.
|
|
- assert(TM->getCodeModel() >= CodeModel::Large);
|
|
+ JTMB->setCodeModel(CodeModel::Large);
|
|
#endif
|
|
|
|
+ std::unique_ptr<TargetMachine> TM = cantFail(JTMB->createTargetMachine());
|
|
+
|
|
// Forcefully disable GlobalISel, it might be enabled on AArch64 without
|
|
// optimizations. In tests on an Apple M1 after the upgrade to LLVM 9, this
|
|
// new instruction selection framework emits branches / calls that expect all
|
|
--
|
|
2.30.2
|
|
|