<html xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:w="urn:schemas-microsoft-com:office:word" xmlns:m="http://schemas.microsoft.com/office/2004/12/omml" xmlns="http://www.w3.org/TR/REC-html40">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<meta name="Generator" content="Microsoft Word 14 (filtered medium)">
<style><!--
/* Font Definitions */
@font-face
{font-family:Calibri;
panose-1:2 15 5 2 2 2 4 3 2 4;}
@font-face
{font-family:Tahoma;
panose-1:2 11 6 4 3 5 4 4 2 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
{margin:0cm;
margin-bottom:.0001pt;
font-size:12.0pt;
font-family:"Times New Roman","serif";}
a:link, span.MsoHyperlink
{mso-style-priority:99;
color:blue;
text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
{mso-style-priority:99;
color:purple;
text-decoration:underline;}
p.MsoAcetate, li.MsoAcetate, div.MsoAcetate
{mso-style-priority:99;
mso-style-link:"Balloon Text Char";
margin:0cm;
margin-bottom:.0001pt;
font-size:8.0pt;
font-family:"Tahoma","sans-serif";}
p.MsoListParagraph, li.MsoListParagraph, div.MsoListParagraph
{mso-style-priority:34;
margin-top:0cm;
margin-right:0cm;
margin-bottom:0cm;
margin-left:36.0pt;
margin-bottom:.0001pt;
font-size:12.0pt;
font-family:"Times New Roman","serif";}
span.EmailStyle17
{mso-style-type:personal-reply;
font-family:"Calibri","sans-serif";
color:windowtext;}
span.BalloonTextChar
{mso-style-name:"Balloon Text Char";
mso-style-priority:99;
mso-style-link:"Balloon Text";
font-family:"Tahoma","sans-serif";
mso-fareast-language:EN-GB;}
.MsoChpDefault
{mso-style-type:export-only;
font-family:"Calibri","sans-serif";
mso-fareast-language:EN-US;}
@page WordSection1
{size:612.0pt 792.0pt;
margin:72.0pt 72.0pt 72.0pt 72.0pt;}
div.WordSection1
{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext="edit" spidmax="1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext="edit">
<o:idmap v:ext="edit" data="1" />
</o:shapelayout></xml><![endif]-->
</head>
<body lang="EN-GB" link="blue" vlink="purple">
<div class="WordSection1">
<p class="MsoNormal">> Fixed in r260531.<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">Thanks<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">> I think we will eventually want to merge the following commits, assuming they don't regress the build, especially with the -gcc-toolchain option.<o:p></o:p></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif""><o:p> </o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif"">Yes, the MIPS release needs these changes to fix the ~24 tests in PR26369.<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif""><o:p> </o:p></span></p>
<div style="border:none;border-left:solid blue 1.5pt;padding:0cm 0cm 0cm 4.0pt">
<div>
<div style="border:none;border-top:solid #B5C4DF 1.0pt;padding:3.0pt 0cm 0cm 0cm">
<p class="MsoNormal"><b><span lang="EN-US" style="font-size:10.0pt;font-family:"Tahoma","sans-serif"">From:</span></b><span lang="EN-US" style="font-size:10.0pt;font-family:"Tahoma","sans-serif""> Eric Fiselier [mailto:eric@efcs.ca]
<br>
<b>Sent:</b> 11 February 2016 16:08<br>
<b>To:</b> Daniel Sanders<br>
<b>Cc:</b> Vasileios Kalintiris; hans@chromium.org; mclow.lists@gmail.com; cfe-commits@lists.llvm.org<br>
<b>Subject:</b> Re: [libcxx] r260235 - Introduce a cmake module to figure out whether we need to link with libatomic.<o:p></o:p></span></p>
</div>
</div>
<p class="MsoNormal"><o:p> </o:p></p>
<div>
<div>
<p class="MsoNormal" style="margin-bottom:12.0pt">> <span style="font-size:9.5pt"> we need to rename HAVE_CXX_ATOMICS_WITH_LIB to something like LIBCXX_HAVE_CXX_ATOMICS_WITH_LIB.</span><o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">Fixed in r260531.<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal">I think we will eventually want to merge the following commits, assuming they don't regress the build, especially with the -gcc-toolchain option.<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal">- [libcxx] r260515 - Re-commit "Introduce a cmake module to figure out whether we need to link with libatomic."<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">- [libcxx] r260524 - Fix r260515 - Correct typos in CMake changes<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">- [libcxx] r260531 - Rename CheckLibcxxAtomic.cmake variable result names so they don't clash with LLVM<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal">@Marshall Any objections?<o:p></o:p></p>
</div>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
<div>
<p class="MsoNormal">On Thu, Feb 11, 2016 at 2:18 AM, Daniel Sanders via cfe-commits <<a href="mailto:cfe-commits@lists.llvm.org" target="_blank">cfe-commits@lists.llvm.org</a>> wrote:<o:p></o:p></p>
<p class="MsoNormal">Hi,<br>
<br>
In my latests rc2+patches build I've also found that we need to rename HAVE_CXX_ATOMICS_WITH_LIB to something like LIBCXX_HAVE_CXX_ATOMICS_WITH_LIB. It's currently re-using the result of LLVM's check which doesn't include 64-bit atomics.<br>
________________________________________<br>
From: Vasileios Kalintiris<br>
Sent: 09 February 2016 23:50<o:p></o:p></p>
<div>
<div>
<p class="MsoNormal">To: <a href="mailto:hans@chromium.org">hans@chromium.org</a><br>
Cc: <a href="mailto:cfe-commits@lists.llvm.org">cfe-commits@lists.llvm.org</a>; Daniel Sanders;
<a href="mailto:mclow.lists@gmail.com">mclow.lists@gmail.com</a><br>
Subject: RE: [libcxx] r260235 - Introduce a cmake module to figure out whether we need to link with libatomic.<br>
<br>
Hi Hans,<br>
<br>
Please wait before merging this as it breaks LLVM bootstraps when using the -gcc-toolchain option and the system's default gcc installation does not provide libatomic. We have to check LIBCXX_GCC_TOOLCHAIN in our test. I'll create a patch tomorrow and I'll
let you know when it's okay to merge them. In the meantime, I reverted it in r260323.<br>
<br>
- Vasileios<br>
________________________________________<br>
From: Vasileios Kalintiris<br>
Sent: 09 February 2016 18:56<br>
To: <a href="mailto:hans@chromium.org">hans@chromium.org</a><br>
Cc: <a href="mailto:cfe-commits@lists.llvm.org">cfe-commits@lists.llvm.org</a>; Daniel Sanders;
<a href="mailto:mclow.lists@gmail.com">mclow.lists@gmail.com</a><br>
Subject: RE: [libcxx] r260235 - Introduce a cmake module to figure out whether we need to link with libatomic.<br>
<br>
Hi Hans,<br>
<br>
Can we merge this on the 3.8 release branch? It fixes libcxx builds on MIPS by linking with libatomic when needed. Also, all the x86_64 and ARM buildbots for libcxx look good.<br>
<br>
Thanks,<br>
Vasileios<br>
<br>
________________________________________<br>
From: cfe-commits [<a href="mailto:cfe-commits-bounces@lists.llvm.org">cfe-commits-bounces@lists.llvm.org</a>] on behalf of Vasileios Kalintiris via cfe-commits [<a href="mailto:cfe-commits@lists.llvm.org">cfe-commits@lists.llvm.org</a>]<br>
Sent: 09 February 2016 17:00<br>
To: <a href="mailto:cfe-commits@lists.llvm.org">cfe-commits@lists.llvm.org</a><br>
Subject: [libcxx] r260235 - Introduce a cmake module to figure out whether we need to link with libatomic.<br>
<br>
Author: vkalintiris<br>
Date: Tue Feb 9 11:00:38 2016<br>
New Revision: 260235<br>
<br>
URL: <a href="http://llvm.org/viewvc/llvm-project?rev=260235&view=rev" target="_blank">
http://llvm.org/viewvc/llvm-project?rev=260235&view=rev</a><br>
Log:<br>
Introduce a cmake module to figure out whether we need to link with libatomic.<br>
<br>
Summary:<br>
This fixes the tests under std/atomics for 32-bit MIPS CPUs where the<br>
8-byte atomic operations call into the libatomic library.<br>
<br>
Reviewers: dsanders, mclow.lists, EricWF, jroelofs, joerg<br>
<br>
Subscribers: cfe-commits<br>
<br>
Differential Revision: <a href="http://reviews.llvm.org/D16613" target="_blank">http://reviews.llvm.org/D16613</a><br>
<br>
Added:<br>
libcxx/trunk/cmake/Modules/CheckLibcxxAtomic.cmake<br>
Modified:<br>
libcxx/trunk/cmake/config-ix.cmake<br>
libcxx/trunk/lib/CMakeLists.txt<br>
libcxx/trunk/test/CMakeLists.txt<br>
libcxx/trunk/test/libcxx/test/target_info.py<br>
libcxx/trunk/test/<a href="http://lit.site.cfg.in" target="_blank">lit.site.cfg.in</a><br>
<br>
Added: libcxx/trunk/cmake/Modules/CheckLibcxxAtomic.cmake<br>
URL: <a href="http://llvm.org/viewvc/llvm-project/libcxx/trunk/cmake/Modules/CheckLibcxxAtomic.cmake?rev=260235&view=auto" target="_blank">
http://llvm.org/viewvc/llvm-project/libcxx/trunk/cmake/Modules/CheckLibcxxAtomic.cmake?rev=260235&view=auto</a><br>
==============================================================================<br>
--- libcxx/trunk/cmake/Modules/CheckLibcxxAtomic.cmake (added)<br>
+++ libcxx/trunk/cmake/Modules/CheckLibcxxAtomic.cmake Tue Feb 9 11:00:38 2016<br>
@@ -0,0 +1,38 @@<br>
+INCLUDE(CheckCXXSourceCompiles)<br>
+<br>
+# Sometimes linking against libatomic is required for atomic ops, if<br>
+# the platform doesn't support lock-free atomics.<br>
+#<br>
+# We could modify LLVM's CheckAtomic module and have it check for 64-bit<br>
+# atomics instead. However, we would like to avoid careless uses of 64-bit<br>
+# atomics inside LLVM over time on 32-bit platforms.<br>
+<br>
+function(check_cxx_atomics varname)<br>
+ set(OLD_CMAKE_REQUIRED_FLAGS ${CMAKE_REQUIRED_FLAGS})<br>
+ set(CMAKE_REQUIRED_FLAGS "-std=c++11")<br>
+ check_cxx_source_compiles("<br>
+#include <cstdint><br>
+#include <atomic><br>
+std::atomic<uintptr_t> x;<br>
+std::atomic<uintmax_t> y;<br>
+int main() {<br>
+ return x + y;<br>
+}<br>
+" ${varname})<br>
+ set(CMAKE_REQUIRED_FLAGS ${OLD_CMAKE_REQUIRED_FLAGS})<br>
+endfunction(check_cxx_atomics)<br>
+<br>
+check_cxx_atomics(HAVE_CXX_ATOMICS_WITHOUT_LIB)<br>
+# If not, check if the library exists, and atomics work with it.<br>
+if(NOT HAVE_CXX_ATOMICS_WITHOUT_LIB)<br>
+ check_library_exists(atomic __atomic_fetch_add_8 "" HAVE_LIBATOMIC)<br>
+ if(HAVE_LIBATOMIC)<br>
+ list(APPEND CMAKE_REQUIRED_LIBRARIES "atomic")<br>
+ check_cxx_atomics(HAVE_CXX_ATOMICS_WITH_LIB)<br>
+ if (NOT HAVE_CXX_ATOMICS_WITH_LIB)<br>
+ message(FATAL_ERROR "Host compiler must support std::atomic!")<br>
+ endif()<br>
+ else()<br>
+ message(FATAL_ERROR "Host compiler appears to require libatomic, but cannot find it.")<br>
+ endif()<br>
+endif()<br>
<br>
Modified: libcxx/trunk/cmake/config-ix.cmake<br>
URL: <a href="http://llvm.org/viewvc/llvm-project/libcxx/trunk/cmake/config-ix.cmake?rev=260235&r1=260234&r2=260235&view=diff" target="_blank">
http://llvm.org/viewvc/llvm-project/libcxx/trunk/cmake/config-ix.cmake?rev=260235&r1=260234&r2=260235&view=diff</a><br>
==============================================================================<br>
--- libcxx/trunk/cmake/config-ix.cmake (original)<br>
+++ libcxx/trunk/cmake/config-ix.cmake Tue Feb 9 11:00:38 2016<br>
@@ -1,5 +1,6 @@<br>
include(CheckLibraryExists)<br>
include(CheckCXXCompilerFlag)<br>
+include(CheckLibcxxAtomic)<br>
<br>
# Check compiler flags<br>
<br>
@@ -17,3 +18,7 @@ check_library_exists(c fopen "" LIBCXX_H<br>
check_library_exists(m ccos "" LIBCXX_HAS_M_LIB)<br>
check_library_exists(rt clock_gettime "" LIBCXX_HAS_RT_LIB)<br>
check_library_exists(gcc_s __gcc_personality_v0 "" LIBCXX_HAS_GCC_S_LIB)<br>
+<br>
+if (NOT HAVE_CXX_ATOMICS_WITHOUT_LIB)<br>
+ set(LIBCXX_HAS_ATOMIC_LIB True)<br>
+endif()<br>
<br>
Modified: libcxx/trunk/lib/CMakeLists.txt<br>
URL: <a href="http://llvm.org/viewvc/llvm-project/libcxx/trunk/lib/CMakeLists.txt?rev=260235&r1=260234&r2=260235&view=diff" target="_blank">
http://llvm.org/viewvc/llvm-project/libcxx/trunk/lib/CMakeLists.txt?rev=260235&r1=260234&r2=260235&view=diff</a><br>
==============================================================================<br>
--- libcxx/trunk/lib/CMakeLists.txt (original)<br>
+++ libcxx/trunk/lib/CMakeLists.txt Tue Feb 9 11:00:38 2016<br>
@@ -79,6 +79,7 @@ add_library_flags_if(LIBCXX_HAS_C_LIB c)<br>
add_library_flags_if(LIBCXX_HAS_M_LIB m)<br>
add_library_flags_if(LIBCXX_HAS_RT_LIB rt)<br>
add_library_flags_if(LIBCXX_HAS_GCC_S_LIB gcc_s)<br>
+add_library_flags_if(LIBCXX_HAS_ATOMIC_LIB atomic)<br>
<br>
# Setup flags.<br>
add_flags_if_supported(-fPIC)<br>
<br>
Modified: libcxx/trunk/test/CMakeLists.txt<br>
URL: <a href="http://llvm.org/viewvc/llvm-project/libcxx/trunk/test/CMakeLists.txt?rev=260235&r1=260234&r2=260235&view=diff" target="_blank">
http://llvm.org/viewvc/llvm-project/libcxx/trunk/test/CMakeLists.txt?rev=260235&r1=260234&r2=260235&view=diff</a><br>
==============================================================================<br>
--- libcxx/trunk/test/CMakeLists.txt (original)<br>
+++ libcxx/trunk/test/CMakeLists.txt Tue Feb 9 11:00:38 2016<br>
@@ -15,6 +15,7 @@ pythonize_bool(LIBCXX_ENABLE_SHARED)<br>
pythonize_bool(LIBCXX_BUILD_32_BITS)<br>
pythonize_bool(LIBCXX_GENERATE_COVERAGE)<br>
pythonize_bool(LIBCXXABI_USE_LLVM_UNWINDER)<br>
+pythonize_bool(LIBCXX_HAS_ATOMIC_LIB)<br>
<br>
# The tests shouldn't link to any ABI library when it has been linked into<br>
# libc++ statically or via a linker script.<br>
<br>
Modified: libcxx/trunk/test/libcxx/test/target_info.py<br>
URL: <a href="http://llvm.org/viewvc/llvm-project/libcxx/trunk/test/libcxx/test/target_info.py?rev=260235&r1=260234&r2=260235&view=diff" target="_blank">
http://llvm.org/viewvc/llvm-project/libcxx/trunk/test/libcxx/test/target_info.py?rev=260235&r1=260234&r2=260235&view=diff</a><br>
==============================================================================<br>
--- libcxx/trunk/test/libcxx/test/target_info.py (original)<br>
+++ libcxx/trunk/test/libcxx/test/target_info.py Tue Feb 9 11:00:38 2016<br>
@@ -172,6 +172,9 @@ class LinuxLocalTI(DefaultTargetInfo):<br>
flags += ['-lunwind', '-ldl']<br>
else:<br>
flags += ['-lgcc_s', '-lgcc']<br>
+ use_libatomic = self.full_config.get_lit_bool('use_libatomic', False)<br>
+ if use_libatomic:<br>
+ flags += ['-latomic']<br>
san = self.full_config.get_lit_conf('use_sanitizer', '').strip()<br>
if san:<br>
# The libraries and their order are taken from the<br>
<br>
Modified: libcxx/trunk/test/<a href="http://lit.site.cfg.in" target="_blank">lit.site.cfg.in</a><br>
URL: <a href="http://llvm.org/viewvc/llvm-project/libcxx/trunk/test/lit.site.cfg.in?rev=260235&r1=260234&r2=260235&view=diff" target="_blank">
http://llvm.org/viewvc/llvm-project/libcxx/trunk/test/lit.site.cfg.in?rev=260235&r1=260234&r2=260235&view=diff</a><br>
==============================================================================<br>
--- libcxx/trunk/test/<a href="http://lit.site.cfg.in" target="_blank">lit.site.cfg.in</a> (original)<br>
+++ libcxx/trunk/test/<a href="http://lit.site.cfg.in" target="_blank">lit.site.cfg.in</a> Tue Feb 9 11:00:38 2016<br>
@@ -20,6 +20,7 @@ config.generate_coverage = "@LIBC<br>
config.target_info = <a href="mailto:%22@LIBCXX_TARGET_INFO@">"@LIBCXX_TARGET_INFO@</a>"<br>
config.executor = "@LIBCXX_EXECUTOR@"<br>
config.llvm_unwinder = "@LIBCXXABI_USE_LLVM_UNWINDER@"<br>
+config.use_libatomic = "@LIBCXX_HAS_ATOMIC_LIB@"<br>
<br>
# Let the main config do the real work.<br>
lit_config.load_config(config, "@LIBCXX_SOURCE_DIR@/test/lit.cfg")<br>
<br>
<br>
_______________________________________________<br>
cfe-commits mailing list<br>
<a href="mailto:cfe-commits@lists.llvm.org">cfe-commits@lists.llvm.org</a><br>
<a href="http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits" target="_blank">http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits</a><br>
_______________________________________________<br>
cfe-commits mailing list<br>
<a href="mailto:cfe-commits@lists.llvm.org">cfe-commits@lists.llvm.org</a><br>
<a href="http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits" target="_blank">http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits</a><o:p></o:p></p>
</div>
</div>
</div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
</div>
</div>
</body>
</html>