[cfe-dev] Clang debug linking options on Windows

Mon Sep 14 10:16:52 PDT 2015

That works, thanks! It turns out to be only needed for the compile step,
and it requires clang-cl - the option to output intermediate code is then
'-c -Xclang -emit-llvm' which will output files with .obj extension, but
these are actually intermediate code files that just have to be renamed to
.ll before machine code generation and linking.

On Mon, Sep 14, 2015 at 6:09 PM, don hinton <hintonda at gmail.com> wrote:

> Took a look at llvm/tools/clang/lib/Driver/Tools.cpp, and it looks like
> all you need to do is pass "/MTd" and clang will add the following:
>
> -D_DEBUG
> -D_MT
> --dependent-lib=libcmtd
>
> You'll probably need it for both the compile and link steps.
>
> I don't have windows, so I can't verify, so I hope this helps...
> don
>
> On Mon, Sep 14, 2015 at 9:53 AM, Russell Wallace via cfe-dev <
> cfe-dev at lists.llvm.org> wrote:
>
>> _MT has no effect; same errors.
>>
>> On Mon, Sep 14, 2015 at 2:19 PM, Joshua Gerrard <joshua.gerrard at roli.com>
>> wrote:
>>
>>> Also try defining _MT
>>>
>>> --
>>> Joshua Gerrard
>>> JUCE Software Developer
>>>
>>> The Seaboard GRAND is a breakthrough new musical instrument called “the
>>> piano of the future” (CNN) and “influential, innovative, and inspiring” and
>>> “ingenious” (Telegraph). With orders from 34 countries, the Seaboard GRAND
>>> is now shipping and available for sale at www.roli.com/seaboard
>>>
>>> Office: +44(0)207 254 2155
>>> Cell:    07885 557494
>>>
>>> ROLI
>>> 2 Glebe Road, London E8 4BD
>>> www.roli.com
>>>
>>> Discover more about us on Facebook - Twitter - Youtube
>>>
>>> ROLI Ltd. is a registered company in England and Wales, and this e-mail
>>> and its attachment(s) are intended for the above named only and are
>>> confidential. If they have come to you in error then you must take no
>>> action based upon them but contact us immediately. Any disclosure, copying,
>>> distribution or any action taken or omitted to be taken in reliance on it
>>> is prohibited and may be unlawful. Although this e-mail and its attachments
>>> are believed to be free of any virus, it is the responsibility of the
>>> recipient to ensure that they are virus free. If you contact us by e-mail
>>> then we will store your name and address to facilitate communications. Any
>>> statements contained herein are those of the individual and not the
>>> organisation.
>>>
>>>
>>> On 11 September 2015 at 23:49, Russell Wallace via cfe-dev <
>>> cfe-dev at lists.llvm.org> wrote:
>>>
>>>> That almost works. What am I still missing? I tried -g but no effect.
>>>>
>>>> C:\t>clang -D_DEBUG -I/llvm/build/include -I/llvm/include -S -emit-llvm
>>>> -fms-compatibility-version=19 a.cpp
>>>>
>>>> C:\t>clang a.ll \llvm\build\Debug\lib\*.lib
>>>> warning: overriding the module target triple with
>>>> x86_64-pc-windows-msvc18.0.0 [-Woverride-module]
>>>> 1 warning generated.
>>>> LINK : warning LNK4098: defaultlib 'LIBCMTD' conflicts with use of
>>>> other libs; use /NODEFAULTLIB:library
>>>> LLVMSupport.lib(TargetParser.obj) : error LNK2001: unresolved external
>>>> symbol _CrtDbgReportW
>>>> libcpmtd.lib(stdthrow.obj) : error LNK2001: unresolved external symbol
>>>> _CrtDbgReportW
>>>> libcpmtd.lib(syserror.obj) : error LNK2001: unresolved external symbol
>>>> _CrtDbgReportW
>>>> LLVMSupport.lib(ConvertUTFWrapper.obj) : error LNK2001: unresolved
>>>> external symbol _CrtDbgReportW
>>>> LLVMSupport.lib(Host.obj) : error LNK2001: unresolved external symbol
>>>> _CrtDbgReportW
>>>> LLVMSupport.lib(MemoryBuffer.obj) : error LNK2001: unresolved external
>>>> symbol _CrtDbgReportW
>>>> etc...
>>>>
>>>> On Fri, Sep 11, 2015 at 9:14 PM, Reid Kleckner <rnk at google.com> wrote:
>>>>
>>>>> You're crossing the relase/debug MSVC CRT streams. Make sure all
>>>>> objects in a given image are using the same CRT.
>>>>>
>>>>> In your specific example, you should add -D_DEBUG to the command line.
>>>>> You want to simulate the effect of passing /MTd to cl.exe or clang-cl.exe.
>>>>>
>>>>> On Fri, Sep 11, 2015 at 10:07 AM, Russell Wallace via cfe-dev <
>>>>> cfe-dev at lists.llvm.org> wrote:
>>>>>
>>>>>> I'm trying to compile some code via IR and link it with the debug
>>>>>> libraries on Windows, but getting errors. A minimal test case:
>>>>>>
>>>>>> C:\t>type a.cpp
>>>>>> #define _HAS_EXCEPTIONS 0
>>>>>> #include <llvm/Support/Signals.h>
>>>>>>
>>>>>> int main() {
>>>>>>   llvm::sys::PrintStackTraceOnErrorSignal();
>>>>>>   return 0;
>>>>>> }
>>>>>>
>>>>>> C:\t>clang -I/llvm/build/include -I/llvm/include -S -emit-llvm
>>>>>> -fms-compatibility-version=19 a.cpp
>>>>>>
>>>>>> C:\t>clang a.ll \llvm\build\Debug\lib\*.lib
>>>>>> warning: overriding the module target triple with
>>>>>> x86_64-pc-windows-msvc18.0.0 [-Woverride-module]
>>>>>> 1 warning generated.
>>>>>> LLVMSupport.lib(Signals.obj) : error LNK2038: mismatch detected for
>>>>>> '_ITERATOR_DEBUG_LEVEL': value '2' doesn't match value '0' in a-db717b.o
>>>>>> LLVMSupport.lib(Signals.obj) : error LNK2038: mismatch detected for
>>>>>> 'RuntimeLibrary': value 'MTd_StaticDebug' doesn't match value
>>>>>> 'MT_StaticRelease' in a-db717b.o
>>>>>> LLVMSupport.lib(Path.obj) : error LNK2038: mismatch detected for
>>>>>> '_ITERATOR_DEBUG_LEVEL': value '2' doesn't match value '0' in a-db717b.o
>>>>>> LLVMSupport.lib(Path.obj) : error LNK2038: mismatch detected for
>>>>>> 'RuntimeLibrary': value 'MTd_StaticDebug' doesn't match value
>>>>>> 'MT_StaticRelease' in a-db717b.o
>>>>>> LLVMSupport.lib(raw_ostream.obj) : error LNK2038: mismatch detected
>>>>>> for '_ITERATOR_DEBUG_LEVEL': value '2' doesn't match value '0' in a-db717b.o
>>>>>> ...etc...
>>>>>>
>>>>>> There's probably an option that I'm missing?
>>>>>>
>>>>>> _______________________________________________
>>>>>> cfe-dev mailing list
>>>>>> cfe-dev at lists.llvm.org
>>>>>> http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-dev
>>>>>>
>>>>>>
>>>>>
>>>>
>>>> _______________________________________________
>>>> cfe-dev mailing list
>>>> cfe-dev at lists.llvm.org
>>>> http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-dev
>>>>
>>>>
>>>
>>
>> _______________________________________________
>> cfe-dev mailing list
>> cfe-dev at lists.llvm.org
>> http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-dev
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/cfe-dev/attachments/20150914/b50c863c/attachment.html>