[cfe-dev] Clang on Windows targeting gcc requirements
Edward Diener
eldlistmailingz at tropicsoft.com
Sat Jul 4 22:43:18 PDT 2015
On 7/3/2015 10:59 AM, Yaron Keren wrote:
> You only need to have the bin directory in your PATH. clang will find
> the include and library directory based on the bin directory. There are
> no fixed paths, you can put in c:\mingw or wherever.
I ran into a problem using mingw-64 and the latest clang. In the
mingw-64 intrin.h for gcc-4.9 on up it uses the gcc version number to
decide whether to define the macro __MINGW_FORCE_SYS_INTRINS. If this
macro is not defined it leads to multiple declarations for __m64 and
some other CPU feature typdefs. Its gcc implementation defines
__MINGW_FORCE_SYS_INTRINS for gcc-4.9 on up but clang still sets the gcc
version number to gcc-4.2.1 so that __MINGW_FORCE_SYS_INTRINS is not
automatically defined.
My workaround for this was to define __MINGW_FORCE_SYS_INTRINS when
using clang. This appears to work fine in my tests for clang but then I
found a link of yours at
http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20141110/118349.html
which appears to claim that defining __MINGW_FORCE_SYS_INTRINS forces
the use of invalid CPU instructions when compiling with clang but
evidently not when compiling with gcc.
Does merely specifying all the x86-x64 compiler intrinsics, as the
intrin.h in gcc-4.9 and up does, cause CPU instruction set problems when
compiling with clang ? If so this means that I cannot use clang with
gcc-4.9 and up unless I locally change their headers to avoid the
problem. I have alerted mingw-64 about the general issue.
>
>
>
> 2015-07-03 16:13 GMT+03:00 Edward Diener
> <eldlistmailingz at tropicsoft.com
> <mailto:eldlistmailingz at tropicsoft.com>>:
>
> On 6/25/2015 2:13 PM, Yaron Keren wrote:
>
> See http://reviews.llvm.org/D5268
> <https://urldefense.proofpoint.com/v2/url?u=http-3A__reviews.llvm.org_D5268&d=AwMFaQ&c=8hUWFZcy2Z-Za5rBPlktOQ&r=CnzuN65ENJ1H9py9XLiRvC_UQz6u3oG6GUNn7_wosSM&m=ecMMp1TUmHLfiiZ_MlnPwTgmGQGZqkJoByqz4GWmOJ8&s=GyhTLhjqmvS-L0bXc8ytxxjhk6AnpKowj6eW2JYk9i4&e=>
> <https://urldefense.proofpoint.com/v2/url?u=http-3A__reviews.llvm.org_D5268&d=AwMFaQ&c=8hUWFZcy2Z-Za5rBPlktOQ&r=CnzuN65ENJ1H9py9XLiRvC_UQz6u3oG6GUNn7_wosSM&m=YgSrUuiAhPt76cF_h_LXb8Mg84W8p9nMvUiL3koHKP4&s=aW_6t6KDGfo5hO6NiTn5Hy9afNhzwfkX-BZHT9PtXmk&e=>
>
>
> I can see that the patch has been applied officially to the latest
> clang source. It's in the latest code when I update llvm/clang.
> Thanks very much !
>
> But how do I use it ?
>
> Am I supposed to be able to build clang using a mingw-64 distro in
> my PATH ?
>
> When I compile with clang how do I tell it to use the mingw-64
> headers and RTL ? Does it matter whether the mingw-64 distro is 32
> bit or 64 bit ?
>
> Do I need to have the mingw-64 distro at c:\mingw ?
>
> I am willing to test this for clang and report back results here,
> since I can easily switch between mingw-64 distros ( 4.8.4, 4.9.2,
> 5.1 ) and I test a number of Boost libraries including my own with
> clang, but I need a little information on how to do it.
>
>
>
>
> 2015-06-25 10:46 GMT+03:00 Edward Diener
> <eldlistmailingz at tropicsoft.com
> <mailto:eldlistmailingz at tropicsoft.com>
> <mailto:eldlistmailingz at tropicsoft.com
> <mailto:eldlistmailingz at tropicsoft.com>>>:
>
>
> For either the latest clang built from souce on Windows or
> a binary
> distribution of clang on Windows it appears that clang
> expects the
> supporting gcc include files and lib files to be in the
> hardcoded
> directory:
>
> c:\mingw
>
> and follow the mingw directory structure.
>
> 1) Can this please be changed in the future so that some
> environment
> variable ( maybe CLANG_GCC ) can be set to the gcc installation
> rather than use a hardcoded path ?
>
> 2) Can clang be updated to support the mingw-64 directory
> structure
> and not just the mingw directory structure ? The mingw-64
> distros
> seem now to be the preferred gcc distributions on Windows and
> support the latest versions of gcc.
>
> Do I need to make these suggestions on the clang bug
> tracker for the
> suggestions to be seriously considered ?u
>
>
>
> _______________________________________________
> cfe-dev mailing list
> cfe-dev at cs.uiuc.edu
> <mailto:cfe-dev at cs.uiuc.edu>
> http://lists.cs.uiuc.edu/mailman/listinfo/cfe-dev
>
>
>
>
> _______________________________________________
> cfe-dev mailing list
> cfe-dev at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/cfe-dev
>
More information about the cfe-dev
mailing list