I have built llvm-13.0.1 from source using Visual Studio 17 2022 on Windows 11. When attempting to build llvm-sys (through external crate llvm-ir). It fails, saying:
error: No suitable version of LLVM was found system-wide or pointed to by LLVM_SYS_130_PREFIX.
I know llvm-config exists because I can use it from the command line, and it returns the correct version. I don't understand why this doesn't work. This might be a stupid question but I am really stuck.
I was also having this problem, trying to use llvm-sys on Windows. Here are some mistakes that I did that made it fail to compile just like yours.
Installation Prefix
Make sure that your LLVM installation path doesn't have any whitespace.
Mine was -DCMAKE_INSTALL_PREFIX=C:\Program Files (x86)\LLVM, which was their default install prefix. The whitespace in the installation path causes some errors when using it with llvm-sys. Later, I changed it to C:\LLVM and it worked.
Set Environment Variable
Set Environment Variable LLVM_SYS_<version>_PREFIX to the installed LLVM folder.
From CMD For LLVM-13.0.1
setx LLVM_SYS_130_PREFIX "<PATH TO LLVM WITHOUT WHITESPACE>" /M
Make sure that the variable is registered by using echo command
echo %LLVM_SYS_130_PREFIX%
It should output your registered path, not %LLVM_SYS_130_PREFIX%.
If it doesn’t output your path, restart the computer and try again and it should be working.
Now you should be good to go .....
Hope this helps :)
Related
I'm new on this Makefile thing, and I'm having some issues trying to install the Makefile Extension from the VS Code Marketplace.
First of all, I'm using VS Code to connect to a remote machine, with RedHat8 kernel. It uses makefile and it works perfectly on a linux terminal, but now we are trying to migrate to VS Code.
So I was searching and I found this Makefile Extension.
On the extension settings, I have my Makefile Path defined, and it works, but I have some problem with the Make Path, because it is expecting a make.exe kind but I don't have any, because it's a linux server.
I know my make path is on /usr/bin/make, but it says it cannot find make path.
What should I put in there?
PS: Sorry for my bad english, and thank you!
You should not have to set the "Make Path" to anything, it should default to make.
By default the "Makefile Path" is ${workspaceFolder}/Makefile, but you can change it to whatever is required.
If you have a separate build folder, then it's better to set the "Make directory" instead.
For example, Debian packages are built under debian/build/main so set the makeDirectory like this:
{
"makefile.makeDirectory": "${workspaceFolder}/debian/build/main"
}
I followed this guide and installed emscripten using emsdk and activated it.Then I configured my $PATH as instructed by the emsdk itself and also sourced emsdk-master/emsdk_env.sh.Now I can access both emcc and em++ in the terminal.The file ~/.emscripten is also present(this is the file that QtCreator will fetch to find the path of compilers for WASM).
The WebAssembly kit for Qt also has been installed by the Qt Maintenance Tool.
Now in the Kit configuration in the QtCreator I get this(QtCreator is opened via terminal after sourcing emsdk_env.sh):
It cannot determine the path of compilers by itself.
In the Compilers tab I manually added a compiler as follows:
But I get this error in the Kits tab after that:
What does that mean? What did I skip? Does anybody ever have the experience of doing this?
Also changing the compiler from em++ to wasm-32-wasi-clang++ or clang++ doesn't change anything.
By the way if I use that kit I get:
Error while parsing file whatever.pro. Giving up.
Project ERROR: Cannot run target compiler 'em++'. Output:
===================
===================
Maybe you forgot to setup the environment?
And please don't tell me that this question is the duplicate of this because it isn't(mine has more details) and there's no useful answer for that after 9 months at this time.
Any help is much appreciated.
Make sure you run:
emsdk install
and
emsdk activate
within emsdk folder, not from a relative path.
You have to setup the emsdk inside QtCreator => Preferences => Devices => WebAssembly. Afterwards the compiler should be auto-detected by Qt Creator.
I had the same problem and the solution was, that python was not correctly recognized.
In my case the symbolic link /usr/bin/python was not set
ln -sf /usr/bin/python3 /usr/bin/python
Afterwards the emsripten Toolchain is recognized automatically in QtCreator.
I have installed gprbuild, xmlada, and gnatcoll. I am now attempting to install gnatcoll_postgres. Which I have downladed from here: https://github.com/AdaCore/gnatcoll-db/
Within the Postgres folder is a Makefile, which I execute like so...
[parallels#localhost postgres]$ ls
gnatcoll_postgres.gpr gnatcoll-sql-postgres-gnade.ads
gnatcoll-sql-postgres.adb gnatcoll-sql-ranges.adb
gnatcoll-sql-postgres.ads gnatcoll-sql-ranges.ads
gnatcoll-sql-postgres-builder.adb Makefile
gnatcoll-sql-postgres-builder.ads postgres_support.c
gnatcoll-sql-postgres-gnade.adb README.md
[parallels#localhost postgres]$ make Makefile
which: no gnatls in (/usr/local/bin:/usr/local/sbin:/usr/bin:/usr/sbin:/bin:/sbin:/var/lib/snapd/snap/bin:/home/parallels/.local/bin:/home/parallels/bin)
make: Nothing to be done for `Makefile'.
[parallels#localhost postgres]$
Would anybody please be able to tell me what this means...
which: no gnatls in (/usr/local/bin:/usr/local/sbin:/usr/bin:/usr/sbin:/bin:/sbin:/var/lib/snapd/snap/bin:/home/parallels/.local/bin:/home/parallels/bin)
make: Nothing to be done for `Makefile'.
Any help would be greatly appreciated.
Please see the xmlada and gnatcoll in my project below, does this look like it's installed correctly? I'm presuming this isn't correct...
Thanks,
Lloyd
It means that your GNAT installation binaries aren’t on your PATH.
The README.txt from the adacore.com site says, in part,
To start using the tools in command-line mode, you will need to add
{install_prefix}/bin
to your PATH environment variable. Alternatively, you can simply launch
{install_prefix}/bin/gps
and GPS will automatically add itself to the PATH - it will also find the
cross compiler, if you have installed everything in the default locations.
Note that GPS will add this at the end of the PATH, meaning that it will find first any other GNAT installations that you have in your PATH.
I strongly suspect that you’ve been doing the latter, so that GPS adds itself (actually, of course, its own location) to the PATH, so that when it launches the compiler it finds the correct one.
When you run make from the terminal, the compiler isn’t on the PATH, so neither are gnatls, gprconfig, gprbuild and the rest of the GNAT tools.
What you need to do is to take the first choice from the README, and add /home/parallel/opt/GNAT/2019/bin to (the front of) your default PATH. How you do that depends on your shell.
You will find xmlada, gnatcoll already installed.
When I try compiling for webasm, I get an error “rust-lld not found”. What should I do?
I have been following the instructions in the rust webasm book with a dummy program, and everything worked up till the initial compile, which failed with this error which apparently means that some linker has not been installed.
My context is Ubuntu on an ARM 64 bit box.
rust-lld is in a non $PATH-path which means you have to use either the full path or add the path to your $PATH environment variable.
In my case it can be found in ~/.rustup/toolchains/stable-x86_64-unknown-linux-gnu/lib/rustlib/x86_64-unknown-linux-gnu/bin/rust-lld
Of course the path may vary by architecture and version.
The VS environment variable is set beforehand with vcvars32.bat, and VS is the only compiler that I use; the OS is win 7.
The error happens with cmake-gui; specifying the compiler with path results in error 'compiler failed to make simple test'; alternatively, cmake -i from command line succeeds initially with presenting an 'options' menu which then fails to allocate a 'cmakelists.txt' file (which is somewhere deeper in the cmake folder).
Where am I going wrong?
(I continued with building the solution files from VS 12, but there's other issues.)
The answer to your question from cmake mailing list
Run cmake-gui FROM the target environment, just like you run cmake...
i.e. : type "cmake-gui" in the cmd prompt that has your stuff set up
in it. Don't just launch it from a short-cut unless you're using the
"Visual Studio *" generators. Those do not need any special
environment. Many of the makefile ones do.