similar to: Automake Assembler Assumptions with LLVM-MC

All, we recently had a user report an issue associated with utilizing LLVM-MC as the desired assembler in place of GNU’s GAS. They are utilizing a tangled mess of automake scripts to build this respective software package and have decided to manually override the “CCAS” macro to point directly to LLVM-MC (rather than using Clang). Apparently, automake assumes that whatever “compiler” (as it is

As one of the OS' without current CMake support, I'm closely watching this discussion. We currently have LLVM 3.4.2 hosted on OpenVMS Itanium (as a host only, x86 target) using configure/make with little hassle. We plan to port CMake to OpenVMS, but that has been trickier than you'd think (others have tried, I haven't found anybody who has done it). Looks like I'll want to

Hello, I'd like to introduce myself, my company, and our upcoming use of LLVM. My name is John Reagan. I've been working on compilers and assemblers since 1983 (yes, 31 years). Most of that time was spent on compilers for VAX/VMS (later renamed to OpenVMS), then OpenVMS on Alpha, and OpenVMS on Itanium. I've also worked with the HP NonStop platform and was directly involved

Yes. We currently build LLVM 3.4.2 on our OpenVMS Itanium box with an older EDG/Intel C++03 compiler to create legacy cross-compilers to our OpenVMS x86 box (well, VirtualBox). We do have a few tweaks to the relocations to access static data always through the GOT (including CodeGen's static data). Our linker sees references to code (which might be in 64-bit space) and creates trampolines

> -----Original Message----- > From: llvm-dev [mailto:llvm-dev-bounces at lists.llvm.org] On Behalf Of JF > Bastien via llvm-dev > Sent: Thursday, June 13, 2019 12:25 PM > To: John Reagan > Cc: llvm-dev at lists.llvm.org > Subject: Re: [llvm-dev] [RFC] Coding Standards: "prefer `int` for, regular > arithmetic, use `unsigned` only for bitmask and when you, intend to

We use .note descriptors with ELF on OpenVMS.  We have self-describing entries with tags and lengths.  The compiler encodes stuff for the linker such as compiler version, compilation date-time, module name, final severity of the compilation, etc.  The linker itself then adds/modifies the .note entries to insert linker version, link date/time, etc.  We also use .notes to record floating point

As part of our port of OpenVMS to x86-64, we are using LLVM with our own frontends on OpenVMS Itanium. We are writing a converter between our old backend's IR and the LLVM IR. We can cross-compile (hosted on OpenVMS Itanium) and link/execute on x86-64 CentOS. At present, we pass over 88% of our C test suite. Porting starts with being able to compile the code. Since we are stuck with a

I'm not sure if they are doing an x86 to IA64 cross compile, but in any event I'm going to guess they may need an ancient version to avoid any C++11 dependencies. In terms of IA64 compilers you have afaik 3 choices HP compiler, Open64 and Intel? (Does gcc still support it and how up-to-date or EOL is the Intel compiler IA64 support?) I really hope nobody decides not to move to a more

Hi all, We're at the point in our port of OpenVMS to x86-64 that we're working on the unwind code.  The current ABI and the current codebase doesn't have enough support for true asynchronous unwinding from any point (most notably in the prologue/epilogue) in the code that OpenVMS needs.  We're working on a set of changes to the compact unwind information to handle the additional

As part of our OpenVMS-specific changes, we're adding an arg-count to all of our calls. We've done this as part of callLowering to put the # of slots used into %ah and leaving %al to hold the standard info about vector registers.  That part is working. However, we're struggling somewhat on the code in the prologue to capture the %ah value to then be obtained by our

I had thought about "used", but not aware of the @llvm.used, et al. I wrote some test programs with __attribute__((used)) but that felt like something you'd put on function definitions to force code to be generated regardless. In the worst case, I'll do some metadata hack (I've ready had to do that for BLISS' GLOBAL BIND feature). Thanks for the response. I'll let

As somebody who's currently hosting LLVM on a platform (OpenVMS Itanium) that has configure but not a working CMake (we're working to fix that but there are some tricky issues), I would appreciate if you didn't scrub the existence of configure from the source or the documentation. Perhaps keep pointers to the older pages and link to them from the downloads pages or something with an

Amy, when you say "implement MSVC's extensions", is that MSVC/LLVM or are you going to add these to clang? OpenVMS has dual-sized pointers as well and want to see them in clang. One of my engineers noticed that address space 4 was a 32-bit pointer address space (or at least it used to be). I haven't seen any formal discussion of that over in the cfe dev list. I'm happy to

On Fri, Aug 25, 2017 at 8:51 AM, Daniel Sanders <daniel_l_sanders at apple.com> wrote: > (removed my @imgtec.com address since it no longer exists) > > Sorry for the slow reply, it's a busy time for me right now. > > > On 23 Aug 2017, at 00:21, George Karpenkov via llvm-dev < > llvm-dev at lists.llvm.org> wrote: > > > > Hi, > > > > As a

Hi John, On 14 March 2018 at 18:28, John Reagan via llvm-dev <llvm-dev at lists.llvm.org> wrote: > However, we're struggling somewhat on the code in the prologue to > capture the %ah value to then be obtained by our "arg_count" compiler > builtin. We found where the 1st parameter is put into a virtual > register in the event that it has to be passed back in %rax

That is basically what we do today to provide mixed sized pointers with our legacy frontends. They generate IR to our old code generator which has ADDR32 and ADDR64 datatypes. We use a 64-bit address data layout and then typecast the 32-bit forms to/from the underlying 64-bit addresses. I have been warned that such rampant typecasting might interfere with certain optimizations or TBAA data. We

I'm defining an external function in the IR that has no uses at all. No calls, no address taken, nada. Such an unused declaration seems to be just dropped on the floor as not needed. Seems reasonable in most cases. However, one of my OpenVMS compilers (BLISS) has a language rule that expects such definitions to get into the ELF symbol table as a way to compel the linker to include certain

Let me see if I understand the response ;-) When you are saying integrated assembler do you mean llc --filetype=obj? If so, we currently have that for an option when running the test-suite. When you say that to test the llvm-mc assembler for your target you don't substitute the gcc assembler invocation for llvm-mc which would expect the resultant executable run to pass. Instead you have to

# Summary As a consequence of comparing the RISC-V LLVM MC assembler to the RISC-V GNU assembler I've noticed that a number of targets have quite different handling for pc-relative jumps/branches with immediate integer operands in llvm-mc vs GNU as. I'll admit that this isn't likely to occur in hand-written code (as you'd almost always prefer to use a label), but thought it was

Hello all, Pardon me if this has already been covered elsewhere, however I have not been able to find such documentation. Is there a consolidated set of documentation that clearly explains what's necessary to port LLVM to other OSes & how to add support for building executables (& libraries) for those OSes? I'm searching through the source in an attempt to understand what needs to