Hi all,
Background
I’ve been working on adding the necessary support to LLVM for clang to be able to support the STDC FENV_ACCESS pragma, which basically allows users to modify rounding mode at runtime and depend on the value of floating-point status flags or to unmask floating point exceptions without unexpected side effects. I’ve committed an initial patch (r293226) that adds constrained intrinsics for the basic FP operations, and I have a patch up for review now (⚙ D32319 Add constrained intrinsics for some libm-equivalent operations) that adds constrained versions of a number of libm-like FP intrinsics.
Current problem
Now I’m trying to make sure I have a good solution for the way in which the optimizer handles recognized calls to libm functions (sqrt, pow, cos, sin, etc.). Basically, I need to prevent all passes from making any modifications to these calls that would make assumptions about rounding mode or improperly affect the FP status flags (either suppressing flags that should be present or setting flags that should not be set). For instance, there are circumstances in which the optimizer will constant fold a call to one of these functions if the value of the arguments are known at compile time, but this constant folding generally assumes the default rounding mode and if the library call would have set a status flag, I need the flag to be set.
Question
My question is, can/should I just rely on the front end setting the “nobuiltin” attribute for the call site in any location where the FP behavior needs to be restricted?
Ideally, I would like to be able to conditionally enable optimizations like constant folding if I am able to prove that the rounding mode, though dynamic, is known for the callsite at compile time (the constrained intrinsics have a mechanism to enable this), but at the moment I am more concerned about correctness and would be willing to sacrifice optimizations to get correct behavior. Long term, I was thinking that maybe I could do something like attach metadata to indicate rounding mode and exception behavior when they were known.
Is there a better way to do this?
Thanks,
Andy
Hi all,
Background
I’ve been working on adding the necessary support to LLVM for clang to be able to support the STDC FENV_ACCESS pragma, which basically allows users to modify rounding mode at runtime and depend on the value of floating-point status flags or to unmask floating point exceptions without unexpected side effects. I’ve committed an initial patch (r293226) that adds constrained intrinsics for the basic FP operations, and I have a patch up for review now (https://reviews.llvm.org/D32319) that adds constrained versions of a number of libm-like FP intrinsics.
Current problem
Now I’m trying to make sure I have a good solution for the way in which the optimizer handles recognized calls to libm functions (sqrt, pow, cos, sin, etc.). Basically, I need to prevent all passes from making any modifications to these calls that would make assumptions about rounding mode or improperly affect the FP status flags (either suppressing flags that should be present or setting flags that should not be set). For instance, there are circumstances in which the optimizer will constant fold a call to one of these functions if the value of the arguments are known at compile time, but this constant folding generally assumes the default rounding mode and if the library call would have set a status flag, I need the flag to be set.
Question
My question is, can/should I just rely on the front end setting the “nobuiltin” attribute for the call site in any location where the FP behavior needs to be restricted?
Unless you want to assume that the frontend knows about all functions that backend
knows about, which I don’t think we do, I think this will essentially mean adding nobuiltin
to all calls. This is what we do if you compile using Clang with -fno-builtin or
-ffreestanding. This will work, although it seems better to have a dedicated attribute for this.
no_fp_opts, no_fp_builtin or whatever.
I considered the idea of an FP-specific attribute anyway, because eventually I think I’ll want to do things with these calls that nobuiltins really should be disallowing. It hadn’t occurred to me that the front end doesn’t know which calls we might be tinkering with. I guess I was assuming that we’d add something to the front end to recognize the FP library calls, but I can understand why we might want to not do that.
Other than that, will this approach solve my problem?
Apart from the existing intrinsics (which I believe the front end uses for builtins?) and LibFunc handling, are there other cases I need to be thinking about?
Thanks,
Andy
Yea, I don’t think we want that coupling. We obviously do know the standard C ones, but the backend can recognize all sorts of things. Yes. I believe that it will. I think that should be everything. -Hal