[PATCH] D73570: [FPEnv][X86] Platform-specific builtin constrained FP enablement
Kevin P. Neal via Phabricator via cfe-commits
cfe-commits at lists.llvm.org
Wed Jan 29 05:50:04 PST 2020
kpn marked 2 inline comments as done.
kpn added inline comments.
================
Comment at: clang/test/CodeGen/fma-builtins-constrained.c:4
+// RUN: %clang_cc1 -ffreestanding %s -triple=x86_64-apple-darwin -target-feature +fma -S -o - | FileCheck --check-prefix=COMMON --check-prefix=CHECK-ASM %s
+// RUN: %clang_cc1 -ffreestanding %s -triple=x86_64-apple-darwin -target-feature +fma -ffp-exception-behavior=strict -S -o - | FileCheck --check-prefix=COMMON --check-prefix=CHECK-ASM %s
+
----------------
craig.topper wrote:
> Technically the fmsub/fnmsub/fnmadd assembly requires optimizations to be enabled. If it appears to work without optimizations its only because fast-isel fell back to SelectionDAG and picked up optimizations due to that. Not something that should be relied on.
Ok. Well, the eventual goal is to ship a product with optimization turned on. So I think it makes sense for me to give this a spin with optimizations and see what if anything needs to be done. Unless that's overkill I'll start this afternoon.
================
Comment at: clang/test/CodeGen/fma-builtins-constrained.c:6
+
+// FIXME: Several of these tests are broken when constrained.
+
----------------
craig.topper wrote:
> Is this just referring to the FIXME-CHECK-ASM?
Yes.
Repository:
rG LLVM Github Monorepo
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D73570/new/
https://reviews.llvm.org/D73570
More information about the cfe-commits
mailing list