'mtfsf' instruction can clear FI incorrectly
Using mtfsf instruction can clear the FPSCR FI bit incorrectly. This code snippet exhibits the issue: -- fpscr.ll = 0x1fffffff; __builtin_mtfsf (0b11111111, fpscr.d); fpscr.d = __builtin_mffs (); -- On POWER9 hardware: mffs : FPSCR = 0x000000007ffff7ff On qemu (git master; "-cpu POWER9"): -- $ ./mtfsf mffs : FPSCR = 0x000000007ffdffff -- Two differences: bit 52: "reserved", so maybe a "don't care" case bit 46: "FI" $ git log -1 master commit 89ea03a7dc83ca36b670ba7f787802791fcb04b1 Merge: 019217c 2531164 Author: Peter Maydell <email@example.com> Date: Mon Sep 9 09:48:34 2019 +0100 I tracked the clear is coming from do_float_check_status, likely the one in gen_mtfsf, but then I get lost figuring out what _should_ be happening. :-/ Test attached.
To upload designs, you'll need to enable LFS and have an admin enable hashed storage. More information