> Aren't they actually the IEEE 754-2019 operations (with different > signaling NaN semantics; C functions such as fmaximum in C23), not the > IEEE 754-2008 operations (C functions such as fmax)? V spec 1.0 says "The > vector floating-point vfmin and vfmax instructions have the same behavior > as the corresponding scalar floating-point instructions in version 2.2 of > the RISC-V F/D/Q extension.". And version 2.2 of F/D/Q (which is *not* > version 2.2 of the instruction set, it's later than that) changed the
Oh, thanks for catching this - I indeed incorrectly assumed this refers to version 2.2 of the RISC-V spec (which contains F/D, .. of version 2.0). Too bad, it appeared too convenient. Then I need to add the same !HONOR_SNANS to all the expanders as well as the tests. Regards Robin