michael-roe

Ok, so this is meant to be a float to int conversion where large (negative or positive) values saturate at MAXINT or MININT. But.... - what about NaN? - Do...

For what it's worth, the behaviour of MIPS cvt.w.s is delightfully dependent on ISA version and mode flags, but will probably do what you want with a sufficiently recent version...

The RISC-V fcvt instruction will also do what you want, with NaN (whose conversion result we don't have strong opinions about) being converted to maxint. So on this architecture, too,...

In gnuradio's float_to_int_impl.c, I see this comment: // Disable the Volk for now. There is a problem for large 32-bit ints that // are not properly represented by the precisions...

I think the right way to do this is that the kernel reads the FDT from the boot rom, and uses that to work out which extensions the CPU supports....

User space programs (which includes cpu_features) ought to use proc/cpuinfo and/or hwcap. But the part of the Linux kernel that creates /proc/cpuinfo (and, probably, hwcap) looks at the FDT to...

See https://github.com/torvalds/linux/blob/master/arch/riscv/kernel/cpufeature.c for the part of the Linux kernel that works out which ISA features are supported. What I think is happening here is that it's getting the information from...

"Me too" --- I would like to have an FMVH instruction. Is there a draft online of the proposed floating point extensions? Is there a recommended opcode from the implenentation-specific...

> How do we determine whether this is compatible with other/legacy apps? It won't be compatible with prior versions of GR. I was worrying someone would say that ... that...

Ok, so the more generic "CRC Check" block will do what I want, with appropriate parameters. I still think crc16_async_bb is buggy. It will produce different results when run on...