OCP FP8 E5M2 #
OCP E5M2 has a nominal format identity and a direct UInt8 runtime carrier. Its IEEE-style
infinity and NaN encodings remain properties of this format package rather than assumptions of
the universal executable carrier.
For addition, subtraction, multiplication, division, and square root, balanced and throughput planning select exhaustive byte tables; latency planning selects the arithmetic kernels. Fused multiply-add uses the proved generic single-rounding kernel, avoiding a 16,777,216-entry ternary table while preserving the same independent specification.
Reference #
- Open Compute Project, 8-bit Floating Point Specification, revision 1.0, E5M2, https://www.opencompute.org/documents/ocp-8-bit-floating-point-specification-ofp8-revision-1-0-2023-12-01-pdf-1.
OCP E5M2 FP8 with direct byte storage.
Instances For
The E5M2 encoding fits in its direct byte carrier.
Certified direct-byte E5M2 addition table.
Instances For
Certified direct-byte E5M2 subtraction table.
Instances For
Certified direct-byte E5M2 multiplication table.
Instances For
Certified direct-byte E5M2 division table.
Instances For
Certified direct-byte E5M2 square-root table.
Instances For
Package the five byte tables with the proved arithmetic FMA baseline.
Construct OCP E5M2 from a natural bit pattern, reduced to eight bits.
Instances For
Read the complete OCP E5M2 encoding byte.