CPU supplier Ampere Computing has announced AmpereOne as its next-generation AArch64 “cloud-native” server CPU brand, replacing the current Neoverse-N1-based Ampere Altra / Ampere Altra Max processors.

Back in November last year, when Ampere Computing added a support patch for the next-generation server processor “Ampere1” to GCC, it confirmed that the Ampere1 processor was using the ARMv8.6-based ISA and other basic functions, and then added Ampere1 CPU support. into LLVM.

While the AmpereOne processor has not yet been officially launched and the new AArch64 core is an original design, Ampere Computing has submitted a support patch to the GCC compiler for its latest “Ampere-1A” variant.

Now the GCC 13 merge window introduces “ampere-1a”, as a new CPU target. According to foreign media Phoronix, Ampere-1A has an updated instruction table, a new fusion pair (A + B + 1 and A – B – 1), and has a different processor ID than Ampere-1 (not A) . Unlike Ampere-1, Ampere-1A supports AArch64’s Memory Tag Extensions (MTE).

The compiler patch was released early to incorporate it into the next annual GCC compiler release, but Ampere Computing has yet to release details of the “Ampere-1A” CPU, which is supposed to target the AMD EPYC Genoa and Bergamo, and upcoming Intel Xeon Sapphire Rapids .