New Hygon Model 8 "Suzhou" x86 CPU Support Appears In The GCC Compiler
A seemingly new generation of Hygon x86 processors are on the way with the GNU Compiler Collection (GCC) seeing support today for the Hygon Model 8 "Suzhou" c86-4g-m8 processors.
A few months back the GCC 17 compiler saw initial support for Hygon C86-4G CPUs with the C86-4G-M4, C86-4G-M6, and C86-4G-M7. Those series were already known to the world and not much of a surprise, but great to see Hygon contributing to the upstream GCC compiler -- they also did similarly for LLVM/Clang too. Now upstreamed to GCC is the C86-4G-M8 support that I haven't heard about previously.
There isn't much in the way of Google search results on the Hygon C86-4G-M8 with the results going back to the GCC patchwork itself or turning up GCC SPARC M8 from back in the day. Going through the commit of the new c86-4g-m8 target for GCC, the HYGON Family 18h model 8 CPU appears to be codenamed Suzhou as the successor to their Model 7 "Chengdu" processors.
The ISA capabilities of Hygon Model 8 Suzhou is the same as with Model 7 and includes AVX-512 and other modern capabilities: "x86-64, MMX, SSE, SSE2, SSE3, SSE4A, CX16, ABM, SSSE3, SSE4.1, SSE4.2, AES, PCLMUL, AVX, AVX2, BMI, BMI2, F16C, FMA, PRFCHW, FXSR, SHA, XSAVE, XSAVEOPT, XSAVEC, FSGSBASE, RDRND, MOVBE, MWAITX, ADX, RDSEED, CLZERO, CLFLUSHOPT, XSAVES, LZCNT, POPCNT, AVX512F, AVX512DQ, AVX512IFMA, AVX512CD,AVX512BW, AVX512VL, AVX512BF16, AVX512VBMI, AVX512VBMI2, GFNI, AVX512VNNI, VAES, AVX512BITALG, AVX512VPOPCNTDQ, AVX512VP2INTERSECT, AVXVNNI, VPCLMULQDQ, WBNOINVD instruction set support."
It will be interesting to learn more about the c86-4g-m8 processors given the current lack of details online and ultimately how well it is able to perform for this Chinese x86 CPU, but for now at least it's nice seeing Hygon has upstreamed the initial enablement into the GCC 17 open-source compiler.
A few months back the GCC 17 compiler saw initial support for Hygon C86-4G CPUs with the C86-4G-M4, C86-4G-M6, and C86-4G-M7. Those series were already known to the world and not much of a surprise, but great to see Hygon contributing to the upstream GCC compiler -- they also did similarly for LLVM/Clang too. Now upstreamed to GCC is the C86-4G-M8 support that I haven't heard about previously.
There isn't much in the way of Google search results on the Hygon C86-4G-M8 with the results going back to the GCC patchwork itself or turning up GCC SPARC M8 from back in the day. Going through the commit of the new c86-4g-m8 target for GCC, the HYGON Family 18h model 8 CPU appears to be codenamed Suzhou as the successor to their Model 7 "Chengdu" processors.
👁 Hygon Model 8 CPU ISA capabilities
The ISA capabilities of Hygon Model 8 Suzhou is the same as with Model 7 and includes AVX-512 and other modern capabilities: "x86-64, MMX, SSE, SSE2, SSE3, SSE4A, CX16, ABM, SSSE3, SSE4.1, SSE4.2, AES, PCLMUL, AVX, AVX2, BMI, BMI2, F16C, FMA, PRFCHW, FXSR, SHA, XSAVE, XSAVEOPT, XSAVEC, FSGSBASE, RDRND, MOVBE, MWAITX, ADX, RDSEED, CLZERO, CLFLUSHOPT, XSAVES, LZCNT, POPCNT, AVX512F, AVX512DQ, AVX512IFMA, AVX512CD,AVX512BW, AVX512VL, AVX512BF16, AVX512VBMI, AVX512VBMI2, GFNI, AVX512VNNI, VAES, AVX512BITALG, AVX512VPOPCNTDQ, AVX512VP2INTERSECT, AVXVNNI, VPCLMULQDQ, WBNOINVD instruction set support."
It will be interesting to learn more about the c86-4g-m8 processors given the current lack of details online and ultimately how well it is able to perform for this Chinese x86 CPU, but for now at least it's nice seeing Hygon has upstreamed the initial enablement into the GCC 17 open-source compiler.