[thomasbuilds]

randomx_get_flags() reports RANDOMX_FLAG_HARD_AES only if HAVE_AES is 1, which it never was on RISC-V: rx_aesenc_vec_i128/rx_aesdec_vec_i128 had no RISC-V implementation. So callers that take their flags from it, such as Monero and randomx-benchmark --auto, used software AES even where hardware AES was detected.

This implements both functions with Zvkned in aes_hash_rv64_zvkned.cpp, the file already compiled with Zvkned, and sets HAVE_AES to 1 on RISC-V. They are only called from RANDOMX_FLAG_HARD_AES VMs, which VmBase::allocate() rejects unless cpu.hasAes(). In those VMs, the interpreter's v2 AES mix now uses them instead of software AES. platformError stays defined on RISC-V, where allocate() uses it, and is now a const pointer so GCC's -Wall doesn't flag it as unused. The "Hardware AES not supported" test still runs on RISC-V whenever cpu.hasAes() is false.

[hyc]

If the CPU has hardware AES but HAVE_AES is not being defined, then the right thing to do is to fix that, not muck around here.

[SChernykh]

Yes, HAVE_AES should be 1 on RISC-V because there are AES instruction extensions there.

[SChernykh]

This change seems completely unnecessary. Why single out RISC-V?

[thomasbuilds]

With HAVE_AES now 1 on RISC-V, the old condition would skip this test on every RISC-V build, including the riscv64 CI job, where it runs today (QEMU's CPU has VLEN 128 so cpu.hasAes() is false). RISC-V is singled out because it's the only platform where allocate() checks cpu.hasAes(). Elsewhere, with HAVE_AES 1, allocate() executes an AES instruction so with a plain !cpu.hasAes() the test would fail in the ppc64le and Win32 CI jobs, where cpu.hasAes() is never set but the CPU has AES.

[SChernykh]

Completely unrelated and unnecessary change.

[thomasbuilds]

platformError is only declared inside the #ifndef HAVE_AES block of intrin_portable.h so with HAVE_AES now defined on RISC-V the original line fails to compile ('platformError' was not declared in this scope). The message text is unchanged.

[SChernykh]

I would rather do #if !defined(HAVE_AES) || defined(__riscv) for platformError definition than changing all places where platformError is used.

[thomasbuilds]

I also made it static const char* const otherwise, when building for RISC-V, GCC -Wall warns 'platformError' defined but not used in every file that includes intrin_portable.h without using it, which would break P2Pool's riscv64 GCC build (side_chain.cpp is built with -Werror).

[SChernykh]

Why remove it? If HAVE_AES is 0 and softAes is false, this will compile to a code that will throw an exception in intrin_portable.h

[thomasbuilds]

I restored it