Version 0.80.0, _mm_aeskeygenassist_si128 inlining mismatch, compile fails

[PyBonnetainNesterenko]

Gentoo Linux, gcc 12.

/usr/lib/gcc/x86_64-pc-linux-gnu/12/include/wmmintrin.h: In function ‘aesni_setup’:
/usr/lib/gcc/x86_64-pc-linux-gnu/12/include/wmmintrin.h:87:1: error: inlining failed in call to ‘always_inline’ ‘_mm_aeskeygenassist_si128’: target specific option mismatch

87 | _mm_aeskeygenassist_si128 (__m128i __X, const int __C)
| ^~~~~~~~~~~~~~~~~~~~~~~~~
ltc/ciphers/aes/aesni.c:27:25: note: called from here
27 | #define setup_mix(t, c) _mm_extract_epi32(_mm_aeskeygenassist_si128(t, 0), c)
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
ltc/ciphers/aes/aesni.c:79:26: note: in expansion of macro ‘setup_mix’
79 | rk[4] = rk[0] ^ setup_mix(temp, 3) ^ rcon[i];
| ^~~~~~~~~
/usr/lib/gcc/x86_64-pc-linux-gnu/12/include/wmmintrin.h:87:1: error: inlining failed in call to ‘always_inline’ ‘_mm_aeskeygenassist_si128’: target specific option mismatch
87 | _mm_aeskeygenassist_si128 (__m128i __X, const int __C)
| ^~~~~~~~~~~~~~~~~~~~~~~~~
ltc/ciphers/aes/aesni.c:27:25: note: called from here
27 | #define setup_mix(t, c) _mm_extract_epi32(_mm_aeskeygenassist_si128(t, 0), c)
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

(and several other errors). Looks somehow like #97.

[thesamesam]

I imagine #99 fixes this.

[karel-m]

ping @sjaeckel