Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fix _mm_loadu_si64 not available in GCC8 #366

Open
wants to merge 1 commit into
base: main
Choose a base branch
from

Conversation

razvanphp
Copy link

This preserves BC while allowing compilation on older systems like Debian 10.

Inspired from ggerganov/llama.cpp#1154

This preserves BC while allowing compilation on older systems like Debian 10.
@@ -956,7 +956,7 @@ static OPUS_INLINE void silk_nsq_del_dec_scale_states_avx2(
inv_gain_Q26 = silk_sar_round_32(inv_gain_Q31, 5);
for (i = 0; i < psEncC->subfr_length; i+=4)
{
__m256i x = _mm256_cvtepi16_epi64(_mm_loadu_si64(&x16[i]));
__m256i x = _mm256_cvtepi16_epi64(_mm_loadl_epi64(&x16[i]));
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

_mm_loadl_epi64() takes a __m128i *, which means the PR would break alignment rules. I suspect there should still be a workaround...

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Both _mm_loadl_epi64 and _mm_loadu_epi64 take void const *
From gcc-14:

extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_loadu_si64 (void const *__P)
{
  return _mm_loadl_epi64 ((__m128i_u *)__P);
}
/* */
extern __inline __m128i
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm_loadu_epi64 (void const *__P)
{
  return (__m128i) (*(__v2di_u *) __P);
}

Maybe just adding an (__m128i_u *) cast would be enough??

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants