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.

Sign up for GitHub

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

Jump to bottom

Fix _mm_loadu_si64 not available in GCC8 #366

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

Fix _mm_loadu_si64 not available in GCC8 #366

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion silk/x86/NSQ_del_dec_avx2.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -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??

x = _mm256_slli_epi64(_mm256_mul_epi32(x, _mm256_set1_epi32(inv_gain_Q26)), 16);
_mm_storeu_si128((__m128i*)&x_sc_Q10[i], silk_cvtepi64_epi32_high(x));
}
Expand Down