- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi all.
I'm serching for a routine to do pointwise multiplication of two double precision vectors (something like a vdMul in VML library)
I already saw a message about this topic, about the __mm_mul_pd instrinsic, but this intrinsic is a SIMD 2 (only works for Pentium4).
Any body have a hint to solve construct a vdMul function?
Thanks.
I'm serching for a routine to do pointwise multiplication of two double precision vectors (something like a vdMul in VML library)
I already saw a message about this topic, about the __mm_mul_pd instrinsic, but this intrinsic is a SIMD 2 (only works for Pentium4).
Any body have a hint to solve construct a vdMul function?
Thanks.
Link Copied
2 Replies
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hello,
Intel compiler should perform well on pointwise multiplication. Did you try the compiler? I wounder if adding this feature to MKL would bringsome benefit compared to the compiler generated code.
Sergey
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I completely concur with Sergey.
Bruce

Reply
Topic Options
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page