[solved] How to load 256 bits to AVX register at once?

flat assembler
Message board for the users of flat assembler.

Index > Main > [solved] How to load 256 bits to AVX register at once?

Author

Thread

ml64

Joined: 17 Jul 2020
Posts: 12

ml64 17 Jul 2020, 05:53

This doesn't work:

Code:

vmovdqa ymm0,ymmword [rsi]

So doesn't this:

Code:

vmovdqa ymm0,ymmword ptr [rsi]

17 Jul 2020, 05:53

revolution
When all else fails, read the source

Joined: 24 Aug 2004
Posts: 20454
Location: In your JS exploiting you and your system

revolution 17 Jul 2020, 06:31

Code:

vmovdqa ymm0,yword[rsi]

17 Jul 2020, 06:31

ml64

Joined: 17 Jul 2020
Posts: 12

ml64 24 Aug 2020, 08:18

Thank You Very Much!
The question is completely solved.
As a post scriptum i'll expand the solution.
If 256-bit variable is considered as 8 single-precision floating-point numbers 32 bits each
then it can be loaded with <V-vector MOV U-unaligned P-packed S-single> instruction:

Code:

vmovups ymm0, yword [A]

The same instruction extracts it back to the RAM:

Code:

vmovups yword [A],ymm0

24 Aug 2020, 08:18

< Last Thread | Next Thread >

Forum Rules:

You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
You cannot attach files in this forum
You can download files in this forum