• james.p.leblanc (17/17) Sep 07 2021 Dear All,
• Tejas (17/36) Sep 07 2021 Yes you are correct (to my understanding)
• Tejas (3/21) Sep 07 2021 Here's the link:
• Johan (5/6) Sep 08 2021 Yes.

james.p.leblanc <james.p.leblanc gmail.com> writes:

Dear All,

In searching through the forum archives (going back to 2013, 2016
etc), and experiments, it **appears to me** that LDC does indeed
respect the standard "align" properties.  (Meaning: proper 
alignment
for using AVX with static arrays can be guaranteed).

Experiments (and forum discussions) also lead me to believe that
DMD does NOT respect this alignment.

So, what is the "official status" of AVX alignment possibilities?

Succinctly:

1) Can we truly rely on LDC's alignment for AVX ?

2) Is DMD presently **not** respecting alignment? (Or have I
    misunderstood?)

This has important impact on performance of numerical 
computations.

Thanks for all illumination!

James

Tejas <notrealemail gmail.com> writes:

On Wednesday, 8 September 2021 at 04:32:50 UTC, james.p.leblanc 
wrote:
 Dear All,

 In searching through the forum archives (going back to 2013, 
 2016
 etc), and experiments, it **appears to me** that LDC does indeed
 respect the standard "align" properties.  (Meaning: proper 
 alignment
 for using AVX with static arrays can be guaranteed).

 Experiments (and forum discussions) also lead me to believe that
 DMD does NOT respect this alignment.

 So, what is the "official status" of AVX alignment 
 possibilities?

 Succinctly:

 1) Can we truly rely on LDC's alignment for AVX ?

 2) Is DMD presently **not** respecting alignment? (Or have I
    misunderstood?)

 This has important impact on performance of numerical 
 computations.

 Thanks for all illumination!

 James

Yes you are correct (to my understanding)

DMD only respects `align` keyword upto the value 16,ie, until 
`align(16)`, the code behaves the way you expect it to. It is 
100% a bug(don't have the link on me right now)

Try the following code on DMD, then LDC. See for yourself

```d
import std.stdio:writeln;

void main()
{
     align(16) int[100] a;
     align(1024) int[100] b;
     writeln(cast(ulong)&a[0] % 16);
     writeln(cast(ulong)&b[0] % 1024);
}
```

Tejas <notrealemail gmail.com> writes:

On Wednesday, 8 September 2021 at 04:43:31 UTC, Tejas wrote:
 On Wednesday, 8 September 2021 at 04:32:50 UTC, james.p.leblanc 
 wrote:
 [...]

 Yes you are correct (to my understanding)

 DMD only respects `align` keyword upto the value 16,ie, until 
 `align(16)`, the code behaves the way you expect it to. It is 
 100% a bug(don't have the link on me right now)

 Try the following code on DMD, then LDC. See for yourself

 ```d
 import std.stdio:writeln;

 void main()
 {
     align(16) int[100] a;
     align(1024) int[100] b;
     writeln(cast(ulong)&a[0] % 16);
     writeln(cast(ulong)&b[0] % 1024);
 }
 ```

Here's the link:
https://issues.dlang.org/show_bug.cgi?id=16098

Johan <j j.nl> writes:

On Wednesday, 8 September 2021 at 04:32:50 UTC, james.p.leblanc 
wrote:
 1) Can we truly rely on LDC's alignment for AVX ?

Yes.
If you find wrong alignment, it's a bug.

-Johan