There have been some complaints about the SIMD bugs in DMD, so I thought I'd have a look and see what can be done. I've fixed several issues (for dmd, excluding the library and enhancement requests), these are the remainder:

https://issues.dlang.org/buglist.cgi?bug_severity=regression&bug_severity=blocker&bug_severity=critical&bug_severity=major&bug_severity=normal&bug_severity=minor&bug_severity=trivial&bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&component=dmd&keywords=simd&keywords_type=allwords&list_id=234152&query_format=advanced

If your bug was filed and was neither fixed nor on that list, please tag with the SIMD keyword.

If your bug was not filed, it won't be fixed, so sad, too bad!

On 24.12.20 02:46, Walter Bright via Digitalmars-d wrote:
> I've fixed several issues

Awesome! Thanks a lot for taking the time and looking into this, Walter!

On Thursday, 24 December 2020 at 01:46:22 UTC, Walter Bright wrote:
> If your bug was filed and was neither fixed nor on that list, please tag with the SIMD keyword.

Please take a look at 20041; it's awful. It would be really nice if this actually worked, but even an ICE would be preferable to what it actually does:

    https://issues.dlang.org/show_bug.cgi?id=20041

On 12/23/2020 7:32 PM, tsbockman wrote:
>      https://issues.dlang.org/show_bug.cgi?id=20041

Ok, I'll have a look. (It's a problem in the front end, not DMD specific, hence affects all D compilers.)

On Thursday, 24 December 2020 at 01:46:22 UTC, Walter Bright wrote:
> I've fixed several issues (for dmd, excluding the library and enhancement requests), these are the remainder:
>

Currently happy with the current state of master, thanks for the recent fixes!
Planning to add actual D_SIMD usage in intel-intrinsics so that DMD is not artificlally behind.
It's a lot of work to do so if anyone is attracted by unusually repetitive and detail-oriented work don't hesitate (particularly for SSSE3, SSE4.1, SSE4.2 any help will help)

On 12/23/2020 7:32 PM, tsbockman wrote:
> Please take a look at 20041; it's awful. It would be really nice if this actually worked, but even an ICE would be preferable to what it actually does:
> 
>      https://issues.dlang.org/show_bug.cgi?id=20041

https://github.com/dlang/dmd/pull/12057

On Sunday, 27 December 2020 at 09:14:28 UTC, Walter Bright wrote:
> On 12/23/2020 7:32 PM, tsbockman wrote:
>> Please take a look at 20041; it's awful. It would be really nice if this actually worked, but even an ICE would be preferable to what it actually does:
>> 
>>      https://issues.dlang.org/show_bug.cgi?id=20041
>
> https://github.com/dlang/dmd/pull/12057

Great, thanks! This will make it *so* much easier to use SIMD in D without accidentally shooting oneself in the foot.

Down to 3 items:

https://issues.dlang.org/buglist.cgi?bug_severity=regression&bug_severity=blocker&bug_severity=critical&bug_severity=major&bug_severity=normal&bug_severity=minor&bug_severity=trivial&bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&component=dmd&keywords=simd&keywords_type=allwords&list_id=234194&query_format=advanced

On Wednesday, 30 December 2020 at 01:56:12 UTC, Walter Bright wrote:
> Down to 3 items:
>
> https://issues.dlang.org/buglist.cgi?bug_severity=regression&bug_severity=blocker&bug_severity=critical&bug_severity=major&bug_severity=normal&bug_severity=minor&bug_severity=trivial&bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&component=dmd&keywords=simd&keywords_type=allwords&list_id=234194&query_format=advanced

Great focused work on related issues!

On 12/31/2020 6:07 AM, M.M. wrote:
> On Wednesday, 30 December 2020 at 01:56:12 UTC, Walter Bright wrote:
>> Down to 3 items:
>>
>> https://issues.dlang.org/buglist.cgi?bug_severity=regression&bug_severity=blocker&bug_severity=critical&bug_severity=major&bug_severity=normal&bug_severity=minor&bug_severity=trivial&bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&component=dmd&keywords=simd&keywords_type=allwords&list_id=234194&query_format=advanced 
>>
> 
> Great focused work on related issues!

It's easier to work on the bunch when they are closely related.