Re: [Patch] libgomp.texi: Document libmemkind + nvptx/gcn specifics

[Jakub Jelinek <jakub@redhat.com>]

On Mon, Aug 29, 2022 at 12:54:33PM +0200, Tobias Burnus wrote:
> I had this patch lying around since about half a year. I did tweak and agumented it
> a bit today, but finally want to get rid of it (locally - by getting it committed) ...
> 
> This patch changes -misa to -march for nvptx (the latter is now an alias
> for the former), it adds a new section about libmemkind and some information
> about interns of our nvptx/gcn implementation. (The latter should be mostly
> correct, but I might have missed some fine print or a more recent update.)
> 
> OK for mainline?
> 
> Tobias
> 
> 
> -----------------
> Siemens Electronic Design Automation GmbH; Anschrift: Arnulfstraße 201, 80634 München; Gesellschaft mit beschränkter Haftung; Geschäftsführer: Thomas Heurung, Frank Thürauf; Sitz der Gesellschaft: München; Registergericht München, HRB 106955

> libgomp.texi: Document libmemkind + nvptx/gcn specifics
> 
> libgomp/ChangeLog:
> 
> 	* libgomp.texi (OpenMP-Implementation Specifics): New; add libmemkind
> 	section; move OpenMP Context Selectors from ...
> 	(Offload-Target Specifics): ... here; add 'AMD Radeo (GCN)' and
> 	'nvptx' sections.
> +All OpenMP and OpenACC levels are used, i.e.
> +@itemize
> +@item OpenMP's simd and OpenACC's vector map to work items (thread)
> +@item OpenMP's threads (``parallel'') and OpenACC's workers map
> +      to wavefronts
> +@item OpenMP's teams and OpenACC's gang use use a threadpool with the

s/use use/use/

> +All OpenMP and OpenACC levels are used, i.e.
> +@itemize
> +@item OpenMP's simd and OpenACC's vector map to threads
> +@item OpenMP's threads (``parallel'') and OpenACC's workers map to warps
> +@item OpenMP's teams and OpenACC's gang use use a threadpool with the

Again.

Otherwise LGTM.

	Jakub