Displaying 4 results from an estimated 4 matches for "parallel_map".
2012 Sep 26
0
[LLVMdev] [RFC] Parallelization metadata and intrinsics in LLVM (for OpenMP, etc.)
...n actual
openmp backend.
I have also tried to make this representation more orthogonal and
general; mirroring a significant subset of openmp directives inside
llvm's IR doesn't feel right. For instance, in the following
proposal, the OpenMP TASK directive is lowered using the more general
parallel_map construct. A pass lowering the intrinsics into an OpenMP
backend may "reverse engineer" the mentioned pattern into tasks when
possible, but, in principle, I think the directives we introduce into
llvm's IR are best left as mutually exclusive as possible. Keeping
the intrinsics simpl...
2012 Aug 14
4
[LLVMdev] [RFC] Parallelization metadata and intrinsics in LLVM (for OpenMP, etc.)
On Tue, 14 Aug 2012 10:22:35 +0300
Pekka Jääskeläinen <pekka.jaaskelainen at tut.fi> wrote:
> On 08/13/2012 10:54 PM, Hal Finkel wrote:
> > I had thought about uses for shared-memory OpenCL implementations,
> > but I don't know enough about the use cases to make a specific
> > proposal. Is your metadata documented anywhere?
>
> It is now a quick "brute
2012 Sep 26
1
[LLVMdev] [RFC] Parallelization metadata and intrinsics in LLVM (for OpenMP, etc.)
...n actual
openmp backend.
I have also tried to make this representation more orthogonal and
general; mirroring a significant subset of openmp directives inside
llvm's IR doesn't feel right. For instance, in the following
proposal, the OpenMP TASK directive is lowered using the more general
parallel_map construct. A pass lowering the intrinsics into an OpenMP
backend may "reverse engineer" the mentioned pattern into tasks when
possible, but, in principle, I think the directives we introduce into
llvm's IR are best left as mutually exclusive as possible. Keeping
the intrinsics simpl...
2012 Oct 02
7
[LLVMdev] [RFC] Parallelization metadata and intrinsics in LLVM (for OpenMP, etc.)
Hal, Andrey, Alexey,
>From the LLVM design viewpoint, there is a fundamental problem with both Hal's approach and the Intel approach: both are quite language-specific. OpenMP is a particular parallel language, with particular constructs (e.g., parallel regions) and semantics. LLVM is a language-neutral IR and infrastructure and OpenMP-specific concepts should not creep into it. I've