[llvm-dev] Memory scope proposal

[Justin Lebar via llvm-dev]

I'm coming at this from a CUDA perspective, so apologies if this
doesn't make a lot of sense:

In CUDA we have a similar problem as OpenCL.  CUDA solves it by having
a bunch of atomic builtins for each of the memory scopes.  These map
to various llvm target-specific intrinsics.

It's not great, because the intrinsics are mostly opaque to the
optimizer.  But atomic ops are already pretty slow on the GPU, so I've
been operating under the assumption that this isn't hurting us too
much.

Am I wrong about that?

On Wed, Aug 17, 2016 at 3:05 PM, Mehdi Amini via llvm-dev
<llvm-dev at lists.llvm.org> wrote:
>
> On Aug 17, 2016, at 2:08 PM, Zhuravlyov, Konstantin
> <Konstantin.Zhuravlyov at amd.com> wrote:
>
>>Why not going with a metadata attachment directly and kill the
>> "singlethread" keyword? Something like:
>>Something like:
>>  cmpxchg i32* %addr, i32 42, i32 0 monotonic monotonic, 3,
>> !memory.scope{!42}
>>  cmpxchg i32* %addr, i32 42, i32 0 monotonic monotonic, 3,
>> !memory.scope{!43}
>
>>...
>
>>!42 = !{"singlethread"}
>>!43 = !{"L2"}
>
>>It is not clear to me if there is any correctness issues to dropping
>> metadata?
>
> Yes, we cannot use the metadata approach since this metadata can be dropped
> during the processing of one module but not dropped in the processing of a
> second module, potentially resulting in inconsistent scopes for
> synchronizing operations leading to data races and subsequently leading to
> correctness issues.
>
>
>
> Right, I saw Sameer's explanation for that earlier, and we shouldn’t move
> forward (without Philip’s opinion on the topic as he expressed concerns).
>
> But you stripped out the second part of my email where I wrote "It seems
> you’re going back to integer, which I don’t really like for reasons
> mentioned earlier in this thread, and that I don’t feel you addressed here”.
> Why can’t `synchscope` take a string literal?
>
>
> —
> Mehdi
>
>
>
>
>
>
>
> ________________________________
> From: Zhuravlyov, Konstantin
> Sent: Wednesday, August 17, 2016 4:29:30 PM
> To: Sameer Sahasrabuddhe; Philip Reames
> Cc: Mehdi Amini; Liu, Yaxun (Sam); Ke Bai; Mekhanoshin, Stanislav; Sumner,
> Brian; llvm-dev at lists.llvm.org; Tye, Tony
> Subject: RE: [llvm-dev] Memory scope proposal
>
> Hi,
>
> I have updated the review here:
> https://reviews.llvm.org/D21723
>
> As Sameer pointed out, the motivation is:
> In OpenCL 2.x, two atomic operations on the same atomic object need to have
> the same scope to prevent a data race. This derives from the definition of
> "inclusive scope" in OpenCL 2.x. Encoding OpenCL 2.x scope as metadata in
> LLVM IR would be a problem because there cannot be a "safe default value" to
> be used when the metadata is dropped. If the "largest" scope is used as the
> default, then the optimizer must guarantee that the metadata is dropped from
> every atomic operation in the whole program, or not dropped at all.
>
> Thanks,
> Konstantin
>
> From: Sameer Sahasrabuddhe [mailto:sameer at sbuddhe.net]
> Sent: Sunday, July 10, 2016 4:06 AM
> To: Philip Reames <listmail at philipreames.com>
> Cc: Mehdi Amini <mehdi.amini at apple.com>; Liu, Yaxun (Sam)
> <Yaxun.Liu at amd.com>; Ke Bai <kebai613 at gmail.com>; Mekhanoshin, Stanislav
> <Stanislav.Mekhanoshin at amd.com>; Sumner, Brian <Brian.Sumner at amd.com>;
> llvm-dev at lists.llvm.org; Zhuravlyov, Konstantin
> <Konstantin.Zhuravlyov at amd.com>; Tye, Tony <Tony.Tye at amd.com>
> Subject: Re: [llvm-dev] Memory scope proposal
>
>
> On Mon, Jul 4, 2016 at 5:09 AM, Philip Reames via llvm-dev
> <llvm-dev at lists.llvm.org> wrote:
>
> I will comment - as one of the few people actually working on llvm's atomic
> implementation with any regularity - that I am opposed to extending the
> instructions without a strong motivating case.  I don't care anywhere near
> as much about metadata based schemes, but extending the instruction
> semantics imposes a much larger burden on the rest of the community.  That
> burden has to be well justified and supported.
>
>
> In OpenCL 2.x, two atomic operations on the same atomic object need to have
> the same scope to prevent a data race. This derives from the definition of
> "inclusive scope" in OpenCL 2.x. Encoding OpenCL 2.x scope as metadata in
> LLVM IR would be a problem because there cannot be a "safe default value" to
> be used when the metadata is dropped. If the "largest" scope is used as the
> default, then the optimizer must guarantee that the metadata is dropped from
> every atomic operation in the whole program, or not dropped at all.
>
> Hence the original attempt to extend LLVM atomic instructions with a broader
> scope field.
>
> Sameer.
>
>
>
> _______________________________________________
> LLVM Developers mailing list
> llvm-dev at lists.llvm.org
> http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev
>