Re: [PATCH] D12221: [RFC] Introduce `__attribute__((nontemporal))`.
mzolotukhin abandoned this revision. mzolotukhin added a comment. We decided to go with an alternative way - with builtins instead of type attributes. The corresponding patch is http://reviews.llvm.org/D12313, and it's already reviewed and committed. http://reviews.llvm.org/D12221 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
Re: [PATCH] D12221: [RFC] Introduce `__attribute__((nontemporal))`.
mzolotukhin added a comment. Hi, I implemented builtin-based version in http://reviews.llvm.org/D12313 - could you please take a look? Thanks, Michael http://reviews.llvm.org/D12221 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
Re: [PATCH] D12221: [RFC] Introduce `__attribute__((nontemporal))`.
mzolotukhin added a comment.
Oh, I see. So, you meant something like this?
void foo(std::vectorfloat * __attribute__((nontemporal)) av, float * b, int
N) {
for (auto a: av) // `a` doesn't have nontemporal attribute here
for (int i = 0; i N; i++)
a[i] = b[i]+1;
}
One can easily work around it by using an explicit type here (`float *
__attribute__((nontemporal))` instead of `auto`), but I agree that disappeared
attribute might be a surprise for the user. Do you think it would be a frequent
case?
BTW, there are other type attributes, which also suffer from the same issue,
e.g. `vector_size`. What was the rationale of making them type attributes?
http://reviews.llvm.org/D12221
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
Re: [PATCH] D12221: [RFC] Introduce `__attribute__((nontemporal))`.
On Fri, Aug 21, 2015 at 12:14 PM, Michael Zolotukhin via cfe-commits
cfe-commits@lists.llvm.org wrote:
mzolotukhin added a comment.
Oh, I see. So, you meant something like this?
void foo(std::vectorfloat * __attribute__((nontemporal)) av, float *
b, int N) {
for (auto a: av) // `a` doesn't have nontemporal attribute here
for (int i = 0; i N; i++)
a[i] = b[i]+1;
}
One can easily work around it by using an explicit type here (`float *
__attribute__((nontemporal))` instead of `auto`), but I agree that
disappeared attribute might be a surprise for the user. Do you think it
would be a frequent case?
BTW, there are other type attributes, which also suffer from the same
issue, e.g. `vector_size`. What was the rationale of making them type
attributes?
vector_size produces a completely different type, with a different size,
alignment, different semantics and constraints for primitive operations,
and so on.
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
Re: [PATCH] D12221: [RFC] Introduce `__attribute__((nontemporal))`.
mzolotukhin added a comment. Thanks for the feedback everyone! I think at this point I'll try to return to builtins then. In my original patch I didn't have type overloading, so I'll need some time to add this. We can return back to type attributes later if we'd like to. And do I understand it correctly, that we are talking about target-independent builtins? http://reviews.llvm.org/D12221 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
Re: [PATCH] D12221: [RFC] Introduce `__attribute__((nontemporal))`.
majnemer added a subscriber: majnemer. majnemer added a comment. What does it mean to have the attribute applied to non-pointer types like `int __attribute__((nontemporal)) i;` ? The ACLE doesn't say but making it erroneous might make sense. Perhaps it would be good to have a semantic test which uses `__attribute__((nontemporal))`. http://reviews.llvm.org/D12221 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
Re: [PATCH] D12221: [RFC] Introduce `__attribute__((nontemporal))`.
On Thu, Aug 20, 2015 at 6:39 PM, Michael Zolotukhin via cfe-commits cfe-commits@lists.llvm.org wrote: mzolotukhin created this revision. mzolotukhin added reviewers: hfinkel, doug.gregor, t.p.northover, ab, mcrosier. mzolotukhin added a subscriber: cfe-commits. Herald added a subscriber: aemerson. Currently there is no way to generate nontemporal memory accesses for some architectures, e.g. for AArch64. In contrast to x86, it doesn't have special intrinsics for this, and the suggested solution is using such attribute (see ARM ACLE 2.0, section 13.1.6). The attribute would result in generating '!nontemporal' attribute in IR, which then will (hopefully) live through optimizations till backend, where it will be lowered to a non-temporal instruction (for AArch64 - to STNP). I have committed a couple of patches for vectorizers to preserve this attribute, and it seems that no other transformation removes it. So, is introducing a new type attribute a right approach for this problem? This seems like a property of an operation, rather than a property of a type. Have you considered adding a __builtin_nontemporal_store builtin as an alternative? Also, since I don't have much experience in front-end, I'd appreciate any help with the patch itself to get it ready to be committed. Specifically, I currently have following questions: 1) What tests should I add (examples would be appreciated)? 2) How does one implements constraints on how the attribute can be used, what should be the constraints in this case, and how to properly implement them? 3) How can I check if I covered all places where this attribute might be used in codegen? I.e. I seem to cover array-subscript and pointer-dereference expressions, which is probaly the only cases I care about, but I easily could miss something. Any other feedback is also welcome! Thanks, Michael http://reviews.llvm.org/D12221 Files: include/clang/AST/Type.h include/clang/Basic/Attr.td lib/AST/Type.cpp lib/AST/TypePrinter.cpp lib/CodeGen/CGExpr.cpp lib/CodeGen/CGValue.h lib/CodeGen/CodeGenFunction.cpp lib/CodeGen/CodeGenFunction.h lib/Sema/SemaType.cpp test/CodeGen/nontemporal.cpp ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
