♟ jlebar

Google developer working on CUDA for Clang. On IRC as jlebar.

+tra in the hopes that perhaps he's comfortable reviewing this (sorry that I'm not).

I defer to the others here.

Just to check, the notion is that it's OK if I report a sm version less than what I end up running on?

I'm not confident that the GUIDS remain stable, either. I have no data to tell one way or another, though.

You don't want to pass in the GPU-deadbeef form from nvidia-smi -L so we don't have this problem?

[CUDA] Add tests that, in C++14 mode, min/max are constexpr.

[CUDA] Add tests to ensure that std::min/max can be called from __host__…

[CUDA] Add tests that, in C++14 mode, min/max are constexpr.

[CUDA] Add tests to ensure that std::min/max can be called from __host__…

[CUDA] Make __host__/__device__ min/max overloads constexpr in C++14.

[CUDA] Make min/max shims host+device.

Looks right to me (other than the missing constexpr in C++14 onwards). Though this is subtle enough that I suspect the only way to know for sure is to try it.

In D47757#1142886, @ahatanak wrote:

I mean ToT clang (without my patch applied) seems to select the non-sized host version 'T::operator delete(void*)'.

@jlebar, is the change I made to call-host-fn-from-device.cu correct?

@rsmith friendly ping on this one.

Revert "[SCEV] Use LLVM_MARK_AS_BITMASK_ENUM in SCEV." -- breaks MSVC builds.

Revert "[SCEV] Simplify some flags expressions." -- dependent revision breaks…

[SCEV] Simplify some flags expressions.

[SCEV] Use LLVM_MARK_AS_BITMASK_ENUM in SCEV.

In D48151#1133954, @rsmith wrote:

LGTM

@rsmith friendly ping on this one.

@rsmith friendly ping on this review.

[SCEV] Simplify zext/trunc idiom that appears when handling bitmasks.

[SCEV] Fix a variable name, NFC.

[SCEV] Simplify trunc-of-add/mul to add/mul-of-trunc under more circumstances.

[SCEV] Fix indentation and combine two if statements in getMulExpr, NFC.

Thank you for the reviews, Sanjoy!

Sorry, bit of a phab phail, I meant to send out this revision differently.

In D48036#1131279, @tra wrote:

Ack.

Last comment in the bug pointed out that those overloads should be constexpr in c++14. Maybe in a separate patch, though.

[SCEV] Add transform zext((A * B * ...)<nuw>) --> (zext(A) * zext(B) * ...

[SCEV] Add nuw/nsw to mul ops in StrengthenNoWrapFlags where safe.

Fix indentation in ScalarEvolution.cpp.

Thank you for the reviews!

Fix all remaining tests.

What could possibly go wrong.

I can't think of any reason why we'd drop the flags in this case.

In D47113#1106541, @bixia wrote:

InstCombiner::visitShl performs similar narrowing without checking user count and can increase the total number to ZExt instructions.

Seems reasonable to me. I dunno if this solves @bixia's problem or not, but even if it doesn't, seems reasonable...

I defer to Art on this one.

[test-suite] Enable CUDA complex tests with libc++ now that D25403 is resolved.

[test-suite] Test CUDA in C++14 mode with C++11 stdlibs.

[test-suite] Enable CUDA complex tests with libc++ now that D25403 is resolved.

[test-suite] Test CUDA in C++14 mode with C++11 stdlibs.

Thanks for the reviews, Art. Submitting with this change...

NVPTX/implicit-def.ll - this isn't testing what it intended to test, but I don't know how to do that...could just delete the file?

jlebar (Justin Lebar)
User

Projects

User Details

Recent Activity
View All

Today

Thu, Aug 9

Tue, Aug 7

Thu, Aug 2

Fri, Jul 27

Tue, Jul 24

Jul 20 2018

Jul 18 2018

Jun 29 2018

Jun 26 2018

Jun 25 2018

Jun 15 2018

Jun 14 2018

Jun 13 2018

Jun 11 2018

Jun 6 2018

Jun 5 2018

Jun 4 2018

May 23 2018

May 21 2018

May 20 2018

May 18 2018

May 17 2018

jlebar (Justin Lebar)User