Fix issues that came up with building cuDF with main #1643
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
miscco
added
libcu++
miscco
force-pushed
the
fix_cuda_pinned_memory_include
branch
from
0330422
to
aebae6f
Compare
libcudacxx/include/cuda/__memory_resource/cuda_pinned_memory_resource.h
Outdated
Show resolved
Hide resolved
miscco
force-pushed
the
fix_cuda_pinned_memory_include
branch
from
aebae6f
to
efc14e7
Compare
miscco
changed the title
cuda_pinned_memory_resource.h
miscco
force-pushed
the
fix_cuda_pinned_memory_include
branch
from
efc14e7
to
a94885d
Compare
miscco
commented
Apr 17, 2024
gevtushenko
requested changes
Apr 18, 2024
miscco
force-pushed
the
fix_cuda_pinned_memory_include
branch
3 times, most recently
from
bbb4c49
to
3814b65
Compare
miscco
commented
Apr 22, 2024
1 task
miscco
force-pushed
the
fix_cuda_pinned_memory_include
branch
from
f6bf8d5
to
ec4e698
Compare
griwes
approved these changes
Apr 29, 2024
miscco
force-pushed
the
fix_cuda_pinned_memory_include
branch
2 times, most recently
from
9226cc3
to
31a334a
Compare
miscco
force-pushed
the
fix_cuda_pinned_memory_include
branch
from
76b9784
to
be5ddf7
Compare
1 task
…tional until we get `<algorithm>`
miscco
force-pushed
the
fix_cuda_pinned_memory_include
branch
from
be5ddf7
to
bc342df
Compare
miscco
force-pushed
the
fix_cuda_pinned_memory_include
branch
from
bc342df
to
3cc958f
Compare
miscco
force-pushed
the
fix_cuda_pinned_memory_include
branch
from
3cc958f
to
fe3a98b
Compare
|
|
gevtushenko
reviewed
May 3, 2024
gevtushenko
approved these changes
May 3, 2024
bdice
approved these changes
May 3, 2024
Comment on lines
+69
to
+71
| // The maximum amount of static shared memory available per thread block | ||
| // Note that in contrast to dynamic shared memory, static shared memory is still limited to 48 KB | ||
| static constexpr std::size_t max_smem_per_block = 48 * 1024; |
There was a problem hiding this comment.
I didn't know that, interesting!
Suggestion: you may want to add a pointer to the documentation confirming that. I found this on the web: https://docs.nvidia.com/cuda/cuda-c-programming-guide/index.html#shared-memory-7-x
Question: is there a way (and if there was should) we could statically assert that this is still the case?
This was referenced May 6, 2024
bernhardmgruber
approved these changes
May 6, 2024
Closed
1 task
1 task
Closed
1 task
Comment on lines
75
to
-77
| template <class T, class U> | ||
| using pair = ::cuda::std::pair<T, U>; | ||
|
|
There was a problem hiding this comment.
I just fell over this again because of a doxygen issue and now wonder, why didn't you just pull pair into the thrust namespace without creating an alias? Like:
using _CUDA_VSTD::pair;This works with CTAD and passes the unit tests.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
We need
cuda_runtime.hfor the C++cudaMallocHostand notcuda_runtime_api.h