Quincey Koziol
Quincey Koziol
It looks like the locking protocol for reentrancy may require a new API context and possibly error stack for all types of reentrancy into the library (with normal API routines...
Closing to develop the concurrency design further.
I have a feeling that it's not in the type conversion macro also (i.e. the memset is masking the root cause)
Ah, I thought the generated H5E headers were included in the repo. I'll take them out.
> Ah, I thought the generated H5E headers were included in the repo. I'll take them out. Removed
Can I get some reviews on this PR? I'd like to get this change to the H5E package in, so I can work on refactoring the H5E_clear_stack() calls out of...
> My perl isn't exactly strong, but everything looks reasonable to me Super, thanks!
This could be reasonable, but why not just enhance the GDS VFD?
> > MacOS doesn't have O_DIRECT, so I commented out this line in cpp/include/kvikio/file_handle.hpp: > > // if (o_direct) { file_flags |= O_DIRECT; } > > After that, ./build.sh libkvikio...
@madsbk - One more change was necessary to get the CPU-only build of legate.io working: ```diff diff --git a/cpp/include/kvikio/shim/cufile_h_wrapper.hpp b/cpp/include/kvikio/shim/cufile_h_wrapper.hpp index 72131de..d73f04b 100644 --- a/cpp/include/kvikio/shim/cufile_h_wrapper.hpp +++ b/cpp/include/kvikio/shim/cufile_h_wrapper.hpp @@ -58,8 +58,7...