Update to track changes in Standard#3635
Conversation
Fix a few typos, but mostly complete the change from macros to functions, as per latest changes to Standard Signed-off-by: Ralph Castain <rhc@pmix.org>
|
@naughtont3 Could you please take a quick look at this? I'm getting flooded with strange Cython errors: Objecting to every call down into the library because of "...without gil". Any idea what is going on? |
Looks like the thread shifting into/out Python interpreter context needs the "Global Interpreter Lock" (GIL). I can poke at it a bit. |
| const char *key, | ||
| const void *data, | ||
| pmix_data_type_t type); | ||
| PMIX_EXPORT void PMIx_Pdata_xfer(pmix_pdata_t *dest, |
There was a problem hiding this comment.
@rhc54 - looks like a bad merge here. I think Cython issues are red herring as they occur for the PMIx_App_construct below?
| const char *key, | ||
| const void *data, | ||
| pmix_data_type_t type); | ||
| PMIX_EXPORT void PMIx_Pdata_xfer(pmix_pdata_t *dest, |
There was a problem hiding this comment.
| PMIX_EXPORT void PMIx_Pdata_xfer(pmix_pdata_t *dest, | |
| PMIX_EXPORT void PMIx_Pdata_xfer(pmix_pdata_t *dest, | |
| pmix_pdata_t *src); |
There was a problem hiding this comment.
Hmmm....thanks for spotting that! I think what I'm going to do is merge the deprecated file contents back into the primary files. Having two copies of these function declarations appears to be a recipe for problems. Because the bindings "harvest" these files, the duplicate entries are getting into the .pxi/.pxd files, and that is where the problem was hit - and why it builds everywhere else.
So I'll do that for this PR to fix the overall situation. Thanks again for chasing it down!
Signed-off-by: Ralph Castain <rhc@pmix.org>
Fix a few typos, but mostly complete the change from macros to functions, as per latest changes to Standard