-
Notifications
You must be signed in to change notification settings - Fork 512
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Update Win32 metadata #3111
Update Win32 metadata #3111
Conversation
Hm, that failure looks similar to one I've seen elsewhere. Rustup stopped modifying |
Here's the real error:
Interestingly, this function doesn't live in OLE32 as documented but is rather exported by an apiset (and ultimately now lives in COMBASE). Investigating further. |
I corrected metadata locally and regenerated libs/bindings, and tests pass. Filed an issue microsoft/win32metadata#1928 to get this fixed on the metadata side. Now to investigate why we're just now pulling that in. |
|
We added it for RAII purposes about a month ago microsoft/win32metadata@d7c12d8#diff-3aa5b50df6faebc7292a072a8060e1ec5d900ed37522da9168f3a35a9f33da7cR696 and with the linker stuffing everything under the sun into the import tables, ... 💥 So think we'll quickly fix up metadata and maybe try again soon. |
Makes total sense - thanks Rafael! |
Part of the issue is the eagerness the Rust compiler has for generating code for functions even when they're not called on the assumption that the optimizer will strip it out in the end. This is why I originally added the |
@ChrisDenton please take it for a spin and let me know if the new pointer handles is a better fit. |
I've been trying it out in a number of projects and so far it's mostly just meant no longer needing |
@ChrisDenton Maybe a dumb question--should we be using that std HANDLE type directly instead of |
We could do except that it's std only so |
This is the highest we can go right now because this change which went into 0.58 absolutely fucks our shit up: microsoft/windows-rs#3111
This is the first update to the Win32 and WDK metadata for some time so there's a fair bit of ripple.
isize
#3093 where many Win32 handle types are now represented by pointers - this has a large impact on samples 70ef5dd and tests c898d58