Andy Fingerhut

@mbudiu-vmw In this comment https://github.com/p4lang/p4-spec/issues/1177#issuecomment-1310741880 you say: "but which is provided by the dataplane and not by the program". I don't think there is a desire for the initial value...

This topic has been discussed again recently in the 2023-Jan-30 Architecture work group meeting. If anyone has ideas for syntax that do not involve using a new direction name (perhaps...

The old report https://github.com/p4lang/p4c/issues/1932 mentioned above did result in p4c changing almost all cases of duplicate names for at least top level names to be a compile-time error. The one...

I only created one other PR which addressed removing structure initializers. My thinking was that this PR can now be updated to focus solely on any equality/inequality changes that might...

Note: If I recall correctly, this was a change made shortly before the 1.0.0 release of the spec in May 2017, and Nate and I made it while reviewing the...

If we do make this change, and if this comment: https://github.com/p4lang/p4-spec/issues/995#issuecomment-996201592 applies after this change, but not before this change, then I think it would be good to add some...

@mbudiu-vmw I'm confused. Doesn't https://github.com/p4lang/p4c/pull/2820 only make changes related to handling side effects in table key expressions? I thought this issue was about side effects in the list of actions...

For anyone curious, there is a very short P4 program for which today's p4c can increase the number of possible behaviors of an input P4 program, because of the LocalCopyPropagation...

Thanks for that link. It looks highly relevant to P4's current definition of the behavior of uninitialized variables, from what I can tell. I haven't read all the linked discussions...

"The main point of his text is that we (the LDWG) should define what "undefined" means, and not specific implementations, like Tofino or BMv2." I agree with this statement, and...