Subteam report: compiler 2016-02-12

Highlights

No firm decisions were reached, but we discussed two topics of general interest:

RFCs #1477 and #1478, which discuss the intrinsics for doing atomic operations. At issue is what the behavior ought to be in the case of attempting to use the intrinsics with unsupported types -- the current behavior is an LLVM abort, which seems suboptimal.
Whether or not to support MIR plugins. Preliminary support landed in a recent PR, but it's unclear if we will ever want to support MIR plugins long-term (and it's certainly too early to offer even the faintest of stability promises).

-- nmatsakis

Issue #5016: implement "dynamic drop" semantics using flags on the stack rather than zeroing
Issue #12634: Replace various intrinsic wrappers with intrinsic reexports
Issue #15023: Better temporary lifetimes (tracking issue for RFC 66)
Issue #15287: Implement Pattern Guards with Bind-By-Move
Issue #20561: Tracking issue for Consistent no-prelude attribute (RFC 501)
Issue #27060: repr(packed) allows invalid unaligned loads

Issue #15536: floating point to floating point casts have undefined behaviour
Issue #22889: Duplicate inherent static methods can be defined in separate impl blocks
Issue #29149: Lifetime bounds in Copy impls are ignored
Issue #29859: Cyclic traits allow arbitrary traits to be synthesized
Issue #30829: Code no longer builds because of RFC 1214
Issue #31299: Bogus error in beta and nightly: recursive type has infinite size
Issue #31504: ICE with nightly in LvalyeTy<'tcx>::protection_ty
Issue #31567: Drop and leaking &mut references