https://github.com/rust-lang/rust http://git.dreamy.place/mirrors/rust/atom?h=1.71.0 2023-06-03T16:17:46+00:00 2023-06-03T16:17:46+00:00 lcnr rust@lcnr.de 2023-05-29T15:15:48+00:00 urn:sha1:29b922d5232c32da11623518606b4285961a19f1 2023-05-27T11:38:31+00:00 Guillaume Gomez guillaume1.gomez@gmail.com 2023-05-27T11:38:31+00:00 urn:sha1:ddb5424569b1d598adf9010cece8d49968c357e6 Remove DesugaringKind::Replace. A simple boolean flag is enough. 2023-05-25T23:49:35+00:00 clubby789 jamie@hill-daniel.co.uk 2023-05-23T01:51:25+00:00 urn:sha1:f97fddab91fbf290ea5b691fe355d6f915220b6e 2023-05-25T17:40:46+00:00 Camille GILLOT gillot.camille@gmail.com 2023-05-25T17:30:23+00:00 urn:sha1:844c1cc5fec38f691a2ffb53ef3366f25cf7b02b 2023-05-17T13:41:53+00:00 Dylan DPC 99973273+Dylan-DPC@users.noreply.github.com 2023-05-17T13:41:53+00:00 urn:sha1:828caa80a942e9e923b75e4806da93c6e85f9752 Don't expect normalization to succeed in elaborate_drops Fixes https://github.com/rust-lang/rust/issues/110682 This was exposed through the changes in https://github.com/rust-lang/rust/pull/109247, which causes more things to be inlined. Inlining can happen before monomorphization, so we can't expect normalization to succeed. In the elaborate_drops analysis we currently have [this call](https://github.com/rust-lang/rust/blob/033aa092ab23ba14cdad27073c5e37ba0eddb428/compiler/rustc_mir_dataflow/src/elaborate_drops.rs#L278) to `normalize_erasing_regions`, which ICEs when normalization fails. The types are used to infer [whether the type needs a drop](https://github.com/rust-lang/rust/blob/033aa092ab23ba14cdad27073c5e37ba0eddb428/compiler/rustc_mir_dataflow/src/elaborate_drops.rs#L374), where `needs_drop` itself [uses `try_normalize_erasing_regions`](https://github.com/rust-lang/rust/blob/033aa092ab23ba14cdad27073c5e37ba0eddb428/compiler/rustc_middle/src/ty/util.rs#L1121). ~[`instance_mir`](https://doc.rust-lang.org/stable/nightly-rustc/rustc_middle/ty/context/struct.TyCtxt.html#method.instance_mir) isn't explicit about whether it expects the instances corresponding to the `InstanceDef`s to be monomorphized (though I think in all other contexts the function is used post-monomorphization), so the use of `instance_mir` in inlining doesn't necessarily seem wrong to me.~ 2023-05-10T20:54:31+00:00 bors bors@rust-lang.org 2023-05-10T20:54:31+00:00 urn:sha1:9a767b6b9e1e53c3cfc25103ed7c88d77ee65d5d Optimize dataflow-const-prop place-tracking infra Optimization opportunities found while investigating https://github.com/rust-lang/rust/pull/110719 Computing places breadth-first ensures that we create short projections before deep projections, since the former are more likely to be propagated. The most relevant is the pre-computation of flooded places. Callgrind showed `flood_*` methods and especially `preorder_preinvoke` were especially hot. This PR attempts to pre-compute the set of `ValueIndex` that `preorder_invoke` would visit. Using this information, we make some `PlaceIndex` inaccessible when they contain no `ValueIndex`, allowing to skip computations for those places. cc `@jachris` as original author 2023-05-10T16:03:52+00:00 b-naber b_naber@gmx.de 2023-05-04T21:03:57+00:00 urn:sha1:e7a2f52ba163a47e751b6e6d666b52c2acdd0949 in DropGlue. 2023-05-09T17:59:35+00:00 Camille GILLOT gillot.camille@gmail.com 2023-04-26T18:40:10+00:00 urn:sha1:38612f5ec7a464709f2dabde4edb843695ec84b2 2023-05-09T17:59:34+00:00 Camille GILLOT gillot.camille@gmail.com 2022-12-04T18:26:09+00:00 urn:sha1:34903755701a5595ee4bab3bb89de15c5469cd3e 2023-05-09T17:27:58+00:00 Camille GILLOT gillot.camille@gmail.com 2023-05-08T08:48:16+00:00 urn:sha1:ccc1da247bf3be7e71932844484847da6e35f185