summary refs log tree commit diff
path: root/compiler/rustc_const_eval/src/interpret
AgeCommit message (Collapse)AuthorLines
2023-09-28Skip reinterning if nothing changedOli Scherer-1/+3
2023-09-28Strip `OpaqueCast` during `RevealAll`.Oli Scherer-1/+1
2023-09-26subst -> instantiatelcnr-3/+4
2023-09-23Remove GeneratorWitness and rename GeneratorWitnessMIR.Camille GILLOT-4/+1
2023-09-21Auto merge of #116010 - RalfJung:more-typed-immediates, r=oli-obkbors-175/+181
interpret: more consistently use ImmTy in operators and casts The diff in src/tools/miri/src/shims/x86/sse2.rs should hopefully suffice to explain why this is nicer. :)
2023-09-21Rollup merge of #115972 - RalfJung:const-consistency, r=oli-obkGuillaume Gomez-2/+2
rename mir::Constant -> mir::ConstOperand, mir::ConstKind -> mir::Const Also, be more consistent with the `to/eval_bits` methods... we had some that take a type and some that take a size, and then sometimes the one that takes a type is called `bits_for_ty`. Turns out that `ty::Const`/`mir::ConstKind` carry their type with them, so we don't need to even pass the type to those `eval_bits` functions at all. However this is not properly consistent yet: in `ty` we have most of the methods on `ty::Const`, but in `mir` we have them on `mir::ConstKind`. And indeed those two types are the ones that correspond to each other. So `mir::ConstantKind` should actually be renamed to `mir::Const`. But what to do with `mir::Constant`? It carries around a span, that's really more like a constant operand that appears as a MIR operand... it's more suited for `syntax.rs` than `consts.rs`, but the bigger question is, which name should it get if we want to align the `mir` and `ty` types? `ConstOperand`? `ConstOp`? `Literal`? It's not a literal but it has a field called `literal` so it would at least be consistently wrong-ish... ``@oli-obk`` any ideas?
2023-09-21rename mir::Constant -> mir::ConstOperand, mir::ConstKind -> mir::ConstRalf Jung-2/+2
2023-09-21try to avoid some layout_of callsRalf Jung-35/+34
2023-09-20stronger consistency check in ImmTy::from_immediateRalf Jung-3/+11
2023-09-20interpret: less debug-printing of typesRalf Jung-39/+29
2023-09-20interpret: more consistently use ImmTy in operators and castsRalf Jung-113/+122
2023-09-19adjust constValue::Slice to work for arbitrary slice typesRalf Jung-18/+22
2023-09-19move ConstValue into mirRalf Jung-17/+13
this way we have mir::ConstValue and ty::ValTree as reasonably parallel
2023-09-18Auto merge of #115748 - RalfJung:post-mono, r=oli-obkbors-79/+32
move required_consts check to general post-mono-check function This factors some code that is common between the interpreter and the codegen backends into shared helper functions. Also as a side-effect the interpreter now uses the same `eval` functions as everyone else to get the evaluated MIR constants. Also this is in preparation for another post-mono check that will be needed for (the current hackfix for) https://github.com/rust-lang/rust/issues/115709: ensuring that all locals are dynamically sized. I didn't expect this to change diagnostics, but it's just cycle errors that change. r? `@oli-obk`
2023-09-15make interpreter type Debug impl independent of Ty debug implRalf Jung-4/+17
2023-09-14don't point at const usage site for resolution-time errorsRalf Jung-12/+2
also share the code that emits the actual error
2023-09-14move required_consts check to general post-mono-check functionRalf Jung-72/+35
2023-09-14make it more clear which functions create fresh AllocIdRalf Jung-10/+7
2023-09-14cleanup op_to_const a bit; rename ConstValue::ByRef → IndirectRalf Jung-1/+1
2023-09-14use AllocId instead of Allocation in ConstValue::ByRefRalf Jung-8/+11
2023-09-13Auto merge of #115820 - matthiaskrgr:rollup-kyglvpu, r=matthiaskrgrbors-32/+7
Rollup of 6 pull requests Successful merges: - #115736 (Remove `verbose_generic_activity_with_arg`) - #115771 (cleanup leftovers of const_err lint) - #115798 (add helper method for finding the one non-1-ZST field) - #115812 (Merge settings.css into rustdoc.css) - #115815 (fix: return early when has tainted in mir pass) - #115816 (Disabled socketpair for Vita) r? `@ghost` `@rustbot` modify labels: rollup
2023-09-12Introduce Machine::POST_MONO_CHECKS.Camille GILLOT-4/+9
2023-09-12add helper method for finding the one non-1-ZST fieldRalf Jung-32/+7
2023-09-12Auto merge of #115705 - cjgillot:const-prop-aggregate, r=oli-obkbors-5/+6
Read from non-scalar constants and statics in dataflow const-prop DataflowConstProp is designed to handle scalar values. When MIR features an assignment from a non-scalar constant, we need to manually decompose it into the custom state space. This PR tweaks interpreter callbacks to allow reusing `eval_mir_constant` without having a stack frame to get a span from. r? `@oli-obk` cc `@jachris`
2023-09-11Use write_immediate.Camille GILLOT-2/+2
2023-09-11Remove cur_span hook.Camille GILLOT-12/+3
2023-09-11Inline callback.Camille GILLOT-0/+1
2023-09-11Return ImmTy in discriminant_for_variant.Camille GILLOT-5/+6
2023-09-11Support non-scalar constants.Camille GILLOT-3/+11
2023-09-09handle/hack for arbitrary-self dyn receiversRalf Jung-32/+44
2023-09-09implement and test ABI compatibility for transparent wrappers around NPO typesRalf Jung-14/+27
2023-09-09implement and test fn ptr ABI compatibility rulesRalf Jung-2/+7
2023-09-09give extra context to ABI mismatch errorsRalf Jung-14/+8
2023-09-09interpret: change ABI-compat test to be type-based, so the test is ↵Ralf Jung-50/+125
consistent across targets
2023-09-08Auto merge of #115372 - RalfJung:abi-assert-eq, r=davidtwcobors-66/+12
add rustc_abi(assert_eq) to test some guaranteed or at least highly expected ABI compatibility guarantees This new repr(transparent) test is super useful, it would have found https://github.com/rust-lang/rust/issues/115336 and found https://github.com/rust-lang/rust/issues/115404, https://github.com/rust-lang/rust/issues/115481, https://github.com/rust-lang/rust/issues/115509.
2023-09-08turns out Layout has some more things to worry about -- move ABI comparison ↵Ralf Jung-6/+2
into helper function like is_bool, and some special magic extra fields
2023-09-08accept some differences for rustc_abi(assert_eq), so that we can test more ↵Ralf Jung-60/+10
things to be compatible
2023-09-06miri: catch function calls where the argument is caller-invalid / the return ↵Ralf Jung-0/+7
value callee-invalid
2023-09-04interpret: make MemPlace, Place, Operand types private to the interpreterRalf Jung-217/+298
2023-08-31miri ABI check: fix handling of 1-ZST; don't accept sign differencesRalf Jung-24/+23
2023-08-30miri function ABI check: specifically look for repr(transparent)Ralf Jung-71/+102
2023-08-30interpret: make sure we accept transparent newtypes as ABI-compatibleRalf Jung-16/+36
also we were missing the case for Vector arguments, so handle those as well
2023-08-30storage_live: avoid computing the layout unless necessaryRalf Jung-14/+65
2023-08-30unify passing of sized and unsized function arguments :-)Ralf Jung-31/+35
2023-08-30a bit of meta-related cleanup on ProjectableRalf Jung-36/+32
2023-08-30move marking-locals-live out of push_stack_frame, so it happens with ↵Ralf Jung-84/+146
argument passing this entirely avoids even creating unsized locals in Immediate::Uninitialized state
2023-08-30interpret: fix projecting into an unsized field of a localRalf Jung-97/+100
new invariant: Place::Local never refers to something unsized
2023-08-30Rollup merge of #115272 - RalfJung:miri-error-print, r=saethlinMatthias Krüger-5/+28
miri/diagnostics: don't forget to print_backtrace when ICEing on unexpected errors This should fix the missing output encountered [here](https://github.com/rust-lang/rust/issues/115145#issuecomment-1694334410). r? `@saethlin`
2023-08-29const_eval and codegen: audit uses of is_zstRalf Jung-12/+17
2023-08-28Auto merge of #115182 - RalfJung:abi-compat-sign, r=b-naberbors-15/+29
miri ABI compatibility check: accept u32 and i32 If only the sign differs, then surely these types are compatible. (We do still check that `arg_ext` is the same, just in case.) Also I made it so that the ABI check must *imply* that size and alignment are the same, but it doesn't actively check that itself. With how crazy ABI constraints get, having equal size and align really shouldn't be used as a signal for anything I think...
generated by cgit-pink 1.4.1-3-g733a5 (git 2.37.1) at 2026-07-03 21:34:39 +0000