bjorn3
98802769a1
Rustup to rustc 1.28.0-nightly ( a3085756e 2018-05-19)
2018-06-03 13:08:51 +02:00
Oliver Schneider
601673d06f
trans -> codegen_backend
2018-06-03 13:08:51 +02:00
bjorn3
b906ce84ec
Rustup to rustc 1.28.0-nightly ( 952f344cd 2018-05-18)
2018-06-03 13:08:51 +02:00
bjorn3
574aa3bc42
Rustup to rustc 1.27.0-nightly ( ff2ac35db 2018-05-12)
2018-06-03 13:08:51 +02:00
Oliver Schneider
20c2e0bede
Test miri on mac os
2018-05-11 11:29:18 +02:00
bjorn3
8d6a893a4a
Rustup to 1.27.0-nightly ( acd3871ba 2018-05-10)
2018-05-11 09:38:13 +02:00
Oliver Schneider
27fe263669
Move back to a normal folder structure
2018-05-09 14:37:00 +02:00
Oliver Schneider
52599adf27
Miri core has moved to rustc::mir::interpret
2017-10-06 17:13:59 +02:00
king6cong
6221bbc5a7
update trans_fulfill_obligation call signature
2017-09-30 14:55:16 +08:00
Ralf Jung
a8980fd5e8
use exchange_malloc lang item for Box statements
2017-09-25 15:55:21 +02:00
Ralf Jung
1ad9709e00
get rid of ad-hoc inhabitedness test
2017-09-22 13:24:55 +02:00
Ralf Jung
8509dbbafe
validation: allow undef integers and raw pointers, as a crude work-around
2017-09-16 12:08:26 +02:00
Ralf Jung
2ea6663440
validation: check that int, float etc. are not undef
2017-09-16 11:45:56 +02:00
Oliver Schneider
ec5820c273
Use rustc's APFloat impl instead of interpreter host floats
2017-09-15 15:06:27 +02:00
Oliver Schneider
8cbfbf77f8
Remove #[linkage(foo)] statics from core miri
2017-09-15 13:03:07 +02:00
Ralf Jung
9db9308a79
Storage{Live,Dead} in generators got fixed
2017-09-15 08:58:12 +02:00
Ralf Jung
91d9b83db2
validation: remove a hack that is no longer necessary
2017-09-14 10:00:37 +02:00
Ralf Jung
59a329d4f7
use catch expression to handle the 'release undef' problem slightly better
2017-09-13 15:46:53 +02:00
Ralf Jung
bac89ae1d8
enable validation for a bunch of more tests
2017-09-13 15:46:53 +02:00
Ralf Jung
aa578de018
identify write locks by lvalues, not regions
...
This makes a new compile-fail test pass.
2017-09-13 14:37:31 +02:00
Oliver Schneider
2cf984d76e
This is unreachable code and it should never ever be reachable even in the future
2017-09-13 14:21:07 +02:00
Oliver Schneider
c47ee6b521
Rustup part 2/2
2017-09-13 13:46:54 +02:00
Ralf Jung
79993e63a0
rustup (part 1/2)
2017-09-13 12:58:25 +02:00
Ralf Jung
a1f71af5b2
restrict cur_frame visibility
2017-09-09 11:37:00 +02:00
Ralf Jung
41f228824f
fix warnings in tests; update validation whitelist
2017-09-09 11:18:02 +02:00
Ralf Jung
1591977432
clean up suspensions when function ends
2017-09-09 11:18:02 +02:00
Oliver Schneider
bc2f9259e6
rustup
2017-09-09 11:18:02 +02:00
Oliver Schneider
e9315a60e4
Rustup
2017-09-06 10:10:15 +02:00
Ralf Jung
9a273e0192
update backtrace to newly released version
2017-09-04 12:37:42 +02:00
Ralf Jung
10101dd9f4
rename variables to match new type name
2017-09-04 12:13:05 +02:00
David Renshaw
bb2bfb4353
update for upstream rename: CodeExtent -> Scope
2017-09-03 22:39:03 -04:00
Oliver Schneider
58a59e9d23
Rustup (generator support)
2017-08-30 13:16:27 +02:00
Oliver Schneider
506a5c7910
Mir optimizations treat single variant enums as structs
2017-08-29 12:26:20 +02:00
Oliver Schneider
88fc45b37c
Get some more rustc tests working
2017-08-29 11:58:22 +02:00
Oliver Schneider
206f0bd6df
Fix panic in subslice patterns of arrays ( fixes #276 )
2017-08-29 10:03:00 +02:00
Oliver Schneider
dd5b0100f3
Merge remote-tracking branch 'origin/master' into memory
2017-08-28 16:13:09 +02:00
Oliver Schneider
2f0dcfba1d
Simplify write_value_to_ptr
2017-08-28 16:06:49 +02:00
Oliver Schneider
129b914bdc
int_align has no dynamic error case
2017-08-28 15:58:58 +02:00
Oliver Schneider
e53e9b9e63
Deduplicate writing null case of struct wrapped nullable pointers
2017-08-28 15:27:50 +02:00
Oliver Schneider
ac7dfcac38
Fix #313 by correctly copying relocations when doing overlapping copies
2017-08-28 14:08:55 +02:00
Oliver Schneider
1fce886ac3
Make some things public that are required by priroda
2017-08-28 14:08:10 +02:00
Oliver Schneider
3464401576
Merge branch 'master' into memory
2017-08-26 23:13:55 +02:00
David Renshaw
8b5f22c3fd
use rustc version of fulfill_obligation now that we can
2017-08-26 13:48:59 -04:00
Ralf Jung
2307d6fc5b
Merge branch 'master' into memory
2017-08-26 13:51:13 +02:00
Ralf Jung
8ce6b06d52
Remove specialize float/bool read/write methods from memory
...
Notice that the write methods were previously unused, so if this makes us handle some things incorrectly, they probably were already broken
2017-08-25 19:33:12 +02:00
Ralf Jung
b1ca65447a
refactor memory write API to match read API
2017-08-25 19:08:04 +02:00
Ralf Jung
f036fe0d32
refactor memory read API: provide only size-based, no type-based methods
2017-08-25 16:20:13 +02:00
Ralf Jung
8539728607
memory: make sure we check non-NULL/undef even fore 0-sized accesses
2017-08-25 14:41:59 +02:00
Oliver Schneider
fad1bc8b20
Merge branch 'master' into drop
2017-08-24 23:52:36 +02:00
Oliver Schneider
19018852c3
Simplify dropping by reusing eval_fn_call
2017-08-24 16:04:50 +02:00
Oliver Schneider
110d50e7f9
Merge branch 'master' into libstd
2017-08-24 15:30:35 +02:00
Oliver Schneider
9ede346ee2
Refactoring: Couple values and types into an object when passed around together
2017-08-24 14:42:42 +02:00
Ralf Jung
e332ab9b13
Handle closures in get_field_ty
...
Fixes #304
2017-08-24 13:34:05 +02:00
Oliver Schneider
f75dd90961
Also reuse function pointer shims and drop glue from rustc
2017-08-23 17:46:36 +02:00
Oliver Schneider
7b8e2c772d
Rustup (Clone is now a builtin trait)
2017-08-23 17:24:38 +02:00
Ralf Jung
7e5d971c56
fix warnings; make test suite pass again
2017-08-12 09:45:44 -07:00
Ralf Jung
1326aed02c
rustfmt
...
Except for error.rs, the result there looks rather ugly
2017-08-10 08:48:41 -07:00
Ralf Jung
8e8c9c862c
turns out we can enable this sanity check now
2017-08-09 22:32:13 -07:00
Ralf Jung
5e018b1deb
analyzing hashmap.rs uncovered a deeper problem; disable validation there for now
2017-08-09 15:49:47 -07:00
Ralf Jung
7b5f8a36ab
try harder to preserve regions when doing inference
...
This is not complete yet, but it is enough to make unsized-tuple-impls work.
2017-08-09 15:49:47 -07:00
Ralf Jung
668491a892
Work on making validation test pass again
...
Turns out that tracking write locks by their lifetime is not precise enough,
but for now, we don't have an alternative. Also, we need to force_allocate
what we acquire or else the memory will not be in the right state.
2017-08-09 15:49:47 -07:00
Ralf Jung
de80bcbdbf
some tests for RangeMap
2017-08-09 15:49:47 -07:00
Oliver Schneider
91db25bdef
Rename destructure method to into_alloc_id_kind
2017-08-09 14:54:37 +02:00
Oliver Schneider
81f5de7f05
rename memory::Kind to memory::MemoryKind
2017-08-09 14:53:22 +02:00
Oliver Schneider
9e787fbd82
Dump all allocs when dumping locals
2017-08-08 17:09:26 +02:00
Oliver Schneider
897b563e70
clarify which function is pushing a global
2017-08-08 17:09:26 +02:00
Oliver Schneider
3ce57d1225
Remove unused function
2017-08-08 17:09:26 +02:00
Oliver Schneider
8ab1eeef51
Implement packed statics
2017-08-08 17:09:26 +02:00
Oliver Schneider
181851fc6b
Always allocate for globals
...
statics are rare and constants are copied anyway.
Reading from a constant should then yield a `ByVal` again if possible.
2017-08-08 17:09:26 +02:00
Oliver Schneider
4ba2b82f31
Split the alloc id address space into functions and normal allocs
...
instead of interleaving them as before.
The next step is to also separate out static memory into its own
address space.
2017-08-08 17:09:25 +02:00
Ralf Jung
847396e412
Re-do memory locking (again). Now with proper support for multiple independent shared borrows of the same mutable reference.
2017-08-07 12:57:32 -07:00
Oliver Schneider
e7d8037ebf
Rustup
2017-08-07 12:37:15 -07:00
Oliver Schneider
b59c0927f6
Merge pull request #283 from RalfJung/backtrace
...
Backtrace performance
2017-08-05 16:29:45 +02:00
Ralf Jung
2a1d7666ed
Merge pull request #286 from RalfJung/mir-validate
...
Update MIR validation and test it
2017-08-04 22:02:52 -07:00
Ralf Jung
fb2ed457c6
consolidate making Undef release a NOP in one place
2017-08-04 16:02:16 -07:00
Ralf Jung
ac49e7c650
only collect backtrace when RUST_BACKTRACE is set; resolve symbols lazily when printing
2017-08-04 10:55:35 -07:00
Oliver Schneider
181bb30d07
Rustc reorders fields, but miri expects them in source order
2017-08-04 13:43:17 +02:00
Ralf Jung
58a17026c9
Let -Zmir-emit-validate also control whether miri even acts on the validation commands
2017-08-03 18:13:23 -07:00
Ralf Jung
3754572b0a
Release of an Undef local is fine, and a NOP
2017-08-03 18:12:48 -07:00
Ralf Jung
85dc3e563e
make execution with full MIR work: need to whitelist Arc drop impl and whatever it calls
2017-08-03 18:12:26 -07:00
Ralf Jung
c2c05014af
with the new 'limited validation when there is unsafe', we can significantly shrink our whitelist
2017-08-03 18:12:26 -07:00
Ralf Jung
b3fa42fc0e
with Validation MIR around drop being fixed, we can remove the hack working around this
2017-08-03 18:12:24 -07:00
Ralf Jung
dd533a9ec6
Revert "disable validation code so that it all compiles against current nightly"
...
This reverts commit 791dbaf584
2017-08-03 18:10:47 -07:00
Oliver Schneider
726b027ba3
Merge pull request #280 from solson/archeology
...
Produce backtraces for miri internals
2017-08-03 23:02:29 +02:00
Ralf Jung
5d27b94b04
fix comment
2017-08-03 11:06:25 -07:00
Oliver Schneider
f7bc6ab162
Reuse the Backtrace object instead of rolling our own
2017-08-03 12:37:52 +02:00
Oliver Schneider
163821b500
Remove accidental text paste
2017-08-03 12:37:24 +02:00
Ralf Jung
c533cf8496
fix size_of_val on unsized tuples
2017-08-02 15:29:13 -07:00
Ralf Jung
8deb9387e9
step.rs: improve comment for clarity
2017-08-02 15:10:25 -07:00
Oliver Schneider
29a3c4d28c
Produce backtraces for miri internals
2017-08-02 16:59:01 +02:00
Oliver Schneider
79f38bf2cd
Process untagged unions
2017-08-01 16:12:29 +02:00
Oliver Schneider
d05784f23c
Address comments
2017-08-01 11:11:57 +02:00
Oliver Schneider
8de85808ec
Properly process ptr_op for const eval
2017-08-01 09:56:21 +02:00
Oliver Schneider
adfea61665
Allow machines to create new memory kinds
2017-08-01 09:56:21 +02:00
Oliver Schneider
f16b9e280b
Move all intrinsics out of interpret and fail CTFE on intrinsic calls
2017-08-01 09:56:21 +02:00
Oliver Schneider
df7c42bcc8
Move a method used everywhere out of intrinsic.rs
2017-08-01 09:56:21 +02:00
Oliver Schneider
e15d374dda
Fix doc text
2017-08-01 09:56:21 +02:00
Oliver Schneider
e5799a6af3
Reduce the chance of accidentally calling functions in CTFE
...
previously miri had a check for const fn and other cases that
CTFE requires. Instead the function call is completely
processed inside the machine. This allows CTFE to have full
control over what is called and miri to not have useless
CTFE-checks in normal mode.
2017-08-01 09:56:21 +02:00
Oliver Schneider
728bb878ac
Move the global_item function to the EvalContext
2017-08-01 09:56:21 +02:00
Oliver Schneider
960dca172d
Rename const_eval::Evaluator to CompileTimeFunctionEvaluator
2017-08-01 09:56:21 +02:00
