Ralf Jung
1326aed02c
rustfmt
...
Except for error.rs, the result there looks rather ugly
2017-08-10 08:48:41 -07:00
Ralf Jung
34685044f9
add a bunch of compile-fail tests for validation
2017-08-09 18:01:10 -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
bff1ad156e
integer-ops needs a rustc patch to work again
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
11f0aedc3d
add some tests making sure we get the alias checking right
2017-08-09 15:49:47 -07:00
Oliver Schneider
8ab1eeef51
Implement packed statics
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
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
Oliver Schneider
181bb30d07
Rustc reorders fields, but miri expects them in source order
2017-08-04 13:43:17 +02:00
Ralf Jung
4957031e3c
selectively disable validation in compile-fail where appropriate. cargo test now passes :)
2017-08-03 18:13:23 -07:00
Ralf Jung
9458111681
for now, disable validation on optimized tests -- inlining breaks validation
2017-08-03 18:13:23 -07:00
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
766a69f656
make test suite MIR validation aware. run-pass passes with and without fullmir, if optimizations are disabled!
2017-08-03 18:13:22 -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
2cf394955b
actually, we can use a plain constant
2017-08-03 12:58:34 -07:00
Ralf Jung
0f43321178
no need to do run-time formatting
2017-08-03 12:39:55 -07:00
Ralf Jung
de1376f923
Add a build script to inform the binary about its profile, and use that in the test suite
2017-08-03 11:38:00 -07:00
Ralf Jung
c533cf8496
fix size_of_val on unsized tuples
2017-08-02 15:29:13 -07:00
Oliver Schneider
40677a5c47
Choose release miri for release tests
2017-08-02 18:28:12 +02:00
Oliver Schneider
f8c61da070
Add a test for untagged unions
2017-08-02 12:56:07 +02:00
Oliver Schneider
adfea61665
Allow machines to create new memory kinds
2017-08-01 09:56:21 +02:00
Oliver Schneider
11e30a65e6
Split up miri into the librustc_mir and bin parts
2017-08-01 09:56:21 +02:00
Ralf Jung
4672cb7bde
make force_allocation handle packed ByValPair
2017-07-26 23:45:22 -07:00
Oliver Schneider
f906c5458c
Merge pull request #272 from oli-obk/mir-validate
...
Mir validate
2017-07-26 23:37:52 +02:00
Ralf Jung
744780e794
more hacks to make test cases pass
2017-07-25 17:31:40 -07:00
Ralf Jung
e5cc8aaf5b
Test that &* of a dangling (and even unaligned) ptr is okay
2017-07-25 17:02:57 -07:00
Ralf Jung
f960f4c537
fix casting integers to pointers
2017-07-21 20:02:48 -07:00
Oliver Schneider
4cfa7c5d12
Prep miri repository for rustc merger
2017-07-21 11:30:56 +02:00
Oliver Schneider
b4e300d577
Merge pull request #266 from RalfJung/align
...
Always test alignment in memory.rs
2017-07-21 09:46:50 +02:00
Ralf Jung
14cb85809b
always test alignment in memory.rs
2017-07-20 13:24:06 -07:00
Oliver Schneider
f2d0101065
Also test optimized MIR
2017-07-20 20:25:43 +02:00
Oliver Schneider
f822ad5c63
Merge pull request #260 from RalfJung/tests
...
Split up test suite
2017-07-20 16:07:08 +02:00
Eduard-Mihai Burtescu
e79bb42ab3
Merge pull request #262 from RalfJung/never
...
remove ad-hoc 'never' type check in read_lvalue
2017-07-20 15:13:21 +03:00
Oliver Schneider
101e5a8ba0
Move rustc tests from tests to a binary, since they are only run manually anyway
2017-07-20 13:11:40 +02:00
Ralf Jung
3b19c83c67
remove ad-hoc 'never' type check in read_lvalue
2017-07-19 20:32:51 -07:00
Ralf Jung
7648ccaae1
add test for misaligned pointer loads
2017-07-19 20:26:30 -07:00
Ralf Jung
e38ee0a30a
split up the test suite
2017-07-19 12:52:20 -07:00
Ralf Jung
e4ffab1754
enable test code that passes just fine
2017-07-19 11:38:15 -07:00
Ralf Jung
72664e42aa
No longer check aligment and non-NULLness on &
...
This breaks creating unaligned raw pointers via `&packed.field as *const _`, which needs to be legal.
Also it doesn't seem like LLVM still relies on this, see
* https://github.com/solson/miri/issues/244#issuecomment-315563640
* https://internals.rust-lang.org/t/rules-for-alignment-and-non-nullness-of-references/5430/16
We probably want to handle this invariant like the others that validation is concerned with, and only
check it on function boundaries for now.
2017-07-19 11:28:35 -07:00
Ralf Jung
ff9192e346
remove reundant dangling checks in {r,d}eallocate
2017-07-18 13:50:54 -07:00
Oliver Schneider
2e562a4d64
Fix static mutation tests
2017-07-14 17:39:19 +02:00
Oliver Schneider
7701ff2f89
Remove duplicate test
2017-07-14 17:39:19 +02:00
Oliver Schneider
192da8819f
Ensure that it is not possible to explicitly free stack memory
2017-07-14 17:36:06 +02:00
Oliver Schneider
56d4de303f
Merge pull request #247 from RalfJung/packed
...
Re-do packed memory accesses
2017-07-14 08:30:30 +02:00
Ralf Jung
0fbbcae92d
packed structs: test unsize coercions
2017-07-13 14:18:26 -07:00
Ralf Jung
287b6be5ca
track alignment also for ByRef values
2017-07-12 21:06:57 -07:00
Ralf Jung
1fe310c8ba
Memory::read_ptr has to check for relocations on the edges
2017-07-12 19:30:37 -07:00
Ralf Jung
c149c3fc6a
Re-do packed memory accesses
...
We now track in the lvalue whether what we computed is expected to be aligend or not, and then set some state in the memory system accordingly to make it (not) do alignment checks
2017-07-12 17:46:56 -07:00
