Commit Graph

776 Commits

Author SHA1 Message Date
Ralf Jung
340267525c exported_symbols_cache: ensure we do not overwrite anything 2021-07-06 10:09:53 +02:00
Smitty
d19376985d Make work after mir-alloc-oom 2021-07-04 09:59:55 -04:00
bors
35af23b6a9 Auto merge of #1833 - hyd-dev:82261, r=RalfJung
Filter out items other than non-generic functions and statics in our version of `exported_symbols`

[`#[no_mangle]` on a `use` item](https://docs.rs/brotli-decompressor/2.3.1/src/brotli_decompressor/ffi/mod.rs.html#3-5) can make Miri ICE when compiling a dependency (rust-lang/rust#86261):
```rs
#[no_mangle]
use std::{thread,panic, io, boxed, any, string};
```

<details>

```
error: internal compiler error: compiler/rustc_middle/src/ty/mod.rs:1650:13: item_name: no name for DefPath { data: [DisambiguatedDefPathData { data: Misc, disambiguator: 14 }], krate: crate0 }

thread 'rustc' panicked at 'Box<dyn Any>', compiler/rustc_errors/src/lib.rs:1007:9
stack backtrace:
   0: std::panicking::begin_panic
   1: std::panic::panic_any
   2: rustc_errors::HandlerInner::bug
   3: rustc_errors::Handler::bug
   4: rustc_middle::ty::context::tls::with_opt
   5: rustc_middle::util:🐛:opt_span_bug_fmt
   6: rustc_middle::util:🐛:bug_fmt
   7: rustc_middle::ty::<impl rustc_middle::ty::context::TyCtxt>::item_name
   8: rustc_symbol_mangling::symbol_name_provider
   9: rustc_query_impl::<impl rustc_query_system::query::config::QueryAccessors<rustc_query_impl::plumbing::QueryCtxt> for rustc_query_impl::queries::symbol_name>::compute
  10: rustc_query_system::query::plumbing::get_query_impl
  11: <rustc_query_impl::Queries as rustc_middle::ty::query::QueryEngine>::symbol_name
  12: rustc_middle::middle::exported_symbols::ExportedSymbol::symbol_name_for_local_instance
  13: rustc_codegen_ssa:🔙:symbol_export::symbol_name_for_instance_in_crate
  14: rustc_codegen_ssa:🔙:linker::exported_symbols
  15: <core::iter::adapters::map::Map<I,F> as core::iter::traits::iterator::Iterator>::fold
  16: rustc_codegen_ssa:🔙:linker::LinkerInfo::new
  17: rustc_codegen_ssa:🔙:write::start_async_codegen
  18: <rustc_codegen_llvm::LlvmCodegenBackend as rustc_codegen_ssa::traits::backend::CodegenBackend>::codegen_crate
  19: rustc_interface::passes::QueryContext::enter
  20: rustc_interface::queries::Queries::ongoing_codegen
  21: rustc_interface::queries::<impl rustc_interface::interface::Compiler>::enter
  22: rustc_span::with_source_map
  23: rustc_interface::interface::create_compiler_and_run
  24: rustc_span::with_session_globals
note: Some details are omitted, run with `RUST_BACKTRACE=full` for a verbose backtrace.

note: the compiler unexpectedly panicked. this is a bug.

note: we would appreciate a bug report: https://github.com/rust-lang/rust/issues/new?labels=C-bug%2C+I-ICE%2C+T-compiler&template=ice.md

note: rustc 1.54.0-nightly (a50d72158 2021-06-08) running on x86_64-unknown-linux-gnu

note: compiler flags: -C embed-bitcode=no -C debuginfo=1 --crate-type lib

note: some of the compiler flags provided by cargo are hidden

query stack during panic:
#0 [symbol_name] computing the symbol for `{misc#14}`
end of query stack
```
</details>

This might be because in #1776, we override the `exported_symbols` query, and our version of `exported_symbols` can return a `use` item which don't have a name if the `use` item is tagged with `#[no_mangle]`, and then:
- `rustc_codegen_ssa:🔙:symbol_export::symbol_name_for_instance_in_crate` is called for for every `exported_symbols`: fb3ea63d9b/compiler/rustc_codegen_ssa/src/back/linker.rs (L1300-L1304)
- it calls `rustc_middle::middle::exported_symbols::ExportedSymbol::symbol_name_for_local_instance`: fb3ea63d9b/compiler/rustc_codegen_ssa/src/back/symbol_export.rs (L412)
- which calls `rustc_symbol_mangling::symbol_name_provider`: fb3ea63d9b/compiler/rustc_middle/src/middle/exported_symbols.rs (L37-L44)
- which calls `item_name`: fb3ea63d9b/compiler/rustc_symbol_mangling/src/lib.rs (L216), which triggers the ICE

It might also be problematic for d39f0c64b8/src/shims/foreign_items.rs (L165) which also uses `item_name`, but Miri cannot compile the dependency, so that code can't be reached.

Therefore, this PR makes `exported_symbols` filter out all items that are not functions or statics, so all items returned will have a name, which avoids the ICE (I have tested it in the https://github.com/jorgecarleitao/arrow2 repository).
(This PR also includes a commit that fixes a small (unrelated) bug for `#[no_mangle]` on associated functions -- I found that because I notice `#[no_mangle]` is supported on associated functions and they should not be filtered out in `exported_symbols`.)

Fixes (when the submodule is bumped) rust-lang/rust#86261.
2021-06-15 10:07:24 +00:00
bors
486b5dfe74 Auto merge of #1832 - hyd-dev:1776-follow-up, r=RalfJung
Report an error if a `#[no_mangle]`/`#[export_name = ...]` function has the same symbol name as a built-in shim

Implements https://github.com/rust-lang/miri/pull/1776#issuecomment-821322605.

The error looks like this:
```
error: found `malloc` symbol definition that clashes with a built-in shim
  --> tests/compile-fail/function_calls/exported_symbol_shim_clashing.rs:12:9
   |
12 |         malloc(0);
   |         ^^^^^^^^^ found `malloc` symbol definition that clashes with a built-in shim
   |
help: the `malloc` symbol is defined here

  --> tests/compile-fail/function_calls/exported_symbol_shim_clashing.rs:2:1
   |
2  | / extern "C" fn malloc(_: usize) -> *mut std::ffi::c_void {
3  | |     //~^ HELP the `malloc` symbol is defined here
4  | |     unreachable!()
5  | | }
   | |_^
   = note: inside `main` at tests/compile-fail/function_calls/exported_symbol_shim_clashing.rs:12:9
```

This does not implement "better error messages than we do currently for arg/ABI mismatches" in https://github.com/rust-lang/miri/pull/1776#issuecomment-821343175 -- I failed to remove all `check_arg_count()` and `check_abi()` (they are still used in `src/shims/intrinsics.rs` and `call_dlsym()`) and they don't receive the name of the shim.
2021-06-15 09:53:05 +00:00
hyd-dev
aaaa142dc1
Rename all link_name_sym to link_name and remove the only remaining let link_name = link_name_sym.as_str() 2021-06-15 01:24:09 +08:00
hyd-dev
9011524454
Remove strip_linker_suffix 2021-06-15 01:16:38 +08:00
hyd-dev
a67a65359f
Only pass Symbol to emulate_foreign_item_by_name 2021-06-15 00:43:15 +08:00
hyd-dev
34603e586f
Add whitespace 2021-06-14 23:38:15 +08:00
hyd-dev
d1e72d0854
Check argument count for CreateThread 2021-06-14 23:01:06 +08:00
hyd-dev
89c722ac32
Add some comments about check_shim 2021-06-14 22:53:17 +08:00
hyd-dev
da2ed6f768
Don't report UB for #[no_mangle] on associated functions 2021-06-14 20:58:54 +08:00
hyd-dev
49a8f002a0
let-bind exp_abi of "exit" | "ExitProcess" 2021-06-11 15:53:58 +08:00
hyd-dev
e46aab5816
Use check_shim() for abort 2021-06-11 15:47:12 +08:00
hyd-dev
c822ec59aa
Implement cache for not found symbols 2021-06-11 15:24:51 +08:00
hyd-dev
ce7040075a
Disallow #[no_mangle]/#[export_name = ...] functions that have the same symbol name as built-in shims 2021-06-11 15:18:47 +08:00
hyd-dev
4f3718ef85
Fix the wrong EmulateByNameResult::NotSupported in syscall shim 2021-06-11 11:44:57 +08:00
bors
e5c3af6f51 Auto merge of #1818 - landaire:feature/panic-on-unsupported-syscalls, r=RalfJung
Add support for panicking in the emulated application when unsupported functionality is encountered

This PR fixes #1807 and allows an optional flag to be specified to panic when an unsupported syscall is encountered. In essence, instead of bubbling up an error in the context of the Miri application Miri will panic within the context of the *emulated* application. This feature is desired to allow CI pipelines to determine if a Miri failure is unsupported functionality or actual UB. Please read [this comment](https://github.com/rust-lang/miri/issues/1807#issuecomment-845425076) for the rationale behind this change.

Note: this change does not cover all cases where unsupported functionality errors may be raised. If you search the repo for `throw_unsup_format!` there are many cases that I think are less likely to occur and may still be problematic for some folks.

TODO:

- [x] README documentation on this new flag
- [x] Add tests
2021-06-09 18:09:06 +00:00
Ralf Jung
a38f02c44c isolated operations return EPERM; tweak isolation hint 2021-06-09 18:24:59 +02:00
Smit Soni
ba64f485c8 Fix parameter of io error helper function
`set_last_error_from_io_error` works with only the error kind, and
discards the payload. Fix its signature to make it explicit.
2021-06-09 06:28:35 -07:00
Smit Soni
892f706ce5 Add a support to execute isolated op without halting
In user interface, added a new flag `-Zmiri-isolation-error` which
takes one of the four values -- hide, warn, warn-nobacktrace, and
abort. This option can be used to configure Miri to either abort or
return an error code upon executing isolated op. If not aborted, Miri
prints a warning, whose verbosity can be configured using this flag.

In implementation, added a new enum `IsolatedOp` to capture all the
settings related to ops requiring communication with the
host. Old `communicate` flag in both miri configs and machine
stats is replaced with a new helper function `communicate()` which
checks `isolated_op` internally.

Added a new helper function `reject_in_isolation` which can be called
by shims to reject ops according to the reject_with settings. Use miri
specific diagnostics function `report_msg` to print backtrace in the
warning. Update it to take an enum value instead of a bool, indicating
the level of diagnostics.

Updated shims related to current dir to use the new APIs. Added a new
test for current dir ops in isolation without halting machine.
2021-06-09 05:50:52 -07:00
Lander Brandt
ae237098f8 Add support for panicking in the emulated application when unsupported syscalls are encountered 2021-06-07 15:21:53 -07:00
hyd-dev
3871c493b2
in_std -> frame_in_std 2021-06-06 11:21:20 +08:00
hyd-dev
879000b133
Detect std by checking if the crate defines #[lang = "start"] rather than string comparison 2021-06-06 11:12:05 +08:00
hyd-dev
e026ad584d
Use unwrap_or_else() 2021-05-31 11:57:15 +08:00
hyd-dev
b054a19f94
We don't need to check SymbolExportLevel 2021-05-31 11:57:14 +08:00
hyd-dev
382295cd47
Move check_abi() into EvalContextExt 2021-05-31 11:15:31 +08:00
hyd-dev
41f33a64f8
Implement calls to exported symbols 2021-05-31 11:15:29 +08:00
hyd-dev
71f4140550
Add -Zmiri-disable-abi-check 2021-05-31 11:14:23 +08:00
hyd-dev
9b2d42587f
unwind is no longer Option<BasicBlock> 2021-05-28 22:08:51 +08:00
Ralf Jung
c60efa0c69 allocate backtrace strings mutably 2021-05-23 18:04:13 +02:00
Ralf Jung
e09c571eec avoid some borrow_mut calls in data_race 2021-05-23 18:04:13 +02:00
Ralf Jung
aba96b82b4 fix write_os_str_to_wide_str 2021-05-19 16:34:14 +02:00
Ralf Jung
e4a27150cb fmt 2021-05-19 16:10:28 +02:00
Ralf Jung
801a1744cd update for Memory API changes 2021-05-19 15:27:58 +02:00
Ralf Jung
4f171d7fe3 stop relying on c_str/wide_str helpers in rustc 2021-05-17 14:31:59 +02:00
Ralf Jung
78f74c69e3 fix rustfmt fallout 2021-05-17 10:57:21 +02:00
Ralf Jung
d6fdfaa047 hand-held formatting for remaining files 2021-05-16 11:51:39 +02:00
Ralf Jung
4e231bab5e format much of Miri 2021-05-16 11:48:28 +02:00
Andy Wang
0d2278c6c6
Prefer remapped filename in backtrace to match rustc behaviour 2021-05-12 15:49:33 +01:00
bors
bdce1ddf88 Auto merge of #1785 - Smittyvb:fast-math-ub, r=RalfJung
Throw UB if f*_fast intrinsic called with non-finite value

Calling these intrinsics with non-finite values is undefined behaviour, since they result in `f*` intrinsics in LLVM with the `fast` flag, and `fast` math on non-finite values results in `poison` values. (technically LLVM only considers it UB upon _using_ the value, but that shouldn't make much of a difference)
2021-05-03 17:32:50 +00:00
Smitty
e0e59f6020 Simplify finiteness checking 2021-05-03 10:08:42 -04:00
Ralf Jung
c3ae8a6f90 adjust for different 'yield' hint on aarch64 2021-05-03 11:25:05 +02:00
Smitty
e591b83185 UB if f*_fast intrinsic called with nonfinite value 2021-05-02 12:25:00 -04:00
Ralf Jung
20e31dbdad fix newer getrandom on Windows 2021-04-04 11:41:04 +02:00
bstrie
4f899ce9ac Replace deprecated collections::Bound 2021-03-17 18:34:44 -04:00
hyd-dev
7ec919daa4
Remove the macro and expand it manually 2021-03-17 22:27:57 +08:00
hyd-dev
bbc348539b
Unsupported foreign non-"C"/"system"-ABI function calls are not UB 2021-03-17 21:20:04 +08:00
bors
0788188c7b Auto merge of #1743 - RalfJung:callee-checks, r=RalfJung
Check callee ABI when Miri calls closures

Fixes https://github.com/rust-lang/miri/issues/1741
2021-03-14 18:05:18 +00:00
Ralf Jung
15465a5881 don't ICE when we unwind despite panic=abort 2021-03-14 17:10:45 +01:00
Ralf Jung
893843fd45 when Miri calls a function ptr, make sure it has the right ABI 2021-03-14 15:30:37 +01:00
Ralf Jung
98f28ac9c8 rustup 2021-03-11 09:07:05 +01:00
Mara Bos
5bd5ea21ea Remove unwrap_none as it won't be stabilized.
This upgrades to the latest rustc to be able to use try_insert()
instead.
2021-03-08 16:59:20 +01:00
Smittyvb
8d43d727f4
Check arg count
Co-authored-by: Ralf Jung <post@ralfj.de>
2021-03-05 07:16:50 -05:00
Smitty
26e97ec54f Support breakpoint intrinsic 2021-03-04 07:48:28 -05:00
Ralf Jung
9f3dbad5d0 rename atomic_op_min_max -> atomic_op 2021-03-02 19:54:45 +01:00
Henry Boisdequin
f8440d6c99 Add in atomic_{min,max}_x intrinsics
Co-authored-by: Greg Bowyer <gbowyer@fastmail.co.uk>
2021-03-02 20:17:05 +05:30
Henry Boisdequin
7d1531f3c1 fmt data_race.rs and intrinsics.rs 2021-03-02 18:06:12 +05:30
Tomasz Miąsko
0eb341417c rustup to e7c23ab933 2021-02-20 00:00:00 +00:00
Ralf Jung
052cd3bff7 rustup; remove some no-longer-needed Windows shims 2021-01-31 14:23:49 +01:00
bors
b38dc837f2 Auto merge of #1670 - unseddd:abi, r=RalfJung
Add ABI check for shims

Resolves #1631
2021-01-29 19:36:30 +00:00
Yuki Okushi
53612ece7c Rustup for rust-lang/rust#79951 2021-01-28 22:31:22 +09:00
Nym Seddon
de4eea9d49
Add ABI check for shims 2021-01-27 22:32:25 +00:00
Thom Chiovoloni
c678bd722e Add random failures to compare_exchange_weak 2021-01-26 00:13:54 -08:00
Ralf Jung
13dd513254 adjust Windows shims for stdlib changes (Windows XP removal) 2021-01-25 10:43:48 +01:00
Ralf Jung
7d8f8c405f macos: support aarch64 link names 2021-01-23 18:14:57 +01:00
Ralf Jung
848be1bf82 implement aarch64 hint 'intrinsic' 2021-01-23 18:11:01 +01:00
Corey Farwell
2aedbf0993 Add shim for libc::sysconf(libc::_SC_NPROCESSORS_CONF) 2021-01-22 09:16:27 -05:00
Ralf Jung
3d01ba11c0 rustup; remove some intrinsics that are gone or moved to rustc 2021-01-18 12:55:30 +01:00
Ralf Jung
65f5c27d61 remove intrinsic that is now implemented in the rustc side 2020-12-22 12:47:59 +01:00
Ralf Jung
7bbd6bca77 rustup 2020-12-10 19:53:45 +01:00
bors
d473242ecd Auto merge of #1617 - JCTyblaidd:data_race_detector, r=RalfJung
Add simple data-race detector

Partially fixes data-race detection, see #1372, based on Dynamic Race Detection for C++11

- This does not explore weak memory behaviour, only exploring one sequentially consistent ordering.
- Data-race detection is only enabled after the first thread is created, so should have minimal overhead for non-concurrent execution.
- ~~Does not attempt to re-use thread id's so creating and joining threads lots of time in an execution will result in the vector clocks growing in size and slowing down program execution~~ It does now
2020-11-29 18:44:28 +00:00
JCTyBlaidd
6c5722933e Fix typos - looked into the papers handling of timestamps, after looking into it again, it seems the paper only increments the timestamp after release operations, so changed to approximation of that implementation. 2020-11-28 17:17:07 +00:00
JCTyBlaidd
55fc552d99 Apply review changes, incrementing the clocks twice is an unnecessary hold-over from earlier versions so fixed. 2020-11-27 19:26:06 +00:00
JCTyBlaidd
3268f56a97 Fix review changes 2020-11-22 17:28:12 +00:00
Ralf Jung
697f6e36bd rustup 2020-11-20 14:07:52 +01:00
Ralf Jung
517728bf97 avoid fallback logic (and we do not need the flag value currently anyway) 2020-11-19 09:36:33 +01:00
Mara Bos
cdb7adb4b3 Make weak syscalls in std work.
std now looks up `getrandom` and `statx` with `dlsym` before attempting
to use `syscall(SYS_.., ..)`.

It also now passes all arguments as a machine-sized word, instead of
their original types.
2020-11-19 09:34:46 +01:00
JCTyBlaidd
69fb6413dd Tidy up comments and function layout, should fix most of the review notes. 2020-11-15 18:30:26 +00:00
Ralf Jung
43673d9a87 rustup 2020-11-11 10:29:10 +01:00
Ralf Jung
f31f00d5af rustup 2020-11-05 17:09:44 +01:00
JCTyBlaidd
9cb6b8da3f Split out vector_clock to separate file, general tidy up of some of the
code & add support for vector index re-use for multiple threads
 after termination.
2020-11-05 03:54:39 +00:00
JCTyBlaidd
95c99b2044 Detect races between atomic and non-atomic accesses of a variable,
previously only data races between two non-atomic accesses were
 detected.
2020-11-04 21:35:48 +00:00
JCTyBlaidd
ed32b260f9 Add tests, and fix bug in atomic RMW relaxed stores 2020-11-02 02:57:06 +00:00
JCTyBlaidd
89814f1b3f Initial data-race detector,
passes all current tests but additional tests are required
2020-11-02 02:57:05 +00:00
est31
1ae157bc9e Replace target.target with target
Rustc removed the target wrapper and exposed the target directly.

Result of running:

find . -type f -exec sed -i -e 's/target\.target\([)\.,;]\)/target\1/g' {} \;

Plus one manual edit of the rust-version file
2020-10-15 21:18:23 +02:00
Aaron Hill
c889eba4b2
Add comment about deprecation 2020-10-11 17:57:49 -04:00
Aaron Hill
6a6767fa2a
Apply suggestions from code review
Co-authored-by: Ralf Jung <post@ralfj.de>
2020-10-11 13:40:00 -04:00
Aaron Hill
be51e6bd07
Add an fn_ptr field to MiriFrame
The `backtrace-rs` crate can use this to implement
`Frame::symbol_address`, which is used to skip frames
above the call to `Backtrace::capture` on the stack.

The function pointer will not be useable for comparison purposes if the
function is generic, as CTFE creates a new function pointer for each
cast of a (monomorphized) generic function. However, this already
affects code running under Miri, and isn't a problem for `backtrace-rs`
(which only casts a non-generic function).

I've added logic to allow `MiriFrame` to have either 4 or 5 fields - if
a 5th field is present, we write the function pointer to it.
2020-10-10 13:07:57 -04:00
Aaron Hill
8e8828259a
Use macro callsite spans in backtrace
This mirrors what we do in the debuginfo used for runtime backtraces.
2020-10-05 17:29:55 -04:00
Ralf Jung
3aaab3dd98
more punctuation 2020-10-04 23:07:19 +02:00
Aaron Hill
c1c82c2503
Properly capitalize PathConversion 2020-10-04 16:28:09 -04:00
Aaron Hill
9e6320f101
Move convert_path_separator to trait and use it in readlink 2020-10-04 12:34:13 -04:00
Aaron Hill
462f58298a
Make helper functions freestanding 2020-10-04 12:06:51 -04:00
Aaron Hill
8b89c7e1da
Use panicking coversions instead of as 2020-10-04 12:06:51 -04:00
Aaron Hill
186d59a3ec
Move some helper functions around 2020-10-04 12:06:51 -04:00
Aaron Hill
78bc89b4fc
Implement readlink
Due to the truncating behavior of `readlink`, I was not able to
directly use any of the existing C-cstring helper functions.
2020-10-04 12:06:51 -04:00
Ralf Jung
2b2a3a0cc1 check that all syscall arguments are scalars 2020-10-03 16:01:53 +02:00
Ralf Jung
b350c80a31 add backtics back in isolation error message 2020-10-03 15:25:30 +02:00
Mara Bos
68776d2921
Add FIXME about type of addr in futex implementation.
Co-authored-by: Ralf Jung <post@ralfj.de>
2020-10-03 14:32:30 +02:00
Mara Bos
c268ee2bcb
Add note about use of force_ptr in futex implementation.
Co-authored-by: Ralf Jung <post@ralfj.de>
2020-10-03 14:21:37 +02:00