Aaron Hill [Sat, 10 Oct 2020 17:07:57 +0000 (13:07 -0400)]
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.
bors [Tue, 6 Oct 2020 11:36:22 +0000 (11:36 +0000)]
Auto merge of #1576 - RalfJung:gha, r=RalfJung
add bors notification jobs
There are probably ways to simplify this... I just copied this from the Rust repo and expanded the anchors and adjusted the `needs` thing. Hopefully this unblocks Miri PR landing.
bors [Sun, 4 Oct 2020 13:14:43 +0000 (13:14 +0000)]
Auto merge of #1563 - lzutao:dummy-actions, r=RalfJung
Add a dummy actions template to enable it on CI
I already had a complete template to transition from Travis and Appveyor.
Since actions will not run on repo haven't enabled it before.
I have this dummy template first to enable it on default branch.
We should address a few todo here, merge it and I will send another patch
to complete the migration.
bors [Sat, 3 Oct 2020 15:49:15 +0000 (15:49 +0000)]
Auto merge of #1570 - RalfJung:syscalls, r=RalfJung
check that all syscall arguments are scalars
`@m-ou-se` do you think this check makes sense?
The `abi` of a layout contains everything needed for the calling convention (as far as I know), so this should ensure that all the ignored arguments are passed like integers and do not otherwise mess up argument passing.
The one thing I am not sure about is what happens when more arguments are passed than fit in registers. The `syscall` man page says that all calling conventions support at least 6 arguments, except for "mips/o32" where it says that
```
[1] The mips/o32 system call convention passes arguments 5 through 8 on the user stack.
```
If we assume that passing these extra arguments to futex is legal even with that calling convention, that would mean that those user stack arguments are just silently ignored as well, which seems plausible to me -- but I know very little about calling conventions.
Auto merge of #1559 - Aaron1011:new-miri-backtrace, r=RalfJung
Add API for capturing backtrace
This PR adds two new Miri-defined extern functions:
`miri_get_backtrace` and `miri_resolve_frame`, which are documented in
the README. Together, they allow obtaining a backtrace for the currently
executing program.
I've added a test showing how these APIs are used. I've also prepared a
companion PR `backtrace-rs`, which will allow
`backtrace::Backtrace::new()` to work automatically under Miri.
Once these two PRs are merged, we will be able to print backtraces from
the normal Rust panic hook (since libstd is now using backtrace-rs).
A few notes:
* Resolving the backtrace frames is *very* slow - you can actually see
each line being printed out one at a time. Some local testing showed
that this is not (primrary) caused by resolving a `Span` - it seems
to be just Miri being slow.
* For the first time, we now interact directly with a user-defined
struct (instead of just executing the user-provided MIR that
manipulates the struct). To allow for future changes, I've added
a 'version' parameter (currently required to be 0). This should allow
us to change the `MiriFrame` struct should the need ever arise.
* I used the approach suggested by `@oli-obk` - a returned backtrace
pointer consists of a base function allocation, with the 'offset'
used to encode the `Span.lo`. This allows losslessly reconstructing
the location information in `miri_resolve_frame`.
* There are a few quirks on the `backtrace-rs` side:
* `backtrace-rs` calls `getcwd()` by default to try to simplify
the filename. This results in an isolation error by default,
which could be annoying when printing a backtrace from libstd.
* `backtrace-rs` tries to remove 'internal' frames (everything between
the call to `Backtrace::new()` and the internal API call made by
backtrace-rs) by comparing the returned frame pointer value to
a Rust function pointer. This doesn't work due to the way we
construct the frame pointers passed to the caller. We could
attempt to support this kind of comparison, or just add a
`#[cfg(miri)]` and ignore the frames ourselves.
Aaron Hill [Tue, 3 Dec 2019 02:15:58 +0000 (21:15 -0500)]
Add API for capturing backtrace
This PR adds two new Miri-defined extern functions:
`miri_get_backtrace` and `miri_resolve_frame`, which are documented in
the README. Together, they allow obtaining a backtrace for the currently
executing program.
I've added a test showing how these APIs are used. I've also prepared a
companion PR `backtrace-rs`, which will allow
`backtrace::Backtrace::new()` to work automatically under Miri.
Once these two PRs are merged, we will be able to print backtraces from
the normal Rust panic hook (since libstd is now using backtrace-rs).
A few notes:
* Resolving the backtrace frames is *very* slow - you can actually see
each line being printed out one at a time. Some local testing showed
that this is not (primrary) caused by resolving a `Span` - it seems
to be just Miri being slow.
* For the first time, we now interact directly with a user-defined
struct (instead of just executing the user-provided MIR that
manipulates the struct). To allow for future changes, I've added
a 'version' parameter (currently required to be 0). This should allow
us to change the `MiriFrame` struct should the need ever arise.
* I used the approach suggested by @oli-obk - a returned backtrace
pointer consists of a base function allocation, with the 'offset'
used to encode the `Span.lo`. This allows losslessly reconstructing
the location information in `miri_resolve_frame`.
* There are a few quirks on the `backtrace-rs` side:
* `backtrace-rs` calls `getcwd()` by default to try to simplify
the filename. This results in an isolation error by default,
which could be annoying when printing a backtrace from libstd.
* `backtrace-rs` tries to remove 'internal' frames (everything between
the call to `Backtrace::new()` and the internal API call made by
backtrace-rs) by comparing the returned frame pointer value to
a Rust function pointer. This doesn't work due to the way we
construct the frame pointers passed to the caller. We could
attempt to support this kind of comparison, or just add a
`#[cfg(miri)]` and ignore the frames ourselves.
Auto merge of #1549 - RalfJung:panic-abort, r=oli-obk
support panic=abort
This adds support for abort-on-panic (https://github.com/rust-lang/miri/issues/1058). To achieve this, we insert `cargo-miri` as `RUSTC` when building the standard library, and patch the rustc flags in a way similar to what bootstrap does.
However, this is currently not supported on Windows as the Windows code uses inline assembly to cause an abort (?!?). I'll submit a rustc PR with some `cffg(miri)` to make that work. (EDIT: that would be https://github.com/rust-lang/rust/pull/76871)