With these changes, exchange allocator headers are never initialized, read or written to. Removing the header will now just involve updating the code in trans using an offset to only do it if the type contained is managed.
The only thing blocking removing the initialization of the last field in the header was ~fn since it uses it to store the dynamic size/types due to captures. I temporarily switched it to a `closure_exchange_alloc` lang item (it uses the same `exchange_free`) and #7496 is filed about removing that.
Since the `exchange_free` call is now inlined all over the codebase, I don't think we should have an assert for null. It doesn't currently ever happen, but it would be fine if we started generating code that did do it. The `exchange_free` function also had a comment declaring that it must not fail, but a regular assert would cause a failure. I also removed the atomic counter because valgrind can already find these leaks, and we have valgrind bots now.
Note that exchange free does not currently print an error an out-of-memory when it aborts, because our `io` code may allocate. We could probably get away with a `#[rust_stack]` call to a `stdio` function but it would be better to make a write system call.
This stems from trying to perform as few allocations as possible throughout the standard libraries.
This specializes the `ToStr` implementation for floats/ints separately because it's known that ints will have a maximum length (whereas floats could be very very large).
I also removed a `FIXME` to remove a malloc from the `to_str()` of floats in `repr.rs` because I think that this should be addressed elsewhere. I think that we may not be able to avoid it easily because floats can have such large representations, but regardless this should be a problem with the implementation of `float_to_str_bytes_common` now and not in the `Repr` module.
* stop using an atomic counter, this has a significant cost and
valgrind will already catch these leaks
* remove the extra layer of function calls
* remove the assert of non-null in free, freeing null is well defined
but throwing a failure from free will not be
* stop initializing the `prev`/`next` pointers
* abort on out-of-memory, failing won't necessarily work
I almost got locked out of my machine because I misunderstood the purpose of the function and called it with a limit of uint::max_value, which turned this function into an almost endless loop.
Currently we pass all "self" arguments by reference, for the pointer
variants this means that we end up with double indirection which causes
a unnecessary performance hit.
The fix itself is pretty straight-forward and just means that "self"
needs to be handled like any other argument, except for by-value "self"
which still needs to be passed by reference. This is because
non-pointer types can't just be stuffed into the environment slot which
is used to pass "self".
What made things tricky is that there was also a bug in the typechecker
where the method map entries are created. For type impls, that stored
the base type instead of the actual self-type in the method map, e.g.
Foo instead of &Foo for &self. That worked with pass-by-reference, but
fails with pass-by-value which needs the real type.
Code that makes use of methods seems to be about 10% faster with this
change. Also, build times are reduced by about 4%.
Fixes#4355, #4402, #5280, #4406 and #7285
This removes usage of `&const` throughout the standard libraries/compiler, and it removes some extraneous fields in the AST now that unique boxes always inherit their mutability.
Currently we pass all "self" arguments by reference, for the pointer
variants this means that we end up with double indirection which causes
a unnecessary performance hit.
The fix itself is pretty straight-forward and just means that "self"
needs to be handled like any other argument, except for by-value "self"
which still needs to be passed by reference. This is because
non-pointer types can't just be stuffed into the environment slot which
is used to pass "self".
What made things tricky is that there was also a bug in the typechecker
where the method map entries are created. For type impls, that stored
the base type instead of the actual self-type in the method map, e.g.
Foo instead of &Foo for &self. That worked with pass-by-reference, but
fails with pass-by-value which needs the real type.
Code that makes use of methods seems to be about 10% faster with this
change. Also, build times are reduced by about 4%.
Fixes#4355, #4402, #5280, #4406 and #7285
The code that tried to revoke the cleanup for the self argument tried
to use "llself" to do so, but the cleanup might actually be registered
with a different ValueRef due to e.g. casting. Currently, this is
worked around by early revocation of the cleanup for self in
trans_self_arg.
To handle this correctly, we have to put the ValueRef for the cleanup
into the MethodData, so trans_call_inner can use it to revoke the
cleanup when it's actually supposed to.
This makes it possible for us to trigger the llvm-clean make-target by checking in a change that touches rustllvm/llvm-auto-clean-stamp. Most developers don't need to see or know about this, but when you push a change that "needs an LLVM rebuild", even if not otherwise obvious, this should give a mechanism to do it.
"self" is always passed as an opaque box, so there's no point in using
the concrete self type when translating the argument. All it does it
causing the value to be casted back to an opaque box right away.
The commit f9a5453 is meant to be a temporary hold-over. Whether or not there is added a way for the compiler to "implicitly borrow" stack closures in this way, there should be a codegen optimization that prevents having to traverse possibly-very-many function pointers to find the function you ultimately wanted to call. I tried to separate out the changes so this particular commit could be straight-up reverted if auto-borrowing happens in the future.
r? @nikomatsakis
@graydon suggested that once closures not be part of the language for 1.0, but that they might be hidden behind a -Z compile flag as an "experimental feature" in case people decide they need them.
Regardless of whether ```-Z once-fns``` is set, this PR will parse the ```once``` keyword and will prevent closures labelled with it from being called more than once. It will also permit moving out of captured vars in heap closures, just to let the runtime writers stop using ```Cell``` sooner. Setting ```-Z once-fns``` only toggles whether the move-out-from-capture privilege is also given for stack closures.
r? @nikomatsakis
