The linker needs to know how big the heap is, otherwise, it could place
some of the global variables in the heap region.

53: Change usage example for statically allocated heap r=adamgreig a=nviennot

The linker needs to know how big the heap is, otherwise, it could place some of the global variables in the heap region.
This commit changes the example code to do that.

Co-authored-by: Nicolas Viennot <nicolas@viennot.biz>

Remove &mut and just use .as_ptr() as the function requires

54: fix: remove usage of HEAP mut ref in example r=adamgreig a=wcampbell0x2a

Remove &mut and just use .as_ptr() as the function requires

Co-authored-by: wcampbell <wcampbell1995@gmail.com>

versions <= 0.10.1 were affected by [CVE-2022-36086][].
note that with 0.10.0 a breaking API change was done which changed the
signature of `init`. this can however be avoided simply by casting back
(to avoid a breaking API change in `alloc-cortex-m` by changing the API
here as well - if wished, this should be done as a separate step).

i've also tried to instead switch to the `init_from_slice` API
(introduced in 0.9.1), however I've failed at getting this to compile
due to lifetimes (i'm sure it's somehow possible and i just missed the
obvious...), but that'd anyway have been a breaking change for
`alloc-cortex-m` and, if done, should be done in a separate step (though
it'd definitely clean the API up and make it nicer!).

[CVE-2022-36086]: GHSA-xg8p-34w2-j49j

57: update `linked_list_allocator` to `0.10.4` r=adamgreig a=rursprung

versions <= 0.10.1 were affected by [CVE-2022-36086][].
note that with 0.10.0 a breaking API change was done which changed the signature of `init`. this can however be avoided simply by casting back (to avoid a breaking API change in `alloc-cortex-m` by changing the API here as well - if wished, this should be done as a separate step).

i've also tried to instead switch to the `init_from_slice` API (introduced in 0.9.1), however I've failed at getting this to compile due to lifetimes (i'm sure it's somehow possible and i just missed the obvious...), but that'd anyway have been a breaking change for `alloc-cortex-m` and, if done, should be done in a separate step (though it'd definitely clean the API up and make it nicer!).

[CVE-2022-36086]: GHSA-xg8p-34w2-j49j

Co-authored-by: Ralph Ursprung <ralph.ursprung@gmail.com>

58: Prepare v0.4.3 r=newAM a=adamgreig

cc #57

Co-authored-by: Adam Greig <adam@adamgreig.com>