- carve dependencies looking at load commands when accelerator info isn’t there (important for iOS 13.4+ caches)
- follow upwards deps as well
- expose metadata about all images (addresses and sizes)
- emit header metadata in JSON format
- make sure libobjc is always loaded, to prevent glitches on the slide estimation logic
- blindly support rebase info v4, as it should work like v2
* Add initial version of MAINTAINERS file
This is a first effort in trying to split the review process to multiple
people that should be specialized in some particular areas of the
project. This will ensure higher review quality and more distributed
process.
* Update MAINTAINERS
* Converts to GH usernames
* Use CODEOWNERS instead of MAINTAINERS
See
https://help.github.com/en/github/creating-cloning-and-archiving-repositories/about-code-owners
* No need to mention CODEOWNERS, it's automatically used by GH
* Add yossizap on some components
* Who owns CODEOWNERS itself? :D
Co-authored-by: Anton Kochkov <xvilka@gmail.com>
* Try to run newshell tests on newshell- PRs
GH actions pull_request.branches only filter the base branch.
* Make it clearer which builds run tests and which do not
Detailed description
I've confirmed #16433 (comment) (including the nasm rax -> eax bug) and thus this pr
makes the x86_64 assembler use the MOV B8+ encoding (MOV r64, imm64) when the immediate
is between 0x80000000 and 0xffffffff for both:
Consistency with GNU assembler. It actually does seem to be the expected encoding.
This does mean that mov eax, 0xffffffff and mov rax, 0xffffffff will have very different semantics.
The string literal `"dr*"` has a length of 3 so the strncmp used should have size argument of 3 instead of 2.
PS: I am not sure if this fix is correct but it seems so. If it is not, please close this PR. However, if the fix is incorrect it seems there could be some bug as next `else if` checks for `"dr"` too.
