Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
libunwind: Use new unwinding APIs under c18n #740
libunwind: Use new unwinding APIs under c18n #740
Changes from all commits
8112874
fd6ea94
File filter
Filter by extension
Conversations
Jump to
There are no files selected for viewing
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
What's this for?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Since the declarations for
dl_c18n_is_tramp
anddl_c18n_pop_trusted_stk
are now inlink_elf.h
, I includelink.h
to avoid having to re-declare both functions.There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Ok, but then you need to detect them in CMake, because otherwise you'll break the build on Linux or current/old versions of CheriBSD.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Also, why isn't the struct declared by system headers then?..
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Hmmm maybe it's easier to just hard-code everything here in libunwind then.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Well, test it? See if you can build against old and new CheriBSD.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
And make sure to test against CHERI-RISC-V too.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
OK. I've successfully built the latest commit with CTSRD-CHERI/cheribsd#2123 for both Morello and CHERI-RISC-V, and it also builds on an older CheriBSD installation through CMake.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
And when built for Morello with 2123 it detected c18n support?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yes, CMake detects c18n support on 2123.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Whilst using a saved register is better anyway, this justification seems undesirable, shouldn't it leave us on the same stack? Then when this function / longjmp returns it can restore the real stack that corresponds to what the compartment should have.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
We can leave the stack unchanged, but then there would be a small window between after
dl_c18n_unwind_trusted_stk
returns and the new stack is installed where the trusted stack says we're in the target compartment but the stack pointer still contains that of the source compartment. This doesn't seem desirable.Under the current implementation, after
dl_c18n_unwind_trusted_stk
returns, we are for all purposes already in the target compartment (i.e. not even in libunwind's compartment anymore). And the finalret
resumes execution of the target compartment without going through a trampoline. (Becauseunw_getcontext
is a trusted function, it has access to the raw return capability to the target compartment.)