CPI first pass impl #8
Conversation
| &sol_ix as *const _ as *const _, | ||
| &sol_account_infos as *const _ as *const _, | ||
| sol_account_infos_len, | ||
| signers as *const _ as *const _, |
There was a problem hiding this comment.
We can't pass the signers rust references directly to the C code, since it expects the length of each individual seed unfortunately. I started adding signer "types" on this PR: https://github.com/febo/pinocchio/pull/11/files#diff-c14b0096da794786d77f37e2667c1f5cf800e7bf7875f8e99059e14857a03cf4R107-R127
There was a problem hiding this comment.
| } | ||
| } | ||
|
|
||
| pub fn invoke_signed<'a, const MAX_ACCOUNTS_INCL_PROGRAM: usize>( |
There was a problem hiding this comment.
I think we need to add the borrow checks: https://github.com/anza-xyz/agave/blob/master/sdk/program/src/program.rs#L255-L269
We could have an unsafe *_unchecked variation without the checks.
There was a problem hiding this comment.
btw do you know why the https://github.com/anza-xyz/agave/blob/master/sdk/program/src/program.rs#L255-L269 is doing this n^2 thing of looking for the corresponding AccountInfo for a given AccountMeta? Aren't both the AccountInfo and AccountMeta array supposed to be in the same order?
There was a problem hiding this comment.
They are not required to be in the same order.
|
@billythedummy What do you think if we use the types from #11? We could merge #11 and then rebase this PR. |
Sure thing, please go ahead with any changes you're planning to make and feel free to merge or adapt all my open PRs; unfortunately I won't have a lot of time to dedicate to this until at least the end of breakpoint, apologies |
|
Closing this as the changes suggested on this PR were added to #11 – thank you for your contribution @billythedummy. |
Tested out with some actual programs CPI-ing system program.
Feel free to critique design