Migrate x86_64-fortanix-unknown-sgx-lvirun-make test to rmake

[Oneirical]

Part of #121876 and the associated Google Summer of Code project.

The final Makefile! Every Makefile test is now claimed.

This is difficult to test due to the uncommon architecture it is specific to. I don't think it is in the CI (I didn't find it in jobs.yml, but if there is a way to test it, please do.

Locally, on Linux, it compiles and panics at the llvm_filecheck part (if I replace the x86_64-fortanix-unknown-sgx with x86_64-unknown-linux-gnu, of course), which is expected.

For this reason, the Makefile and associated script have been kept, but with a leading underscore.

[rustbot]

r? @jieyouxu

rustbot has assigned @jieyouxu.
They will have a look at your PR within the next two weeks and either review your PR or reassign to another reviewer.

Use r? to explicitly pick a reviewer

[rustbot]

This PR modifies tests/run-make/. If this PR is trying to port a Makefile
run-make test to use rmake.rs, please update the
run-make port tracking issue
so we can track our progress. You can either modify the tracking issue
directly, or you can comment on the tracking issue and link this PR.

cc @jieyouxu

[jieyouxu]

I'm going to be real with you, I have absolutely no clue about sgx. So I'm going to request help from our sgx target maintainers after I take a look and try to guess what the test is trying to test.

[workingjubilee]

This is about verifying that LLVM cross-compiles SGX code with load value injection defenses.

It should be possible to make it work after building the cross-compile target.

[workingjubilee]

This target is tier 2 so we should be building the cross-compilation artifacts every time anyways.

[workingjubilee]

So: per policy, tests don't have to pass on SGX, a tier 2 target, in order to merge this. As this is an SGX-only test, we can just merge this through.

I do know, however, that the SGX maintainers do in fact run the CI regularly and add ignore-sgx, etc. and are reasonably responsive, usually.

@jethrogb@raoulstrackx@mzohreva Can we get your review?

[raoulstrackx]

Thanks for tagging @workingjubilee ! I have a day off today, but I'll try to look at it on Monday.

[jieyouxu]

Thanks for the port, I have several questions (that we will need help from the sgx maintainers).

[jieyouxu]

Thanks for tagging @workingjubilee ! I have a day off today, but I'll try to look at it on Monday.

@raoulstrackx Thank you very much for helping out! ❤️

[rustbot]

The run-make-support library was changed

cc @jieyouxu

[raoulstrackx]

At the moment this PR breaks CI. See comments. Can you please check it? You shouldn't have to have an SGX machine as the enclaves don't need to be executed.

[raoulstrackx]

There are other issues as well. Please apply:

diff --git a/tests/run-make/x86_64-fortanix-unknown-sgx-lvi/rmake.rs b/tests/run-make/x86_64-fortanix-unknown-sgx-lvi/rmake.rs index 84c292d019d..506b9424d39 100644 --- a/tests/run-make/x86_64-fortanix-unknown-sgx-lvi/rmake.rs +++ b/tests/run-make/x86_64-fortanix-unknown-sgx-lvi/rmake.rs @@ -41,13 +41,13 @@ fn main() { check("std::io::stdio::_print::[[:alnum:]]+", "print.with_frame_pointers.checks"); - check("st_plus_one_global_asm", "rust_plus_one_global_asm.checks"); + check("rust_plus_one_global_asm", "rust_plus_one_global_asm.checks"); - check("_plus_one_c", "cc_plus_one_c.checks"); - check("_plus_one_c_asm", "cc_plus_one_c_asm.checks"); - check("_plus_one_cxx", "cc_plus_one_cxx.checks"); - check("_plus_one_cxx_asm", "cc_plus_one_cxx_asm.checks"); - check("_plus_one_asm", "cc_plus_one_asm.checks"); + check("cc_plus_one_c", "cc_plus_one_c.checks"); + check("cc_plus_one_c_asm", "cc_plus_one_c_asm.checks"); + check("cc_plus_one_cxx", "cc_plus_one_cxx.checks"); + check("cc_plus_one_cxx_asm", "cc_plus_one_cxx_asm.checks"); + check("cc_plus_one_asm", "cc_plus_one_asm.checks"); check("cmake_plus_one_c", "cmake_plus_one_c.checks"); check("cmake_plus_one_c_asm", "cmake_plus_one_c_asm.checks"); @@ -66,9 +66,11 @@ fn check(func_re: &str, mut checks: &str) { .stdout_utf8(); let re = regex::Regex::new(&format!("[[:blank:]]+{func_re}")).unwrap(); let func = re.find_iter(&dump).map(|m| m.as_str().trim()).collect::<Vec<&str>>().join(","); + assert!(!func.is_empty()); + let dump = llvm_objdump() .input("enclave/target/x86_64-fortanix-unknown-sgx/debug/enclave") - .arg(&format!("--disassemble-symbols={func}")) + .args(&["--demangle", &format!("--disassemble-symbols={func}")]) .run() .stdout_utf8(); let dump = dump.as_bytes();

[Oneirical]

There are other issues as well. Please apply:

Oh dear, I should have watched where my editor's multiple cursors were going. Thank you so much once again! Changes applied.

[raoulstrackx]

Now the tests all pass. :)

[Oneirical]

Now the tests all pass. :)

I deleted both Makefile and script.sh as they are no longer needed, let me know if you'd like to keep them.

Thank you so much for all your help!! This would have been much harder without a fortanix-rust maintainer.

[jieyouxu]

Now the tests all pass. :)

@raoulstrackx thank you so much for helping to review and test 💚

[jieyouxu]

I deleted both Makefile and script.sh as they are no longer needed, let me know if you'd like to keep them.

They're still in the change history, and now the rmake.rs version reimplements them, so we shouldn't keep them.

[jieyouxu]

Thanks, this LGTM as well. Feel free to r=me after PR CI is green.

[jieyouxu]

@bors delegate+

[bors]

✌️ @Oneirical, you can now approve this pull request!

If @jieyouxu told you to "r=me" after making some further change, please make that change, then do @bors r=@jieyouxu

[Oneirical]

@bors r=@jieyouxu

[bors]

📌 Commit e276d22 has been approved by jieyouxu

It is now in the queue for this repository.