add safe compilation options

[lxy19980601]

Add two options when building rustc : strip and stack protector.
If set strip = true, rustc will be stripped of symbols using -Cstrip=symbols.
Also can set stack-protector and then rustc will be compiled with stack protectors.

[rustbot]

Thanks for the pull request, and welcome! The Rust team is excited to review your changes, and you should hear from @onur-ozkan (or someone else) soon.

Please see the contribution instructions for more information. Namely, in order to ensure the minimum review times lag, PR authors and assigned reviewers should ensure that the review label (S-waiting-on-review and S-waiting-on-author) stays updated, invoking these commands when appropriate:

• @rustbot author: the review is finished, PR author should check the comments and take action accordingly
• @rustbot review: the author is ready for a review, this PR will be queued again in the reviewer's queue

[rustbot]

This PR modifies src/bootstrap/src/core/config. If appropriate, please also update CONFIG_CHANGE_HISTORY in src/bootstrap/src/lib.rs and change-id in config.example.toml.

This PR changes config.example.toml. If appropriate, please also update CONFIG_CHANGE_HISTORY in src/bootstrap/src/lib.rs and change-id in config.example.toml.

[davidtwco]

Some context: these options are useful to us at Huawei to imply with some internal requirements, if we can configure them via config.toml then that saves us having to maintain an (admittedly small) patch.

[Mark-Simulacrum]

When is this necessary, if debuginfo is not enabled in the first place? Do we really need another option here?

[lxy19980601]

-Cstrip=symbols will strip all symbols, not only the debuginfo. The artifacts will be smaller and have fewer symbols if we use strip = true.

[lxy19980601]

I compare the size of binary artifacts, it will be smaller if use -Cstrip=symbols compare to not using. For example, the size of librustc_driver.so is 194.5MB without using -Cstrip=symbols, but 141.3MB when using -Cstrip=symbols.

[Mark-Simulacrum]

I see. Thanks!

[Mark-Simulacrum]

I'd prefer to avoid duplicating the names of the values for this unstable option here from rustc. Can we just pass this through as an opaque string?

[lxy19980601]

Modified according to comments.

[Mark-Simulacrum]

@rustbot author

[lxy19980601]

@Mark-Simulacrum

[davidtwco]

Suggested change

	# Indicates whether stack protectors should be used
	# via the unstable option `-Zstack-protector`.
	# Valid options are : `none`(default),`basic`,`strong`, or `all`.
	# But `strong` and `basic` options may be buggy and are not suggested here.(rust-lang/rust #114903)
	#stack-protector = "none"
	# Indicates whether stack protectors should be used
	# via the unstable option `-Zstack-protector`.
	#
	# Valid options are: `none`(default), `basic`, `strong`, or `all`.
	# `strong` and `basic` may be buggy and are not recommended, see rust-lang/rust#114903.
	#stack-protector = "none"

[davidtwco]

Something like this might have been more what Mark had in mind, assuming that rust_stack_protector is made an Option.

Suggested change

	match self.config.rust_stack_protector.as_str() {
	"none" => rustflags.arg("-Zstack-protector=none"),
	"basic" => rustflags.arg("-Zstack-protector=basic"),
	"strong" => rustflags.arg("-Zstack-protector=strong"),
	"all" => rustflags.arg("-Zstack-protector=all"),
	other => unreachable!("{:?} is not recognized as valid stack protectors", other),
	};
	if let Some(stack_protector) = self.config.rust_stack_protector {
	rustflags.arg(format!("-Zstack-protector={stack_protector}")));
	}

[lxy19980601]

@Mark-Simulacrum

[Mark-Simulacrum]

This is in my queue to review, I just haven't had the chance to do so yet.

[Mark-Simulacrum]

@bors r+ rollup

[bors]

📌 Commit 3f8487a has been approved by Mark-Simulacrum

It is now in the queue for this repository.