Enable link-arg link kind inside of #[link] attribute

[zetanumbers]

#99427 (comment)

...
This would help to make link-arg usable in #[link] attributes and e.g. wrap libc and libgcc into a group (*) in the libc crate like

#[link(kind = "link-arg", name = "--start-group")]
#[link(kind = "static", name = "c")]
#[link(kind = "static", name = "gcc")]
#[link(kind = "link-arg", name = "--end-group")]

(*) to address cyclic dependencies between them

This is an analogue of CMake's LINKER: prefix (https://cmake.org/cmake/help/git-stage/command/target_link_options.html#handling-compiler-driver-differences), and was discussed as a possible future extension in the link modifier RFC (https:/rust-lang/rfcs/blob/master/text/2951-native-link-modifiers.md#support-linkarg--string-in-addition-to-the-modifiers).

[rustbot]

Thanks for the pull request, and welcome! The Rust team is excited to review your changes, and you should hear from @cjgillot (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

[petrochenkov]

r? @petrochenkov

[zetanumbers]

Add unstable book entry

[bjorn3]

I thought we had this ability in the past but removed it because it depends on the exact linker that is used.

[petrochenkov]

@bjorn3
The previous version (#[link_args]) was removed in #83820.
It was different than this one and more similar to the -C link-arg flag - its relative order with -l flags and #[link] attribute wasn't respected, an when used in a rlib it wasn't recorded in rlib metadata and wasn't later used during the final link.

This version is an attribute version of -l link-arg, so the order is respected, and it's preserved in metadata, and it otherwise behaves like a native library.
This new setup was described as a future possibility and potential replacement for #[link_args] in https:/rust-lang/rfcs/blob/master/text/2951-native-link-modifiers.md#support-linkarg--string-in-addition-to-the-modifiers.

depends on the exact linker that is used

The portability here is relatively good because -Wl, is added automatically if the current linker flavor requires it (-verbatim is the default).
There are also some relevant docs at enum LinkerFlavor - basically in practice for each given target there is a basic linker flavor with its own interface and two "modifiers" cc and lld that can be added to it. lld doesn't change the interface, and cc requires adding -Wl, which can be done automatically, so use cases like --(start,end)-group in #99427 (comment) should work.

[petrochenkov]

r=me after squashing commits into one.
@rustbot author

[zetanumbers]

@bors r=petrochenkov

[bors]

@zetanumbers: 🔑 Insufficient privileges: Not in reviewers

[petrochenkov]

@bors r+

[bors]

📌 Commit 08bfe49 has been approved by petrochenkov

It is now in the queue for this repository.

[bors]

⌛ Testing commit 08bfe49 with merge 25432b7...

[petrochenkov]

@bors r+

[bors]

📌 Commit f7617c1 has been approved by petrochenkov

It is now in the queue for this repository.

[bors]

⌛ Testing commit f7617c1 with merge 0079625...

[bors]

☀️ Test successful - checks-actions
Approved by: petrochenkov
Pushing 0079625 to master...

[rust-timer]

Finished benchmarking commit (0079625): comparison URL.

Overall result: ❌✅ regressions and improvements - no action needed

@rustbot label: -perf-regression

Instruction count

This is a highly reliable metric that was used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	0.1%	[0.1%, 0.3%]	4
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-0.2%	[-0.2%, -0.2%]	1
All ❌✅ (primary)	-	-	0

Max RSS (memory usage)

Results

This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-1.6%	[-1.6%, -1.6%]	1
All ❌✅ (primary)	-	-	0

Cycles

This benchmark run did not return any relevant results for this metric.

Binary size

This benchmark run did not return any relevant results for this metric.

Bootstrap: 672.048s -> 673.226s (0.18%)
Artifact size: 313.42 MiB -> 313.37 MiB (-0.01%)