Allow storing format_args!() in variable

[m-ou-se]

Fixes #92698

Tracking issue for super let: #139076

Tracking issue for format_args: #99012

This change allows:

let name = "world";
let f = format_args!("hello {name}!"); // New: Store format_args!() for later!

println!("{f}");

This will need an FCP.

This implementation makes use of super let, which is unstable and might not exist in the future in its current form. However, it is entirely reasonable to assume future Rust will always have a way of expressing temporary lifetimes like this, since the (stable) pin!() macro needs this too. (This was also the motivation for merging #139114.)

(This is a second version of #139135)

[m-ou-se]

@bors try @rust-timer queue

[bors]

⌛ Trying commit d261de6 with merge beb8d4e...

[bors]

☀️ Try build successful - checks-actions
Build commit: beb8d4e (beb8d4ea463c3ae10548aabc62ac2cff505be075)

[rust-timer]

Finished benchmarking commit (beb8d4e): comparison URL.

Overall result: ❌✅ regressions and improvements - please read the text below

Benchmarking this pull request likely means that it is perf-sensitive, so we're automatically marking it as not fit for rolling up. While you can manually mark this PR as fit for rollup, we strongly recommend not doing so since this PR may lead to changes in compiler perf.

Next Steps: If you can justify the regressions found in this try perf run, please indicate this with @rustbot label: +perf-regression-triaged along with sufficient written justification. If you cannot justify the regressions please fix the regressions and do another perf run. If the next run shows neutral or positive results, the label will be automatically removed.

@bors rollup=never
@rustbot label: -S-waiting-on-perf +perf-regression

Instruction count

This is the most reliable metric that we have; it was used to determine the overall result at the top of this comment. However, even this metric can sometimes exhibit noise.

	mean	range	count
Regressions ❌ (primary)	0.4%	[0.2%, 1.1%]	24
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-2.4%	[-4.5%, -0.2%]	2
Improvements ✅ (secondary)	-1.8%	[-1.9%, -1.7%]	6
All ❌✅ (primary)	0.2%	[-4.5%, 1.1%]	26

Max RSS (memory usage)

Results (primary 0.4%, secondary -0.2%)

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)	1.7%	[0.4%, 12.3%]	27
Regressions ❌ (secondary)	3.8%	[3.8%, 3.8%]	1
Improvements ✅ (primary)	-2.2%	[-6.1%, -0.5%]	13
Improvements ✅ (secondary)	-2.3%	[-3.2%, -1.3%]	2
All ❌✅ (primary)	0.4%	[-6.1%, 12.3%]	40

Cycles

Results (primary 0.1%, secondary 2.4%)

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)	0.7%	[0.4%, 2.0%]	14
Regressions ❌ (secondary)	2.4%	[2.4%, 2.4%]	1
Improvements ✅ (primary)	-2.6%	[-5.4%, -0.5%]	3
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	0.1%	[-5.4%, 2.0%]	17

Binary size

Results (primary 0.1%, secondary 0.1%)

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)	0.1%	[0.0%, 0.8%]	87
Regressions ❌ (secondary)	0.1%	[0.0%, 0.2%]	15
Improvements ✅ (primary)	-0.1%	[-0.4%, -0.0%]	6
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	0.1%	[-0.4%, 0.8%]	93

Bootstrap: 770.356s -> 770.437s (0.01%)
Artifact size: 365.20 MiB -> 365.11 MiB (-0.02%)

[nikomatsakis]

@rfcbot reviewed

Exciting!

[traviscross]

This seems right. Thanks to @m-ou-se for pushing this forward.

@rfcbot reviewed

[joshtriplett]

I'm super excited to let format_args! work better!

@rfcbot reviewed

[tmandry]

I'm really happy we get to support this. Sounds like the extended lifetime behavior is straightforward and supported by all proposals for super let-like functionality.

@rfcbot reviewed

[scottmcm]

I was potentially worried about whether there's any specific details of super let that we might get locked into by this (maybe how far outside a scope it could go in tail position or something), but the conversation in the lang meeting sounded like there probably isn't, and we're already probably stuck with any of those from pin! anyway, so seems fine.

@rfcbot reviewed

[rfcbot]

🔔 This is now entering its final comment period, as per the review above. 🔔