Make incremental compilation the default for all profiles.

[michaelwoerister]

This PR makes incremental compilation the default for all profiles, that is, also release and bench. rustc performs ThinLTO by default for incremental release builds for a while now and the data we've gathered so far indicates that the generated binaries exhibit roughly the same runtime performance as non-incrementally compiled ones. At the same time, incremental release builds can be 2-5 times as fast as non-incremental ones.

Making incremental compilation (temporarily) the default in cargo would be a simple way of gathering more data about runtime performance via lolbench.rs. If the results look acceptable, we can just leave it on and give a massive compile time reduction to everyone. If not, we can revert the change and think about a plan B.

This strategy assumes that lolbench will actually use the nightly cargo version. Is that true, @anp?

r? @alexcrichton

[dwijnand]

Interesting!

How does one enable incremental compilation for the release and bench profiles right now?

[michaelwoerister]

By setting the CARGO_INCREMENTAL flag as in:

CARGO_INCREMENTAL=1 cargo build --release

[anp]

lolbench consumes toolchains via rustup, so as long as that works for you you'll get the data you want. Gosh, I'm a little sad that not having had time for a try feature could now be a part of the geology of master :P.

EDIT: I think rust-lang/rust-central-station#82 (comment) will be a prereq of try-jobs if someone wants to help me out there.

[alexcrichton]

I'm personally fine with this, but this may want some more buy-in perhaps? Should we pull in the compiler team and/or FCP with a set of folks for this?

[michaelwoerister]

We discussed this in the @rust-lang/compiler triage meeting. The team seems to be in favor of enabling this now, in order to collect data via lolbench, and then unconditionally disabling it again at the beginning of the all hands.

[alexcrichton]

@bors: r+

Ok!

[bors]

📌 Commit 2c40858 has been approved by alexcrichton

[bors]

⌛️ Testing commit 2c40858 with merge d5109ad...

[bors]

☀️ Test successful - checks-travis, status-appveyor
Approved by: alexcrichton
Pushing d5109ad to master...

[japaric]

I think this change broke release builds for embedded projects. The embedded template used by the community has this in its Cargo.toml:

[profile.release]
codegen-units = 1 # better optimizations
debug = true # symbols are nice and they don't increase the size on Flash
lto = true # better optimizations

These settings produce a the following error: "can't perform LTO when compiling incrementally" with the latest nightly. Adding incremental = false to Cargo.toml fixes the problem.

[michaelwoerister]

@japaric Since we are enabling this just for data collection at the moment and will revert at the end of this week, do think we should do something here now?

[japaric]

@michaelwoerister would it be possible to not enable incremental if lto is set to true in the Cargo.toml? I think that would avoid the breakage. If that's too much trouble then we'll just have to deal with the breakage reports until the weekend.

[alexcrichton]

I've gone ahead and posted the revert for this at #6610, and @michaelwoerister we've got a nightly or two now with the incremental bit turned on so I think we're good to go for benchmarking, right?

@japaric this is definitely something we'll handle before turning it back on by default for real!

@michaelwoerister do you perhaps want to make a tracking issue for turning on incremental by default in release mode?

[Eh2406]

Which is the first nightly when this was on? (for looking at lolbench)

[ehuss]

@Eh2406 nightly-2019-01-28 is the first one.

[michaelwoerister]

Sorry for the breakage, @japaric!

@michaelwoerister do you perhaps want to make a tracking issue for turning on incremental by default in release mode?

Will do.

If I understand things correctly #6610 reverts the change in Cargo but the change still takes some time to bubble up to nightlies? I think we have some good numbers already, so I'm OK with ending the experiment.

[michaelwoerister]

@michaelwoerister do you perhaps want to make a tracking issue for turning on incremental by default in release mode?

Done: rust-lang/rust#57968

[ehuss]

the change still takes some time to bubble up to nightlies?

Lately I've been doing updates around once a week. The last update was 2 days ago, so I would probably do it this coming weekend. Do you feel like this issue is important enough to update now? (Which is totally fine, it's usually not difficult, just depends on how long you want the experiment to last.)

[michaelwoerister]

The original plan was to keep it running throughout the week. It depends on how much trouble this creates for the embedded folks.