New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
cargo-vendor: 0.1.13 -> 0.1.23 #57017
Conversation
9e62121
to
ce373b8
Compare
@GrahamcOfBorg build cargo-vendor |
@@ -14,7 +14,7 @@ in | |||
buildInputs = [ openssl zlib curl ] | |||
++ stdenv.lib.optionals stdenv.isDarwin [ CoreFoundation libiconv ]; | |||
# TODO: buildRustCrate seems to use incorrect default inference | |||
crateBin = [ { name = "cargo"; path = "src/bin/cargo.rs"; } ]; | |||
crateBin = [ { name = "cargo"; path = "src/bin/cargo/main.rs"; } ]; |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
You can remove the path
attribute. Since fc5e595 we are doing the right thing(tm) within buildRustCrate
.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
You are right, fixed
Unfortunately with big crate deps list (like a |
ce373b8
to
0a4fcdb
Compare
I am only seeing
When running EDIT: Now I see the segfault in dmesg: |
@andir This assertion is you catch because Looks like:
|
FWIW this seems to be working for me? If I'm the only one NOT seeing segfaults I can investigate further... I am building on top of staging's rust bump to 1.33-- no clue if that's relevant :). |
@dtzWill how did you test it? I had to run |
On Fri, 08 Mar 2019 12:06:09 -0800, Andreas Rammhold ***@***.***> wrote:
@dtzWill how did you test it? I had to run `--check` on the `cargoDeps` attribute to actually re-run the whole cargo-vendor thing.
Yep, with `--check` using the command for `parity` you posted.
I also used it in packaging up a newer (git) version of `fractal` \o/
which was what led me to discovering this PR :).
Let's see what behavior this has for other folks, maybe?
…
--
You are receiving this because you were mentioned.
Reply to this email directly or view it on GitHub:
#57017 (comment) part: text/html
|
@GrahamcOfBorg build parity.cargoDeps Does ofBorg also have --check? |
@dtzWill I tried to again on a newly installed NixOS packet.net machine. Cloned nixpkgs and ran What differences are there between our systems? |
FWIW I'm not seeing any failures with the command andir posted. |
I was able to reproduce this after a second try:
dmesg:
|
To be clear, I've been using this on top of my normal branch and not the exact tree here, and there are many reasons I'd be seeing different behavior-- but MOSTLY the bits that turn out well I send upstream so I don't think I'm sitting on anything relevant. The most obvious difference is I'm using the rust 1.33 updates from our staging branch-- and I don't know what changed in 1.33 but "fixing things that cause crashes" sounds plausible enough to check it out ;). When my builders have a chance I'll try the tree here and maybe on a few machines... heisenbug is lame! |
0a4fcdb
to
26435b7
Compare
As I see it fixed in latest |
@GrahamcOfBorg build cargo-vendor |
This would be really nice to have, just ran into issues with the old version of |
What's the current status of this? It looks like I can't update Incidentally, I can't build this on macOS. I tried merging it into current |
@@ -13,15 +13,15 @@ in | |||
cargo = attrs: { | |||
buildInputs = [ openssl zlib curl ] | |||
++ stdenv.lib.optionals stdenv.isDarwin [ CoreFoundation libiconv ]; |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
It looks like my macOS build error can be fixed with
++ stdenv.lib.optionals stdenv.isDarwin [ CoreFoundation libiconv ]; | |
++ stdenv.lib.optionals stdenv.isDarwin [ CoreFoundation Security libiconv ]; |
Of course, then I get the same error for cargo-vendor itself, so the same fix needs to be applied there too.
@@ -13,15 +13,15 @@ in | |||
cargo = attrs: { | |||
buildInputs = [ openssl zlib curl ] | |||
++ stdenv.lib.optionals stdenv.isDarwin [ CoreFoundation libiconv ]; | |||
# TODO: buildRustCrate seems to use incorrect default inference | |||
crateBin = [ { name = "cargo"; path = "src/bin/cargo.rs"; } ]; | |||
}; | |||
|
|||
cargo-vendor = attrs: { | |||
buildInputs = [ openssl zlib curl ]; |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
buildInputs = [ openssl zlib curl ]; | |
buildInputs = [ openssl zlib curl ] | |
++ stdenv.lib.optionals stdenv.isDarwin [ Security ]; |
I've been trying to reproduce the segfault we did observe with this rebased onto currentmaster. After 6 rebuilds there hasn't been a single segfault. Currently trying to reproduce it on another machine but that also succeeded for a few times. I hope this isn't due to the host being on nixos-19.03 now... @lilyball can you apply the mentioned fix to cargo to figure out if this package (and the users) work on master for Darwin? I do not own any of that hardware :/ |
@andir I already applied the two fixes I suggested and successfully built |
This is supposedly fixing the build of the cargo crate on Drawin [1]. [1] NixOS#57017 (review)
This is supposedly fixing the build of the cargo crate on Drawin [1]. [1] NixOS#57017 (review)
26435b7
to
1bb989c
Compare
@lilyball does this look alright now? |
@andir Yes, it builds just fine on Darwin now. |
This is supposedly fixing the build of the cargo crate on Drawin [1]. [1] NixOS#57017 (review)
This is supposedly fixing the build of the cargo crate on Drawin [1]. [1] NixOS#57017 (review)
Motivation for this change
Upgrade cargo-vendor to 0.1.23 is required for support build edition 2018 crates.
Things done
sandbox
innix.conf
on non-NixOS)nix-shell -p nox --run "nox-review wip"
./result/bin/
)nix path-info -S
before and after)