target-cpu=native is broken on latest nightly
#53322
Labels
A-LLVM
Area: Code generation parts specific to LLVM. Both correctness bugs and optimization-related issues.
P-high
High priority
regression-from-stable-to-nightly
Performance or correctness regression from stable to nightly.
T-compiler
Relevant to the compiler team, which will review and decide on the PR/issue.
Comments
|
I'm seeing this with |
petrochenkov
added
A-LLVM
|
|
|
petrochenkov
added
I-nominated
T-compiler
|
And.. What happened and why? Is it a broken LLVM version or what? |
|
Problem still present with $ rustc --print cfg
debug_assertions
target_arch="x86_64"
target_endian="little"
target_env=""
target_family="unix"
target_feature="fxsr"
target_feature="mmx"
target_feature="sse"
target_feature="sse2"
target_feature="sse3"
target_feature="ssse3"
target_has_atomic="128"
target_has_atomic="16"
target_has_atomic="32"
target_has_atomic="64"
target_has_atomic="8"
target_has_atomic="cas"
target_has_atomic="ptr"
target_os="macos"
target_pointer_width="64"
target_thread_local
target_vendor="apple"
unix
$ rustc --print cfg -C target-cpu=native -C opt-level=3
target_arch="x86_64"
target_endian="little"
target_env=""
target_family="unix"
target_feature="aes"
target_feature="avx"
target_feature="avx2"
target_feature="bmi1"
target_feature="bmi2"
target_feature="fma"
target_feature="fxsr"
target_feature="lzcnt"
target_feature="mmx"
target_feature="pclmulqdq"
target_feature="popcnt"
target_feature="rdrand"
target_feature="rdseed"
target_feature="sse"
target_feature="sse2"
target_feature="sse3"
target_feature="sse4.1"
target_feature="sse4.2"
target_feature="ssse3"
target_feature="xsave"
target_feature="xsavec"
target_feature="xsaveopt"
target_feature="xsaves"
target_has_atomic="128"
target_has_atomic="16"
target_has_atomic="32"
target_has_atomic="64"
target_has_atomic="8"
target_has_atomic="cas"
target_has_atomic="ptr"
target_os="macos"
target_pointer_width="64"
target_thread_local
target_vendor="apple"
unixThis is especially weird as it accepts it for printing the configuration, but then doesn't when actually compiling. 😫 |
|
Still seeing this with |
|
cc @nagisa @alexcrichton -- anybody have any ideas what's going on here? Could this be related to an LLVM upgrade or something? |
|
Marking as P-high for now. This is a regression, but on the other hand, if the message is coming from LLVM, there may not be a lot we can do here? Would be nice to get some more insight into what might be happening here. |
|
I've bisected this to #53031 and can reproduce this with a simple cc @michaelwoerister (as a heads up) I'll look into seeing if we can extract the name of the CPU from LLVM and see if we can set that instead. |
|
This should be fixed in #53642 |
alexcrichton
added a commit
to alexcrichton/rust
that referenced
this issue
Aug 24, 2018
This fixes a regression from rust-lang#53031 where specifying `-C target-cpu=native` is printing a lot of warnings from LLVM about `native` being an unknown CPU. It turns out that `native` is indeed an unknown CPU and we have to perform a mapping to an actual CPU name, but this mapping is only performed in one location rather than all locations we inform LLVM about the target CPU. This commit centralizes the mapping of `native` to LLVM's value of the native CPU, ensuring that all locations we inform LLVM about the `target-cpu` it's never `native`. Closes rust-lang#53322
bors
added a commit
that referenced
this issue
Aug 27, 2018
Fix warnings about the `native` target-cpu This fixes a regression from #53031 where specifying `-C target-cpu=native` is printing a lot of warnings from LLVM about `native` being an unknown CPU. It turns out that `native` is indeed an unknown CPU and we have to perform a mapping to an actual CPU name, but this mapping is only performed in one location rather than all locations we inform LLVM about the target CPU. This commit centralizes the mapping of `native` to LLVM's value of the native CPU, ensuring that all locations we inform LLVM about the `target-cpu` it's never `native`. Closes #53322
alexcrichton
added a commit
to alexcrichton/rust
that referenced
this issue
Aug 27, 2018
This fixes a regression from rust-lang#53031 where specifying `-C target-cpu=native` is printing a lot of warnings from LLVM about `native` being an unknown CPU. It turns out that `native` is indeed an unknown CPU and we have to perform a mapping to an actual CPU name, but this mapping is only performed in one location rather than all locations we inform LLVM about the target CPU. This commit centralizes the mapping of `native` to LLVM's value of the native CPU, ensuring that all locations we inform LLVM about the `target-cpu` it's never `native`. Closes rust-lang#53322
bors
added a commit
that referenced
this issue
Aug 28, 2018
Fix warnings about the `native` target-cpu This fixes a regression from #53031 where specifying `-C target-cpu=native` is printing a lot of warnings from LLVM about `native` being an unknown CPU. It turns out that `native` is indeed an unknown CPU and we have to perform a mapping to an actual CPU name, but this mapping is only performed in one location rather than all locations we inform LLVM about the target CPU. This commit centralizes the mapping of `native` to LLVM's value of the native CPU, ensuring that all locations we inform LLVM about the `target-cpu` it's never `native`. Closes #53322
alexcrichton
added a commit
to alexcrichton/rust
that referenced
this issue
Aug 28, 2018
This fixes a regression from rust-lang#53031 where specifying `-C target-cpu=native` is printing a lot of warnings from LLVM about `native` being an unknown CPU. It turns out that `native` is indeed an unknown CPU and we have to perform a mapping to an actual CPU name, but this mapping is only performed in one location rather than all locations we inform LLVM about the target CPU. This commit centralizes the mapping of `native` to LLVM's value of the native CPU, ensuring that all locations we inform LLVM about the `target-cpu` it's never `native`. Closes rust-lang#53322
bors
added a commit
that referenced
this issue
Aug 29, 2018
Fix warnings about the `native` target-cpu This fixes a regression from #53031 where specifying `-C target-cpu=native` is printing a lot of warnings from LLVM about `native` being an unknown CPU. It turns out that `native` is indeed an unknown CPU and we have to perform a mapping to an actual CPU name, but this mapping is only performed in one location rather than all locations we inform LLVM about the target CPU. This commit centralizes the mapping of `native` to LLVM's value of the native CPU, ensuring that all locations we inform LLVM about the `target-cpu` it's never `native`. Closes #53322
|
I still got the following error when I was compiling alacritty: Here are some of my configurations that may be useful: |
|
@alexcrichton Can you take a look at this? Thanks. |
|
@EFanZh: The error is unclear, can you do a verbose build? |
|
Here is the log from verbose build: |
|
Please file an issue at https://github.com/brendanzab/gl-rs/ for this. This issue has been solved. It may seem related to yours but it is independent. |
|
I do think this is a Rust issue, or maybe an LLVM issue. My guess is that Rust translates I have tested the following commands to determine my CPU’s capability: |
|
What does |
|
I have listed my CPU info above. |
|
|
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Reproduce by creating a crate with
cargo initand .cargo/config set toIt produces the following errors:
It was working fine on
nightly-2018-08-06-x86_64-unknown-linux-gnu, but broke withnightly-2018-08-14-x86_64-unknown-linux-gnu(rustup doesn't show any nightlies available between those dates.Hardware is Skylake Xeon and correctly detected by
rustc --print:The text was updated successfully, but these errors were encountered: