For most formulae, the bottles need a minimum libstdc++ rather than a
minimum GCC version. This is particularly important when building on
Ubuntu where the default compiler version is older than libstdc++.
So, checking the host libstdc++ version is a more accurate way to
determine whether brew GCC is needed at runtime. This can be improved in
the future to check symbol versions (e.g. GLIBCXX, CXXABI, GLIBC) which
can allow some bottles to be installed even with older glibc/libstdc++.
On the [beta Bluefin LTS](https://docs.projectbluefin.io/lts), `brew config` reports:
```console
OS: CentOS Stream release 10 (Coughlan)
```
This is due to the `/etc/redhat-release` file being prioritized over `::OS_VERSION` (`PRETTY_NAME` from `/etc/os-release`).
On Bluefin LTS, `PRETTY_NAME` is `Bluefin LTS`.
This commit reverses the order so `brew config` has better detection:
```
OS: Bluefin LTS
```
I tested on RHEL 10 and the OS line only changed in that the word
"release" was omitted:
```
$ cat /etc/redhat-release
Red Hat Enterprise Linux release 10.0 (Coughlan)
$ grep PRETTY_NAME /etc/os-release
PRETTY_NAME="Red Hat Enterprise Linux 10.0 (Coughlan)"
```
- This is an improvement on my previous PR https://github.com/Homebrew/brew/pull/15788
On Linux, we occasionally see `EMFILE` ("too many open files") errors
especially when installing a large formula like `llvm`. Currently, this
can be reliably reproduced in a Homebrew/brew GitHub codespace (where
`ulimit -n` seems to be 1024 by default) with `brew install geeqie`,
with the following error message:
Error: Too many open files @ rb_sysopen - /home/linuxbrew/.linuxbrew/Cellar/llvm/20.1.8/bin/tblgen-lsp-server
The reason is that each instance of `PatchELF::Patcher` keeps the ELF
file open. We prepend the `ELFShim` module to the `Pathname` class and
cache the patcher as an instance variable, which means that the ELF file
remains open so long as the `Pathname` instance is still alive even if
we don't need to access the ELF metadata anymore. When performing
certain checks (e.g., linkage), we also store these `Pathname`
instances, so the number of open file descriptors simply keeps
increasing.
We can fix that by not caching the patcher and only use it when
necessary. We create a patcher instance whenever we need to read or
write ELF metadata, and reading of metadata is consolidated into the
existing `ELFShim::Metadata` class so that we don't repeatedly create
patcher instances.
A fix for a file descriptor leak issue in patchelf.rb has been submitted
at https://github.com/david942j/patchelf.rb/pull/48. Together with that,
this fixes#19177, #19866, #20223, #20302.
We have multiple formulae (e.g. `ola`, `openvino`, `opencv`, `or-tools`,
`pytorch`) that seem to be broken by `patchelf.rb` on x86_64 Linux.[^1] The
common thread seems to be the presence of a `protodesc_cold` section in
the ELF header.
Let's avoid breaking these bottles by skipping relocation when a binary
file has a `protodesc_cold` section. It will probably hurt
relocatability of these bottles, but that's better shipping broken
binaries.
[^1]: https://github.com/Homebrew/homebrew-core/pull/210860#issue-2918569212
The library_paths method was using readlines which could leave file
descriptors open due to Ruby's garbage collection behavior. When
processing many packages during 'brew upgrade' or 'brew linkage',
this caused "Too many open files" errors on Linux systems.
Changes:
- Replace readlines with explicit file.open block to ensure proper closure
- Add caching to avoid repeatedly reading /etc/ld.so.conf during a session
- Cache included files as well to optimize recursive include processing
Fixes: #19866, #20302, #19177, #20223🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
Will fix or at least partly address:
```
/home/linuxbrew/.linuxbrew/Homebrew/Library/Homebrew/os/linux/elf.rb:225: warning: The class Pathname reached 8 shape variations, instance variables accesses will be slower and memory usage increased.
35
It is recommended to define instance variables in a consistent order, for instance by eagerly defining them all in the #initialize method.
```
Apply suggestions from code review
Co-authored-by: Douglas Eichelberger <697964+dduugg@users.noreply.github.com>
feat: add linux appdir
Apply suggestions from code review
Co-authored-by: Douglas Eichelberger <697964+dduugg@users.noreply.github.com>
We're currently returning the first match for `com.apple.dt.Xcode`,
which could be any version if a user has multiple installed.
Instead, let's try to find the newest if all our results have an
`Info.plist` that we can interrogate for the version.
Maybe resolves#18736?
Some formulae include these files, and they can't always be removed.
However, they can cause spurious linkage failures, so let's skip them
when checking for linkage. See, for example, faust at
Homebrew/homebrew-core#191308.
I think this is a bug in Ruby, but I've no idea how to track it down. I
can reproduce it intermittently in a codespace when `brew install`ing a
large number of formulae.
To work around this:
- cache the return value of `brewed_ld_so_diagnostics` so that we can
limit the number of calls to `IO.popen`
- retry once when we see a `TypeError`
Closes#17828.
