It's not uncommon for simple binding to wrap structs without
any Ruby object references. Hence with no `mark` function.
Might as well mark them as protected by a write barrier.
This instructs make to delete target files if the recipe fails midway, like
when make itself is interrupted. This is mostly for development since it
protects against corrupt builds that need a `make clean` to repair. Release
builds normally don't fail mid-recipe.
GNU make and BSD make support this.
From GNU make's manual:
> This is almost always what you want make to do, but it is not historical
> practice; so for compatibility, you must explicitly request it.
See https://innolitics.com/articles/make-delete-on-error/
[Bug #19415]
If multiple threads attemps to load the same file concurrently
it's not a circular dependency issue.
So we check that the existing ThreadShield is owner by the current
fiber before warning about circular dependencies.
Create SHAPE_MAX_NUM_IVS (currently 50) and limit all shapes of
T_OBJECTS to that number of IVs. When a shape with a T_OBJECT has more than 50 IVs, fall back to the
obj_too_complex shape which uses hash lookup for ivs.
Note that a previous version of this commit
78fcc9847a was reverted in
88f2b94065 because it did not account for
non-T_OBJECTS
The previous code loads bundler's gemspec which does not include the generated
gemspec file, and thus the test was passing where it should indeed fail.
With this change, the test properly fails now.
2098ea0d75
Fixes the same issue at https://github.com/ruby/ruby/pull/5417
`ruby` is not always available in certain build environments and
configure options (e.g. --program-suffix)
This patch tries to choose an appropriate command line for spawning a
fresh Ruby process, based on EnvUtil implementation in ruby/ruby's test
suite.
Plus when this library is directly mirrored into ruby/ruby, prefer EnvUtil
available there over the implementation in this library's test suite.
278327d2e9
Currently, the instructions and placeholder contradict each other.
The commit that introduced this placeholder (42bc4715d920e836c8499883d)
makes it clear that the placeholder should be replaced AFTER publishing
(i.e. once the author successfully claims the gem name on Rubygems), so
the placeholder should match.
23e2dea828
We set the block address as soon as we make the block, so there is no
point in making it `Option<CodePtr>`. No memory saving, unfortunately,
as `mem::size_of::<Block>() = 176` before and after this change. Still
a simplification for the logic, though.
They are needed very often but it's hard to remember. I thought it'd be
useful to just copy that to /.vscode and edit that.
Usage:
cp -r misc/.vscode .vscode
Don't symlink it because you'd edit it but not want to commit it.
Because `optflags` is pasted into the invocation line after `CC`, we were
building with -O1 unintentionally. You can see this in the configuration
summary: 6727044423 (step):9:753
The check actually fails with -O2. To make it pass, upstream suggest
that we use the annocheck GCC plugin. Since it requires building from
source as the debian package for it isn't ready yet, punt on it for now
and use `--skip-gaps`.
Co-authored-by: Jun Aruga <jaruga@redhat.com>
