[PATCH v5 0/4] Add system-wide tunables

Yury Khrustalev yury.khrustalev@arm.com
Tue Feb 17 10:03:13 GMT 2026


On Thu, Feb 12, 2026 at 03:19:36PM -0500, DJ Delorie wrote:
> 
> > One thing I forgot to ask, currently only the tests that need the rseq 
> > tunable disabled use GLIBC_TUNABLES to explictly set the value because 
> > we expect the default to always be enabled.
> >
> > Since the default could now be disabled by a system-wide policy, should 
> > I add GLIBC_TUNABLES to the environment of all rseq tests like this:
> >
> > tst-rseq-nptl-ENV = GLIBC_TUNABLES=glibc.pthread.rseq=1
> >
> > Or would that be considered overzealous?
> 
> Given your concerns about non-overridable settings, I think a check
> inside the test to ensure the tunable is set the way the test expects,
> and UNSUPPORTED if not, should be sufficient.

I agree. Either that or use test container.

> If (it's a big if) developers find their machines are usually in a
> non-standard overridable configuration due to system-wide tunables they
> have no control over...  we can revisit this then ;-)

If the dev machine's setup interfere with Glibc tests, the tests need to
be fixed really :) Otherwise, what are we even testing?



More information about the Libc-alpha mailing list