[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [PATCH hurd v1] ext2fs: raise RLIMIT_AS if running as the root fs
From: |
Samuel Thibault |
Subject: |
Re: [PATCH hurd v1] ext2fs: raise RLIMIT_AS if running as the root fs |
Date: |
Mon, 9 Jun 2025 15:48:00 +0200 |
Samuel Thibault, le dim. 08 juin 2025 21:57:14 +0200, a ecrit:
> Diego Nieto Cid, le jeu. 29 mai 2025 23:30:58 +0100, a ecrit:
> > On Tue, May 27, 2025 at 04:26:38AM +0200, Samuel Thibault wrote:
> > >
> > > Normally it works but possibly it got broken by some upstream change.
> > > git bisect welcome to determine what broke it.
> > >
> >
> > Ok, I finally could do a bisection for this issue and the result is the
> > following[1]:
> >
> > demo@debian:~/dev/hurd/upstream/glibc$ git bisect view
> > commit 25d37948c9f3f483d63302b9967b975d3bd74a4e (HEAD, tag: first-bad)
> > Author: Wilco Dijkstra <wilco.dijkstra@arm.com>
> > Date: Thu May 1 19:58:38 2025 +0000
> >
> > malloc: Improve malloc initialization
> >
> > Move malloc initialization to __libc_early_init. Use a hidden
> > __ptmalloc_init
> > for initialization and a weak call to avoid pulling in the system
> > malloc in a
> > static binary. All previous initialization checks can now be
> > removed.
> >
> > Reviewed-by: Florian Weimer <fweimer@redhat.com>
>
> Thanks!
>
> I tracked that down to __ptmalloc_init using random_bits whcih uses
> clock_gettime, thus an RPC, but at this stage we haven't called
> __mach_init() for libc yet. I'll fix that.
I have pushed it.
Samuel