Hi all,
this patch introduces a regression in some versions of qemu-aarch64 (at least as built by debian): https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1087822
It doesn't look like it still is a problem with newer versions of qemu so I'm not sure if this should be reverted on master, but it took me a bit of time to track this down to this commit as my reproducer isn't great, so it might make sense to revert this commit on stable branches?
(I don't remember the policies on "don't break userspace", but qemu-user is a bit of a special case here so I'll leave that up to Greg)
I've confirmed that this bug occurs on top of the latest v6.1.118 and goes away reverting this. (I've also checked the problem also occurs on master and reverting the patch also works around the issue there at this point)
Thank you, Dominique.
[leaving subject below for context, no more text after this] Greg Kroah-Hartman wrote on Tue, Aug 27, 2024 at 04:38:03PM +0200:
From: Kees Cook keescook@chromium.org
[ Upstream commit 44c76825d6eefee9eb7ce06c38e1a6632ac7eb7d ]
In commit c1d171a00294 ("x86: randomize brk"), arch_randomize_brk() was defined to use a 32MB range (13 bits of entropy), but was never increased when moving to 64-bit. The default arch_randomize_brk() uses 32MB for 32-bit tasks, and 1GB (18 bits of entropy) for 64-bit tasks.
Update x86_64 to match the entropy used by arm64 and other 64-bit architectures.
Reported-by: y0un9n132@gmail.com Signed-off-by: Kees Cook keescook@chromium.org Signed-off-by: Thomas Gleixner tglx@linutronix.de Acked-by: Jiri Kosina jkosina@suse.com Closes: https://lore.kernel.org/linux-hardening/CA+2EKTVLvc8hDZc+2Yhwmus=dzOUG5E4gV7... Link: https://lore.kernel.org/r/20240217062545.1631668-1-keescook@chromium.org Signed-off-by: Sasha Levin sashal@kernel.org