Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 16 Oct 2022 20:39:53 +0200
From:      Paul Floyd <paulf2718@gmail.com>
To:        freebsd-hackers <freebsd-hackers@freebsd.org>
Subject:   Re: AMD64 14.0-CURRENT memory layout changes
Message-ID:  <a6472308-335e-67cc-d9d6-339a5b06a07e@gmail.com>
In-Reply-To: <18733815c5cf1af1e313168629916ee5@gundo.com>
References:  <bcfa260b-7a94-1414-a5ae-c281e96c76ec@gmail.com> <18733815c5cf1af1e313168629916ee5@gundo.com>

next in thread | previous in thread | raw e-mail | index | archive | help


On 10/16/22 17:29, Pau Amma wrote:

> Jumoing in with some quarterly reports I saw pass that may be related to 
> either/both:
> - 
> https://www.freebsd.org/status/report-2022-04-2022-06/#_shared_page_address_randomization (not sure this is about rtld)
> - 
> https://www.freebsd.org/status/report-2021-07-2021-09/#_stack_gap_handling_improvements (this one mentions a switch-off sysctl).

Hi Pau Amma and Mark

Thanks for the info.

I now understand the problem.

Valgrind needs to intercept sysctl(byname) kern.stacktop and return its 
own value. I'm already doing this for kern.usrstack.

Without interception the guest is getting the stacktop of the host. The 
host obviously refuses to let the guest mmap over its own stack, 
resulting in an assert and guest crash.



A+
Paul



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?a6472308-335e-67cc-d9d6-339a5b06a07e>