linux-kernel - Re: [PATCH] ARM: kexec: Fix panic after TLB are invalidated

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20210205094458.GF1463@shell.armlinux.org.uk>
Date:   Fri, 5 Feb 2021 09:44:59 +0000
From:   Russell King - ARM Linux admin <linux@...linux.org.uk>
To:     Giancarlo Ferrari <giancarlo.ferrari89@...il.com>
Cc:     Mark Rutland <mark.rutland@....com>, linux-kernel@...r.kernel.org,
        penberg@...nel.org, geert@...ux-m68k.org,
        linux-arm-kernel@...ts.infradead.org, akpm@...ux-foundation.org,
        rppt@...nel.org, giancarlo.ferrari@...ia.com
Subject: Re: [PATCH] ARM: kexec: Fix panic after TLB are invalidated

On Fri, Feb 05, 2021 at 12:40:54AM +0000, Giancarlo Ferrari wrote:
> Russell,
> 
> On Fri, Feb 05, 2021 at 12:18:33AM +0000, Russell King - ARM Linux admin wrote:
> > On Thu, Feb 04, 2021 at 11:48:42PM +0000, Giancarlo Ferrari wrote:
> > > Can I ask about having it integrated ?
> > 
> > Thanks for testing. Are you willing for me to add:
> > 
> > Tested-by: Giancarlo Ferrari <giancarlo.ferrari89@...il.com>
> > 
> > to the commit log?
> > 
> 
> Sure.
> 
> I have a question regarding the patch. I saw that the structure start at
> the end of the relocation code:
> 
> data = reboot_code_buffer + relocate_new_kernel_size;
> 
> which means it overlap with the global symbol relocate_new_kernel_size.
> I think is minor comment as the variable is only used in the fncpy()
> then thrown away.

The same is true of the rest of the kernel image if that's how you'd
like to look at it.

relocate_new_kernel_size is just there to tell the C code the size of
the function, nothing more nothing less. It isn't there to be copied.
The copied code doesn't use it.

> Something like:
> 
> data = reboot_code_buffer + relocate_new_kernel_size + sizeof(long);

No. That will place the structure after the size variable, which we
don't want, and...

> and accordingly in the instruction (arch/arm/kernel/relocate_kernel.S):
> 
> adr	r7, relocate_new_kernel_end

... we will then need to follow this with:
	add	r7, r7, #4

or replace it with:
	adr	r7, relocate_new_kernel_end + 4

so that r7 points at "data".

-- 
RMK's Patch system: https://www.armlinux.org.uk/developer/patches/
FTTP is here! 40Mbps down 10Mbps up. Decent connectivity at last!