netdev - Re: [PATCH 1/2] net: ethernet: 3c515: Fix cast from pointer to integer of different size

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] [day] [month] [year] [list]

Date:   Tue, 7 Jan 2020 09:35:48 +0100
From:   Krzysztof Kozlowski <krzk@...nel.org>
To:     David Miller <davem@...hat.com>
Cc:     netdev@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 1/2] net: ethernet: 3c515: Fix cast from pointer to
 integer of different size

On Mon, Jan 06, 2020 at 01:31:55PM -0800, David Miller wrote:
> From: Krzysztof Kozlowski <krzk@...nel.org>
> Date: Sat,  4 Jan 2020 15:33:05 +0100
> 
> > Pointer passed as integer should be cast to unsigned long to
> > avoid warning (compile testing on alpha architecture):
> > 
> >     drivers/net/ethernet/3com/3c515.c: In function ‘corkscrew_start_xmit’:
> >     drivers/net/ethernet/3com/3c515.c:1066:8: warning:
> >         cast from pointer to integer of different size [-Wpointer-to-int-cast]
> > 
> > Signed-off-by: Krzysztof Kozlowski <krzk@...nel.org>
> > 
> > ---
> > 
> > Only compile tested
> 
> Sorry, I'm not applying these two.
> 
> It is clear that these drivers only work properly on 32-bit architectures
> where virtual address equals the DMA address.
> 
> Making this warning goes away creates a false sense that they are in
> fact 64-bit clean and capable, they are not.

The existing casts are clearly wrong - the convention is that pointer
should be cast to unsigned long, not int. In the second case it is even
weirder - the buffer array is actually unsigned long so the cast is
confusing.

However I understand your argument that these casts serve as a
documentation purpose of only 32-bit support, so let it be.

Thanks!

Best regards,
Krzysztof