linux-kernel - Re: [PATCH 03/58] serial_core: Un-ifdef sysrq SUPPORT

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, 17 Dec 2019 14:51:36 +0100
From:   Greg Kroah-Hartman <gregkh@...uxfoundation.org>
To:     Dmitry Safonov <dima@...sta.com>
Cc:     linux-kernel@...r.kernel.org,
        Dmitry Safonov <0x7f454c46@...il.com>,
        Jiri Slaby <jslaby@...e.com>,
        Vasiliy Khoruzhick <vasilykh@...sta.com>,
        linux-serial@...r.kernel.org
Subject: Re: [PATCH 03/58] serial_core: Un-ifdef sysrq SUPPORT_SYSRQ

On Fri, Dec 13, 2019 at 12:06:02AM +0000, Dmitry Safonov wrote:
> The SUPPORT_SYSRQ is messy: every .c source should define it before
> including "serial_core.h" if sysrq is supported or struct uart_port will
> differ in sizes. Also this prevents moving to serial_core.c functions:
> uart_handle_sysrq_char(), uart_prepare_sysrq_char(),
> uart_unlock_and_check_sysrq().
> 
> It doesn't save many bytes in the structure, and a better way to reduce
> it's size would be making rs485 and iso7816 pointers.
> 
> Introduce `has_sysrq` member to be used by serial line drivers further.
> 
> Signed-off-by: Dmitry Safonov <dima@...sta.com>
> ---
>  include/linux/serial_core.h | 77 +++++++++++++++++++++----------------
>  1 file changed, 43 insertions(+), 34 deletions(-)
> 
> diff --git a/include/linux/serial_core.h b/include/linux/serial_core.h
> index bbbe57bf5163..5f761c399282 100644
> --- a/include/linux/serial_core.h
> +++ b/include/linux/serial_core.h
> @@ -240,14 +240,13 @@ struct uart_port {
>  	resource_size_t		mapsize;
>  	struct device		*dev;			/* parent device */
>  
> -#if defined(CONFIG_SERIAL_CORE_CONSOLE) || defined(SUPPORT_SYSRQ)
>  	unsigned long		sysrq;			/* sysrq timeout */
>  	unsigned int		sysrq_ch;		/* char for sysrq */
> -#endif
> +	unsigned char		has_sysrq;
>  
>  	unsigned char		hub6;			/* this should be in the 8250 driver */
>  	unsigned char		suspended;
> -	unsigned char		unused[2];

Ugh, the samsung driver was using both of these fields to overload
things for it's own use :(

It's not your fault, I'll go fix up that driver first before applying
this one.

bah, what a horrid hack they did...

greg k-h