linux-kernel - Re: [PATCH 0/6] Clean up the sdio_uart driver and fix the tty code

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: <20091103120618.26b669be@lxorguk.ukuu.org.uk>
Date:	Tue, 3 Nov 2009 12:06:18 +0000
From:	Alan Cox <alan@...rguk.ukuu.org.uk>
To:	Nicolas Pitre <nico@...xnic.net>
Cc:	Alan Cox <alan@...ux.intel.com>, linux-mmc@...r.kernel.org,
	lkml <linux-kernel@...r.kernel.org>, dhowells@...hat.com
Subject: Re: [PATCH 0/6] Clean up the sdio_uart driver and fix the tty code

> With this folded in, the card does work with the full series applied.  
> However the kernel is now crashing when the card is pulled out while 
> some process is reading from the device.  This used to behave well 
> before. I don't have time to investigate that one right now though.

Going through it I found one assumption in the tty_port code that wanted
fixing. We would release the port and then try to change the modem lines.

The dtr_rts method didn't claim the function which in conjunction with
that made it crash.

Also the hangup checked your old port->opened which now never got set.

I am scratching my head over some of the other stuff I found however. In
particular port->func can be set to NULL when the device is removed.

The claim method takes the mutex, checks if it is NULL and acts
accordingly but it releases the mutex, which makes it useless as the code
then uses port->func. If I move the release of the mutex to the
release_func method then that fixes almost all cases.

The one I'm stuck on is this

		CPU1				CPU2

	sdio_uart_irq
					sdio_uart_port_remove
					port->func = NULL;
	sdio_in
	BUG_ON

I'm not sure what the required rules on the irq disable/remove are meant
to be here ?

Alan
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/