linux-kernel - Re:[PATCH] 8250_pci.c: Only check for communication class in serial_pci_guess

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

Message-ID: <1547187487-10056-1-git-send-email-guan.yung.tseng@ni.com>
Date:   Fri, 11 Jan 2019 14:18:07 +0800
From:   Guan Yung Tseng <guan.yung.tseng@...com>
To:     <gregkh@...uxfoundation.org>
CC:     <linux-serial@...r.kernel.org>, <linux-kernel@...r.kernel.org>,
        "Guan Yung Tseng" <guan.yung.tseng@...com>
Subject: Re:[PATCH] 8250_pci.c: Only check for communication class in serial_pci_guess_board

On Thu, Jan 10, 2019 at 04:12:12PM +0800, Guan Yung Tseng wrote:
> Some multiport serial cards, such as the NI PXI-8430/2, NI PXI-8430/8,
> and NI PXI-8432/4 use PCI_CLASS_COMMUNICATION_OTHER and this fail the
> serial_pci_is_class_communication test added in the commit 7d8905d06405
> ("serial: 8250_pci: Enable device after we check black list").
> 
> Since these devices are correctly listed in serial_pci_tbl, we
> shouldn't need to check the PCI class IDs. This change relocates the
> class checking solely into "serial_pci_guess_board" where it had been
> before so that the class-check doesn't hinder initialization.
> 
> Signed-off-by: Guan Yung Tseng <guan.yung.tseng@...com>
> ---
>  drivers/tty/serial/8250/8250_pci.c | 28 +++++++++-------------------
>  1 file changed, 9 insertions(+), 19 deletions(-)
> 
> diff --git a/drivers/tty/serial/8250/8250_pci.c b/drivers/tty/serial/8250/8250_pci.c
> index 4986b4a..e33a869 100644
> --- a/drivers/tty/serial/8250/8250_pci.c
> +++ b/drivers/tty/serial/8250/8250_pci.c
> @@ -3382,21 +3382,6 @@ static const struct pci_device_id blacklist[] = {
>  	{ PCI_VDEVICE(COMMTECH, PCI_ANY_ID), },
>  };
>  
> -static int serial_pci_is_class_communication(struct pci_dev *dev)
> -{
> -	/*
> -	 * If it is not a communications device or the programming
> -	 * interface is greater than 6, give up.
> -	 */
> -	if ((((dev->class >> 8) != PCI_CLASS_COMMUNICATION_SERIAL) &&
> -	     ((dev->class >> 8) != PCI_CLASS_COMMUNICATION_MULTISERIAL) &&
> -	     ((dev->class >> 8) != PCI_CLASS_COMMUNICATION_MODEM)) ||
> -	    (dev->class & 0xff) > 6)
> -		return -ENODEV;
> -
> -	return 0;
> -}

>
> Noted.
>

Please leave this as a function, no need to move it inside another
function at this point in time.


>  static int serial_pci_is_blacklisted(struct pci_dev *dev)
>  {
>  	const struct pci_device_id *bldev;
> @@ -3427,6 +3412,15 @@ serial_pci_guess_board(struct pci_dev *dev, struct pciserial_board *board)
>  	int num_iomem, num_port, first_port = -1, i;
>  
>  	/*
> +	 * If it is not a communications device or the programming
> +	 * interface is greater than 6, give up.
> +	 *
> +	 */
> +	if ((((dev->class >> 8) != PCI_CLASS_COMMUNICATION_SERIAL) &&
> +	     ((dev->class >> 8) != PCI_CLASS_COMMUNICATION_MODEM)) ||
> +	    (dev->class & 0xff) > 6)
> +		return -ENODEV;
> +	/*

>
> It was my mistake. I should not remove it.
> I will fix them and submit another patch.
>

what happened to the multiserial check?

You should just call the function here.

thanks,

greg k-h