linux-kernel - Re: [PATCH RESEND v2] usb: typec: tcpm: Fix altmode re-registration causes sysfs create fail

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:   Fri, 13 Jan 2023 08:33:07 +0800
From:   ChiYuan Huang <cy_huang@...htek.com>
To:     <linux@...ck-us.net>, <heikki.krogerus@...ux.intel.com>,
        <matthias.bgg@...il.com>
CC:     <gregkh@...uxfoundation.org>, <gene_chen@...htek.com>,
        <linux-usb@...r.kernel.org>, <linux-kernel@...r.kernel.org>,
        <linux-arm-kernel@...ts.infradead.org>,
        <linux-mediatek@...ts.infradead.org>, <stable@...r.kernel.org>
Subject: Re: [PATCH RESEND v2] usb: typec: tcpm: Fix altmode re-registration
 causes sysfs create fail

On Mon, Jan 09, 2023 at 03:19:50PM +0800, cy_huang@...htek.com wrote:
> From: ChiYuan Huang <cy_huang@...htek.com>
> 
> There's the altmode re-registeration issue after data role
> swap (DR_SWAP).
> 
> Comparing to USBPD 2.0, in USBPD 3.0, it loose the limit that only DFP
> can initiate the VDM command to get partner identity information.
> 
> For a USBPD 3.0 UFP device, it may already get the identity information
> from its port partner before DR_SWAP. If DR_SWAP send or receive at the
> mean time, 'send_discover' flag will be raised again. It causes discover
> identify action restart while entering ready state. And after all
> discover actions are done, the 'tcpm_register_altmodes' will be called.
> If old altmode is not unregistered, this sysfs create fail can be found.
> 
> In 'DR_SWAP_CHANGE_DR' state case, only DFP will unregister altmodes.
> For UFP, the original altmodes keep registered.
> 
> This patch fix the logic that after DR_SWAP, 'tcpm_unregister_altmodes'
> must be called whatever the current data role is.
> 
> Reviewed-by: Macpaul Lin <macpaul.lin@...iatek.com>
> Fixes: ae8a2ca8a221 ("usb: typec: Group all TCPCI/TCPM code together)
> Reported-by: TommyYl Chen <tommyyl.chen@...iatek.com>
> Cc: stable@...r.kernel.org
> Signed-off-by: ChiYuan Huang <cy_huang@...htek.com>
> ---
> Since v2:
> - Correct the mail sent from Richtek.
> - Add 'Reviewed-by' tag.
>

Sorry, it seems I focus on the email testing and forget to attach the issue
log in the v2 patch appendix.

If anyone check this issue. Please refer to the v1 patch link.
https://lore.kernel.org/lkml/1671096096-20307-1-git-send-email-u0084500@gmail.com/
Inside this, the issue log show how it happened.

This can help better analyze this issue. 

ChiYuan Huang.
> Hi, Greg:
> 
>   Please check this one. I have strongly requested our MIS to remove the confidential string.
> 
> ChiYuan Huang.
> ---
>  drivers/usb/typec/tcpm/tcpm.c | 7 +++----
>  1 file changed, 3 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/usb/typec/tcpm/tcpm.c b/drivers/usb/typec/tcpm/tcpm.c
> index 904c7b4..59b366b 100644
> --- a/drivers/usb/typec/tcpm/tcpm.c
> +++ b/drivers/usb/typec/tcpm/tcpm.c
> @@ -4594,14 +4594,13 @@ static void run_state_machine(struct tcpm_port *port)
>  		tcpm_set_state(port, ready_state(port), 0);
>  		break;
>  	case DR_SWAP_CHANGE_DR:
> -		if (port->data_role == TYPEC_HOST) {
> -			tcpm_unregister_altmodes(port);
> +		tcpm_unregister_altmodes(port);
> +		if (port->data_role == TYPEC_HOST)
>  			tcpm_set_roles(port, true, port->pwr_role,
>  				       TYPEC_DEVICE);
> -		} else {
> +		else
>  			tcpm_set_roles(port, true, port->pwr_role,
>  				       TYPEC_HOST);
> -		}
>  		tcpm_ams_finish(port);
>  		tcpm_set_state(port, ready_state(port), 0);
>  		break;
> -- 
> 2.7.4
>