netdev - Re: [PATCH net-next 1/7] net: phy: always call phy_process_state

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: <99667e5c-588d-1eea-dd42-a11384d286dd@samsung.com>
Date: Mon, 18 Sep 2023 15:05:41 +0200
From: Marek Szyprowski <m.szyprowski@...sung.com>
To: Andrew Lunn <andrew@...n.ch>
Cc: "Russell King (Oracle)" <rmk+kernel@...linux.org.uk>, Heiner Kallweit
	<hkallweit1@...il.com>, chenhao418@...wei.com, "David S. Miller"
	<davem@...emloft.net>, Eric Dumazet <edumazet@...gle.com>, Jakub Kicinski
	<kuba@...nel.org>, Jijie Shao <shaojijie@...wei.com>, lanhao@...wei.com,
	liuyonglong@...wei.com, netdev@...r.kernel.org, Paolo Abeni
	<pabeni@...hat.com>, shenjian15@...wei.com, wangjie125@...wei.com,
	wangpeiyang1@...wei.com
Subject: Re: [PATCH net-next 1/7] net: phy: always call
 phy_process_state_change() under lock

On 18.09.2023 14:55, Andrew Lunn wrote:
>> This probably need to be fixed somewhere in drivers/net/usb/asix* but at
>> the first glance I don't see any obvious place that need a fix.
> static int __asix_mdio_read(struct net_device *netdev, int phy_id, int loc,
>                              bool in_pm)
> {
>          struct usbnet *dev = netdev_priv(netdev);
>          __le16 res;
>          int ret;
>
>          mutex_lock(&dev->phy_mutex);
>
> Taking this lock here is the problem. Same for write.
>
> There is some funky stuff going on in asix_devices.c. It using both
> phylib and the much older mii code.

This must be something different. Removing those calls to phy_mutex from 
__asix_mdio_read/write doesn't fix this deadlock (I intentionally 
ignored the fact that some kind of synchronization is probably required 
there).

Best regards
-- 
Marek Szyprowski, PhD
Samsung R&D Institute Poland