netdev - Re: [PATCH v2] net: sh_eth: fix mdio access in sh_eth

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]

Message-ID: <2ae6d8c6-372f-4708-8bd4-f461b403b986@cogentembedded.com>
Date:   Wed, 5 Jun 2019 21:19:08 +0300
From:   Sergei Shtylyov <sergei.shtylyov@...entembedded.com>
To:     Yoshihiro Shimoda <yoshihiro.shimoda.uh@...esas.com>,
        davem@...emloft.net
Cc:     netdev@...r.kernel.org, linux-renesas-soc@...r.kernel.org
Subject: Re: [PATCH v2] net: sh_eth: fix mdio access in sh_eth_close() for
 R-Car Gen2 and RZ/A1 SoCs

Hello!

   I've been meaning to look around this issue before replying to this respin
but didn't notify DaveM, so he finally merged it...

On 05/28/2019 07:10 AM, Yoshihiro Shimoda wrote:

> The sh_eth_close() resets the MAC and then calls phy_stop()
> so that mdio read access result is incorrect without any error
> according to kernel trace like below:
> 
> ifconfig-216   [003] .n..   109.133124: mdio_access: ee700000.ethernet-ffffffff read  phy:0x01 reg:0x00 val:0xffff
> 
> According to the hardware manual, the RMII mode should be set to 1
> before operation the Ethernet MAC. However, the previous code was not
> set to 1 after the driver issued the soft_reset in sh_eth_dev_exit()
> so that the mdio read access result seemed incorrect. To fix the issue,
> this patch adds a condition and set the RMII mode register in
> sh_eth_dev_exit() for R-Car Gen2 and RZ/A1 SoCs.

   I told you RZ/G1, not RZ/A1. The latter has its own data structure
 and different register layout.

> Note that when I have tried to move the sh_eth_dev_exit() calling
> after phy_stop() on sh_eth_close(), but it gets worse (kernel panic
> happened and it seems that a register is accessed while the clock is
> off).

  I've reproduced it but still don't know why that happens.

> Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@...esas.com>
> ---
>  Changes from v1 (https://patchwork.kernel.org/patch/10944265/):
>  - Revise the subject, commit log and the comment of the code.
>  - Move the RMII setting to right after soft_reset.
> 
>  drivers/net/ethernet/renesas/sh_eth.c | 4 ++++
>  1 file changed, 4 insertions(+)
> 
> diff --git a/drivers/net/ethernet/renesas/sh_eth.c b/drivers/net/ethernet/renesas/sh_eth.c
> index 6354f19..7ba35a0 100644
> --- a/drivers/net/ethernet/renesas/sh_eth.c
> +++ b/drivers/net/ethernet/renesas/sh_eth.c
> @@ -1594,6 +1594,10 @@ static void sh_eth_dev_exit(struct net_device *ndev)
>  	sh_eth_get_stats(ndev);
>  	mdp->cd->soft_reset(ndev);
>  
> +	/* Set the RMII mode again if required */

   When I asked for more details, I was meaning this comment. Thanks for telling me
about the gory details anyway. :-)

> +	if (mdp->cd->rmiimode)
> +		sh_eth_write(ndev, 0x1, RMIIMODE);
> +
>  	/* Set MAC address again */
>  	update_mac_address(ndev);
>  }

MBR, Sergei