linux-kernel - Re: [PATCH] net: macb: Clean 64b dma addresses if they are not detected

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]

Date:   Thu, 20 Sep 2018 08:23:58 +0200
From:   Michal Simek <michal.simek@...inx.com>
To:     "Edgar E. Iglesias" <edgar.iglesias@...inx.com>,
        Michal Simek <michal.simek@...inx.com>
CC:     <linux-kernel@...r.kernel.org>, <monstr@...str.eu>,
        "Edgar E. Iglesias" <edgari@...inx.com>,
        "David S. Miller" <davem@...emloft.net>, <netdev@...r.kernel.org>,
        <u-boot@...ts.denx.de>, Joe Hershberger <joe.hershberger@...com>,
        Nicolas Ferre <nicolas.ferre@...rochip.com>
Subject: Re: [PATCH] net: macb: Clean 64b dma addresses if they are not
 detected

On 19.9.2018 20:08, Edgar E. Iglesias wrote:
> On Wed, Sep 19, 2018 at 06:08:18PM +0200, Michal Simek wrote:
>> Clear ADDR64 dma bit in DMACFG register in case that HW_DMA_CAP_64B
>> is not detected on 64bit system.
>> The issue was observed when bootloader(u-boot) does not check macb
>> feature at DCFG6 register (DAW64_OFFSET) and enabling 64bit dma support
>> by default. Then macb driver is reading DMACFG register back and only
>> adding 64bit dma configuration but not cleaning it out.
>>
>> This is also align with other features which are also cleared if they are not
>> present.
> 
> Hi Michal,
> 
>>
>> Signed-off-by: Michal Simek <michal.simek@...inx.com>
>> ---
>>
>>  drivers/net/ethernet/cadence/macb_main.c | 2 ++
>>  1 file changed, 2 insertions(+)
>>
>> diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c
>> index 16e4ef7d7185..79707dff3f13 100644
>> --- a/drivers/net/ethernet/cadence/macb_main.c
>> +++ b/drivers/net/ethernet/cadence/macb_main.c
>> @@ -2163,6 +2163,8 @@ static void macb_configure_dma(struct macb *bp)
>>  #ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT
>>  		if (bp->hw_dma_cap & HW_DMA_CAP_64B)
>>  			dmacfg |= GEM_BIT(ADDR64);
>> +		else
>> +			dmacfg &= ~GEM_BIT(ADDR64);
>>  #endif
> 
> I think you might want to do this clearing outside of the #ifdef.
> If CONFIG_ARCH_DMA_ADDR_T_64BIT is not defined, we'd want to make
> sure the ADDR64 is cleared. E.g something like:
> 
>              dmacfg &= ~GEM_BIT(ADDR64);
> #ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT
>              if (bp->hw_dma_cap & HW_DMA_CAP_64B)
>                      dmacfg |= GEM_BIT(ADDR64);
> #endif
> 
> 
> Same thing for the USE_HWSTAMP/PTP flags below.

Origin patch, which introduce this read with mask,
macfg = gem_readl(bp, DMACFG) & ~GEM_BF(RXBS, -1L);
was done in 2011 and from that time this function was extended a little
bit. I am even not quite sure if make sense to read this reg and apply
setting on the top of it.

Nicolas: Isn't it better simply compose that reg from scratch?

Thanks,
Michal