netdev - Re: [PATCH net-next 5/6] ionic: use mc sync for multicast filters

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: <694cb96c-1a88-289e-4f5b-25e29ae1a433@pensando.io>
Date:   Thu, 5 Nov 2020 10:50:00 -0800
From:   Shannon Nelson <snelson@...sando.io>
To:     Saeed Mahameed <saeed@...nel.org>, netdev@...r.kernel.org,
        davem@...emloft.net, kuba@...nel.org
Subject: Re: [PATCH net-next 5/6] ionic: use mc sync for multicast filters

On 11/4/20 5:18 PM, Saeed Mahameed wrote:
> On Wed, 2020-11-04 at 14:33 -0800, Shannon Nelson wrote:
>> We should be using the multicast sync routines for the
>> multicast filters.
>>
>> Fixes: 1800eee16676 ("net: ionic: Replace in_interrupt() usage.")
>> Signed-off-by: Shannon Nelson <snelson@...sando.io>
>> ---
>>   drivers/net/ethernet/pensando/ionic/ionic_lif.c | 10 +++++++++-
>>   1 file changed, 9 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/net/ethernet/pensando/ionic/ionic_lif.c
>> b/drivers/net/ethernet/pensando/ionic/ionic_lif.c
>> index 28044240caf2..a58bb572b23b 100644
>> --- a/drivers/net/ethernet/pensando/ionic/ionic_lif.c
>> +++ b/drivers/net/ethernet/pensando/ionic/ionic_lif.c
>> @@ -1158,6 +1158,14 @@ static void ionic_dev_uc_sync(struct
>> net_device *netdev, bool from_ndo)
>>   
>>   }
>>   
>> +static void ionic_dev_mc_sync(struct net_device *netdev, bool
>> from_ndo)
>> +{
>> +	if (from_ndo)
>> +		__dev_mc_sync(netdev, ionic_ndo_addr_add,
>> ionic_ndo_addr_del);
>> +	else
>> +		__dev_mc_sync(netdev, ionic_addr_add, ionic_addr_del);
>> +}
>> +
> I don't see any point of this function since it is used in one place.
> just unfold it in the caller and you will endup with less code.
>
> also keep in mind passing boolean to functions is usually a bad idea,
> and only complicates things, keep things simple and explicit, let the
> caller do what is necessary to be done, so if you must do this if
> condition, do it at the caller level.
>
> and for a future patch i strongly recommend to remove this from_ndo
> flag, it is really straight forward to do for this function
> 1) you can just pass _addr_add/del function pointers directly
> to ionic_set_rx_mode
> 2) remove _ionic_lif_rx_mode from ionic_set_rx_mode and unfold it in
> the caller since the function is basically one giant if condition which
> is called only from two places.

This was specifically following work that was done a couple of weeks ago 
by Thomas Gleixner et al to clean up questionable uses of 
in_interrupt(), similar to how they used booleans to patch mlx5 and 
other drivers.  They split this out, but later I noticed this issue with 
how multicast got handled.  I agree, I'm not thrilled with the new 
booleans either, which is part of the reason for patch 6/6 in this series.

Yes, I can pull these back into ionic_set_rx_mode() for a v2 patch, 
which will clean up some of this.

I'll look at those future patch ideas: (2) is easy enough and I might 
just add it to this patch series, but I'm not sure about (1) yet, partly 
because I like the current separation of knowledge.

Thanks,
sln

>
>>   static void ionic_set_rx_mode(struct net_device *netdev, bool
>> from_ndo)
>>   {
>>   	struct ionic_lif *lif = netdev_priv(netdev);
>> @@ -1189,7 +1197,7 @@ static void ionic_set_rx_mode(struct net_device
>> *netdev, bool from_ndo)
>>   	}
>>   
>>   	/* same for multicast */
>> -	ionic_dev_uc_sync(netdev, from_ndo);
>> +	ionic_dev_mc_sync(netdev, from_ndo);
>>   	nfilters = le32_to_cpu(lif->identity->eth.max_mcast_filters);
>>   	if (netdev_mc_count(netdev) > nfilters) {
>>   		rx_mode |= IONIC_RX_MODE_F_ALLMULTI;