From mboxrd@z Thu Jan  1 00:00:00 1970
From: Alexander Duyck <alexander.duyck@gmail.com>
Date: Fri, 22 May 2015 11:29:07 -0700
Subject: [Intel-wired-lan] [PATCH] igb: report unsupported ethtool
 settings in set_coalesce
In-Reply-To: <9B4A1B1917080E46B64F07F2989DADD65354DCC3@ORSMSX114.amr.corp.intel.com>
References: <20150522174950.16145.78447.stgit@htfujina-fc.jf.intel.com>
 <555F70BE.6090904@redhat.com>
 <9B4A1B1917080E46B64F07F2989DADD65354DCC3@ORSMSX114.amr.corp.intel.com>
Message-ID: <555F7573.9040704@gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
To: intel-wired-lan@osuosl.org
List-ID: <intel-wired-lan.osuosl.org>

I think I see what you are talking about.  They used s32 in the ethtool 
do_scoalesce call, and pointed the void pointer wanted_val at it which 
is how they did the conversion.

You might want to just == ~0 or ~(ec->...) instead to test the values.

- Alex

On 05/22/2015 11:10 AM, Fujinaka, Todd wrote:
> Those are unsigned? The code for ethtool set them all to -1.
>
> Todd Fujinaka
> Software Application Engineer
> Networking Division (ND)
> Intel Corporation
> todd.fujinaka at intel.com
> (503) 712-4565
>
> -----Original Message-----
> From: Alexander Duyck [mailto:alexander.h.duyck at redhat.com]
> Sent: Friday, May 22, 2015 11:09 AM
> To: Fujinaka, Todd; intel-wired-lan at lists.osuosl.org
> Subject: Re: [Intel-wired-lan] [PATCH] igb: report unsupported ethtool settings in set_coalesce
>
>
>
> On 05/22/2015 10:49 AM, Todd Fujinaka wrote:
>> There are many settings possible using ethtool -C/--coalesce, but not
>> all of them are supported. Report failure when an unsupported option
>> is set.
>>
>> Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com>
>> ---
>>    drivers/net/ethernet/intel/igb/igb_ethtool.c |   21 +++++++++++++++++++++
>>    1 file changed, 21 insertions(+)
>>
>> diff --git a/drivers/net/ethernet/intel/igb/igb_ethtool.c
>> b/drivers/net/ethernet/intel/igb/igb_ethtool.c
>> index 109cad9..13560fe 100644
>> --- a/drivers/net/ethernet/intel/igb/igb_ethtool.c
>> +++ b/drivers/net/ethernet/intel/igb/igb_ethtool.c
>> @@ -2159,6 +2159,27 @@ static int igb_set_coalesce(struct net_device *netdev,
>>    	struct igb_adapter *adapter = netdev_priv(netdev);
>>    	int i;
>>
>> +	if ((ec->rx_max_coalesced_frames != -1) ||
>> +	    (ec->rx_coalesce_usecs_irq != -1) ||
>> +	    (ec->rx_max_coalesced_frames_irq != -1) ||
>> +	    (ec->tx_max_coalesced_frames != -1) ||
>> +	    (ec->tx_coalesce_usecs_irq != -1) ||
>> +	    (ec->stats_block_coalesce_usecs != -1) ||
>> +	    (ec->use_adaptive_rx_coalesce != -1) ||
>> +	    (ec->use_adaptive_tx_coalesce != -1) ||
>> +	    (ec->pkt_rate_low != -1) ||
>> +	    (ec->rx_coalesce_usecs_low != -1) ||
>> +	    (ec->rx_max_coalesced_frames_low != -1) ||
>> +	    (ec->tx_coalesce_usecs_low != -1) ||
>> +	    (ec->tx_max_coalesced_frames_low != -1) ||
>> +	    (ec->pkt_rate_high != -1) ||
>> +	    (ec->rx_coalesce_usecs_high != -1) ||
>> +	    (ec->rx_max_coalesced_frames_high != -1) ||
>> +	    (ec->tx_coalesce_usecs_high != -1) ||
>> +	    (ec->tx_max_coalesced_frames_high != -1) ||
>> +	    (ec->rate_sample_interval != -1))
>> +		return -ENOTSUPP;
>> +
>>    	if ((ec->rx_coalesce_usecs > IGB_MAX_ITR_USECS) ||
>>    	    ((ec->rx_coalesce_usecs > 3) &&
>>    	     (ec->rx_coalesce_usecs < IGB_MIN_ITR_USECS)) ||
>>
> Shouldn't these tests all give you a signed/unsigned mismatch since you are comparing an unsigned 32 bit value versus a signed value?
>
> - Alex
> _______________________________________________
> Intel-wired-lan mailing list
> Intel-wired-lan at lists.osuosl.org
> http://lists.osuosl.org/mailman/listinfo/intel-wired-lan