[ewg] Re: [ofa-general] build failure on nightly tarball -- bonding

[Moni Shoua]

Vladimir Sokolovsky wrote:
> Moni Shoua wrote:
>> Moni Shoua wrote:
>>  
>>> Andrew Friedley wrote:
>>>    
>>>> The chelsio build errors from yesterday appear to be gone, though now
>>>> I'm seeing errors building the IB bonding code with the 3/2 alpha
>>>> tarball -- error below.  I'm wondering, is there a way to selectively
>>>> avoid building things like this that seem to be optional, as a tarball
>>>> user?
>>>>
>>>> Andrew
>>>>       
>>> For the error messages.... It seems to me that the problem is one
>>> that I have already fixed.
>>> The corrected source RPM is in my home dir.
>>>     
>> Hi Vlad,
>>
>> 1. Is there a way to tell OFED build/install not to compile bonding?
>>   
> Yes, see build_env.sh:
> 
> 216 # Add bonding support for RH4.0U3 and SLES10
> 217 case ${K_VER} in
> 218     2.6.16.*-*-*|2.6.9-34.EL*)
> 219         IB_ALL_PACKAGES="$IB_ALL_PACKAGES ib-bonding"
> 220     ;;
> 221 esac
> ...
> 2049                     ib_bonding)
> 2050                         # Add bonding support for RH4.0U3 and SLES10
> 2051                         case ${K_VER} in
> 2052                             2.6.16.*-*-*|2.6.9-34.EL*)
> 2053                                 EXTRA_PACKAGES=$(echo
> "$EXTRA_PACKAGES ib-bonding" | tr -s ' ' '\n' | sort -n | uniq)
> 2054                             ;;
> 2055                             *)
> 2056                                 warn_echo "ib-bonding is not
> supported on this kernel"
> 2057                             ;;
> 2058                         esac
> 2059                     ;;
> 
So I understand that ib-bonding will compile and install if kernel is in the list
of supported kernels. Am I right?
Isn't there a a flag in ofed.conf (or something like that)?
>> 2. Do you automatically take the source RPM that's in the latest.txt
>> when you create an OFED source package?
>>
> Yes, OFED's build script takes ib-bonding from
> http://www.openfabrics.org/~monis/ofed_1_2/latest.txt
> 
Great! Thanks.
> Regards,
> Vladimir
>