[ofa-general] RE: [ofw] Re: saquery & osm vendor IBAL - ca_names missing from osm_vendor_t ?

[Smith, Stan]

Sasha Khapyorsky wrote:
> Hello Stan,
>
> On 13:16 Mon 09 Feb     , Smith, Stan wrote:
>>
>> My proposal for the Windows OpenSM code base is to add
>> ca_names[UMAD_MAX_DEVICES][UMAD_CA_NAME_LEN] to OpenSM vendor-ibal
>> definition of osm_vendor_t and a call to umad_get_cas_names() to
>> populate the osm_vendor_t.ca_names struct for.
>>
>> Comments?
>
> Assuming WinOF already has libibumad implementation with preserved API
> would it be reasonable to switch from vendor-ibal to vendor-ibumad in
> WinOF?
>
> Sasha

Hello,

Path of least resistance thinking would point towards not doing a switch as the vendor-ibal to vendor-ibumad would be part of the Windows OpenSM migration to OFED 1.4x OpenSM.
My thinking is that making a switch to vendor-ibumad would be a good deal more work/involved just to get saquery working.
Not knowing the Windows OpenSM code base, moving part of it forward seems like a task 'which' could blossom into a good deal more work for the small return of saquery working?
Frankly, I'd rather see work put into getting OFED OpenSM 1.4 working on Windows.

Just my $0.02 worth.

Stan.