[ewg] RE: What is the branch that should be used formanagement packagefor OFED 1.4.1?

[Yevgeny Kliteynik]

Hi Sasha,

1 is rare, that's true, but it may happen in any single switch
fabric, such as some evaluation fabric (the last place you want
to see bugs).

3 is limited to cache, but it's completing num. 2.

4 and 6 are logging issues (debugging issues), so we can leave
them aside (although the fixes are very "local").

5 jumps right at you when you run opensm with -Q.
It doesn't break anything, but from a user perspective it's
not minor at all.

And the last - the new bug #1515.
This is a serious bug, but we can discuss this bug severity
(as well as the fix) on the openib list.

-- Yevgeny

> -----Original Message-----
> From: Sasha Khapyorsky [mailto:sashak at voltaire.com] 
> 
> Hi Tziporet,
> 
> On 17:05 Tue 17 Feb     , Tziporet Koren wrote:
> > Yevgeny Kliteynik wrote:
> >> Hi Sasha,
> >>
> >> Below is the list of the OpenSM commits that should go 
> into OFED 1.4.1:
> >>
> >> 1. Commit aa25fcba61e75ac94af2b950c200cdde433f110c
> >>    Don't clear sw->need_update if port 0 is active.
> >>    In a single switch subnet, after a fast switch reset, OpenSM
> >>    doesn't configure LFT of the switch and the routing is lost.
> >>    Bug #1507.
> >>
> >> 2. Commit 595f2e30026f9ec85674e6d70661d26de6f77a6b
> >>    Update LFTs when entering master state
> >>    Bug #1469.
> >>
> >> 3. Commit b44c398e7363e9c9605e9ee1d5893f872ef41feb
> >>    Invalidate routing cache when entering master state
> >>    Bug #1508
> >>
> >> 4. Commit 22da81f8cd432195f6c78cfe5f508f04b081aef7
> >>    Fix full topology dump in ftree routing.
> >>    Topology dump was missing all the leaf switches
> >>    Bug #1509
> >>
> >> 5. Commit c07d245651035cc7ba3b89ac0e81d54bff829f4d
> >>    Don't complain about invalid QoS options when its
> >>    default values are used.
> >>    Bug #1451.
> >>
> >> 6. Commit 72a2fa241dffacc2eea3c2b0c131f8bcb4792162
> >>    Dump SA MAD before sending it, not after.
> >>    Dumping of the SA MAD was done after the MAD was sent
> >>    and returned to the pool, resulting in garbage in
> >>    the OpenSM log.
> >>    Bug #1510.
> >>
> >>   
> >
> > Sasha
> > Please open the branch and include the above pathces.
> > We also wish the new bug fixed (#1515)
> 
> In order to minimize amount of changes which should go to dot 
> release I would suggest to review the list above - 4-6 are 
> minor IMHO, 1 is rear (although the fix is safe), 3 is 
> limited by cache use.
>
> Sasha
>