On 2/8/2012 9:50 AM, Eric Dumazet wrote:
Le mercredi 08 février 2012 à 07:29 +, Hefty, Sean a écrit :
I tested this with Dave's patch and Eric's first patch against Linus' latest
tree 3.3-rc2+, and things look good so far.
Thanks for testing, I'll resend my (updated) patch today.
same
This is OFED bug 1974, it's been open for a couple of years now. Ira, I think
you're the ib-diags guy now, right?
In any case, it's a trivial issue: ibping -S treats receive errors as if they
were valid incoming MAD packets and tries to send responses to them. The fix is
easy, just check the st
On Wed, 8 Feb 2012 06:34:47 -0800
Mike Heinz wrote:
> This is OFED bug 1974, it's been open for a couple of years now. Ira, I think
> you're the ib-diags guy now, right?
>
> In any case, it's a trivial issue: ibping -S treats receive errors as if they
> were valid incoming MAD packets and trie
Absolutely.
Sorry, I knocked the patch out quickly because I came across it while looking
at another problem, next time I'll be sure to do the full procedure.
-Original Message-
From: Ira Weiny [mailto:wei...@llnl.gov]
Sent: Wednesday, February 08, 2012 1:20 PM
To: Mike Heinz
Cc: linux-r
Shlomo Pongratz reported GRO L2 header check was suited for Ethernet
only, and failed on IB/ipoib traffic.
He provided a patch faking a zeroed header to let GRO aggregates frames.
Roland Dreier, Herbert Xu, and others suggested we change GRO L2 header
check to be more generic, ie not assuming L2
From: Roland Dreier
Date: Tue, 7 Feb 2012 16:51:21 -0800
> From: Roland Dreier
>
> Commit a0417fa3a18a ("net: Make qdisc_skb_cb upper size bound
> explicit.") made it possible for a netdev driver to use skb->cb
> between its header_ops.create method and its .ndo_start_xmit
> method. Use this
From: Eric Dumazet
Date: Wed, 08 Feb 2012 19:51:50 +0100
> Shlomo Pongratz reported GRO L2 header check was suited for Ethernet
> only, and failed on IB/ipoib traffic.
>
> He provided a patch faking a zeroed header to let GRO aggregates frames.
>
> Roland Dreier, Herbert Xu, and others suggeste
On Wed, Feb 8, 2012 at 10:50 PM, David Miller wrote:
> From: Eric Dumazet
> > Shlomo Pongratz reported GRO L2 header check was suited for Ethernet
> > only, and failed on IB/ipoib traffic.
> > He provided a patch faking a zeroed header to let GRO aggregates frames.
> >
> > Roland Dreier, Herbert
On Thu, Feb 2, 2012 at 7:08 PM, Roland Dreier wrote:
> On Thu, Feb 2, 2012 at 3:12 AM, Eric Dumazet wrote:
>> Any news on this patch ?
> Sorry, just dropped it in the shuffle. I'll get it into 3.4, thanks.
Roland,
I noted that you typically use the for-next branch of the infiniband
tree for
From: Or Gerlitz
Date: Wed, 8 Feb 2012 23:08:26 +0200
> On Wed, Feb 8, 2012 at 10:50 PM, David Miller wrote:
>> From: Eric Dumazet
>> > Shlomo Pongratz reported GRO L2 header check was suited for Ethernet
>> > only, and failed on IB/ipoib traffic.
>> > He provided a patch faking a zeroed header
On Wed, Feb 8, 2012 at 1:26 PM, Or Gerlitz wrote:
> I noted that you typically use the for-next branch of the infiniband
> tree for fixes during
> the 1 < kernN-rc < (say) 6 time and for features during (kernN-rc > 6)
> till kern(N+1)-rc1
>
> This means that the window of time when features are ac
On Wed, Feb 8, 2012 at 11:31 PM, David Miller wrote:
> From: Or Gerlitz
>> Hi Dave, for correct operation / future bisection, you should 1st
>> apply Roland's patch which reduces the hard header len advertized by
>> ipoib to be only the size of the ipoib header without that 20 bytes
>> headroom,
On Wed, Feb 8, 2012 at 11:35 PM, Roland Dreier wrote:
> On Wed, Feb 8, 2012 at 1:26 PM, Or Gerlitz wrote:
>> [...] This means that the window of time when features are actually accepted
>> into your tree is kind of very limited. Would it be possible to
>> maintain two branches: for-next and (say
So, take a set of 3 or 4 hosts on a fabric, and run
# ibping -S
on each.
Then, on each, run the ibping client in a loop so that each host sends a few
packets to each other host. For example:
# while true; do echo; date; echo; ibping -c 3 -L 3; ibping -c 3 -L 5; ibping
-c 3 -L 1; sleep 1; don
Hi Roland,
referring to the function below, I can't find a strong reason to check
again after acquiring the cq spinlock. Any idea?
static int mlx4_wq_overflow(struct mlx4_ib_wq *wq, int nreq, struct
ib_cq *ib_cq)
{
unsigned cur;
struct mlx4_ib_cq *cq;
cur = wq->head - wq-
On Thu, Feb 9, 2012 at 12:41 AM, Mike Heinz wrote:
> This behavior has been demonstrated on 1.5.4 and 1.5.4.1, and on RHEL6.0, 6.1
> and 5.7
tried kernel.org? this list deals with kernels from that fabric.
Or.
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body
From: Or Gerlitz
Date: Wed, 8 Feb 2012 23:49:43 +0200
> On Wed, Feb 8, 2012 at 11:31 PM, David Miller wrote:
>> From: Or Gerlitz
>
>>> Hi Dave, for correct operation / future bisection, you should 1st
>>> apply Roland's patch which reduces the hard header len advertized by
>>> ipoib to be only
On Tue, 7 Feb 2012, Hugh Dickins wrote:
> On Mon, 6 Feb 2012, Roland Dreier wrote:
> > Which I think explains why the code is the way it is. But clearly
> > we could do better if we had a better way of telling GUP our real
> > intentions -- ie the FOLL_READONLY_COW flag.
>
> You've persuaded me.
On Thu, Feb 9, 2012 at 1:09 AM, David Miller wrote:
> I don't think this is an appropriate bug fix at all.
I'm not sure to understand
> Apparently this problem has existed since day one and the world has
> kept on spinning meanwhile.
Dave, the bug was introduced on 2.6.38 when LRO was removed f
From: Or Gerlitz
Date: Thu, 9 Feb 2012 01:20:46 +0200
> On Thu, Feb 9, 2012 at 1:09 AM, David Miller wrote:
>> I don't think this is an appropriate bug fix at all.
>
> I'm not sure to understand
>
>> Apparently this problem has existed since day one and the world has
>> kept on spinning meanwh
> Subject: Possibly serious bug in ib_mad: processing packets from ibping can
> consume 100% of CPU and may leave user processes locked in umad_recv
Is there a chance that the low level driver is asking to send the MAD to itself
over and over again?
--
To unsubscribe from this list: send the line
21 matches
Mail list logo
