Hi.
Bharath Ramesh wrote:
I am getting this error when I am trying to do a bunch of send/receives.
I have registered the receive buffer. I printed the address of the
buffers and their respective lkeys, they all match but I am still
getting this error.
The code snippet looks as follows:
struct
> Quoting Hoang-Nam Nguyen <[EMAIL PROTECTED]>:
> Subject: ofed-1.3 daily build package's content
>
> Hello Vlad and Michael!
> Just downloaded daily build package OFED-1.3-20070917-0600 and saw
> in SRPMS:
> localhost:/home/nguyen/tmp/OFED-1.3-20070917-0600
> Quoting Roland Dreier <[EMAIL PROTECTED]>:
> Subject: Re: mlx4 violating radix tree API locking rules?
>
> By the way, in the past we've gotten push-back against using RCU in
> dual GPL/BSD code.
I don't know what official Mellanox position is on this.
I think that in this case it's OK, simply
OSM Simulation Regression Summary
[Generated mail - please do NOT reply]
OpenSM binary date = 2007-09-17
OpenSM git rev = Sun_Sep_16_18:47:46_2007
[8224cc5e3f6e5ce03d783e674b4eaa6e1cf37acd]
ibutils git rev = Tue_Sep_4_17:57:34_2007
[4bf283f6a0d7c0264c3a1d2de92745e457585fdb]
Total=520 P
I am getting this error when I am trying to do a bunch of send/receives.
I have registered the receive buffer. I printed the address of the
buffers and their respective lkeys, they all match but I am still
getting this error.
The code snippet looks as follows:
struct ibv_mr *mr;
struct ibv_sge sg
Hi Sasha,
On Tue, 2007-09-18 at 02:58 +0200, Sasha Khapyorsky wrote:
> Hi Hal,
>
> On 11:04 Tue 11 Sep , Hal Rosenstock wrote:
> > OpenSM: Improve QP0 and QP1 counter accounting
> >
> > Note: Patch is based on OFED 1.2
>
> Well, you know :)
>
> The question is below.
>
> >
> > Signed-off
On 16:59 Mon 17 Sep , Hal Rosenstock wrote:
> OpenSM/main.c: Fix compile error
>
> Signed-off-by: Hal Rosenstock <[EMAIL PROTECTED]>
Nice catch.
Applied. Thanks.
Sasha
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics
Hi Hal,
On 11:04 Tue 11 Sep , Hal Rosenstock wrote:
> OpenSM: Improve QP0 and QP1 counter accounting
>
> Note: Patch is based on OFED 1.2
Well, you know :)
The question is below.
>
> Signed-off-by: Hal Rosenstock <[EMAIL PROTECTED]>
>
> diff --git a/osm/include/opensm/osm_sa.h b/osm/incl
OpenSM/main.c: Fix compile error
Signed-off-by: Hal Rosenstock <[EMAIL PROTECTED]>
diff --git a/opensm/opensm/main.c b/opensm/opensm/main.c
index 9597cf1..0005531 100644
--- a/opensm/opensm/main.c
+++ b/opensm/opensm/main.c
@@ -1043,7 +1043,7 @@ int main(int argc, char *argv[])
Roland Dreier <[EMAIL PROTECTED]> wrote:
>Actually, thinking about this some more... would it be cleaner to more
>the knowledge about bonding out of the ipoib driver? in other words,
>export something similar to
>
> > +static int ipoib_slave_detach(struct net_device *dev)
> > +{
> > + int ret =
On 11:03 Tue 11 Sep , Hal Rosenstock wrote:
> OpenSM/(osm_console main).c: Support loopback option to -console for
> local only telnet support
>
> Note: Patch is based on OFED 1.2
>
> Signed-off-by: Hal Rosenstock <[EMAIL PROTECTED]>
Applied. Thanks.
Please next time generate patches again
Roland Dreier wrote:
Thanks for the explanation...
> But basically, with CONFIG_PREEMPT_RT enabled, the lock points, such as
> aqcuiring a spinlock, potentially become places where the current task
> may be context switched out / preempted.
>
> Therefore, when a call is made to lock a spin
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
> Conceptually, I see your point and I'm ok with doing it either
> way. My only question is, would this change would make the ipoib module
> dependent upon having the bonding module loaded (to resolve all of the
> symbols)?
Yes, I guess so, if that function is in bonding. Hmm, that woul
On 17:36 Thu 13 Sep , Hal Rosenstock wrote:
> ibnetdiscover: Support Xsigo chassis grouping
>
> I think this also fixes a bug with grouping of multiple non Voltaire
> chassis as well.
>
> Note: this patch is against OFED 1.2
>
> Signed-off-by: Hal Rosenstock <[EMAIL PROTECTED]>
Applied. Tha
On 15:38 Mon 17 Sep , Hal Rosenstock wrote:
> On Tue, 2007-09-18 at 00:40 +0200, Sasha Khapyorsky wrote:
> > On 15:08 Mon 17 Sep , Hal Rosenstock wrote:
> > > Hi Sasha,
> > >
> > > On Mon, 2007-09-17 at 23:51 +0200, Sasha Khapyorsky wrote:
> > > > Hi Hal,
> > > >
> > > > On 17:36 Thu 13 S
obviously OK...applied.
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
> The IGMP enabling patch posted by me on September 2nd isn't on your list
> http://lists.openfabrics.org/pipermail/general/2007-September/040250.html
> can you add it?
Yes, I lost that somehow. I will add it to my list of things to take
a look at (no opinion yet).
- R.
_
On Tue, 2007-09-18 at 00:40 +0200, Sasha Khapyorsky wrote:
> On 15:08 Mon 17 Sep , Hal Rosenstock wrote:
> > Hi Sasha,
> >
> > On Mon, 2007-09-17 at 23:51 +0200, Sasha Khapyorsky wrote:
> > > Hi Hal,
> > >
> > > On 17:36 Thu 13 Sep , Hal Rosenstock wrote:
> > > > ibnetdiscover: Support Xs
On 00:23 Tue 18 Sep , Yevgeny Kliteynik wrote:
> Checking memory allocation status is qos
> policy parser to prevent seg. fault.
>
> Signed-off-by: Yevgeny Kliteynik <[EMAIL PROTECTED]>
Applied. Thanks.
Sasha
___
general mailing list
general@lists.
On 15:08 Mon 17 Sep , Hal Rosenstock wrote:
> Hi Sasha,
>
> On Mon, 2007-09-17 at 23:51 +0200, Sasha Khapyorsky wrote:
> > Hi Hal,
> >
> > On 17:36 Thu 13 Sep , Hal Rosenstock wrote:
> > > ibnetdiscover: Support Xsigo chassis grouping
> > >
> > > I think this also fixes a bug with groupi
>My first reaction would be to stick them somewhere in debugfs. (I'm
>assuming this feature is for diagnostics etc)
I'm looking into using debugfs, since the counters are mainly for diagnostics,
but are these much different than exposing counters through:
/sys/class/infiniband/mthca0/ports/1/cou
Actually, thinking about this some more... would it be cleaner to more
the knowledge about bonding out of the ipoib driver? in other words,
export something similar to
> +static int ipoib_slave_detach(struct net_device *dev)
> +{
> +int ret = 0;
> +if (dev->flags & IFF_SLAVE) {
> +
On 21:06 Mon 17 Sep , Yevgeny Kliteynik wrote:
> M_Key lease period description should
> be in [sec] instead of [msec].
>
> Signed-off-by: Yevgeny Kliteynik <[EMAIL PROTECTED]>
Applied. Thanks.
Sasha
___
general mailing list
general@lists.openfabri
By the way, in the past we've gotten push-back against using RCU in
dual GPL/BSD code. I have no problem relicensing mlx4 to GPL-only and
then sticking in the rcu_read_lock() stuff to handle this I guess.
- R.
___
general mailing list
general@lists.ope
Sy!m b-oool
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
Hi Sasha,
Should the default for valgrind be off (in libibumad) ? It seems to be
the other way around:
src/umad.c:59:6: warning: #warning "Valgrind support requested, but
VALGRIND_MAKE_MEM_DEFINED not available"
-- Hal
___
general mailing list
general@
>One thing though, if the subnet local = 0 and the incoming hop limit
><= 1 a GRH will still not be used. Your patch should probably also
>force the hop limit to 2 in this case and include a note for later
>fixup..
I think this case would be better handled by rejecting the REQ.
- Sean
___
TrapRepress always had mkey 0, which was copied from trap
notice's mkey (which is always 0).
As a result, TrapRepress was failing for port with mkey != 0
Signed-off-by: Yevgeny Kliteynik <[EMAIL PROTECTED]>
---
opensm/opensm/osm_trap_rcv.c | 10 ++
1 files changed, 10 insertions(+), 0 d
Copied from web link. I didn't have this in my inbox anymore.
I should also mention that I'm not up on all of the IPoIB RFCs, so if
some of my comments don't apply to an RFC, just ignore them. :)
Most of these comments are about code organization, to make the with and
without srq code cleaner
Checking memory allocation status is qos
policy parser to prevent seg. fault.
Signed-off-by: Yevgeny Kliteynik <[EMAIL PROTECTED]>
---
opensm/opensm/osm_qos_parser.y |7 ---
1 files changed, 4 insertions(+), 3 deletions(-)
diff --git a/opensm/opensm/osm_qos_parser.y b/opensm/opensm/osm_q
[EMAIL PROTECTED] wrote on 09/17/2007 02:47:42 PM:
> > > IPoIB CM handles this properly by gathering together single pages in
> > > skbs' fragment lists.
>
> > Then can we reuse IPoIB CM code here?
>
> Yes, if possible, refactoring things so that the rx skb allocation
> code becomes common
Overall idea looks good... one comment:
> +if (n->dev->flags & IFF_MASTER) {
> +/* n->dev is not an IPoIB device and we have
> +to take priv from elsewhere */
> +neigh = *to_ipoib_neigh(n);
> +if (neigh) {
> +pri
OK with me.
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
Looks fine overall, with one minor nitpick:
> -if (unlikely(memcmp(&neigh->dgid.raw,
> +if (unlikely((memcmp(&neigh->dgid.raw,
> skb->dst->neighbour->ha + 4,
> -sizeof(union
I tried to look at the ipoib stuff in this series... this patch looks
fine but it doesn't actually touch ipoib, so the subject line is a bit
misleading...
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/
Jim Hall wrote:
Is it valid to have a CM request message with subnet local = 1 and hop
limit > 1?
I don't see this explicitly called out as invalid, even if it doesn't
make any sense. As a related note, there are a reject codes for invalid
hop limit and TC. I guess we have a choice between
> Note that some people only run
> backported drivers, so making it easier to read and maintain
> *the backport* is also important.
The philosophy of the kernel has always been that the backport needs
to bear the cost, and we don't want to add extra #ifdefs to the
standard kernel that don't do
Hal Rosenstock wrote:
Hi Yevgeny,
Is the failure below a simulator or OpenSM issue ? Thanks.
Not sure yet. I'll try to recreate the failure and update you with more
info.
-- Yevgeny
-- Hal
Forwarded Message
From: [EMAIL PROTECTED]
To: [EMAIL PROTECTED]
Cc: general@lists
> Why not just call synchronize_rcu instead?
Not sure I understand. Where would you put the synchronize_rcu and
what would it protect against? RCU is being used to protect the radix
tree internals, not the mlx4 data structures.
> > I guess CQ spinlock implies rcu_read_lock - is that right?
>
Hi Sasha,
On Mon, 2007-09-17 at 23:51 +0200, Sasha Khapyorsky wrote:
> Hi Hal,
>
> On 17:36 Thu 13 Sep , Hal Rosenstock wrote:
> > ibnetdiscover: Support Xsigo chassis grouping
> >
> > I think this also fixes a bug with grouping of multiple non Voltaire
> > chassis as well.
>
> Could you pr
On 21:23 Mon 17 Sep , Yevgeny Kliteynik wrote:
> QoS policy optimization: replaced node types with
> single bitmask, and node_type_self is implemented
> as an additional guid in group's port map.
>
> Signed-off-by: Yevgeny Kliteynik <[EMAIL PROTECTED]>
Applied. Thanks.
Sasha
On 21:22 Mon 17 Sep , Yevgeny Kliteynik wrote:
> QoS policy optimization: replacing partition list and guid
> ranges in a port group by port map indexed by port guid.
> The port map is filled at parse time, thus checking whether
> some port belongs to a group becomes a single map query.
>
> Si
Hi Yevgeny,
Small comment is below.
On 21:22 Mon 17 Sep , Yevgeny Kliteynik wrote:
> QoS policy optimization: replacing partition list and guid
> ranges in a port group by port map indexed by port guid.
> The port map is filled at parse time, thus checking whether
> some port belongs to a gro
On Mon, Sep 17, 2007 at 02:29:15PM -0500, Jim Hall wrote:
> Is it valid to have a CM request message with subnet local = 1 and hop
> limit > 1?
Oop, I hadn't considered the subnet local field. That should probably
unconditionally control the GRH, like your patch.
Hmm, you know, there is complian
> > IPoIB CM handles this properly by gathering together single pages in
> > skbs' fragment lists.
> Then can we reuse IPoIB CM code here?
Yes, if possible, refactoring things so that the rx skb allocation
code becomes common between CM and non-CM would definitely make sense.
_
Hi Hal,
On 17:36 Thu 13 Sep , Hal Rosenstock wrote:
> ibnetdiscover: Support Xsigo chassis grouping
>
> I think this also fixes a bug with grouping of multiple non Voltaire
> chassis as well.
Could you provide more details about this bug. Should this be a separate
patch?
> Note: this patch
Thanks for the explanation...
> But basically, with CONFIG_PREEMPT_RT enabled, the lock points, such as
> aqcuiring a spinlock, potentially become places where the current task
> may be context switched out / preempted.
>
> Therefore, when a call is made to lock a spinlock for example, the
On 22:20 Mon 17 Sep , Sasha Khapyorsky wrote:
> Hi Hal,
>
> On 12:47 Mon 17 Sep , Hal Rosenstock wrote:
> > Hi Sasha,
> >
> > Should the default for valgrind be off (in libibumad) ? It seems to be
> > the other way around:
> >
> > src/umad.c:59:6: warning: #warning "Valgrind support requ
Hi Hal,
On 12:47 Mon 17 Sep , Hal Rosenstock wrote:
> Hi Sasha,
>
> Should the default for valgrind be off (in libibumad) ? It seems to be
> the other way around:
>
> src/umad.c:59:6: warning: #warning "Valgrind support requested, but
> VALGRIND_MAKE_MEM_DEFINED not available"
Hmm, it is "o
On 21:09 Mon 17 Sep , Yevgeny Kliteynik wrote:
> TrapRepress always had mkey 0, which was copied from trap
> notice's mkey (which is always 0).
> As a result, TrapRepress was failing for port with mkey != 0
>
> Signed-off-by: Yevgeny Kliteynik <[EMAIL PROTECTED]>
Applied. Thanks.
Sasha
_
Is it valid to have a CM request message with subnet local = 1 and hop limit
> 1?
If so, then it's not clear in the spec on how each side of the QP creation
are to determine if they use GRHs or not.
Can Mellanox be configured to accept both (LRH,LRH_GRH) on a RC QP at the
same time?
- Jim
---
QoS policy optimization: replaced node types with
single bitmask, and node_type_self is implemented
as an additional guid in group's port map.
Signed-off-by: Yevgeny Kliteynik <[EMAIL PROTECTED]>
---
opensm/include/opensm/osm_qos_policy.h |9 +
opensm/opensm/osm_qos_parser.y |
QoS policy optimization: replacing partition list and guid
ranges in a port group by port map indexed by port guid.
The port map is filled at parse time, thus checking whether
some port belongs to a group becomes a single map query.
Signed-off-by: Yevgeny Kliteynik <[EMAIL PROTECTED]>
---
opensm/
On Mon, Sep 17, 2007 at 11:39:46AM -0700, Sean Hefty wrote:
> >I'm with Hal on this - why does this cause a problem? There is no IB
> >packet verification check that tests if a GRH is present, only if it
> >is presen it must be valid - so how can an extra correctly filled in
> >GRH cause anything b
M_Key lease period description should
be in [sec] instead of [msec].
Signed-off-by: Yevgeny Kliteynik <[EMAIL PROTECTED]>
---
opensm/opensm/osm_subnet.c |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/opensm/opensm/osm_subnet.c b/opensm/opensm/osm_subnet.c
index 3895732.
The issue is between the active CM (in this case Solaris) and passive OFED.
The SM doesn't look to be involved.
- Original Message -
From: "Sean Hefty" <[EMAIL PROTECTED]>
To: "Jim Hall" <[EMAIL PROTECTED]>
Cc: "Sean Hefty" <[EMAIL PROTECTED]>;
Sent: Monday, September 17, 2007 1:01 PM
Fix iba/*.h installation path.
Signed-off-by: Sasha Khapyorsky <[EMAIL PROTECTED]>
---
opensm/include/Makefile.am |2 ++
1 files changed, 2 insertions(+), 0 deletions(-)
diff --git a/opensm/include/Makefile.am b/opensm/include/Makefile.am
index fc2d7ca..5c41126 100644
--- a/opensm/include/M
>But using this hop limit this was is just a convention we have
>adopted. There is nothing in IBA that allows the active side to send a
>REQ with invalid GIDs/etc in any situation. The passive side should
>always be able to form a correct GRH based on any REQ.
I don't think TC and FL need to be va
>I'm with Hal on this - why does this cause a problem? There is no IB
>packet verification check that tests if a GRH is present, only if it
>is presen it must be valid - so how can an extra correctly filled in
>GRH cause anything but degraded performance?
ib_init_ah_from_path() uses the hop_limit
The problem arises when the active Solaris client is sending a
connection request to a passive OFED server instance. Solaris will set
the hop_limit field to 0xFF and will not expect or enable GRH routing.
The subsequent exchange of RC messages are therefore silently dropped
since one side expec
On Mon, Sep 17, 2007 at 11:01:47AM -0700, Sean Hefty wrote:
> >The problem arises when the active Solaris client is sending a
> >connection request to a passive OFED server instance. Solaris will set
> >the hop_limit field to 0xFF and will not expect or enable GRH routing.
> >The subsequent exch
Ti ck!eeer F:D.E G
Last 0.04
Ta rg et 0.12
Jim,
On 9/17/07, Jim Hall <[EMAIL PROTECTED]> wrote:
>
> Hi Sean,
>
> The problem arises when the active Solaris client is sending a connection
> request to a passive OFED server instance. Solaris will set the hop_limit
> field to 0xFF and will not expect or enable GRH routing. The subsequent
> ex
Hi Sean,
The problem arises when the active Solaris client is sending a connection
request to a passive OFED server instance. Solaris will set the hop_limit field
to 0xFF and will not expect or enable GRH routing. The subsequent exchange of
RC messages are therefore silently dropped since one s
> Subject: Re: [ofa-general] [PATCH] [WORKAROUND] CONFIG_PREEMPT_RT and
ib_umad_close() issue
> From: Roland Dreier <[EMAIL PROTECTED]>
> Date: Mon, 17 Sep 2007 08:56:01 -0700
> To: John Blackwood <[EMAIL PROTECTED]>
> CC: [EMAIL PROTECTED], [EMAIL PROTECTED],
general@lists.openfabrics.org, Sven
(I don't think this made it to the mailng list, so re-posting.)
I don't disagree with the concept here, but can you explain the problem that
you're seeing? Is it that the path is assumed to be routed based on the
hop_limit (set in ib_init_ah_from_path)? Are any changes needed for active side
On Mon, 2007-09-17 at 08:56 -0700, Roland Dreier wrote:
> > When using OFED-1.2.5 based infiniband kernel modules on 2.6.22 based
> > kernels with the Ingo Molnar CONFIG_PREEMPT_RT applied, then commands
> > such as ibnetdiscvoer, smpquery, sminfo, etc. will hang. The problem
> > is with the do
Hi Sasha,
On 9/17/07, Sasha Khapyorsky <[EMAIL PROTECTED]> wrote:
> Hi Hal,
>
> On 05:42 Mon 17 Sep , Hal Rosenstock wrote:
> > Hi Sasha,
> >
> > On Sat, 2007-09-15 at 21:35 +0300, Sasha Khapyorsky wrote:
> > > This merges all subdirectories configure.in scripts into one toplevel
> > > directo
From: jamal <[EMAIL PROTECTED]>
Date: Mon, 17 Sep 2007 08:51:40 -0400
> On Sun, 2007-16-09 at 20:13 -0700, David Miller wrote:
>
> > This only makes sense for devices which can 1) scatter-gather
> > and 2) checksum on transmit.
>
> If you have knowledge there are enough descriptors in the driv
On 12:37 Mon 17 Sep , Hal Rosenstock wrote:
> Hi Sasha,
>
> On 9/17/07, Sasha Khapyorsky <[EMAIL PROTECTED]> wrote:
> > Hi Hal,
> >
> > On 05:42 Mon 17 Sep , Hal Rosenstock wrote:
> > > Hi Sasha,
> > >
> > > On Sat, 2007-09-15 at 21:35 +0300, Sasha Khapyorsky wrote:
> > > > This merges all
Hi Hal,
On 05:42 Mon 17 Sep , Hal Rosenstock wrote:
> Hi Sasha,
>
> On Sat, 2007-09-15 at 21:35 +0300, Sasha Khapyorsky wrote:
> > This merges all subdirectories configure.in scripts into one toplevel
> > directory script. Separate configuring per subdirectory is not needed
> > anymore.
>
>
On Mon, Sep 17, 2007 at 09:09:06AM -0700, Sean Hefty ([EMAIL PROTECTED]) wrote:
> >>>+addr = kmalloc(sizeof *addr, GFP_KERNEL);
> >>
> >>As a small nitpick: this wants to be sizeof(struct in_ifaddr)
>
> See chapter 14 of CodingStyle document. kmalloc(sizeof *addr... is correct.
Come on, do n
+addr = kmalloc(sizeof *addr, GFP_KERNEL);
As a small nitpick: this wants to be sizeof(struct in_ifaddr)
See chapter 14 of CodingStyle document. kmalloc(sizeof *addr... is correct.
- Sean
___
general mailing list
general@lists.openfabrics.org
h
Evgeniy Polyakov wrote:
Hi Steve.
On Sat, Sep 15, 2007 at 10:56:46AM -0500, Steve Wise ([EMAIL PROTECTED]) wrote:
The iWARP driver must translate all listens on address 0.0.0.0 to the
set of rdma-only ip addresses for the device in question. This prevents
incoming connect requests to the TCP
> When using OFED-1.2.5 based infiniband kernel modules on 2.6.22 based
> kernels with the Ingo Molnar CONFIG_PREEMPT_RT applied, then commands
> such as ibnetdiscvoer, smpquery, sminfo, etc. will hang. The problem
> is with the downgrade_write() rw semaphore usage in the
> ib_umad_close() ro
When using OFED-1.2.5 based infiniband kernel modules on 2.6.22 based
kernels with the Ingo Molnar CONFIG_PREEMPT_RT applied, then commands
such as ibnetdiscvoer, smpquery, sminfo, etc. will hang. The problem is
with the downgrade_write() rw semaphore usage in the ib_umad_close()
routine.
Th
Sorry, forgot put this to the list.
Nam
-- Forwarded Message --
Subject: [PATCH] ehca patches inclusion for ofed-1.3-alpha
Date: Monday 17 September 2007 16:36
From: Hoang-Nam Nguyen <[EMAIL PROTECTED]>
To: Vladimir Sokolovsky <[EMAIL PROTECTED]>
Cc: [EMAIL PROTECTED], [EMAIL PR
Hello Vlad and Michael!
Just downloaded daily build package OFED-1.3-20070917-0600 and saw
in SRPMS:
localhost:/home/nguyen/tmp/OFED-1.3-20070917-0600/SRPMS # ls -l
ofa_kernel-1.3-ofed2007091*
-rw-r--r-- 1 1011 1011 1967453 2007-09-10 15:27
ofa_kernel-1.3-ofed20070910.src.rpm
-rw-r--r-- 1 1011
After this is pushed, can you build and publish a new ofed-1.2.5.x
tarball? Or at least a daily build of the full ofed-1.2.5 kit?
Thanks,
Steve.
Michael S. Tsirkin wrote:
Done. I'll push soon.
Quoting Steve Wise <[EMAIL PROTECTED]>:
Subject: [GIT PULL ofed_1_2_c] cxgb3 bug fixes
Vlad (Mic
Hi Sasha,
On Sat, 2007-09-15 at 21:35 +0300, Sasha Khapyorsky wrote:
> This merges all subdirectories configure.in scripts into one toplevel
> directory script. Separate configuring per subdirectory is not needed
> anymore.
How is the requirement for separate OpenSM libraries (complib,
libosmvend
On Sun, 2007-16-09 at 20:13 -0700, David Miller wrote:
> What Herbert and I want to do is basically turn on TSO for
> devices that can't do it in hardware, and rely upon the GSO
> framework to do the segmenting in software right before we
> hit the device.
Sensible.
> This only makes sense for
Hi Sasha,
On Sat, 2007-09-15 at 22:51 +0300, Sasha Khapyorsky wrote:
> On 12:25 Mon 10 Sep , Hal Rosenstock wrote:
> > On 9/10/07, Sasha Khapyorsky <[EMAIL PROTECTED]> wrote:
> > > Hi Hal,
> > >
> > > On 12:13 Mon 10 Sep , Hal Rosenstock wrote:
> > > > Hi Sasha,
> > > >
> > > > On 9/10/07,
On Fri, 2007-09-14 at 11:01 -0700, Sean Hefty wrote:
> I didn't notice any issues with this patch, or anything missing from it.
>
> Reviewed-by: Sean Hefty <[EMAIL PROTECTED]>
I'll ditto the above and mention it was tested in old (coexistence)
mode.
Reviewed-by: Hal Rosenstock <[EMAIL PROTECTED]
On Thu, 2007-09-13 at 11:20 -0700, Sean Hefty wrote:
> > - My user_mad P_Key index support patch. I'll test the ioctl to
> > change to the new mode and merge this I guess, since Hal and Sean
> > have tested this out.
>
> I can give this patch a reviewed-by: too, and I will also try to review
Hi Yevgeny,
Is the failure below a simulator or OpenSM issue ? Thanks.
-- Hal
Forwarded Message
From: [EMAIL PROTECTED]
To: [EMAIL PROTECTED]
Cc: general@lists.openfabrics.org
Subject: [ofa-general] nightly osm_sim report 2007-09-15:normal
completion
Date: 15 Sep 2007 07:32:15
Hi Krishna,
Krishna Kumar2 wrote:
What I will do today/tomorrow is to run the rev5 (which I didn't run
for mthca) on both ehca and mthca and get statistics and send it out.
Otherwise what you stated is correct as far as rev4 goes. After giving
latest details, I will appreciate any help from Mell
/vlad/tmp/ofa_1_3_kernel-20070917-0200_linux-2.6.18_powerpc_check/drivers/infiniband/hw/ehca/ehca_main.c:936:
error: invalid type argument of '->'
/home/vlad/tmp/ofa_1_3_kernel-20070917-0200_linux-2.6.18_powerpc_check/drivers/infiniband/hw/ehca/ehca_main.c:939:
error: invalid typ
Sasha Khapyorsky <[EMAIL PROTECTED]> writes:
> On 17:30 Mon 10 Sep , Arne Redlich wrote:
>> "Hal Rosenstock" <[EMAIL PROTECTED]> writes:
>>
>> > On 9/7/07, Arne Redlich <[EMAIL PROTECTED]> wrote:
>> >> resolve_ca_name() passes a wrong "max" argument to umad_get_cas_names.
>> >>
>> >> Signed-o
89 matches
Mail list logo
