On Mon, Sep 19, 2011 at 11:20 PM, Hefty, Sean wrote:
> It would be easier to maintain opensm on windows if it truly shared the same
> code base. For now, I'd just like to start with a common ib_types.h file.
> (There are currently hundreds, if not thousands, of lines that differ.)
> ib_types
The following patches add a new tool ibccquery to infiniband-diags. It
supports the querying of various congestion control settings. Related
updates to libibmad are also included.
Al
--
Albert Chu
ch...@llnl.gov
Computer Scientist
High Performance Systems Division
Lawrence Livermore National L
It would be easier to maintain opensm on windows if it truly shared the same
code base. For now, I'd just like to start with a common ib_types.h file.
(There are currently hundreds, if not thousands, of lines that differ.)
ib_types.h uses #if defined(__WIN__) to separate linux from windows co
On Mon, Sep 19, 2011 at 02:26:13PM -0400, Hal Rosenstock wrote:
> Right.
>
> > Why carry that same concept into a software interface? ??It seems complex.
>
> So as not to break existing verbs apps by inventing new values to
> existing fields which wouldn't be understood.
Every app I've ever wo
On Mon, Sep 19, 2011 at 03:55:24PM -0400, Hal Rosenstock wrote:
> >> Is this just to make the set of guidinfo simple ? Longer term, this
> >> should support a GUIDCap > 1 but I suppose that could be a follow on
> >> patch to this.
> >
> > It is to make the existing ibsim compliant with the spec.
>
On Mon, Sep 19, 2011 at 11:59 AM, Jason Gunthorpe
wrote:
> On Mon, Sep 19, 2011 at 10:07:57AM -0400, Hal Rosenstock wrote:
>> On Fri, Sep 16, 2011 at 5:06 PM, Jason Gunthorpe
>> wrote:
>> > Hardwire PortInfo.GUIDCap to 1 for both the switch and HCA case,
>>
>> Why hardwire this to 1 ? This is not
The dapl-2.0.33 distribution was built without FCA configuration so it was
missing some of the optional headers for FCA support.
Please pull updated dapl-2.0.33.tar.gz. The git tree is unaffected.
md5sum: 1539095f223d11a3c4a69c5e3775f1cc dapl-2.0.33.tar.gz
Arlin
--
To unsubscribe from
Hal Rosenstock wrote:
> Hefty, Sean wrote:
>> That's was an issue with the wire protocol format, correct?
> Right.
>> Why carry that same concept into a software interface? It seems complex.
> So as not to break existing verbs apps by inventing new values to
> existing fields which wouldn't
On Mon, Sep 19, 2011 at 2:17 PM, Hefty, Sean wrote:
>> It maps to what was done in the PortInfo attribute to add the new
>> extended speeds. There was no room for expansion in the existing
>> original link speed fields so a "parallel" set of fields had to be
>> added there..
>
> That's was an issu
> It maps to what was done in the PortInfo attribute to add the new
> extended speeds. There was no room for expansion in the existing
> original link speed fields so a "parallel" set of fields had to be
> added there..
That's was an issue with the wire protocol format, correct? Why carry that
s
On Mon, Sep 19, 2011 at 2:00 PM, Hefty, Sean wrote:
>> Index: b/drivers/infiniband/core/verbs.c
>> ===
>> --- a/drivers/infiniband/core/verbs.c 2011-09-13 13:34:19.660539000 +0300
>> +++ b/drivers/infiniband/core/verbs.c 2011-09-13 16
> Index: b/drivers/infiniband/core/verbs.c
> ===
> --- a/drivers/infiniband/core/verbs.c 2011-09-13 13:34:19.660539000 +0300
> +++ b/drivers/infiniband/core/verbs.c 2011-09-13 16:42:39.713754400 +0300
> @@ -77,6 +77,23 @@ enum ib_rate
> Index: b/drivers/infiniband/core/sysfs.c
> ===
> --- a/drivers/infiniband/core/sysfs.c 2011-09-14 13:49:58.0 +0300
> +++ b/drivers/infiniband/core/sysfs.c 2011-09-14 13:50:43.731775900 +0300
> @@ -209,7 +209,7 @@ static ssize
An Extended Port Info packet is sent to each hw port during HCA init
and if returns without error it is assumed the port supports
extended port capabilities.
If the port supports extended capabilities and the active speed is
QDR, FDR-10 link encoding is set according to Extended Port Info.
Signe
mlx4_ib: Add extended link speeds support
This patch implements IB spec extension of the PortInfo attribute:
Component : LinkSpeedExtActive
Used By: (All X's except base SP0)
Access : RO
Length : 4
Offset : 496
Description: If PortInfo:CapabilityMask.IsExtendedSpeedsSupported
Added support for new link encodings beyond the regular 8 to 10 bit encoding.
For now only FDR-10 (64 to 66 bit encoding) is implemented.
Signed-off-by: Marcel Apfelbaum
Reviewed-by: Hal Rosenstock
---
drivers/infiniband/core/sysfs.c |2 +-
include/rdma/ib_verbs.h |1 +
2 files
This patch implements IB spec extension of the PortInfo attribute:
Component : LinkSpeedExtActive
Used By: (All X's except base SP0)
Access : RO
Length : 4
Offset : 496
Description: If PortInfo:CapabilityMask.IsExtendedSpeedsSupported, currently
active extended link s
The purpose of the RFC is to introduce FDR/FDR-10:
- FDR10 is a proprietary link speed which is 10.3125 Gbps at 64/66 encoding
rather than 8b10b encoding.
- FDR represents an extended speed: 14.0625 Gbps
- EDR represents an extended speed: 25.78125 Gbps (EDR)
The aim is to show correct speeds
On Mon, Sep 19, 2011 at 10:07:57AM -0400, Hal Rosenstock wrote:
> On Fri, Sep 16, 2011 at 5:06 PM, Jason Gunthorpe
> wrote:
> > Hardwire PortInfo.GUIDCap to 1 for both the switch and HCA case,
>
> Why hardwire this to 1 ? This is not what most defaults are.
>
> > then simply verify that the inco
Hi,
Sorry, there is a problem with patches format, I'll send them again after fix.
Marcel
On Mon, Sep 19, 2011 at 6:05 PM, Marcel Apfelbaum
wrote:
> -- Forwarded message --
> From: Marcel Apfelbaum
> Date: Mon, Sep 19, 2011 at 6:01 PM
> Subject: [PATCH RFC 0/4] adding support f
The problem with creating a new protocol is that we would need to
a) design the protocol
b) get the IBTA to accept it
c) get the application developers to accept it.
I simply cannot see getting this process done in less than two years.
Meanwhile, the problem with SA/SM performance is an issue r
Added support for new link encodings beyond the regular 8 to 10 bit encoding.
For now only FDR-10 (64 to 66 bit encoding) is implemented.
Signed-off-by: Marcel Apfelbaum
Reviewed-by: Hal Rosenstock
---
drivers/infiniband/core/sysfs.c | 2 +-
include/rdma/ib_verbs.h | 1 +
2 files cha
An Extended Port Info packet is sent to each hw port during HCA init
and if returns without error it is assumed the port supports
extended port capabilities.
If the port supports extended capabilities and the active speed is
QDR, FDR-10 link encoding is set according to Extended Port Info.
Signe
mlx4_ib: Add extended link speeds support
This patch implements IB spec extension of the PortInfo attribute:
Component : LinkSpeedExtActive
Used By : (All X's except base SP0)
Access : RO
Length : 4
Offset : 496
Description: If PortInfo:CapabilityMask.IsExtendedSpeedsSupported
This patch implements IB spec extension of the PortInfo attribute:
Component : LinkSpeedExtActive
Used By : (All X's except base SP0)
Access : RO
Length : 4
Offset : 496
Description: If PortInfo:CapabilityMask.IsExtendedSpeedsSupported, currently
active extended link sp
The purpose of the RFC is to introduce FDR/FDR-10:
- FDR10 is a proprietary link speed which is 10.3125 Gbps at 64/66
encoding rather than 8b10b encoding.
- FDR represents an extended speed: 14.0625 Gbps
- EDR represents an extended speed: 25.78125 Gbps (EDR)
The aim is to show correct speeds a
On Fri, Sep 16, 2011 at 5:06 PM, Jason Gunthorpe
wrote:
> Hardwire PortInfo.GUIDCap to 1 for both the switch and HCA case,
Why hardwire this to 1 ? This is not what most defaults are.
> then simply verify that the incoming SubnSet matches the RO data
> in block 0.
Is this just to make the set o
27 matches
Mail list logo
