ib_verbs.h (Roland's branch): Add MAD pointer to send WR ud structure
This is to allow the MAD layer to poke at the MAD header without using
bus_to_virt().
Index: ib_verbs.h
===
--- ib_verbs.h (revision 939)
+++ ib_verbs.h
On Wed, 2004-10-06 at 10:36, Hal Rosenstock wrote:
On Tue, 2004-10-05 at 16:34, Sean Hefty wrote:
if (solicited) {
/* Routing is based on high 32 bits of transaction ID of MAD
*/
- hi_tid = mad-mad_hdr.tid 32;
+ hi_tid =
thanks, applied.
___
openib-general mailing list
[EMAIL PROTECTED]
http://openib.org/mailman/listinfo/openib-general
To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
From: Hal Rosenstock [mailto:[EMAIL PROTECTED]
Sent: Wednesday, October 06, 2004 7:36 AM
The problem with this is that when this is done on a little endian
machine it shows up byte swapped on the network and not in network
endian.
So if hi_tid = 1 and user tid = 0x9abcdef0
then the
ib_mad.h: Remove network endian conversion of QP1 QKey
Index: ib_mad.h
===
--- ib_mad.h(revision 935)
+++ ib_mad.h(working copy)
@@ -58,7 +58,7 @@
#define IB_QP0 0
#define IB_QP1 cpu_to_be32(1)
-#define
On Wed, 2004-10-06 at 12:01, Sean Hefty wrote:
Index: include/ib_verbs.h
===
--- include/ib_verbs.h (revision 935)
+++ include/ib_verbs.h (working copy)
@@ -539,6 +539,7 @@
} atomic;
struct
On Wed, 2004-10-06 at 12:42, Hal Rosenstock wrote:
ib_mad.h: Remove network endian conversion of QP1 QKey
I forgot to say that with this change GMPs including request/response
matching is now working :-)
-- Hal
___
openib-general mailing list
[EMAIL
From: Hal Rosenstock [mailto:[EMAIL PROTECTED]
Sent: Wednesday, October 06, 2004 9:42 AM
ib_mad.h: Remove network endian conversion of QP1 QKey
I don't get the point. Is IB_QP1_QKEY going to be treated in host order,
and then swapped by someone at some point when posting? Does mthca expect
This is the formal announcement on the availability of
the first release of the OpenIB MAD layer with SMI/SMA
(located in the openib-candidate branch).
The OpenIB access layer is comprised of the MAD layer and the SMI/SMA.
These are currently implemented as separate modules. These are layered
on
On Wed, 2004-10-06 at 12:47, Sean Hefty wrote:
I _think_ that the user only needs to reference something like the
following:
union ib_mad_hdrs {
struct ib_mad_hdr mad_hdr;
struct {
struct ib_mad_hdr mad_hdr;
struct ib_rmpp_hdr rmpp_hdr;
On Wed, 2004-10-06 at 12:46, Fab Tillier wrote:
From: Hal Rosenstock [mailto:[EMAIL PROTECTED]
Sent: Wednesday, October 06, 2004 9:42 AM
ib_mad.h: Remove network endian conversion of QP1 QKey
I don't get the point. Is IB_QP1_QKEY going to be treated in host order,
and then swapped by
On Tue, 2004-10-05 at 14:46 +0200, von Wyl wrote:
Hi,
I'm trying to compile a kernel module which use the VAPI and (maybe) the
CMAPI. And when I compile a file with an inclusion of vapi.h I got this :
Please use the new mthca driver instead of the old mellanox driver.
-tduffy
--
When
On Wed, 2004-10-06 at 12:46, Fab Tillier wrote:
From: Hal Rosenstock [mailto:[EMAIL PROTECTED]
Sent: Wednesday, October 06, 2004 9:42 AM
ib_mad.h: Remove network endian conversion of QP1 QKey
I don't get the point. Is IB_QP1_QKEY going to be treated in host order,
and then swapped by
Great! I will start merging this onto my tree as soon as I get my
IPoIB driver into shape to commit. (I've currently torn it apart
getting rid of the pseudo-ethernet layer, and I have it mostly working
again).
- Roland
___
openib-general mailing list
ib_mad: Eliminate ib_tid union
Index: access/ib_mad_priv.h
===
--- access/ib_mad_priv.h(revision 942)
+++ access/ib_mad_priv.h(working copy)
@@ -119,7 +119,7 @@
struct list_head agent_send_list;
struct
On Wed, 2004-10-06 at 12:04, Fab Tillier wrote:
From: Hal Rosenstock [mailto:[EMAIL PROTECTED]
Sent: Wednesday, October 06, 2004 7:36 AM
The problem with this is that when this is done on a little endian
machine it shows up byte swapped on the network and not in network
endian.
ib_mad: Use pointer to madhdr rather than mad in send WR ud structure
We'll go the rest of the way on this when RMPP is implemented.
Index: access/ib_smi.c
===
--- access/ib_smi.c (revision 942)
+++ access/ib_smi.c (working
ib_verbs.h: Use pointer to madhdr rather than mad in send WR ud
structure
We'll go the rest of the way on this when RMPP is implemented.
(This patch applies to Roland's branch).
Index: ib_verbs.h
===
--- ib_verbs.h (revision 944)
Hi,
Has mthca been tested with Arbel (PCI Express) ? Is this in
compatibility mode or native mode or both ?
Thanks.
-- Hal
___
openib-general mailing list
[EMAIL PROTECTED]
http://openib.org/mailman/listinfo/openib-general
To unsubscribe, please
Hal Hi, Has mthca been tested with Arbel (PCI Express) ? Is this
Hal in compatibility mode or native mode or both ?
I've tested mthca with Arbel in compatibility mode. I haven't had
access to native mode firmware or up-to-date documentation, so I have
not even started on native mode
Here are some modifications to support timing out send MADs in the access layer. I
haven't tested this code beyond building it, but wanted to make it available for
review. There are a few race conditions that need to be avoided when handling
timeouts, so if it looks like something was missed,
Hal Hi,
Has mthca been tested with Arbel (PCI Express) ? Is this in
compatibility mode or native mode or both ?
Thanks.
-- Hal
Does the PCI-Express HCA require any software changes to the HCA driver
?
i.e., will it run with an existing PCI-X tavor driver ?
I have heard that it does not
Robert Does the PCI-Express HCA require any software changes to
Robert the HCA driver ? i.e., will it run with an existing PCI-X
Robert tavor driver ? I have heard that it does not require any
Robert changes, but I just wanted to confim that.
As Hal mentioned, there are two
It seems you are using the system-wide keventd queue. This isn't
necessarily a problem per se, but it would probably be better to use a
MAD-layer private workqueue (I suggested a single-threaded workqueue
per MAD port earlier). This avoids two problems. First, keventd is
subject to arbitrary
24 matches
Mail list logo