On Thu, Aug 04, 2005 at 11:52:08PM +, Tom Duffy wrote:
> > From: Libor Michalek [mailto:[EMAIL PROTECTED]
> > No. I don't have any private patches to commit, and it does not look
> > like I'll have time to look at and test all the patches that have been
> &g
On Mon, Aug 01, 2005 at 09:35:29AM +0300, Michael S. Tsirkin wrote:
> in_atomic isnt a reliable way to check that we are in an atomic context.
> Just schedule work, always, since most cq polling is currently done under
> a spinlock, anyway.
I agree this patch is correct, but did you see any perf
On Mon, Aug 01, 2005 at 04:14:45PM +0300, Michael S. Tsirkin wrote:
> Quoting r. Michael S. Tsirkin <[EMAIL PROTECTED]>:
> > Subject: sdp: cancel read with no iocb
> >
> > Libor, I'm seeing these messages:
> >
> > ib_sdp WARN: Cancel read with no IOCB. <2:0:0005>
> >
> > It seems that this w
On Thu, Aug 04, 2005 at 02:46:47PM -0700, Tom Duffy wrote:
> On Thu, 2005-08-04 at 14:26 -0700, Libor Michalek wrote:
> > At the end of this week I will be leaving Cisco and will no longer
> > have the time, or access to the equipment, needed to maintain the SDP
> > or uC
At the end of this week I will be leaving Cisco and will no longer
have the time, or access to the equipment, needed to maintain the SDP
or uCM code. It would seem ideal for Sean Hefty to take over the uCM
code if he has the time and desire. I also think that Tom Duffy would
be the right person
On Wed, Aug 03, 2005 at 10:09:24AM +0300, Michael S. Tsirkin wrote:
> Quoting r. Tom Duffy <[EMAIL PROTECTED]>:
> > Perhaps you need my sdp_inet_port_put() patch?
>
> Could be.
> I'll give it a spin next week. Thanks!
Michael,
I remember this problem from the last time Hal mentioned it, I had
On Mon, Aug 01, 2005 at 04:26:16PM +0300, Michael S. Tsirkin wrote:
> Libor, in sdp_iocb.h we have:
>
> sdp_iocb.h:#define SDP_IOCB_SIZE_MAX (128*1024) /* matches AIO max kvec size.
> */
>
> What does the comment mean?
This is left over from the the old 2.4 AIO code. The AIO layer use to
brea
On Wed, Jul 27, 2005 at 02:13:18PM -0700, Tom Duffy wrote:
> [ Libor, sorry for the dup's, my sendmail is fucked ]
>
> No need for linux/version.h in SDP, either.
No problem, I've removed the include.
-Libor
___
openib-general mailing list
openib-gen
On Thu, Jul 21, 2005 at 03:03:08PM -0700, Arlin Davis wrote:
> Libor Michalek wrote:
> >On Tue, Jul 19, 2005 at 11:10:56AM -0700, Arlin Davis wrote:
> >
> >>Hi Libor,
> >>
> >>I am running uCM and uAT with uDAPL and occasionally hit the following.
>
On Mon, Jul 25, 2005 at 11:18:20AM -0700, Tom Duffy wrote:
> This patch fixes an oops that I introduced in my conversion to use linux
> lists for binds (committed in revision 2874). If two sockets tried to
> use the same port, after failing to get the port (again), it would
> attempt a put and the
On Thu, Jul 21, 2005 at 10:51:54AM +0300, Michael S. Tsirkin wrote:
> Hello, Libor!
> Here's a patch updated against the latest bits.
> I'd like it to get merged, if possible.
>
> The existing code play with VMA's is the #1 source of sdp problems for me.
> The patch below works with no problems fo
On Fri, Jul 22, 2005 at 12:08:49AM +0300, Michael S. Tsirkin wrote:
>
> But, what about merging get_user_pages first? I'm kind of tired of
> maintaining it.
Yes, I'll do that.
-Libor
___
openib-general mailing list
openib-general@openib.org
http://op
On Wed, Jul 20, 2005 at 02:27:11AM +0300, Michael S. Tsirkin wrote:
> Second attempt to solve race where AIO completed asynchronously
> is bypassed by other completions.
> Solution: set a bit in connection, and defer any cq polling
> till work queue handles the AIO. We have to pass the connection
>
On Tue, Jul 19, 2005 at 11:10:56AM -0700, Arlin Davis wrote:
> Hi Libor,
>
> I am running uCM and uAT with uDAPL and occasionally hit the following.
> Can you take a look?
>
> Jul 19 11:10:18 iclust-19 kernel: UCM: Write. cmd <1> in <4> out <0> len
> <12>
> Jul 19 11:10:18 iclust-19 kernel: UCM:
On Wed, Jul 20, 2005 at 05:17:09PM -0700, Fab Tillier wrote:
> > From: Libor Michalek [mailto:[EMAIL PROTECTED]
> > Sent: Wednesday, July 20, 2005 5:02 PM
> >
> > My question is do we really want this, since an application will likely
> > have yet another tab
On Thu, Jun 30, 2005 at 07:01:22PM -0700, Arlin Davis wrote:
>
> The uDAPL code is now connecting properly but I am having difficulty
> setting the QP states properly without the ib_cm_init_qp_attr() call.
> Any chance of providing this call in uCM?
Sean, To do this I'll need to add a kernel
On Tue, Jul 12, 2005 at 10:20:33PM -0700, Fab Tillier wrote:
> > From: Michael S. Tsirkin [mailto:[EMAIL PROTECTED]
> > Quoting r. Sean Hefty <[EMAIL PROTECTED]>:
> > >
> > > I believe that this is purely a userspace issue. I can't see why
> > > using a mutex wouldn't work, but I believe that get_
On Tue, Jul 12, 2005 at 01:45:43PM -0700, Sean Hefty wrote:
> >> one way to handle this is to have destruction flush events first, then
> >> perform the actual destroy...
> >
> >Right, what I was trying to say in the other thread was: walk the event
> >list and kill events for the id that is destro
On Tue, Jul 12, 2005 at 11:53:06AM -0700, Sean Hefty wrote:
> This fell out of the uCM connection ID discussion...
>
> There's an issue reporting events to userspace clients for an object that
> a user may have destroyed. The problem exists with user verbs, but is
> much more likely to be seen by
On Thu, Jul 21, 2005 at 02:40:50AM +0300, Michael S. Tsirkin wrote:
> Quoting r. Libor Michalek <[EMAIL PROTECTED]>:
> > Subject: Re: [PATCH] sdp_conn_put/sdp_conn_hold race
> >
> > On Wed, Jul 20, 2005 at 09:44:18PM +0300, Michael S. Tsirkin wrote:
> > > Q
On Wed, Jul 20, 2005 at 09:44:18PM +0300, Michael S. Tsirkin wrote:
> Quoting r. Michael S. Tsirkin <[EMAIL PROTECTED]>:
> > The current sdp_conn_put/sdp_conn_hold implementation
> > seems to be subject to the following race condition:
>
> Libor, did you have the time to review this patch?
Yes,
On Wed, Jul 20, 2005 at 06:47:31AM -0400, Hal Rosenstock wrote:
> On Wed, 2005-07-20 at 03:35, Michael S. Tsirkin wrote:
> > Quoting r. Hal Rosenstock <[EMAIL PROTECTED]>:
> > > On Tue, 2005-07-19 at 20:14, Libor Michalek wrote:
> > > > On Wed, Jul 13, 2005
On Tue, Jul 12, 2005 at 03:00:59PM -0700, Sean Hefty wrote:
> >This would also be a good time to get NFSoRDMA ported to kDAPL as the
> >API is unlikely to change too much in the future.
>
> I'm not as sure about the API. Is it necessary to abstract away QPs,
> CQs, etc. from the user? (The answe
On Thu, Jul 14, 2005 at 02:04:52PM -0400, James Lentini wrote:
> On Wed, 13 Jul 2005, Christoph Hellwig wrote:
> > On Wed, Jul 13, 2005 at 10:18:41AM -0400, James Lentini wrote:
> >> In addition to the streamlined connection management interface, kDAPL
> >> also provides a unified event model. This
On Wed, Jul 13, 2005 at 10:17:18AM -0700, Michael Krause wrote:
> At 06:39 AM 7/13/2005, James Lentini wrote:
> >On Tue, 12 Jul 2005, xg wang wrote:
> >
> >> Frankly speaking, I can not distinguish the function of SDP and
> >> DAPL. Since Lustre is a file system, it runs on kernel. So I think
off-by: Hal Rosenstock <[EMAIL PROTECTED]>
>
> --- ucm.c.kfree 2005-07-13 08:10:00.0 -0400
> +++ ucm.c 2005-07-13 19:39:04.0 -0400
> @@ -49,6 +49,11 @@ MODULE_AUTHOR("Libor Michalek");
> MODULE_DESCRIPTION("InfiniBand userspace Connectio
On Wed, Jul 13, 2005 at 07:14:24PM -0400, Hal Rosenstock wrote:
> libibcm: Fix some typos
Thanks, applied and commited.
-Libor
___
openib-general mailing list
openib-general@openib.org
http://openib.org/mailman/listinfo/openib-general
To unsubscribe,
On Wed, Jul 13, 2005 at 08:18:34AM -0400, Hal Rosenstock wrote:
> ucm: Eliminate NULL checks prior to calling kfree
>
Thanks, I didn't realize it made the null check internally. Applied
and commited.
-Libor
___
openib-general mailing list
openib-gene
On Wed, Jul 20, 2005 at 02:09:23AM +0300, Michael S. Tsirkin wrote:
> Quoting r. Libor Michalek <[EMAIL PROTECTED]>:
> > Subject: Re: [PATCH] SDP: use linux/list.h for binds
> >
> > On Thu, Jul 14, 2005 at 11:34:06AM -0700, Tom Duffy wrote:
> > > This patch
On Thu, Jul 14, 2005 at 03:30:12PM -0700, Tom Duffy wrote:
> This patch changes the NULL terminated port list in the sdp hca object
> to a Linux list_head. It also removes the sdev_hca *next member from
> the sdev_hca struct as this is never used, AFAICT.
Thanks, applied. The unused member was
On Wed, Jul 06, 2005 at 08:59:49PM +0300, Michael S. Tsirkin wrote:
> The current sdp_conn_put/sdp_conn_hold implementation
> seems to be subject to the following race condition:
>
> - thread A calls sdp_conn_put, atomic dec and test returns 0
> - thread B looks up the connection and calls sdp_con
On Thu, Jul 14, 2005 at 11:34:06AM -0700, Tom Duffy wrote:
> This patch replaces the null terminated linked list for connections
> bound to a port and instead uses the list.h library from core Linux.
>
> I have tested this on a few machines doing several ttcp's at once, but
> please inspect to ver
On Fri, Jul 15, 2005 at 01:26:11PM -0700, Arlin Davis wrote:
> Hi Roland,
>
> I have a question about uverbs cq_events. Currently, uverbs has one
> event (fd) handle that processes all CQ events. You mentioned earlier
> that you would possibly add support for more CQ events. Are you thinking
>
On Thu, Jun 30, 2005 at 07:01:22PM -0700, Arlin Davis wrote:
> Libor Michalek wrote:
>
> > The listen id is in the req rcvd event. (event->param.req_rcvd.listen_id)
> >Do you mean that it is not being set correctly?
> >
> >
> Ok, I didn't look deep en
On Tue, Jun 28, 2005 at 11:18:57PM +0300, Michael S. Tsirkin wrote:
> Libor, here's a stub at solving an old problem.
> Most of the patch is passing sdp_opt to iocb complete and cancel functions.
>
> I didnt yet test it, and I wont have the time today,
> but maybe you could tell me whether I'm goi
On Thu, Jun 30, 2005 at 09:13:28AM -0700, Sean Hefty wrote:
> Libor Michalek wrote:
> > Assume that the userspace 'struct ib_cm_event' contains the cm_id as
> > well as a new 'u64 context' which is inherited from the cm_id, and is
> > set at the tim
On Wed, Jun 29, 2005 at 12:42:09PM -0700, Arlin Davis wrote:
> Libor Michalek wrote:
> >On Tue, Jun 28, 2005 at 01:59:01PM -0700, Arlin Davis wrote:
> >>
> >>I have a couple of uCM questions regarding create_id and events...
> >>
> >>Is it possible fo
On Wed, Jun 29, 2005 at 11:16:19AM -0700, Sean Hefty wrote:
> Libor Michalek wrote:
> >>Is it possible for a consumer of uCM to provide a context with the
> >>create_id that could be returned with the event? I will have some scale
> >>up issues if I have to walk a l
On Tue, Jun 28, 2005 at 01:48:55PM -0700, Nishanth Aravamudan wrote:
>
> Using schedule_timeout() without setting the state first is broken and
> causes schedule_timeout() to return immediately (effectively you call
> schedule() without changing your state and are thus going to run again).
> In eac
On Tue, Jun 28, 2005 at 01:59:01PM -0700, Arlin Davis wrote:
> Hi Libor,
>
> I have a couple of uCM questions regarding create_id and events...
>
> Is it possible for a consumer of uCM to provide a context with the
> create_id that could be returned with the event? I will have some scale
> up i
On Tue, Jun 28, 2005 at 09:15:21AM +0200, Arne Redlich wrote:
> Am Montag, den 27.06.2005, 11:35 -0700 schrieb Libor Michalek:
> > On Wed, Jun 22, 2005 at 10:39:43AM +0200, Arne Redlich wrote:
> > > Hi,
> > >
> > > I'm trying to use SDP from within the
On Mon, Jun 27, 2005 at 04:06:11PM -0700, Libor Michalek wrote:
> On Mon, Jun 27, 2005 at 02:27:54PM -0700, Tom Duffy wrote:
> > On Mon, 2005-06-27 at 11:17 -0700, Libor Michalek wrote:
> > > The problem is that each call to sk_alloc() is grabbing a reference to
> > >
On Sun, Jun 19, 2005 at 10:33:36AM +0300, Michael S. Tsirkin wrote:
> Here's an update to the latest bits: Libor, does this make sense?
>
> Make sdp_cm_actv_error and sdp_conn_inet_error accept standard, negative
> error values (esp. for sdp_cm_actv_error it was sufficiently confusing to
> be pas
On Mon, Jun 27, 2005 at 02:27:54PM -0700, Tom Duffy wrote:
> On Mon, 2005-06-27 at 11:17 -0700, Libor Michalek wrote:
> > The problem is that each call to sk_alloc() is grabbing a reference to
> > the module, but it checks to make sure that there already is at least one
> >
On Wed, Jun 22, 2005 at 10:39:43AM +0200, Arne Redlich wrote:
> Hi,
>
> I'm trying to use SDP from within the kernel. My problem is that the
> code relies on sk_data_ready() (this callback is modified to wake up a
> Rx thread before executing the original function), but sk_data_ready()
> apparentl
On Thu, Jun 23, 2005 at 01:06:47PM -0700, Tom Duffy wrote:
> I am still getting the panic when you try to connect to a machine and
> it is not listening (but has ib_sdp loaded):
>
> [EMAIL PROTECTED] ~]# --- [cut here ] - [please
> bite here ] -
> Kernel BUG at "/build1/tdu
On Tue, Jun 21, 2005 at 08:35:15PM +0300, Michael S. Tsirkin wrote:
> Quoting r. Tom Duffy <[EMAIL PROTECTED]>:
> > Subject: Re: [PATCH] SDP: use container_of in place of cast to sdp_opt
> >
> > On Tue, 2005-06-21 at 09:21 +0300, Michael S. Tsirkin wrote:
> > > As a side note, wouldnt sk_to_sdp an
On Mon, Jun 20, 2005 at 08:40:47PM -0700, Roland Dreier wrote:
> Kevin> Maybe somebody could help me understand the proper way to
> Kevin> map between an IP address assigned to a port to the "device
> Kevin> name" and "port number" in the gen2 architecture. If I
> Kevin> have an IP
On Mon, Jun 20, 2005 at 11:06:12AM -0700, Tom Duffy wrote:
> Here is the updated patch based off r2663 and building off of 2.6.12.
>
> Signed-off-by: Tom Duffy <[EMAIL PROTECTED]>
Thanks a lot Tom for tracking the changes this whole time.
Committed revision 2665.
-Libor
_
On Mon, Jun 20, 2005 at 11:06:12AM -0700, Tom Duffy wrote:
> Here is the updated patch based off r2663 and building off of 2.6.12.
>
> Index: linux-2.6.12-openib/drivers/infiniband/ulp/sdp/sdp_inet.c
> ===
> --- linux-2.6.12-openib/dri
On Fri, Jun 17, 2005 at 04:25:19PM -0700, Roland Dreier wrote:
> Tom> Oh, it came across the [EMAIL PROTECTED] list which tracks
> Tom> the git commits now.
>
> Got it... I guess it takes a while for everything to percolate through
> all the mirror scripts.
>
> The svn repo is down right
Following patch simplifies some code paths, including some error
handling in the data paths.
3 files changed, 143 insertions(+), 377 deletions(-)
-Libor
Index: sdp_send.c
===
--- sdp_send.c (revision 2628)
+++ sdp_send.c (work
On Thu, Jun 16, 2005 at 03:44:11PM -0700, Roland Dreier wrote:
> Troy> WARN: : Failed to find lock pointer. WARN: : Error <-22>
> Troy> initializing memory locking. WARN: : Error <-22>
> Troy> initializing SDP IOCB table. WARN: : INIT: Error <-22>
> Troy> initializing connection
On Wed, Jun 15, 2005 at 04:30:47PM -0500, Troy Benjegerdes wrote:
> I'm getting a 'SDP: device mthca0 does nto support fast memory regions'
> error on PPC64 systems.. Is there something that needs to be done for
> PPC, or could I have an older version of the mthca module hanging
> around?
I'm no
On Wed, Jun 15, 2005 at 02:14:21PM -0700, Grant Grundler wrote:
> On Wed, Jun 15, 2005 at 03:47:15PM -0500, John Partridge wrote:
> > I am trying to get SDP working on SGI Altix. I know that the InfiniBand is
> > working as I have ipoib and ibv_pingpong etc all working fine.
> >
> > My question is
On Wed, Jun 15, 2005 at 03:47:15PM -0500, John Partridge wrote:
> I am trying to get SDP working on SGI Altix. I know that the InfiniBand is
> working as I have ipoib and ibv_pingpong etc all working fine.
>
> My question is how do I prove that an nttcp (or whatever I use) is using SDP
> rather th
On Wed, Jun 15, 2005 at 11:01:19PM +0300, Michael S. Tsirkin wrote:
> Quoting r. Libor Michalek <[EMAIL PROTECTED]>:
> > Subject: Re: [PATCH] sdp_post_rdma_iocb_snk check iocb
> >
> > On Wed, Jun 15, 2005 at 12:44:08PM +0300, Michael S. Tsirkin wrote:
> >
On Wed, Jun 15, 2005 at 12:44:08PM +0300, Michael S. Tsirkin wrote:
> In sdp_post_rdma_iocb_snk, if sdp_iocb_q_get_head returns NULL, iocb is then
> used without checking.
Michael,
This is done in a number of places, and the check for NULL iocb should
not be necessary. Earlier in each function
On Fri, May 13, 2005 at 05:06:55PM +0300, Michael S. Tsirkin wrote:
> Quoting r. Libor Michalek <[EMAIL PROTECTED]>:
> > Subject: Re: [PATCH] Re: 0 op factor
> >
> > On Tue, May 10, 2005 at 05:47:00PM -0700, Libor Michalek wrote:
> > > On Sat, May 07, 2005 a
On Tue, Jun 14, 2005 at 11:28:01AM -0400, Hal Rosenstock wrote:
> Hi Libor,
>
> In the ucm kernel module, in the routine ib_ucm_event, after the user:
> label, if various errors occur relating to copying back to userspace,
> all that is currently done is that the mutex is released and the result
>
On Fri, Jun 03, 2005 at 05:06:19PM -0400, Jordan, Bill wrote:
>
> Patch to change sys class name of userspace cm to match
> format of userpace verbs and userspace mad class names.
Thanks, applied and committed.
-Libor
___
openib-general mailing list
op
On Fri, Jun 10, 2005 at 09:34:10PM +0200, Bernhard Fischer wrote:
> Hi,
>
> Upon loading ib_sdp with debugging enabled, i see:
>
> - remove duplicate ': ' and prefix messages from sdp with ib_sdp
> like ib_mthca does.
> - fix odd "INIT: : INIT: SDP module unload." debug message.
Thanks, applie
On Fri, Jun 10, 2005 at 09:04:55AM -0400, Hal Rosenstock wrote:
> On Thu, 2005-06-09 at 15:47, Sean Hefty wrote:
> > > I'm seeing an unusual problem when both halves of a connection
> > >actively disconnect at the same time. Each connection peer issues
> > >a DREQ at the same time, next each rece
On Thu, Jun 09, 2005 at 11:51:32AM -0700, Tom Duffy wrote:
>
> I am still trying to track down a bug with this patch that occurs when
> you try to initiate an active connection and there is no passive
> listener ready. This can cause a panic on the passive side if ib_sdp is
> loaded.
Do you wa
Sean,
I'm seeing an unusual problem when both halves of a connection
actively disconnect at the same time. Each connection peer issues
a DREQ at the same time, next each receive the DREQ and responds
with a DREP, and finally each connection gets a callback for the
transition to the idle state. H
On Wed, Jun 08, 2005 at 05:53:10PM -0400, William Jordan wrote:
> On 6/8/05, Libor Michalek <[EMAIL PROTECTED]> wrote:
> > On Wed, Jun 08, 2005 at 01:29:42PM -0400, William Jordan wrote:
> > > I'm trying to run netperf over SDP using libsdp. I'm having
> >
On Wed, Jun 08, 2005 at 01:29:42PM -0400, William Jordan wrote:
> I'm trying to run netperf over SDP using libsdp. I'm having
> intermittent success, but usually, I get a failure. The client does a
> shutdown (SHUT_WR), but the server gets an ECONNRESET, "Connection
> reset by peer", on the recv in
On Thu, Jun 02, 2005 at 12:55:34PM -0700, Sean Hefty wrote:
> The ucm code in ib_ucm_event_handler() reads the cm_id->state information in
> response to an event occurrence. The cm_id state can change dynamically in
> response to an event, and in general should not be accessed. (I.e. the
> curren
On Fri, Jun 03, 2005 at 04:50:02PM -0400, William Jordan wrote:
> On 6/3/05, Roland Dreier <[EMAIL PROTECTED]> wrote:
> >William> Should the device nodes /dev/infiniband_mad and
> >William> /dev/infiniband_verbs exist at all? Are they artifacts of
> >William> the way the classes work, o
On Wed, Jun 01, 2005 at 05:52:47PM -0400, Bill Jordan wrote:
> Libor,
>
> Here is a patch to remove the devfs usage from ib_ucm.
Thanks Bill, applied and commited.
> The udev support already exists, but requires a
> rule entry of KERNEL="ucm", NAME="infiniband/%k".
> Is there an appropriate pl
On Tue, May 31, 2005 at 02:03:06PM -0700, Tom Duffy wrote:
> On Sat, 2005-05-28 at 09:13 +0200, Christoph Hellwig wrote:
> > On Fri, May 27, 2005 at 03:56:58PM -0700, Bob Woodruff wrote:
> > > kDAPL is intended as a kernel-level API
> > > for RDMA enabled fabrics. As it was initially written,
> > >
On Tue, May 31, 2005 at 05:54:38PM -0400, William Jordan wrote:
> Why does the userlevel CM use devfs to create device nodes? Userlevel
> verbs and mad layers appear to rely on udev.
No reason except that I went with what I thought was the simpler
model at the time. Unlike the verbs and mad laye
On Wed, May 25, 2005 at 11:21:28PM +0300, Michael S. Tsirkin wrote:
> Quoting r. Libor Michalek <[EMAIL PROTECTED]>:
> > Subject: Re: Re: [PATCH] (repost) sdp: replace mlock with get_user_pages
> >
> > On Fri, May 13, 2005 at 04:51:45PM +0300, Michael S. Tsirkin wro
On Fri, May 20, 2005 at 10:49:40AM -0400, Hal Rosenstock wrote:
> Hi Libor,
>
> A possible minor change to ucm.c::ib_ucm_event_handler line 440 would
> be:
>
> uevent->cm_id = (id == IB_UCM_CM_ID_INVALID ?
> cm_id : NULL);
> rather than:
> uevent->cm_id =
On Thu, May 19, 2005 at 09:34:16AM -0700, Fab Tillier wrote:
> > From: Caitlin Bestler [mailto:[EMAIL PROTECTED]
> > Sent: Thursday, May 19, 2005 9:25 AM
> >
> > So being predictably unreliable for one implementation
> > stage is certainly something you can get away with.
> > Even when you add sup
On Thu, May 19, 2005 at 11:12:38AM +0300, Michael S. Tsirkin wrote:
> Libor, I'm looking at sdp_link.c, and I dont see any lock
> protecting the info_list linked list.
> What prevents sdp_path_info_lookup from being called while
> sdp_path_info_create or sdp_path_info_destroy is in progress?
Micha
On Tue, May 17, 2005 at 07:08:16PM -0700, Grant Grundler wrote:
> On Tue, May 17, 2005 at 06:32:38PM -0700, Jeff Carr wrote:
> > >>>But IPoIB can't really implement NAPI since it's sending work to
> > >>>a shared HCA.
> >
> > Hmm. I'm not knowledgeable to know why; I'll have to take your word for
On Wed, May 18, 2005 at 01:14:07AM +0300, Michael S. Tsirkin wrote:
> Quoting r. Libor Michalek <[EMAIL PROTECTED]>:
> > Subject: Re: [PATCH] (repost) sdp: replace mlock with get_user_pages
> >
> > On Sun, May 15, 2005 at 11:05:03AM +0300, Michael S. Tsirkin wrote:
>
On Tue, May 17, 2005 at 04:48:25PM -0400, Hal Rosenstock wrote:
> On Tue, 2005-05-17 at 16:44, Libor Michalek wrote:
> > On Mon, May 16, 2005 at 12:58:05PM -0400, Hal Rosenstock wrote:
> > > Hi again Libor,
> > >
> > > When I run the example I get the fol
On Sun, May 15, 2005 at 11:05:03AM +0300, Michael S. Tsirkin wrote:
> Quoting r. Libor Michalek <[EMAIL PROTECTED]>:
> > + /*
> > +* valid result can be 0 or 1 for complete so
> > +* we ignore the value.
> > +*/
> > + (void)aio_complete(io
On Sat, May 14, 2005 at 10:19:30PM +0300, Michael S. Tsirkin wrote:
> Quoting r. Libor Michalek <[EMAIL PROTECTED]>:
> >
> > > > Another alternative would be to always complete aios asynchronously,
> > > > which should preserve the order. I guess this w
On Mon, May 16, 2005 at 12:58:05PM -0400, Hal Rosenstock wrote:
> Hi again Libor,
>
> When I run the example I get the following:
> server:
> CM ID <3> Event <1> State <3>
> client:
> CM ID <2> Event <0> State <0>
> server:
> CM ID <3> Event <2> State <0>
>
> The server side looks like it first ge
On Mon, May 16, 2005 at 03:54:18PM -0700, Jeff Carr wrote:
> Libor Michalek wrote:
> > On Mon, May 16, 2005 at 03:26:57PM -0700, Jeff Carr wrote:
> >
> >>It seems to me it would be useful to have a simple ib_send_page() function.
> >>
> >>This is essent
On Mon, May 16, 2005 at 03:26:57PM -0700, Jeff Carr wrote:
> It seems to me it would be useful to have a simple ib_send_page() function.
>
> This is essentially what I'm going to end up writing for what I need IB
> to do. If there is anyone else that has similar needs or interests I'd
> be happy
On Fri, May 13, 2005 at 10:22:53AM -0400, Hal Rosenstock wrote:
> Hi Libor,
>
> A couple of minor libibcm building comments:
>
> 1. When running autogen.sh, it warns about:
> Makefile.am:19: variable `ucm_version_script' not defined
>
> 2. example/simple is not built by the current makefile gene
On Fri, May 13, 2005 at 10:41:57AM -0400, Hal Rosenstock wrote:
> libibcm/examples/simple.c: Eliminate a couple of warnings
>
> Signed-off-by: Hal Rosenstock <[EMAIL PROTECTED]>
Applied, but including netinet/in.h required me to rename connect()
because of the type conflict in socket.h. Also I
On Fri, May 13, 2005 at 09:30:32AM -0700, Sean Hefty wrote:
> Tom Duffy wrote:
> > This patch changes some functions to return void rather than an int
> > since they were always returning 0, thus making checking return values
> > rather pointless. The patch touches both SDP and CM.
>
> Thanks, To
On Thu, May 12, 2005 at 10:00:47PM -0400, James Lentini wrote:
>
> This patch affects ib_cm.h and cm.c, which appear to be maintained
> by Sean, and ucm.c, which appears to be maintained by Libor.
>
> The patch adds the const qualifier to private data pointers passed
> into the IB layer.
Sean,
On Fri, May 13, 2005 at 05:06:00PM +0300, Michael S. Tsirkin wrote:
> Use of virt_to_phys looks wrong in sdp_recv and sdp_send.
> Must not the DMA API be used to get the physical address
> to be put in ib_sge?
It is incorrect. The regular send/recv path uses the DMA API correctly,
but RDMA to in
On Fri, May 13, 2005 at 04:51:45PM +0300, Michael S. Tsirkin wrote:
> Quoting r. Roland Dreier <[EMAIL PROTECTED]>:
> >
> > Libor> Always call aio_complete() immediately when
> > Libor> iocb_complete() is called, and only spawn the work thread
> > Libor> to unlock the memory after th
On Tue, May 10, 2005 at 05:47:00PM -0700, Libor Michalek wrote:
> On Sat, May 07, 2005 at 07:47:18PM +0200, Bernhard Fischer wrote:
>
> > - remove expect from _sdp_cm_path_complete().
>
> When I said that the SDP_EXPECT should eventually be removed, I meant
> that the funct
On Tue, May 10, 2005 at 05:32:32PM +0300, Michael S. Tsirkin wrote:
> Quoting r. Libor Michalek <[EMAIL PROTECTED]>:
> > On Thu, May 05, 2005 at 02:01:58PM +0300, Michael S. Tsirkin wrote:
> > >
> > > The following patch replaces the mlock hack with call
&
On Tue, May 10, 2005 at 01:00:21PM +0300, Michael S. Tsirkin wrote:
> struct sdpc_buff has a copy of ib_sge and uses casts to turn that into
> a real ib_sge. The right thing is obviously to keep ib_sge
> as part of sdpc_buff.
Thanks, applied and checked in.
-Libor
__
The IO address passed ib_fmr_pool_map_phys must be page aligned, here
is a patch to make AIO with unaligned buffers work. By default ttcp.aio.c
aligns buffers on page boundaries, which is why it was not noticed before.
Signed-off-by: Libor Michalek <[EMAIL PROTECTED]>
Index: sdp_
On Wed, May 04, 2005 at 10:00:35PM +0300, Michael S. Tsirkin wrote:
> Quoting r. Libor Michalek <[EMAIL PROTECTED]>:
> > Subject: [PATCH][SDP] AIO buffer corruption
> >
> > Patch to fix the problem a few people reported as ttcp.aio.c
> > aborting with a
On Tue, May 10, 2005 at 06:35:03PM -0700, Tom Duffy wrote:
> On Tue, 2005-05-10 at 18:06 -0700, Libor Michalek wrote:
> > Thanks, applied and checked in.
>
> One minor nit: you changed total to maximuml in conn_size description,
> which is fine, but you misspelled maximum.
On Sat, May 07, 2005 at 09:33:28AM -0700, Tom Duffy wrote:
> On Fri, 2005-05-06 at 02:30 +0200, Bernhard Fischer wrote:
> > in src/linux-kernel/infiniband/ulp/sdp/sdp_inet.c
> > module_param seem to be missing MODULE_PARM_DESC
>
> Signed-off-by: Tom Duffy <[EMAIL PROTECTED]>
Thanks, applied and
On Mon, May 09, 2005 at 03:32:20PM -0400, Hal Rosenstock wrote:
> On Mon, 2005-05-09 at 14:39, Libor Michalek wrote:
> > On Mon, May 09, 2005 at 01:22:17PM -0400, Hal Rosenstock wrote:
> > > Hi Libor,
> > >
> > > I have a couple of questions pertaining to the
On Sat, May 07, 2005 at 07:47:18PM +0200, Bernhard Fischer wrote:
> On Thu, May 05, 2005 at 05:21:22PM -0700, Libor Michalek wrote:
> > No point other then the obvious, changing the sign of the variables
> >value. It's just a style convention that I have no problem changi
On Sat, May 07, 2005 at 04:41:09PM -0700, Tom Duffy wrote:
> sk_alloc() already takes care of setting sk_family to proto. No need to
> do it a second time in sdp.
Thanks, applied and commited.
-Libor
___
openib-general mailing list
openib-general@ope
1 - 100 of 304 matches
Mail list logo
