[openib-general] [DAPL] missing files in dapl_gamma_3.0

Hi James, Eitan, I found 'DAPL Gamma 3.0' at the DAPL sourceforge page. After trying to build that with VERBS=openib_gen_one option, especially kDAPL portion, now I think that at lest two files 'dapl_srv.c' and 'dapl_srv.h' are missing. I checked also IBGD-1.6.1 package but couldn't find them. W

[openib-general] [PATCH]IPOIB: adding spin_lock in path_free()

Please review this patch. diff -urN infiniband/ulp/ipoib/ipoib_main.c infiniband-path/ulp/ipoib/ipoib_main.c --- infiniband/ulp/ipoib/ipoib_main.c        2005-02-17 17:24:56.0 + +++ infiniband-path/ulp/ipoib/ipoib_main.c        2005-02-18 01:29:52.0 + @@ -215,15 +215,17 @

[openib-general] [PATCH]IPOIB: delete neigh from the list before kfree(neigh)

This patch sets caller's ah to NULL when calling ipoib_put_ah(), and deletes neigh from the list before kfree(neigh). Please review this patch. diff -urN infiniband/ulp/ipoib/ipoib_main.c infiniband-ah/ulp/ipoib/ipoib_main.c --- infiniband/ulp/ipoib/ipoib_main.c        2005-02-17 17:24:56.0

[openib-general] [PATCH] IPOIB: initialization patch in ipoib

Here is the patch for : 1. initialize path->ah = NULL 2. move neigh->neighbour->ops before kfree(neigh) 3. set neighbour->ops->destructor to NULL while kfree(neigh) diff -urN infiniband/ulp/ipoib/ipoib_main.c infiniband-patch/ulp/ipoib/ipoib_main.c --- infiniband/ulp/ipoib/ipoib_main.c        2

[openib-general] [PATCH] [CM] add stale connection test to active side

This patch adds testing for a stale connection on the active side of the connection protocol. It also fixes a bug where a duplicate REP would have been dropped, rather than forcing a resend of the RTU. Signed-off-by: Sean Hefty <[EMAIL PROTECTED]> Index: infiniband/core/cm.c ===

[openib-general] Re: RFC on SDP checkin

On Thu, Feb 17, 2005 at 04:03:59PM +0200, Michael S. Tsirkin wrote: > Quoting r. Libor Michalek <[EMAIL PROTECTED]>: > > Possible Issues > > > > [...] > > > > - Memory locking for AIO requires a call to do_mlock() which is not a > > kernel exported function, the method for calling the fu

Re: [openib-general] Re: SDP_CONN_LOCK

On Thu, Feb 17, 2005 at 04:24:13PM -0800, Libor Michalek wrote: > OK. I had used macros since in the fast path the functions wouldn't > get called, but since it sounds like it makes things less readable I > should be able to quickly try out replacing the macros with functions, > and see what if a

Re: [openib-general] [PATCH]deadlock problem in ipoib

Sorry, made a mistake in the previous patch. Here is the right one. diff -urN infiniband/ulp/ipoib/ipoib_main.c infiniband-lock/ulp/ipoib/ipoib_main.c --- infiniband/ulp/ipoib/ipoib_main.c        2005-02-17 17:24:56.0 + +++ infiniband-lock/ulp/ipoib/ipoib_main.c        2005-02-18 00:

Re: [openib-general] Re: SDP_CONN_LOCK

On Thu, Feb 17, 2005 at 03:49:31PM -0800, Sean Hefty wrote: > Michael S. Tsirkin wrote: > > Quoting r. Roland Dreier <[EMAIL PROTECTED]>: > >> > >>BTW, since mthca currently calls completion handlers directly from > >>interrupt context (rather than BH/tasklet context), it might be worth > >>renamin

Re: [openib-general] SDP_CONN_LOCK

On Fri, Feb 18, 2005 at 01:25:39AM +0200, Michael S. Tsirkin wrote: > Quoting r. Libor Michalek <[EMAIL PROTECTED]>: > > On Fri, Feb 18, 2005 at 12:13:30AM +0200, Michael S. Tsirkin wrote: > > > Quoting r. Libor Michalek <[EMAIL PROTECTED]>: > > > > On Thu, Feb 17, 2005 at 11:29:37PM +0200, Michael

Re: [openib-general] Re: SDP_CONN_LOCK

Michael S. Tsirkin wrote: Quoting r. Roland Dreier <[EMAIL PROTECTED]>: Subject: Re: SDP_CONN_LOCK BTW, since mthca currently calls completion handlers directly from interrupt context (rather than BH/tasklet context), it might be worth renaming all the SDP locking macros so they're not confusingly

[openib-general] Re: SDP_CONN_LOCK

Quoting r. Roland Dreier <[EMAIL PROTECTED]>: > Subject: Re: SDP_CONN_LOCK > > BTW, since mthca currently calls completion handlers directly from > interrupt context (rather than BH/tasklet context), it might be worth > renaming all the SDP locking macros so they're not confusingly named > with _B

Re: [openib-general] SDP_CONN_LOCK

Quoting r. Libor Michalek <[EMAIL PROTECTED]>: > Subject: Re: [openib-general] SDP_CONN_LOCK > > On Fri, Feb 18, 2005 at 12:13:30AM +0200, Michael S. Tsirkin wrote: > > Quoting r. Libor Michalek <[EMAIL PROTECTED]>: > > > On Thu, Feb 17, 2005 at 11:29:37PM +0200, Michael S. Tsirkin wrote: > > > >

Re: [openib-general] SDP_CONN_LOCK

On Fri, Feb 18, 2005 at 12:13:30AM +0200, Michael S. Tsirkin wrote: > Quoting r. Libor Michalek <[EMAIL PROTECTED]>: > > On Thu, Feb 17, 2005 at 11:29:37PM +0200, Michael S. Tsirkin wrote: > > > Quoting r. Libor Michalek <[EMAIL PROTECTED]>: > > > > On Thu, Feb 17, 2005 at 05:35:17PM +0200, Michael

[openib-general] [PATCH][SDP][33/33] Fix comment style in sdp_write.c

Signed-off-by: Tom Duffy <[EMAIL PROTECTED]> --- drivers/infiniband/ulp/sdp-orig/sdp_write.c 2005-02-15 16:33:46.194005000 -0800 +++ drivers/infiniband/ulp/sdp/sdp_write.c 2005-02-17 11:57:50.886008000 -0800 @@ -1,35 +1,46 @@ /* - This software is available to you under a choice of one o

[openib-general] [PATCH][SDP][32/33] Fix comment style in sdp_wall.c

Signed-off-by: Tom Duffy <[EMAIL PROTECTED]> --- drivers/infiniband/ulp/sdp-orig/sdp_wall.c 2005-02-10 18:42:56.553008000 -0800 +++ drivers/infiniband/ulp/sdp/sdp_wall.c 2005-02-17 11:56:59.912002000 -0800 @@ -1,35 +1,46 @@ /* - This software is available to you under a choice of one o

[openib-general] [PATCH][SDP][31/33] Fix comment style in sdp_sock.h

Signed-off-by: Tom Duffy <[EMAIL PROTECTED]> --- drivers/infiniband/ulp/sdp-orig/sdp_sock.h 2005-02-11 17:28:16.827018000 -0800 +++ drivers/infiniband/ulp/sdp/sdp_sock.h 2005-02-17 11:53:24.583003000 -0800 @@ -1,28 +1,40 @@ /* - This software is available to you under a choice of one o

[openib-general] [PATCH][SDP][30/33] Fix comment style in sdp_sent.c

Signed-off-by: Tom Duffy <[EMAIL PROTECTED]> --- drivers/infiniband/ulp/sdp-orig/sdp_sent.c 2005-02-15 16:33:46.247005000 -0800 +++ drivers/infiniband/ulp/sdp/sdp_sent.c 2005-02-17 11:52:21.451002000 -0800 @@ -1,35 +1,43 @@ /* - This software is available to you under a choice of one o

[openib-general] [PATCH][SDP][29/33] Fix comment style in sdp_send.c

Signed-off-by: Tom Duffy <[EMAIL PROTECTED]> --- drivers/infiniband/ulp/sdp-orig/sdp_send.c 2005-02-14 13:58:45.860001000 -0800 +++ drivers/infiniband/ulp/sdp/sdp_send.c 2005-02-17 11:38:10.806001000 -0800 @@ -1,35 +1,46 @@ /* - This software is available to you under a choice of one o

[openib-general] [PATCH][SDP][28/33] Fix comment style in sdp_recv.c

Signed-off-by: Tom Duffy <[EMAIL PROTECTED]> --- drivers/infiniband/ulp/sdp-orig/sdp_recv.c 2005-02-14 13:58:45.877005000 -0800 +++ drivers/infiniband/ulp/sdp/sdp_recv.c 2005-02-17 12:03:28.285005000 -0800 @@ -1,34 +1,46 @@ /* - This software is available to you under a choice of one o

[openib-general] [PATCH][SDP][27/33] Fix comment style in sdp_read.c

Signed-off-by: Tom Duffy <[EMAIL PROTECTED]> --- drivers/infiniband/ulp/sdp-orig/sdp_read.c 2005-02-15 16:33:46.22000 -0800 +++ drivers/infiniband/ulp/sdp/sdp_read.c 2005-02-17 11:30:04.047019000 -0800 @@ -1,36 +1,46 @@ /* - This software is available to you under a choice of one o

Re: [openib-general] SDP_CONN_LOCK

Quoting r. Libor Michalek <[EMAIL PROTECTED]>: > Subject: Re: [openib-general] SDP_CONN_LOCK > > On Thu, Feb 17, 2005 at 11:29:37PM +0200, Michael S. Tsirkin wrote: > > Quoting r. Libor Michalek <[EMAIL PROTECTED]>: > > > Subject: Re: [openib-general] SDP_CONN_LOCK > > > > > > On Thu, Feb 17, 20

[openib-general] [PATCH][SDP][26/33] Fix comment style in sdp_rcvd.c

Signed-off-by: Tom Duffy <[EMAIL PROTECTED]> --- drivers/infiniband/ulp/sdp-orig/sdp_rcvd.c 2005-02-15 16:33:46.210001000 -0800 +++ drivers/infiniband/ulp/sdp/sdp_rcvd.c 2005-02-17 11:26:31.11000 -0800 @@ -1,35 +1,43 @@ /* - This software is available to you under a choice of one o

[openib-general] [PATCH][SDP][25/33] Fix comment style in sdp_queue.h

Signed-off-by: Tom Duffy <[EMAIL PROTECTED]> --- drivers/infiniband/ulp/sdp-orig/sdp_queue.h 2005-02-10 18:42:56.256023000 -0800 +++ drivers/infiniband/ulp/sdp/sdp_queue.h 2005-02-17 11:22:36.106017000 -0800 @@ -1,25 +1,36 @@ /* - This software is available to you under a choice of one o

[openib-general] [PATCH][SDP][24/33] Fix comment style in sdp_queue.c

Signed-off-by: Tom Duffy <[EMAIL PROTECTED]> --- drivers/infiniband/ulp/sdp-orig/sdp_queue.c 2005-02-10 18:42:57.12000 -0800 +++ drivers/infiniband/ulp/sdp/sdp_queue.c 2005-02-17 11:21:40.495015000 -0800 @@ -1,37 +1,48 @@ /* - This software is available to you under a choice of one o

[openib-general] [PATCH][SDP][23/33] Fix comment style in sdp_proto.h

Signed-off-by: Tom Duffy <[EMAIL PROTECTED]> --- drivers/infiniband/ulp/sdp-orig/sdp_proto.h 2005-02-10 18:42:56.417024000 -0800 +++ drivers/infiniband/ulp/sdp/sdp_proto.h 2005-02-17 11:17:48.805008000 -0800 @@ -1,25 +1,36 @@ /* - This software is available to you under a choice of one o

[openib-general] [PATCH][SDP][22/33] Fix comment style in sdp_proc.h

Signed-off-by: Tom Duffy <[EMAIL PROTECTED]> --- drivers/infiniband/ulp/sdp-orig/sdp_proc.h 2005-02-10 18:42:56.751003000 -0800 +++ drivers/infiniband/ulp/sdp/sdp_proc.h 2005-02-17 11:13:23.59802 -0800 @@ -1,25 +1,36 @@ /* - This software is available to you under a choice of one o

[openib-general] [PATCH][SDP][21/33] Fix comment style in sdp_proc.c

Signed-off-by: Tom Duffy <[EMAIL PROTECTED]> --- drivers/infiniband/ulp/sdp-orig/sdp_proc.c 2005-02-14 11:13:15.600961000 -0800 +++ drivers/infiniband/ulp/sdp/sdp_proc.c 2005-02-17 11:12:47.182006000 -0800 @@ -1,38 +1,49 @@ /* - This software is available to you under a choice of one o

[openib-general] [PATCH][SDP][20/33] Fix comment style in sdp_post.c

Signed-off-by: Tom Duffy <[EMAIL PROTECTED]> --- drivers/infiniband/ulp/sdp-orig/sdp_post.c 2005-02-10 18:42:57.157027000 -0800 +++ drivers/infiniband/ulp/sdp/sdp_post.c 2005-02-17 11:11:01.646018000 -0800 @@ -1,30 +1,42 @@ /* - This software is available to you under a choice of one o

[openib-general] [PATCH][SDP][19/33] Fix comment style in sdp_pass.c

Signed-off-by: Tom Duffy <[EMAIL PROTECTED]> --- drivers/infiniband/ulp/sdp-orig/sdp_pass.c 2005-02-15 16:33:46.238009000 -0800 +++ drivers/infiniband/ulp/sdp/sdp_pass.c 2005-02-17 11:08:43.171002000 -0800 @@ -1,30 +1,42 @@ /* - This software is available to you under a choice of one o

[openib-general] [PATCH][SDP][18/33] Fix comment style in sdp_msgs.h

Signed-off-by: Tom Duffy <[EMAIL PROTECTED]> --- drivers/infiniband/ulp/sdp-orig/sdp_msgs.h 2005-02-10 18:42:56.449031000 -0800 +++ drivers/infiniband/ulp/sdp/sdp_msgs.h 2005-02-17 11:07:20.058017000 -0800 @@ -1,25 +1,36 @@ /* - This software is available to you under a choice of one o

[openib-general] [PATCH][SDP][17/33] Fix comment style in sdp_main.h

Signed-off-by: Tom Duffy <[EMAIL PROTECTED]> --- drivers/infiniband/ulp/sdp-orig/sdp_main.h 2005-02-14 11:20:29.04899 -0800 +++ drivers/infiniband/ulp/sdp/sdp_main.h 2005-02-17 11:04:28.550001000 -0800 @@ -1,25 +1,36 @@ /* - This software is available to you under a choice of one o

[openib-general] [PATCH][SDP][16/33] Fix comment style in sdp_link.h

Signed-off-by: Tom Duffy <[EMAIL PROTECTED]> --- drivers/infiniband/ulp/sdp-orig/sdp_link.h 2005-02-10 18:42:56.37000 -0800 +++ drivers/infiniband/ulp/sdp/sdp_link.h 2005-02-17 11:04:05.652004000 -0800 @@ -1,25 +1,36 @@ /* - This software is available to you under a choice of one o

[openib-general] [PATCH][SDP][15/33] Fix comment style in sdp_link.c

Signed-off-by: Tom Duffy <[EMAIL PROTECTED]> --- drivers/infiniband/ulp/sdp-orig/sdp_link.c 2005-02-10 18:42:56.318026000 -0800 +++ drivers/infiniband/ulp/sdp/sdp_link.c 2005-02-17 11:03:42.673005000 -0800 @@ -1,25 +1,37 @@ /* - This software is available to you under a choice of one o

[openib-general] [PATCH][SDP][14/33] Fix comment style in sdp_kvec.c

Signed-off-by: Tom Duffy <[EMAIL PROTECTED]> --- drivers/infiniband/ulp/sdp-orig/sdp_kvec.c 2005-02-10 18:42:56.228029000 -0800 +++ drivers/infiniband/ulp/sdp/sdp_kvec.c 2005-02-17 10:58:33.716019000 -0800 @@ -1,35 +1,46 @@ /* - This software is available to you under a choice of one o

[openib-general] [PATCH][SDP][13/33] Fix comment style in sdp_iocb.h

Signed-off-by: Tom Duffy <[EMAIL PROTECTED]> --- drivers/infiniband/ulp/sdp-orig/sdp_iocb.h 2005-02-10 18:42:56.974015000 -0800 +++ drivers/infiniband/ulp/sdp/sdp_iocb.h 2005-02-17 10:55:51.305015000 -0800 @@ -1,25 +1,36 @@ /* - This software is available to you under a choice of one o

[openib-general] [PATCH][SDP][12/33] Fix comment style in sdp_iocb.c

Signed-off-by: Tom Duffy <[EMAIL PROTECTED]> --- drivers/infiniband/ulp/sdp-orig/sdp_iocb.c 2005-02-14 11:13:15.582961000 -0800 +++ drivers/infiniband/ulp/sdp/sdp_iocb.c 2005-02-17 12:21:44.892003000 -0800 @@ -1,42 +1,53 @@ /* - This software is available to you under a choice of one o

[openib-general] [PATCH][SDP][11/33] Fix comment style in sdp_inet.c

Signed-off-by: Tom Duffy <[EMAIL PROTECTED]> --- drivers/infiniband/ulp/sdp-orig/sdp_inet.c 2005-02-11 17:50:31.98000 -0800 +++ drivers/infiniband/ulp/sdp/sdp_inet.c 2005-02-17 10:51:06.250001000 -0800 @@ -1,29 +1,41 @@ /* - This software is available to you under a choice of one o

[openib-general] [PATCH][SDP][10/33] Fix comment style in sdp_event.c

Signed-off-by: Tom Duffy <[EMAIL PROTECTED]> --- drivers/infiniband/ulp/sdp-orig/sdp_event.c 2005-02-15 16:33:46.254005000 -0800 +++ drivers/infiniband/ulp/sdp/sdp_event.c 2005-02-17 10:43:35.172004000 -0800 @@ -1,35 +1,46 @@ /* - This software is available to you under a choice of one o

[openib-general] [PATCH][SDP][9/33] Fix comment style in sdp_dev.h

Signed-off-by: Tom Duffy <[EMAIL PROTECTED]> --- drivers/infiniband/ulp/sdp-orig/sdp_dev.h 2005-02-11 17:27:53.697025000 -0800 +++ drivers/infiniband/ulp/sdp/sdp_dev.h2005-02-17 10:40:47.692004000 -0800 @@ -1,25 +1,36 @@ /* - This software is available to you under a choice of one o

[openib-general] [PATCH][SDP][8/33] Fix comment style in sdp_conn.h

Signed-off-by: Tom Duffy <[EMAIL PROTECTED]> --- drivers/infiniband/ulp/sdp-orig/sdp_conn.h 2005-02-14 11:13:15.571961000 -0800 +++ drivers/infiniband/ulp/sdp/sdp_conn.h 2005-02-17 12:44:49.618014000 -0800 @@ -1,25 +1,36 @@ /* - This software is available to you under a choice of one o

[openib-general] [PATCH][SDP][7/33] Fix comment style in sdp_conn.c

Signed-off-by: Tom Duffy <[EMAIL PROTECTED]> --- drivers/infiniband/ulp/sdp-orig/sdp_conn.c 2005-02-14 12:51:45.936001000 -0800 +++ drivers/infiniband/ulp/sdp/sdp_conn.c 2005-02-17 10:38:44.836003000 -0800 @@ -1,25 +1,36 @@ /* - This software is available to you under a choice of one o

[openib-general] [PATCH][SDP][6/33] Fix comment style in sdp_buff_p.h

Signed-off-by: Tom Duffy <[EMAIL PROTECTED]> --- drivers/infiniband/ulp/sdp-orig/sdp_buff_p.h2005-02-10 18:42:56.714018000 -0800 +++ drivers/infiniband/ulp/sdp/sdp_buff_p.h 2005-02-17 10:28:48.95000 -0800 @@ -1,25 +1,36 @@ /* - This software is available to you under a choice

[openib-general] [PATCH][SDP][5/33] Fix comment style in sdp_buff.h

Signed-off-by: Tom Duffy <[EMAIL PROTECTED]> --- drivers/infiniband/ulp/sdp-orig/sdp_buff.h 2005-02-11 07:59:56.184009000 -0800 +++ drivers/infiniband/ulp/sdp/sdp_buff.h 2005-02-17 10:28:23.514011000 -0800 @@ -1,25 +1,36 @@ /* - This software is available to you under a choice of one o

[openib-general] [PATCH][SDP][4/33] Fix comment style in sdp_buff.c

Signed-off-by: Tom Duffy <[EMAIL PROTECTED]> --- drivers/infiniband/ulp/sdp-orig/sdp_buff.c 2005-02-11 08:02:05.883004000 -0800 +++ drivers/infiniband/ulp/sdp/sdp_buff.c 2005-02-17 10:27:46.25000 -0800 @@ -1,25 +1,36 @@ /* - This software is available to you under a choice of one o

[openib-general] [PATCH][SDP][3/33] Fix comment style in sdp_advt.h

Signed-off-by: Tom Duffy <[EMAIL PROTECTED]> --- drivers/infiniband/ulp/sdp-orig/sdp_advt.h 2005-02-10 18:42:56.640003000 -0800 +++ drivers/infiniband/ulp/sdp/sdp_advt.h 2005-02-17 12:45:25.967016000 -0800 @@ -1,38 +1,50 @@ /* - This software is available to you under a choice of one o

[openib-general] [PATCH][SDP][2/33] Fix comment style in sdp_advt.c

Signed-off-by: Tom Duffy <[EMAIL PROTECTED]> --- drivers/infiniband/ulp/sdp-orig/sdp_advt.c 2005-02-10 18:42:56.537025000 -0800 +++ drivers/infiniband/ulp/sdp/sdp_advt.c 2005-02-17 10:22:17.245012000 -0800 @@ -1,44 +1,53 @@ /* - This software is available to you under a choice of one of

[openib-general] [PATCH][SDP][1/33] Fix comment style in sdp_actv.c

Signed-off-by: Tom Duffy <[EMAIL PROTECTED]> --- drivers/infiniband/ulp/sdp-orig/sdp_actv.c 2005-02-15 16:33:46.230002000 -0800 +++ drivers/infiniband/ulp/sdp/sdp_actv.c 2005-02-17 10:19:08.821002000 -0800 @@ -1,35 +1,46 @@ /* - This software is available to you under a choice of one of

[openib-general] [PATCH][SDP][0/33] Fix comment style in SDP

These patches clean up the comment formatting in sdp. They also update the copyright header to match the rest of OpenIB files. Signed-off-by: Tom Duffy <[EMAIL PROTECTED]> ___ openib-general mailing list openib-general@openib.org http://openib.org/mail

Re: [openib-general] SDP_CONN_LOCK

BTW, since mthca currently calls completion handlers directly from interrupt context (rather than BH/tasklet context), it might be worth renaming all the SDP locking macros so they're not confusingly named with _BH suffixes. - R. ___ openib-general mail

Re: [openib-general] SDP_CONN_LOCK

On Thu, Feb 17, 2005 at 11:29:37PM +0200, Michael S. Tsirkin wrote: > Quoting r. Libor Michalek <[EMAIL PROTECTED]>: > > Subject: Re: [openib-general] SDP_CONN_LOCK > > > > On Thu, Feb 17, 2005 at 05:35:17PM +0200, Michael S. Tsirkin wrote: > > > Hi, Libor! > > > Could you please explain what are

Re: [openib-general] SDP_CONN_LOCK

Quoting r. Libor Michalek <[EMAIL PROTECTED]>: > Subject: Re: [openib-general] SDP_CONN_LOCK > > On Thu, Feb 17, 2005 at 05:35:17PM +0200, Michael S. Tsirkin wrote: > > Hi, Libor! > > Could you please explain what are the SDP_CONN_LOCK > > and friends doing in SDP? > > > > It seems they just imp

Re: [openib-general] Solaris IPoIB MTU with OpenSM

Hi Nitin, On Wed, 2005-02-16 at 17:33, Nitin Hande wrote: > On Wed, 2005-02-16 at 13:26, Hal Rosenstock wrote: > > On Wed, 2005-02-16 at 16:08, Nitin Hande wrote: > > > Hal, > [snip..] > > > > > > > > > Here is the trace of 256 sized MTU: > > > > > > Outgoing MAD: > > > BaseVersion: 0x

[openib-general] [PATCH] [CM] add stale connection checks on passive side

This patch adds checks for a stale connection to the passive side of the connection protocol. Checks on the active side are still needed. Signed-off-by: Sean Hefty <[EMAIL PROTECTED]> Index: infiniband/core/cm.c === --- infiniband/c

Re: [openib-general] [RFC] CM reject handling and redirection

On Thu, 2005-02-17 at 15:10, Roland Dreier wrote: > It is orthogonal in a sense, but consider something like the > following. I set up a connection with a primary path and a secondary > path, using the same primary path for CM communication. Some time > later the primary path fails, so APM causes

Re: [openib-general] [RFC] CM reject handling and redirection

Hal Rosenstock wrote: Code 25 (port redirection only) raises a separate issue that I think is orthogonal to REJ handling. The CM spec always allows the CM to How much complexity is added to the CM to support this ? Do we really need to (and if so what is the application which requires this) ? I t

Re: [openib-general] [RFC] CM reject handling and redirection

Hal> I'm not sure I get the "connection" between the CM path and Hal> data path. Doesn't failover and APM apply only to the data Hal> path ? Is it purely the detected of a failed path between CMs Hal> (which seems orthogonal to this to me) ? That's not to say Hal> that this is

Re: [openib-general] [RFC] CM reject handling and redirection

On Thu, 2005-02-17 at 12:51, Sean Hefty wrote: > It looks like one way to handle this is to add a new field to struct > ib_cm_req_param: struct ib_sa_path_rec *cm_path. A user could easily > set this to the same path record referenced by the *primary_path field. > This would only be a small c

Re: [openib-general] [RFC] CM reject handling and redirection

On Wed, 2005-02-16 at 22:50, Roland Dreier wrote: > Code 25 (port redirection only) raises a separate issue that I think > is orthogonal to REJ handling. The CM spec always allows the CM to > run on a different port from the destination of a connection so if we > want to implement the whole spec w

Re: [openib-general] [PATCH][IPOIB] bring interface down panic

Sorry, it's an oops not a panic. Thanks Shirley Ma IBM Linux Technology Center 15300 SW Koll Parkway Beaverton, OR 97006-6063 Phone(Fax): (503) 578-7638___ openib-general mailing list openib-general@openib.org http://openib.org/mailman/listinfo/openib-

Re: [openib-general] SDP_CONN_LOCK

On Thu, Feb 17, 2005 at 05:35:17PM +0200, Michael S. Tsirkin wrote: > Hi, Libor! > Could you please explain what are the SDP_CONN_LOCK > and friends doing in SDP? > > It seems they just implement exclusive access to socket, > but if so, why is a simple semaphore or mutex not used? They do imple

Re: [openib-general] [PATCH]deadlock problem in ipoib

Hi, Roland, I've created a new patch. Please review this one. diff -urN infiniband/ulp/ipoib/ipoib_main.c infiniband-lock/ulp/ipoib/ipoib_main.c --- infiniband/ulp/ipoib/ipoib_main.c        2005-02-17 17:24:56.0 + +++ infiniband-lock/ulp/ipoib/ipoib_main.c        2005-02-17 17:46:43

Re: [openib-general] [RFC] CM reject handling and redirection

Roland Dreier wrote: Code 25 (port redirection only) raises a separate issue that I think is orthogonal to REJ handling. The CM spec always allows the CM to run on a different port from the destination of a connection so if we want to implement the whole spec we do need a way for the consumer to s

RE: [openib-general] 2005 OpenIB Developers Workshop presentations

James Wrote, > We are proposing to maintain the code in both locations. The OpenIB > specific implementation would be on openib.org while a > platform/transport independent implementation would continue on > our SourceForge site. > I also agree this seems like the right approach. I don't th

[openib-general] [PATCH] SDP headers using unsafe macros

Some SDP hdeaders defines unsafe macros, which may be dangerous to use, e.g. inside an if or while statement. Signed-off-by: Michael S. Tsirkin <[EMAIL PROTECTED]> Index: sdp_proto.h === --- sdp_proto.h (revision 1810) +++ sdp_proto

[openib-general] SDP_CONN_LOCK

Hi, Libor! Could you please explain what are the SDP_CONN_LOCK and friends doing in SDP? It seems they just implement exclusive access to socket, but if so, why is a simple semaphore or mutex not used? Thanks, -- MST - Michael S. Tsirkin ___ openib-ge

[openib-general] Re: ib_mthca 0000:04:00.0: CQ overrun on CQN 00000082, [patchlet] for opensm --guid

On Thu, 2005-02-10 at 11:22, Bernhard Fischer wrote: > small nitpick to opensm is attached. I'm not keen on implementing > the menu stuff, just listing them is good enough for me. Please fix > properly. Thanks. Applied with minor format changes. -- Hal __

Re: [openib-general] Incorrect endian in GUID comparison/SM master selection

Hi Brian, On Wed, 2005-02-16 at 21:37, Brian Eng wrote: > Per the IBA spec, the selection of a master SM is determined by the > GUID (lowest wins). When OpenSM does its comparison, it does not adjust > for the endian of the host. For our x86 systems, this means the > comparison is based on the

[openib-general] [RFC] Diagnostic tree structure

Hi, The current userspace diagnostics tree structure has host and net subdirectories. The distinction between the two is blurring so we would like to flatten the tree and just have all the tools under diags. Any objections ? Thanks. -- Hal ___ openib-

[openib-general] Re: RFC on SDP checkin

Quoting r. Libor Michalek <[EMAIL PROTECTED]>: > Subject: Re: RFC on SDP checkin > Possible Issues > > > [...] > > - Memory locking for AIO requires a call to do_mlock() which is not a > kernel exported function, the method for calling the function is not > standard. Libor, in my

RE: [openib-general] question on opensm error

Hi, > > There is a sys fail red light on the CPU on the 96-port switch that the > > opensm host attaches to. > > > > What's weird is none of the ib admin tools found anything. ibnetdiscover > > happily walked the whole subnet. The only problem was that opensm would > > not run, but the errors were

[openib-general] Re:$B??M}$G$9!#(B

$B!ZA0!Y!'??M}(B $B!XG/Np!Y!'(B37$B:P(B $B!X?&6H!Y!'2qhttp://cutybee.hosting-geomax.jp/secret.php?pr=ext458 $B"(8=:_!"??N$$5$s$+$i$N;XL>%a!<%k$O5.J}MM$X$N0lDL$N$_$H$J$C$F$*$j$^$9!#(B $B%U%j!<%a!<%k%"%I%l%9$+$i$NEPO?$K$bBP1~$7$F$$$^$9!#(B (B (B___