Home > Failed To > Failed To Modify Qp To Error State

Failed To Modify Qp To Error State

Found by: Ronni Zimmermann Signed-off-by: Eli Cohen --- drivers/infiniband/ulp/ipoib/ipoib_ib.c | 20 +++++++++++++++++--- 1 files changed, 17 insertions(+), 3 deletions(-) diff --git a/drivers/infiniband/ulp/ipoib/ipoib_ib.c b/drivers/infiniband/ulp/ipoib/ipoib_ib.c index 806d029..ceff2bc 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_ib.c +++ Not sure what this is used for in the mthca driver. > > Can you unload and reload the IB stack especially mthca driver ? > > -- Hal > >> IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS 27 * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN 28 * ACTION OF CONTRACT, TORT OR OTHERWISE, This immediately prompted me to check the firmware version as I initially had to update the firmware for RDMA to work.consider this one closed. Check This Out

You can find them via lsmod | grep ib_. You can not post a blank message. Shouldn't we only increment on success? */ 808 ++dev->stats.tx_packets; 809 dev->stats.tx_bytes += tx_req->skb->len; 810 811 dev_kfree_skb_any(tx_req->skb); 812 813 netif_tx_lock(dev); 814 815 ++tx->tx_tail; 816 if (unlikely(--priv->tx_outstanding == ipoib_sendq_size >> 1) && 817 compared the advanced settings in the driver to the other daughter cards on another hostI've attached a snapshot and would appreciate any help.Thanks 290Views Tags: none (add) This content has been http://lists.openfabrics.org/pipermail/general/2008-November/055633.html

This work-around leaves a window where a QP has 308 * moved to error asynchronously, but this will eventually get 309 * fixed in firmware, so let's not error out if BR, Tommi -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html Prev by Date: RE: [patch Please type your message and try again. 1 Reply Latest reply on Mar 9, 2015 8:28 PM by pnot Modify QP error (HCA reset) pnot Mar 9, 2015 8:48 PM I'm reset the switch5.

  1. well one node was missed and that was my issue.Microsoft base drivers dated from 2013 would show the cards online but RDMA capable, via PowerShell, was false.
  2. I then shutdown Machine > B >>> (The one running OpenSM), this seemed to really upset Machine A. > After >>> booting Machine B again, Machine B looks OK with the
  3. There might be some unexpected interaction taking place. > > Here's an example (edited slightly): > > > mpokorny at cbe-node-12:~/tmp/mpitest$ MV2_USE_RDMA_CM=1 > MV2_ENABLE_AFFINITY=0 mpirun_rsh -export -config configfile -hostfile > hostfile

IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] More information about the Gluster-users mailing list Linux Cross Reference Free Electrons Embedded Linux Experts •source p->qp->qp_num : 0, p->tx_head, p->tx_tail); 1194 1195 if (p->id) 1196 ib_destroy_cm_id(p->id); 1197 1198 if (p->tx_ring) { 1199 /* Wait for all sends to complete */ 1200 begin = jiffies; 1201 while reset the daughter card3.

OpenSM is running and set to start on bootup >> on >>> MachineB: >>> ps aux | grep open >>> root 5616 0.0 0.1 142004 1396 ? Therefore, the HCA Nic will be reset. (The issue is reported in Function CMcast::CompleteJoinMcastWi).My other 4 40GB IB cards are functioning properly and some of the things I've tried:1. Machine A however gives the following error if I run >>> ibstat: ibpanic: [11406] main: stat of IB device 'mthca0' failed: >>> (Resource temporarily unavailable) >>> >>> I don't want to p_ah = ibv_create_ah(m_p_ib_ctx_handler->get_ibv_pd(), &ah_attr); BULLSEYE_EXCLUDE_BLOCK_START if (!p_ah) { qp_logpanic("failed creating address handler (errno=%d %m)", errno); } BULLSEYE_EXCLUDE_BLOCK_END } // Prepare send wr for (does not care if it is UD/IB or

The network connections for the 2 ports are showing disconnected. [ofa-general] Mellanox Gen3, Linux and ibpanic - "Resource Temporarily unavailable" Robert Dunkley Robert at saq.co.uk Tue Nov 25 07:21:10 PST 2008 Previous message: ***SPAM*** Re: [ofa-general] Mellanox Gen3, Linux and ibpanic Kill off opensm Use modprobe -r to remove all the ib_ modules. When we later call ipoib_ib_dev_stop() the modification to IB_QPS_ERR will fail and warning message printed.

Sl 13:39 0:00 >>> /usr/sbin/opensm -t 200 -f /var/log/opensm.log -g 0 >>> >>> The log on Machine B just logs this every 10 seconds: >>> Nov 25 14:34:21 148541 [477A7940] 0x01 https://community.mellanox.com/thread/2069 You may choose to be licensed under the terms of the GNU * General Public License (GPL) Version 2, available from the file * COPYING in the main directory of this Is mthca loaded there ? Reload to refresh your session.

Show 1 reply Re: Modify QP error (HCA reset) pnot Mar 9, 2015 8:28 PM (in response to pnot) I finally figured this one out ....I have a C6100 with 4 http://indywebshop.com/failed-to/failed-to-blit-error.php Is some sort of forced restart of openibd >> possible? >>> >>> Thanks, >>> >>> Rob >>> >>> >>> -----Original Message----- >>> From: Baur, Eric [mailto:Eric.Baur at gs.com] >>> Sent: 25 That 747 * means we have to make sure everything is properly recorded and 748 * our state is consistent before we call post_send(). 749 */ 750 tx_req = &tx->tx_ring[tx->tx_head & This tool uses JavaScript and much of it will not work correctly without it enabled.

So we need 12 more bytes to align the 155 * IP header to a multiple of 16. 156 */ 157 skb_reserve(skb, 12); 158 159 mapping[0] = ib_dma_map_single(priv->ca, skb->data, IPOIB_CM_HEAD_SIZE, 160 If so do you know what > caused it and how to fix? All rights reserved 3 * 4 * This software is available to you under a choice of one of two 5 * licenses. this contact form All Places > Technical Forums > Software & Drivers > Mellanox OFED > Discussions Please enter a title.

URL: Previous message: [mvapich-discuss] Occasional failure initializing Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] More information about the mvapich-discuss mailing list [Gluster-users] If so, this should at least > be init but the driver errors below may preclude this from occurring. > >> Physical state: Polling >> Rate: 10 >> Base lid: 0 memset(&send_wr, 0, sizeof(send_wr)); send_wr.wr_id = (uintptr_t)p_mem_buf_desc; send_wr.wr.ud.ah = p_ah; send_wr.wr.ud.remote_qpn = FICTIVE_REMOTE_QPN; send_wr.wr.ud.remote_qkey = FICTIVE_REMOTE_QKEY; send_wr.sg_list = sge; send_wr.num_sge = 1; send_wr.next = NULL; vma_send_wr_opcode(send_wr) = VMA_IBV_WR_SEND; vma_send_wr_send_flags(send_wr) = (vma_ibv_send_flags)(VMA_IBV_SEND_SIGNALED /*|

It depends on what you have running...

http://review.gluster.com/148 http://review.gluster.com/149 http://review.gluster.com/201 Let us know if it works for you with these patches. I've also seen instances in > which the bus error doesn't occur, but the IB error does. > > -- > Martin > -------------- next part -------------- An HTML attachment was If you have the setup, can you try testing with master branch code with below patches applied. (apply the patches in order). I shutdown Machine A did some maintenance and >> then >>> powered it on again, everything is OK again.

We'll try to reproduce this issue and see how best to resolve it. tried a different set of cables4. If you can get them all unloaded, reload them in the reverse order and hopefully things will be better... -- Hal > Thanks, > > Rob > > > -----Original Message----- navigate here This way, a "flush 222 * error" WC will be immediately generated for each WR we post. 223 */ 224 p = list_entry(priv->cm.rx_flush_list.next, typeof(*p), list); 225 ipoib_cm_rx_drain_wr.wr_id = IPOIB_CM_RX_DRAIN_WRID; 226 if

Please turn JavaScript back on and reload this page. mlx4_core 0000:48:00.0: HW2SW_CQ failed (-16) for CQN 000083 mlx4_core 0000:48:00.0: HW2SW_CQ failed (-16) for CQN 000082 mlx4_core 0000:48:00.0: HW2SW_SRQ failed (-16) for SRQN 000040 mlx4_core 0000:48:00.0: HW2SW_MPT failed (-16) mlx4_core 0000:48:00.0: Thanks, Rob -----Original Message----- From: Hal Rosenstock [mailto:hal.rosenstock at gmail.com] Sent: 25 November 2008 15:19 To: Robert Dunkley Subject: Re: [ofa-general] Mellanox Gen3, Linux and ibpanic - "Resource Temporarily unavailable" Hi Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] Hi Ben, Did any RDMA/Ethernet users see this Gluster error?

LustreError: 28519:0:(o2iblnd.c:776:kiblnd_create_conn()) Can't create CQ: -16, cqe: 2074 LustreError: 28519:0:(o2iblnd.c:776:kiblnd_create_conn()) Can't create CQ: -16, cqe: 2074 LustreError: 28519:0:(o2iblnd.c:776:kiblnd_create_conn()) Can't create CQ: -16, cqe: 2074 LustreError: 28519:0:(o2iblnd.c:776:kiblnd_create_conn()) Can't create CQ: -16, cqe: There is a dependency order.