mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: "Bryan O'Sullivan" <bos@pathscale.com>
To: rdreier@cisco.com
Cc: openib-general@openib.org, linux-kernel@vger.kernel.org
Subject: [PATCH 18 of 23] IB/ipath - put a limit on the number of QPs that can be created
Date: Fri, 25 Aug 2006 11:24:43 -0700	[thread overview]
Message-ID: <02d9f3ef2291dea0f4d8.1156530283@eng-12.pathscale.com> (raw)
In-Reply-To: <patchbomb.1156530265@eng-12.pathscale.com>

Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com>

diff --git a/drivers/infiniband/hw/ipath/ipath_qp.c b/drivers/infiniband/hw/ipath/ipath_qp.c
--- a/drivers/infiniband/hw/ipath/ipath_qp.c	Fri Aug 25 11:19:45 2006 -0700
+++ b/drivers/infiniband/hw/ipath/ipath_qp.c	Fri Aug 25 11:19:45 2006 -0700
@@ -833,9 +833,21 @@ struct ib_qp *ipath_create_qp(struct ib_
 		}
 	}
 
+	spin_lock(&dev->n_qps_lock);
+	if (dev->n_qps_allocated == ib_ipath_max_qps) {
+		spin_unlock(&dev->n_qps_lock);
+		ret = ERR_PTR(-ENOMEM);
+		goto bail_ip;
+	}
+
+	dev->n_qps_allocated++;
+	spin_unlock(&dev->n_qps_lock);
+
 	ret = &qp->ibqp;
 	goto bail;
 
+bail_ip:
+	kfree(qp->ip);
 bail_rwq:
 	vfree(qp->r_rq.wq);
 bail_qp:
@@ -864,6 +876,9 @@ int ipath_destroy_qp(struct ib_qp *ibqp)
 	spin_lock_irqsave(&qp->s_lock, flags);
 	qp->state = IB_QPS_ERR;
 	spin_unlock_irqrestore(&qp->s_lock, flags);
+	spin_lock(&dev->n_qps_lock);
+	dev->n_qps_allocated--;
+	spin_unlock(&dev->n_qps_lock);
 
 	/* Stop the sending tasklet. */
 	tasklet_kill(&qp->s_task);
diff --git a/drivers/infiniband/hw/ipath/ipath_verbs.c b/drivers/infiniband/hw/ipath/ipath_verbs.c
--- a/drivers/infiniband/hw/ipath/ipath_verbs.c	Fri Aug 25 11:19:45 2006 -0700
+++ b/drivers/infiniband/hw/ipath/ipath_verbs.c	Fri Aug 25 11:19:45 2006 -0700
@@ -72,6 +72,10 @@ module_param_named(max_qp_wrs, ib_ipath_
 module_param_named(max_qp_wrs, ib_ipath_max_qp_wrs, uint,
 		   S_IWUSR | S_IRUGO);
 MODULE_PARM_DESC(max_qp_wrs, "Maximum number of QP WRs to support");
+
+unsigned int ib_ipath_max_qps = 16384;
+module_param_named(max_qps, ib_ipath_max_qps, uint, S_IWUSR | S_IRUGO);
+MODULE_PARM_DESC(max_qps, "Maximum number of QPs to support");
 
 unsigned int ib_ipath_max_sges = 0x60;
 module_param_named(max_sges, ib_ipath_max_sges, uint, S_IWUSR | S_IRUGO);
@@ -958,7 +962,7 @@ static int ipath_query_device(struct ib_
 	props->sys_image_guid = dev->sys_image_guid;
 
 	props->max_mr_size = ~0ull;
-	props->max_qp = dev->qp_table.max;
+	props->max_qp = ib_ipath_max_qps;
 	props->max_qp_wr = ib_ipath_max_qp_wrs;
 	props->max_sge = ib_ipath_max_sges;
 	props->max_cq = ib_ipath_max_cqs;
@@ -1420,6 +1424,7 @@ int ipath_register_ib_device(struct ipat
 	spin_lock_init(&idev->n_pds_lock);
 	spin_lock_init(&idev->n_ahs_lock);
 	spin_lock_init(&idev->n_cqs_lock);
+	spin_lock_init(&idev->n_qps_lock);
 	spin_lock_init(&idev->n_srqs_lock);
 	spin_lock_init(&idev->n_mcast_grps_lock);
 
diff --git a/drivers/infiniband/hw/ipath/ipath_verbs.h b/drivers/infiniband/hw/ipath/ipath_verbs.h
--- a/drivers/infiniband/hw/ipath/ipath_verbs.h	Fri Aug 25 11:19:45 2006 -0700
+++ b/drivers/infiniband/hw/ipath/ipath_verbs.h	Fri Aug 25 11:19:45 2006 -0700
@@ -482,6 +482,8 @@ struct ipath_ibdev {
 	spinlock_t n_ahs_lock;
 	u32 n_cqs_allocated;	/* number of CQs allocated for device */
 	spinlock_t n_cqs_lock;
+	u32 n_qps_allocated;	/* number of QPs allocated for device */
+	spinlock_t n_qps_lock;
 	u32 n_srqs_allocated;	/* number of SRQs allocated for device */
 	spinlock_t n_srqs_lock;
 	u32 n_mcast_grps_allocated; /* number of mcast groups allocated */
@@ -792,6 +794,8 @@ extern unsigned int ib_ipath_max_cqs;
 
 extern unsigned int ib_ipath_max_qp_wrs;
 
+extern unsigned int ib_ipath_max_qps;
+
 extern unsigned int ib_ipath_max_sges;
 
 extern unsigned int ib_ipath_max_mcast_grps;

  parent reply	other threads:[~2006-08-25 18:27 UTC|newest]

Thread overview: 36+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-08-25 18:24 [PATCH 0 of 23] IB/ipath - updates for 2.6.19 Bryan O'Sullivan
2006-08-25 18:24 ` [PATCH 1 of 23] IB/ipath - More changes to support InfiniPath on PowerPC 970 systems Bryan O'Sullivan
2006-08-25 19:45   ` Roland Dreier
2006-08-25 20:19     ` Bryan O'Sullivan
2006-08-29 23:04       ` Brendan Cully
2006-08-25 20:01   ` Roland Dreier
2006-08-25 18:24 ` [PATCH 2 of 23] IB/ipath - lock resource limit counters correctly Bryan O'Sullivan
2006-08-25 18:24 ` [PATCH 3 of 23] IB/ipath - fix for crash on module unload, if cfgports < portcnt Bryan O'Sullivan
2006-08-25 18:24 ` [PATCH 4 of 23] IB/ipath - fix handling of kpiobufs Bryan O'Sullivan
2006-08-25 18:24 ` [PATCH 5 of 23] IB/ipath - drop requirement that PIO buffers be mmaped write-only Bryan O'Sullivan
2006-08-25 18:24 ` [PATCH 6 of 23] IB/ipath - merge ipath_core and ib_ipath drivers Bryan O'Sullivan
2006-08-25 18:24 ` [PATCH 7 of 23] IB/ipath - simplify layering code Bryan O'Sullivan
2006-08-25 18:24 ` [PATCH 8 of 23] IB/ipath - simplify debugging code after ipath_core and ib_ipath merger Bryan O'Sullivan
2006-08-25 18:24 ` [PATCH 9 of 23] IB/ipath - remove stale references to userspace SMA Bryan O'Sullivan
2006-08-25 18:24 ` [PATCH 10 of 23] IB/ipath - trivial cleanups Bryan O'Sullivan
2006-08-25 18:24 ` [PATCH 11 of 23] IB/ipath - add new minor device to allow sending of diag packets Bryan O'Sullivan
2006-08-25 19:50   ` Roland Dreier
2006-08-25 20:20     ` Bryan O'Sullivan
2006-08-25 18:24 ` [PATCH 12 of 23] IB/ipath - do not allow use of CQ entries with invalid counts Bryan O'Sullivan
2006-08-25 18:24 ` [PATCH 13 of 23] IB/ipath - account for attached QPs correctly Bryan O'Sullivan
2006-08-25 18:24 ` [PATCH 14 of 23] IB/ipath - support new QLogic product naming scheme Bryan O'Sullivan
2006-08-25 18:24 ` [PATCH 15 of 23] IB/ipath - add serial number to hardware freeze error message Bryan O'Sullivan
2006-08-25 18:24 ` [PATCH 16 of 23] IB/ipath - be more strict about testing the modify QP verb Bryan O'Sullivan
2006-08-25 18:24 ` [PATCH 17 of 23] IB/ipath - validate path_mig_state properly Bryan O'Sullivan
2006-08-25 18:24 ` Bryan O'Sullivan [this message]
2006-08-25 18:24 ` [PATCH 19 of 23] IB/ipath - handle sq_sig_all field correctly Bryan O'Sullivan
2006-08-25 18:24 ` [PATCH 20 of 23] IB/ipath - allow SMA to be disabled Bryan O'Sullivan
2006-08-25 18:24 ` [PATCH 21 of 23] IB/ipath - fix return value from ipath_poll Bryan O'Sullivan
2006-08-25 18:24 ` [PATCH 22 of 23] IB/ipath - print warning if LID not acquired within one minute Bryan O'Sullivan
2006-08-25 19:35   ` Roland Dreier
2006-08-25 20:40     ` [openib-general] " Robert Walsh
2006-08-26 19:31       ` Michael S. Tsirkin
2006-08-27 22:25         ` Roland Dreier
2006-08-28  1:41           ` Robert Walsh
2006-08-25 18:24 ` [PATCH 23 of 23] IB/ipath - control receive polarity inversion Bryan O'Sullivan
2006-08-25 20:01   ` Roland Dreier

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=02d9f3ef2291dea0f4d8.1156530283@eng-12.pathscale.com \
    --to=bos@pathscale.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=openib-general@openib.org \
    --cc=rdreier@cisco.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®