* [PATCH 08/10] RDMA/cxgb4: Use vmalloc for debugfs qp dump. Allows dumping thousands of qps.
From: Vipul Pandya @ 2012-05-18 9:59 UTC (permalink / raw)
To: linux-rdma, netdev; +Cc: roland, davem, divy, dm, kumaras, swise, Vipul Pandya
In-Reply-To: <1337335173-3226-1-git-send-email-vipul@chelsio.com>
Log active open failures of interest.
Signed-off-by: Vipul Pandya <vipul@chelsio.com>
Signed-off-by: Steve Wise <swise@opengridcomputing.com>
---
drivers/infiniband/hw/cxgb4/cm.c | 18 ++++++++++++++++++
drivers/infiniband/hw/cxgb4/device.c | 4 ++--
2 files changed, 20 insertions(+), 2 deletions(-)
diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c
index 6ce401a..55ab284 100644
--- a/drivers/infiniband/hw/cxgb4/cm.c
+++ b/drivers/infiniband/hw/cxgb4/cm.c
@@ -1413,6 +1413,24 @@ static int act_open_rpl(struct c4iw_dev *dev, struct sk_buff *skb)
return 0;
}
+ /*
+ * Log interesting failures.
+ */
+ switch (status) {
+ case CPL_ERR_CONN_RESET:
+ case CPL_ERR_CONN_TIMEDOUT:
+ break;
+ default:
+ printk(KERN_INFO MOD "Active open failure - "
+ "atid %u status %u errno %d %pI4:%u->%pI4:%u\n",
+ atid, status, status2errno(status),
+ &ep->com.local_addr.sin_addr.s_addr,
+ ntohs(ep->com.local_addr.sin_port),
+ &ep->com.remote_addr.sin_addr.s_addr,
+ ntohs(ep->com.remote_addr.sin_port));
+ break;
+ }
+
connect_reply_upcall(ep, status2errno(status));
state_set(&ep->com, DEAD);
diff --git a/drivers/infiniband/hw/cxgb4/device.c b/drivers/infiniband/hw/cxgb4/device.c
index bdb398f..8545629 100644
--- a/drivers/infiniband/hw/cxgb4/device.c
+++ b/drivers/infiniband/hw/cxgb4/device.c
@@ -121,7 +121,7 @@ static int qp_release(struct inode *inode, struct file *file)
printk(KERN_INFO "%s null qpd?\n", __func__);
return 0;
}
- kfree(qpd->buf);
+ vfree(qpd->buf);
kfree(qpd);
return 0;
}
@@ -145,7 +145,7 @@ static int qp_open(struct inode *inode, struct file *file)
spin_unlock_irq(&qpd->devp->lock);
qpd->bufsize = count * 128;
- qpd->buf = kmalloc(qpd->bufsize, GFP_KERNEL);
+ qpd->buf = vmalloc(qpd->bufsize);
if (!qpd->buf) {
ret = -ENOMEM;
goto err1;
--
1.7.1
^ permalink raw reply related
* Re: [V2 PATCH 2/9] macvtap: zerocopy: fix truesize underestimation
From: Jason Wang @ 2012-05-18 10:10 UTC (permalink / raw)
To: Shirley Ma; +Cc: eric.dumazet, mst, netdev, linux-kernel, ebiederm, davem
In-Reply-To: <1337268512.10741.53.camel@oc3660625478.ibm.com>
On 05/17/2012 11:28 PM, Shirley Ma wrote:
> On Thu, 2012-05-17 at 10:59 +0800, Jason Wang wrote:
>> Didn't see how this affact skb->len. And for truesize, I think they
>> are
>> different, when the offset were not zero, the data in this vector
>> were
>> divided into two parts. First part is copied into skb directly, and
>> the
>> second were pinned from a whole userspace page by
>> get_user_pages_fast(),
>> so we need count the whole page to the socket limit to prevent evil
>> application.
> What I meant that the code for skb->truesize has double added the first
> offset if any left from that vector (partically copied into skb
> directly, and then count pagesize which includes the offset (truesize +=
> PAGE_SIZE)).
Yes, I get you mean. There's no difference between first frag and
others: it's also possible for other frags that didn't occupy the whole
page. Since we pin the whole user page, better to count the whole page
size to prevent evil application.
> Thanks
> Shirley
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
^ permalink raw reply
* [PATCH 09/10] RDMA/cxgb4: remove kfifo usage
From: Vipul Pandya @ 2012-05-18 9:59 UTC (permalink / raw)
To: linux-rdma-u79uwXL29TY76Z2rM5mHXA, netdev-u79uwXL29TY76Z2rM5mHXA
Cc: roland-BHEL68pLQRGGvPXPguhicg, davem-fT/PcQaiUtIeIZ0/mPfg9Q,
divy-ut6Up61K2wZBDgjK7y7TUQ, dm-ut6Up61K2wZBDgjK7y7TUQ,
kumaras-ut6Up61K2wZBDgjK7y7TUQ,
swise-7bPotxP6k4+P2YhJcF5u+vpXobYPEAuW, Vipul Pandya
In-Reply-To: <1337335173-3226-1-git-send-email-vipul-ut6Up61K2wZBDgjK7y7TUQ@public.gmane.org>
Using kfifos for ID management was limiting the number of QPs and
preventing NP384 MPI jobs. So replace it with a simple bitmap
allocator.
Remove IDs from the IDR tables before deallocating them. This bug was
causing the BUG_ON() in insert_handle() to fire because the ID was
getting reused before being removed from the IDR table.
Signed-off-by: Vipul Pandya <vipul-ut6Up61K2wZBDgjK7y7TUQ@public.gmane.org>
Signed-off-by: Steve Wise <swise-7bPotxP6k4+P2YhJcF5u+vpXobYPEAuW@public.gmane.org>
---
drivers/infiniband/hw/cxgb4/Makefile | 2 +-
drivers/infiniband/hw/cxgb4/device.c | 37 +++++---
drivers/infiniband/hw/cxgb4/id_table.c | 112 ++++++++++++++++++++++++
drivers/infiniband/hw/cxgb4/iw_cxgb4.h | 35 ++++++--
drivers/infiniband/hw/cxgb4/mem.c | 10 +--
drivers/infiniband/hw/cxgb4/provider.c | 9 +--
drivers/infiniband/hw/cxgb4/resource.c | 148 ++++++++------------------------
7 files changed, 203 insertions(+), 150 deletions(-)
create mode 100644 drivers/infiniband/hw/cxgb4/id_table.c
diff --git a/drivers/infiniband/hw/cxgb4/Makefile b/drivers/infiniband/hw/cxgb4/Makefile
index 46b878c..e11cf72 100644
--- a/drivers/infiniband/hw/cxgb4/Makefile
+++ b/drivers/infiniband/hw/cxgb4/Makefile
@@ -2,4 +2,4 @@ ccflags-y := -Idrivers/net/ethernet/chelsio/cxgb4
obj-$(CONFIG_INFINIBAND_CXGB4) += iw_cxgb4.o
-iw_cxgb4-y := device.o cm.o provider.o mem.o cq.o qp.o resource.o ev.o
+iw_cxgb4-y := device.o cm.o provider.o mem.o cq.o qp.o resource.o ev.o id_table.o
diff --git a/drivers/infiniband/hw/cxgb4/device.c b/drivers/infiniband/hw/cxgb4/device.c
index 8545629..c8fd1d8 100644
--- a/drivers/infiniband/hw/cxgb4/device.c
+++ b/drivers/infiniband/hw/cxgb4/device.c
@@ -252,25 +252,26 @@ static int stats_show(struct seq_file *seq, void *v)
{
struct c4iw_dev *dev = seq->private;
- seq_printf(seq, " Object: %10s %10s %10s\n", "Total", "Current", "Max");
- seq_printf(seq, " PDID: %10llu %10llu %10llu\n",
+ seq_printf(seq, " Object: %10s %10s %10s %10s\n", "Total", "Current",
+ "Max", "Fail");
+ seq_printf(seq, " PDID: %10llu %10llu %10llu %10llu\n",
dev->rdev.stats.pd.total, dev->rdev.stats.pd.cur,
- dev->rdev.stats.pd.max);
- seq_printf(seq, " QID: %10llu %10llu %10llu\n",
+ dev->rdev.stats.pd.max, dev->rdev.stats.pd.fail);
+ seq_printf(seq, " QID: %10llu %10llu %10llu %10llu\n",
dev->rdev.stats.qid.total, dev->rdev.stats.qid.cur,
- dev->rdev.stats.qid.max);
- seq_printf(seq, " TPTMEM: %10llu %10llu %10llu\n",
+ dev->rdev.stats.qid.max, dev->rdev.stats.qid.fail);
+ seq_printf(seq, " TPTMEM: %10llu %10llu %10llu %10llu\n",
dev->rdev.stats.stag.total, dev->rdev.stats.stag.cur,
- dev->rdev.stats.stag.max);
- seq_printf(seq, " PBLMEM: %10llu %10llu %10llu\n",
+ dev->rdev.stats.stag.max, dev->rdev.stats.stag.fail);
+ seq_printf(seq, " PBLMEM: %10llu %10llu %10llu %10llu\n",
dev->rdev.stats.pbl.total, dev->rdev.stats.pbl.cur,
- dev->rdev.stats.pbl.max);
- seq_printf(seq, " RQTMEM: %10llu %10llu %10llu\n",
+ dev->rdev.stats.pbl.max, dev->rdev.stats.pbl.fail);
+ seq_printf(seq, " RQTMEM: %10llu %10llu %10llu %10llu\n",
dev->rdev.stats.rqt.total, dev->rdev.stats.rqt.cur,
- dev->rdev.stats.rqt.max);
- seq_printf(seq, " OCQPMEM: %10llu %10llu %10llu\n",
+ dev->rdev.stats.rqt.max, dev->rdev.stats.rqt.fail);
+ seq_printf(seq, " OCQPMEM: %10llu %10llu %10llu %10llu\n",
dev->rdev.stats.ocqp.total, dev->rdev.stats.ocqp.cur,
- dev->rdev.stats.ocqp.max);
+ dev->rdev.stats.ocqp.max, dev->rdev.stats.ocqp.fail);
seq_printf(seq, " DB FULL: %10llu\n", dev->rdev.stats.db_full);
seq_printf(seq, " DB EMPTY: %10llu\n", dev->rdev.stats.db_empty);
seq_printf(seq, " DB DROP: %10llu\n", dev->rdev.stats.db_drop);
@@ -292,11 +293,17 @@ static ssize_t stats_clear(struct file *file, const char __user *buf,
mutex_lock(&dev->rdev.stats.lock);
dev->rdev.stats.pd.max = 0;
+ dev->rdev.stats.pd.fail = 0;
dev->rdev.stats.qid.max = 0;
+ dev->rdev.stats.qid.fail = 0;
dev->rdev.stats.stag.max = 0;
+ dev->rdev.stats.stag.fail = 0;
dev->rdev.stats.pbl.max = 0;
+ dev->rdev.stats.pbl.fail = 0;
dev->rdev.stats.rqt.max = 0;
+ dev->rdev.stats.rqt.fail = 0;
dev->rdev.stats.ocqp.max = 0;
+ dev->rdev.stats.ocqp.fail = 0;
dev->rdev.stats.db_full = 0;
dev->rdev.stats.db_empty = 0;
dev->rdev.stats.db_drop = 0;
@@ -350,8 +357,8 @@ void c4iw_release_dev_ucontext(struct c4iw_rdev *rdev,
entry = list_entry(pos, struct c4iw_qid_list, entry);
list_del_init(&entry->entry);
if (!(entry->qid & rdev->qpmask)) {
- c4iw_put_resource(&rdev->resource.qid_fifo, entry->qid,
- &rdev->resource.qid_fifo_lock);
+ c4iw_put_resource(&rdev->resource.qid_table,
+ entry->qid);
mutex_lock(&rdev->stats.lock);
rdev->stats.qid.cur -= rdev->qpmask + 1;
mutex_unlock(&rdev->stats.lock);
diff --git a/drivers/infiniband/hw/cxgb4/id_table.c b/drivers/infiniband/hw/cxgb4/id_table.c
new file mode 100644
index 0000000..f95e5df
--- /dev/null
+++ b/drivers/infiniband/hw/cxgb4/id_table.c
@@ -0,0 +1,112 @@
+/*
+ * Copyright (c) 2011 Chelsio Communications. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses. 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 source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ * Redistribution and use in source and binary forms, with or
+ * without modification, are permitted provided that the following
+ * conditions are met:
+ *
+ * - Redistributions of source code must retain the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer.
+ *
+ * - Redistributions in binary form must reproduce the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer in the documentation and/or other materials
+ * provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. 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, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#include <linux/kernel.h>
+#include <linux/random.h>
+#include "iw_cxgb4.h"
+
+#define RANDOM_SKIP 16
+
+/*
+ * Trivial bitmap-based allocator. If the random flag is set, the
+ * allocator is designed to:
+ * - pseudo-randomize the id returned such that it is not trivially predictable.
+ * - avoid reuse of recently used id (at the expense of predictability)
+ */
+u32 c4iw_id_alloc(struct c4iw_id_table *alloc)
+{
+ unsigned long flags;
+ u32 obj;
+
+ spin_lock_irqsave(&alloc->lock, flags);
+
+ obj = find_next_zero_bit(alloc->table, alloc->max, alloc->last);
+ if (obj >= alloc->max)
+ obj = find_first_zero_bit(alloc->table, alloc->max);
+
+ if (obj < alloc->max) {
+ if (alloc->flags & C4IW_ID_TABLE_F_RANDOM)
+ alloc->last += random32() % RANDOM_SKIP;
+ else
+ alloc->last = obj + 1;
+ if (alloc->last >= alloc->max)
+ alloc->last = 0;
+ set_bit(obj, alloc->table);
+ obj += alloc->start;
+ } else
+ obj = -1;
+
+ spin_unlock_irqrestore(&alloc->lock, flags);
+ return obj;
+}
+
+void c4iw_id_free(struct c4iw_id_table *alloc, u32 obj)
+{
+ unsigned long flags;
+
+ obj -= alloc->start;
+ BUG_ON((int)obj < 0);
+
+ spin_lock_irqsave(&alloc->lock, flags);
+ clear_bit(obj, alloc->table);
+ spin_unlock_irqrestore(&alloc->lock, flags);
+}
+
+int c4iw_id_table_alloc(struct c4iw_id_table *alloc, u32 start, u32 num,
+ u32 reserved, u32 flags)
+{
+ int i;
+
+ alloc->start = start;
+ alloc->flags = flags;
+ if (flags & C4IW_ID_TABLE_F_RANDOM)
+ alloc->last = random32() % RANDOM_SKIP;
+ else
+ alloc->last = 0;
+ alloc->max = num;
+ spin_lock_init(&alloc->lock);
+ alloc->table = kmalloc(BITS_TO_LONGS(num) * sizeof(long),
+ GFP_KERNEL);
+ if (!alloc->table)
+ return -ENOMEM;
+
+ bitmap_zero(alloc->table, num);
+ if (!(alloc->flags & C4IW_ID_TABLE_F_EMPTY))
+ for (i = 0; i < reserved; ++i)
+ set_bit(i, alloc->table);
+
+ return 0;
+}
+
+void c4iw_id_table_free(struct c4iw_id_table *alloc)
+{
+ kfree(alloc->table);
+}
diff --git a/drivers/infiniband/hw/cxgb4/iw_cxgb4.h b/drivers/infiniband/hw/cxgb4/iw_cxgb4.h
index 6818659..2d5b06b 100644
--- a/drivers/infiniband/hw/cxgb4/iw_cxgb4.h
+++ b/drivers/infiniband/hw/cxgb4/iw_cxgb4.h
@@ -45,7 +45,6 @@
#include <linux/kref.h>
#include <linux/timer.h>
#include <linux/io.h>
-#include <linux/kfifo.h>
#include <asm/byteorder.h>
@@ -79,13 +78,22 @@ static inline void *cplhdr(struct sk_buff *skb)
return skb->data;
}
+#define C4IW_ID_TABLE_F_RANDOM 1 /* Pseudo-randomize the id's returned */
+#define C4IW_ID_TABLE_F_EMPTY 2 /* Table is initially empty */
+
+struct c4iw_id_table {
+ u32 flags;
+ u32 start; /* logical minimal id */
+ u32 last; /* hint for find */
+ u32 max;
+ spinlock_t lock;
+ unsigned long *table;
+};
+
struct c4iw_resource {
- struct kfifo tpt_fifo;
- spinlock_t tpt_fifo_lock;
- struct kfifo qid_fifo;
- spinlock_t qid_fifo_lock;
- struct kfifo pdid_fifo;
- spinlock_t pdid_fifo_lock;
+ struct c4iw_id_table tpt_table;
+ struct c4iw_id_table qid_table;
+ struct c4iw_id_table pdid_table;
};
struct c4iw_qid_list {
@@ -107,6 +115,7 @@ struct c4iw_stat {
u64 total;
u64 cur;
u64 max;
+ u64 fail;
};
struct c4iw_stats {
@@ -253,7 +262,7 @@ static inline int _insert_handle(struct c4iw_dev *rhp, struct idr *idr,
if (lock)
spin_lock_irq(&rhp->lock);
ret = idr_get_new_above(idr, handle, id, &newid);
- BUG_ON(newid != id);
+ BUG_ON(!ret && newid != id);
if (lock)
spin_unlock_irq(&rhp->lock);
} while (ret == -EAGAIN);
@@ -755,14 +764,20 @@ static inline int compute_wscale(int win)
return wscale;
}
+u32 c4iw_id_alloc(struct c4iw_id_table *alloc);
+void c4iw_id_free(struct c4iw_id_table *alloc, u32 obj);
+int c4iw_id_table_alloc(struct c4iw_id_table *alloc, u32 start, u32 num,
+ u32 reserved, u32 flags);
+void c4iw_id_table_free(struct c4iw_id_table *alloc);
+
typedef int (*c4iw_handler_func)(struct c4iw_dev *dev, struct sk_buff *skb);
int c4iw_ep_redirect(void *ctx, struct dst_entry *old, struct dst_entry *new,
struct l2t_entry *l2t);
void c4iw_put_qpid(struct c4iw_rdev *rdev, u32 qpid,
struct c4iw_dev_ucontext *uctx);
-u32 c4iw_get_resource(struct kfifo *fifo, spinlock_t *lock);
-void c4iw_put_resource(struct kfifo *fifo, u32 entry, spinlock_t *lock);
+u32 c4iw_get_resource(struct c4iw_id_table *id_table);
+void c4iw_put_resource(struct c4iw_id_table *id_table, u32 entry);
int c4iw_init_resource(struct c4iw_rdev *rdev, u32 nr_tpt, u32 nr_pdid);
int c4iw_init_ctrl_qp(struct c4iw_rdev *rdev);
int c4iw_pblpool_create(struct c4iw_rdev *rdev);
diff --git a/drivers/infiniband/hw/cxgb4/mem.c b/drivers/infiniband/hw/cxgb4/mem.c
index 2a87379..57e07c6 100644
--- a/drivers/infiniband/hw/cxgb4/mem.c
+++ b/drivers/infiniband/hw/cxgb4/mem.c
@@ -131,8 +131,7 @@ static int write_tpt_entry(struct c4iw_rdev *rdev, u32 reset_tpt_entry,
stag_idx = (*stag) >> 8;
if ((!reset_tpt_entry) && (*stag == T4_STAG_UNSET)) {
- stag_idx = c4iw_get_resource(&rdev->resource.tpt_fifo,
- &rdev->resource.tpt_fifo_lock);
+ stag_idx = c4iw_get_resource(&rdev->resource.tpt_table);
if (!stag_idx)
return -ENOMEM;
mutex_lock(&rdev->stats.lock);
@@ -171,8 +170,7 @@ static int write_tpt_entry(struct c4iw_rdev *rdev, u32 reset_tpt_entry,
sizeof(tpt), &tpt);
if (reset_tpt_entry) {
- c4iw_put_resource(&rdev->resource.tpt_fifo, stag_idx,
- &rdev->resource.tpt_fifo_lock);
+ c4iw_put_resource(&rdev->resource.tpt_table, stag_idx);
mutex_lock(&rdev->stats.lock);
rdev->stats.stag.cur -= 32;
mutex_unlock(&rdev->stats.lock);
@@ -695,8 +693,8 @@ int c4iw_dealloc_mw(struct ib_mw *mw)
mhp = to_c4iw_mw(mw);
rhp = mhp->rhp;
mmid = (mw->rkey) >> 8;
- deallocate_window(&rhp->rdev, mhp->attr.stag);
remove_handle(rhp, &rhp->mmidr, mmid);
+ deallocate_window(&rhp->rdev, mhp->attr.stag);
kfree(mhp);
PDBG("%s ib_mw %p mmid 0x%x ptr %p\n", __func__, mw, mmid, mhp);
return 0;
@@ -798,12 +796,12 @@ int c4iw_dereg_mr(struct ib_mr *ib_mr)
mhp = to_c4iw_mr(ib_mr);
rhp = mhp->rhp;
mmid = mhp->attr.stag >> 8;
+ remove_handle(rhp, &rhp->mmidr, mmid);
dereg_mem(&rhp->rdev, mhp->attr.stag, mhp->attr.pbl_size,
mhp->attr.pbl_addr);
if (mhp->attr.pbl_size)
c4iw_pblpool_free(&mhp->rhp->rdev, mhp->attr.pbl_addr,
mhp->attr.pbl_size << 3);
- remove_handle(rhp, &rhp->mmidr, mmid);
if (mhp->kva)
kfree((void *) (unsigned long) mhp->kva);
if (mhp->umem)
diff --git a/drivers/infiniband/hw/cxgb4/provider.c b/drivers/infiniband/hw/cxgb4/provider.c
index 8d58736..fe98a0a 100644
--- a/drivers/infiniband/hw/cxgb4/provider.c
+++ b/drivers/infiniband/hw/cxgb4/provider.c
@@ -188,8 +188,7 @@ static int c4iw_deallocate_pd(struct ib_pd *pd)
php = to_c4iw_pd(pd);
rhp = php->rhp;
PDBG("%s ibpd %p pdid 0x%x\n", __func__, pd, php->pdid);
- c4iw_put_resource(&rhp->rdev.resource.pdid_fifo, php->pdid,
- &rhp->rdev.resource.pdid_fifo_lock);
+ c4iw_put_resource(&rhp->rdev.resource.pdid_table, php->pdid);
mutex_lock(&rhp->rdev.stats.lock);
rhp->rdev.stats.pd.cur--;
mutex_unlock(&rhp->rdev.stats.lock);
@@ -207,14 +206,12 @@ static struct ib_pd *c4iw_allocate_pd(struct ib_device *ibdev,
PDBG("%s ibdev %p\n", __func__, ibdev);
rhp = (struct c4iw_dev *) ibdev;
- pdid = c4iw_get_resource(&rhp->rdev.resource.pdid_fifo,
- &rhp->rdev.resource.pdid_fifo_lock);
+ pdid = c4iw_get_resource(&rhp->rdev.resource.pdid_table);
if (!pdid)
return ERR_PTR(-EINVAL);
php = kzalloc(sizeof(*php), GFP_KERNEL);
if (!php) {
- c4iw_put_resource(&rhp->rdev.resource.pdid_fifo, pdid,
- &rhp->rdev.resource.pdid_fifo_lock);
+ c4iw_put_resource(&rhp->rdev.resource.pdid_table, pdid);
return ERR_PTR(-ENOMEM);
}
php->pdid = pdid;
diff --git a/drivers/infiniband/hw/cxgb4/resource.c b/drivers/infiniband/hw/cxgb4/resource.c
index 1b948d1..cdef4d7 100644
--- a/drivers/infiniband/hw/cxgb4/resource.c
+++ b/drivers/infiniband/hw/cxgb4/resource.c
@@ -30,96 +30,25 @@
* SOFTWARE.
*/
/* Crude resource management */
-#include <linux/kernel.h>
-#include <linux/random.h>
-#include <linux/slab.h>
-#include <linux/kfifo.h>
#include <linux/spinlock.h>
-#include <linux/errno.h>
#include <linux/genalloc.h>
#include <linux/ratelimit.h>
#include "iw_cxgb4.h"
-#define RANDOM_SIZE 16
-
-static int __c4iw_init_resource_fifo(struct kfifo *fifo,
- spinlock_t *fifo_lock,
- u32 nr, u32 skip_low,
- u32 skip_high,
- int random)
-{
- u32 i, j, entry = 0, idx;
- u32 random_bytes;
- u32 rarray[16];
- spin_lock_init(fifo_lock);
-
- if (kfifo_alloc(fifo, nr * sizeof(u32), GFP_KERNEL))
- return -ENOMEM;
-
- for (i = 0; i < skip_low + skip_high; i++)
- kfifo_in(fifo, (unsigned char *) &entry, sizeof(u32));
- if (random) {
- j = 0;
- random_bytes = random32();
- for (i = 0; i < RANDOM_SIZE; i++)
- rarray[i] = i + skip_low;
- for (i = skip_low + RANDOM_SIZE; i < nr - skip_high; i++) {
- if (j >= RANDOM_SIZE) {
- j = 0;
- random_bytes = random32();
- }
- idx = (random_bytes >> (j * 2)) & 0xF;
- kfifo_in(fifo,
- (unsigned char *) &rarray[idx],
- sizeof(u32));
- rarray[idx] = i;
- j++;
- }
- for (i = 0; i < RANDOM_SIZE; i++)
- kfifo_in(fifo,
- (unsigned char *) &rarray[i],
- sizeof(u32));
- } else
- for (i = skip_low; i < nr - skip_high; i++)
- kfifo_in(fifo, (unsigned char *) &i, sizeof(u32));
-
- for (i = 0; i < skip_low + skip_high; i++)
- if (kfifo_out_locked(fifo, (unsigned char *) &entry,
- sizeof(u32), fifo_lock))
- break;
- return 0;
-}
-
-static int c4iw_init_resource_fifo(struct kfifo *fifo, spinlock_t * fifo_lock,
- u32 nr, u32 skip_low, u32 skip_high)
-{
- return __c4iw_init_resource_fifo(fifo, fifo_lock, nr, skip_low,
- skip_high, 0);
-}
-
-static int c4iw_init_resource_fifo_random(struct kfifo *fifo,
- spinlock_t *fifo_lock,
- u32 nr, u32 skip_low, u32 skip_high)
-{
- return __c4iw_init_resource_fifo(fifo, fifo_lock, nr, skip_low,
- skip_high, 1);
-}
-
-static int c4iw_init_qid_fifo(struct c4iw_rdev *rdev)
+static int c4iw_init_qid_table(struct c4iw_rdev *rdev)
{
u32 i;
- spin_lock_init(&rdev->resource.qid_fifo_lock);
-
- if (kfifo_alloc(&rdev->resource.qid_fifo, rdev->lldi.vr->qp.size *
- sizeof(u32), GFP_KERNEL))
+ if (c4iw_id_table_alloc(&rdev->resource.qid_table,
+ rdev->lldi.vr->qp.start,
+ rdev->lldi.vr->qp.size,
+ rdev->lldi.vr->qp.size, 0))
return -ENOMEM;
for (i = rdev->lldi.vr->qp.start;
- i < rdev->lldi.vr->qp.start + rdev->lldi.vr->qp.size; i++)
+ i < rdev->lldi.vr->qp.start + rdev->lldi.vr->qp.size; i++)
if (!(i & rdev->qpmask))
- kfifo_in(&rdev->resource.qid_fifo,
- (unsigned char *) &i, sizeof(u32));
+ c4iw_id_free(&rdev->resource.qid_table, i);
return 0;
}
@@ -127,44 +56,42 @@ static int c4iw_init_qid_fifo(struct c4iw_rdev *rdev)
int c4iw_init_resource(struct c4iw_rdev *rdev, u32 nr_tpt, u32 nr_pdid)
{
int err = 0;
- err = c4iw_init_resource_fifo_random(&rdev->resource.tpt_fifo,
- &rdev->resource.tpt_fifo_lock,
- nr_tpt, 1, 0);
+ err = c4iw_id_table_alloc(&rdev->resource.tpt_table, 0, nr_tpt, 1,
+ C4IW_ID_TABLE_F_RANDOM);
if (err)
goto tpt_err;
- err = c4iw_init_qid_fifo(rdev);
+ err = c4iw_init_qid_table(rdev);
if (err)
goto qid_err;
- err = c4iw_init_resource_fifo(&rdev->resource.pdid_fifo,
- &rdev->resource.pdid_fifo_lock,
- nr_pdid, 1, 0);
+ err = c4iw_id_table_alloc(&rdev->resource.pdid_table, 0,
+ nr_pdid, 1, 0);
if (err)
goto pdid_err;
return 0;
-pdid_err:
- kfifo_free(&rdev->resource.qid_fifo);
-qid_err:
- kfifo_free(&rdev->resource.tpt_fifo);
-tpt_err:
+ pdid_err:
+ c4iw_id_table_free(&rdev->resource.qid_table);
+ qid_err:
+ c4iw_id_table_free(&rdev->resource.tpt_table);
+ tpt_err:
return -ENOMEM;
}
/*
* returns 0 if no resource available
*/
-u32 c4iw_get_resource(struct kfifo *fifo, spinlock_t *lock)
+u32 c4iw_get_resource(struct c4iw_id_table *id_table)
{
u32 entry;
- if (kfifo_out_locked(fifo, (unsigned char *) &entry, sizeof(u32), lock))
- return entry;
- else
+ entry = c4iw_id_alloc(id_table);
+ if (entry == (u32)(-1))
return 0;
+ return entry;
}
-void c4iw_put_resource(struct kfifo *fifo, u32 entry, spinlock_t *lock)
+void c4iw_put_resource(struct c4iw_id_table *id_table, u32 entry)
{
PDBG("%s entry 0x%x\n", __func__, entry);
- kfifo_in_locked(fifo, (unsigned char *) &entry, sizeof(u32), lock);
+ c4iw_id_free(id_table, entry);
}
u32 c4iw_get_cqid(struct c4iw_rdev *rdev, struct c4iw_dev_ucontext *uctx)
@@ -181,8 +108,7 @@ u32 c4iw_get_cqid(struct c4iw_rdev *rdev, struct c4iw_dev_ucontext *uctx)
qid = entry->qid;
kfree(entry);
} else {
- qid = c4iw_get_resource(&rdev->resource.qid_fifo,
- &rdev->resource.qid_fifo_lock);
+ qid = c4iw_get_resource(&rdev->resource.qid_table);
if (!qid)
goto out;
mutex_lock(&rdev->stats.lock);
@@ -252,8 +178,7 @@ u32 c4iw_get_qpid(struct c4iw_rdev *rdev, struct c4iw_dev_ucontext *uctx)
qid = entry->qid;
kfree(entry);
} else {
- qid = c4iw_get_resource(&rdev->resource.qid_fifo,
- &rdev->resource.qid_fifo_lock);
+ qid = c4iw_get_resource(&rdev->resource.qid_table);
if (!qid)
goto out;
mutex_lock(&rdev->stats.lock);
@@ -311,9 +236,9 @@ void c4iw_put_qpid(struct c4iw_rdev *rdev, u32 qid,
void c4iw_destroy_resource(struct c4iw_resource *rscp)
{
- kfifo_free(&rscp->tpt_fifo);
- kfifo_free(&rscp->qid_fifo);
- kfifo_free(&rscp->pdid_fifo);
+ c4iw_id_table_free(&rscp->tpt_table);
+ c4iw_id_table_free(&rscp->qid_table);
+ c4iw_id_table_free(&rscp->pdid_table);
}
/*
@@ -326,16 +251,14 @@ u32 c4iw_pblpool_alloc(struct c4iw_rdev *rdev, int size)
{
unsigned long addr = gen_pool_alloc(rdev->pbl_pool, size);
PDBG("%s addr 0x%x size %d\n", __func__, (u32)addr, size);
- if (!addr)
- printk_ratelimited(KERN_WARNING MOD "%s: Out of PBL memory\n",
- pci_name(rdev->lldi.pdev));
+ mutex_lock(&rdev->stats.lock);
if (addr) {
- mutex_lock(&rdev->stats.lock);
rdev->stats.pbl.cur += roundup(size, 1 << MIN_PBL_SHIFT);
if (rdev->stats.pbl.cur > rdev->stats.pbl.max)
rdev->stats.pbl.max = rdev->stats.pbl.cur;
- mutex_unlock(&rdev->stats.lock);
- }
+ } else
+ rdev->stats.pbl.fail++;
+ mutex_unlock(&rdev->stats.lock);
return (u32)addr;
}
@@ -401,13 +324,14 @@ u32 c4iw_rqtpool_alloc(struct c4iw_rdev *rdev, int size)
if (!addr)
printk_ratelimited(KERN_WARNING MOD "%s: Out of RQT memory\n",
pci_name(rdev->lldi.pdev));
+ mutex_lock(&rdev->stats.lock);
if (addr) {
- mutex_lock(&rdev->stats.lock);
rdev->stats.rqt.cur += roundup(size << 6, 1 << MIN_RQT_SHIFT);
if (rdev->stats.rqt.cur > rdev->stats.rqt.max)
rdev->stats.rqt.max = rdev->stats.rqt.cur;
- mutex_unlock(&rdev->stats.lock);
- }
+ } else
+ rdev->stats.rqt.fail++;
+ mutex_unlock(&rdev->stats.lock);
return (u32)addr;
}
--
1.7.1
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply related
* [PATCH 06/10] RDMA/cxgb4: disable interrupts in c4iw_ev_dispatch().
From: Vipul Pandya @ 2012-05-18 9:59 UTC (permalink / raw)
To: linux-rdma-u79uwXL29TY76Z2rM5mHXA, netdev-u79uwXL29TY76Z2rM5mHXA
Cc: roland-BHEL68pLQRGGvPXPguhicg, davem-fT/PcQaiUtIeIZ0/mPfg9Q,
divy-ut6Up61K2wZBDgjK7y7TUQ, dm-ut6Up61K2wZBDgjK7y7TUQ,
kumaras-ut6Up61K2wZBDgjK7y7TUQ,
swise-7bPotxP6k4+P2YhJcF5u+vpXobYPEAuW, Vipul Pandya
In-Reply-To: <1337335173-3226-1-git-send-email-vipul-ut6Up61K2wZBDgjK7y7TUQ@public.gmane.org>
Use GFP_ATOMIC in _insert_handle() if ints are disabled.
Don't panic if we get an abort with no endpoint found. Just log a
warning.
Signed-off-by: Vipul Pandya <vipul-ut6Up61K2wZBDgjK7y7TUQ@public.gmane.org>
Signed-off-by: Steve Wise <swise-7bPotxP6k4+P2YhJcF5u+vpXobYPEAuW@public.gmane.org>
---
drivers/infiniband/hw/cxgb4/cm.c | 5 ++++-
drivers/infiniband/hw/cxgb4/ev.c | 8 ++++----
drivers/infiniband/hw/cxgb4/iw_cxgb4.h | 2 +-
3 files changed, 9 insertions(+), 6 deletions(-)
diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c
index 4c7c62f..6ce401a 100644
--- a/drivers/infiniband/hw/cxgb4/cm.c
+++ b/drivers/infiniband/hw/cxgb4/cm.c
@@ -1362,7 +1362,10 @@ static int abort_rpl(struct c4iw_dev *dev, struct sk_buff *skb)
ep = lookup_tid(t, tid);
PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid);
- BUG_ON(!ep);
+ if (!ep) {
+ printk(KERN_WARNING MOD "Abort rpl to freed endpoint\n");
+ return 0;
+ }
mutex_lock(&ep->com.mutex);
switch (ep->com.state) {
case ABORTING:
diff --git a/drivers/infiniband/hw/cxgb4/ev.c b/drivers/infiniband/hw/cxgb4/ev.c
index 397cb36..cf2f6b4 100644
--- a/drivers/infiniband/hw/cxgb4/ev.c
+++ b/drivers/infiniband/hw/cxgb4/ev.c
@@ -84,7 +84,7 @@ void c4iw_ev_dispatch(struct c4iw_dev *dev, struct t4_cqe *err_cqe)
struct c4iw_qp *qhp;
u32 cqid;
- spin_lock(&dev->lock);
+ spin_lock_irq(&dev->lock);
qhp = get_qhp(dev, CQE_QPID(err_cqe));
if (!qhp) {
printk(KERN_ERR MOD "BAD AE qpid 0x%x opcode %d "
@@ -93,7 +93,7 @@ void c4iw_ev_dispatch(struct c4iw_dev *dev, struct t4_cqe *err_cqe)
CQE_OPCODE(err_cqe), CQE_STATUS(err_cqe),
CQE_TYPE(err_cqe), CQE_WRID_HI(err_cqe),
CQE_WRID_LOW(err_cqe));
- spin_unlock(&dev->lock);
+ spin_unlock_irq(&dev->lock);
goto out;
}
@@ -109,13 +109,13 @@ void c4iw_ev_dispatch(struct c4iw_dev *dev, struct t4_cqe *err_cqe)
CQE_OPCODE(err_cqe), CQE_STATUS(err_cqe),
CQE_TYPE(err_cqe), CQE_WRID_HI(err_cqe),
CQE_WRID_LOW(err_cqe));
- spin_unlock(&dev->lock);
+ spin_unlock_irq(&dev->lock);
goto out;
}
c4iw_qp_add_ref(&qhp->ibqp);
atomic_inc(&chp->refcnt);
- spin_unlock(&dev->lock);
+ spin_unlock_irq(&dev->lock);
/* Bad incoming write */
if (RQ_TYPE(err_cqe) &&
diff --git a/drivers/infiniband/hw/cxgb4/iw_cxgb4.h b/drivers/infiniband/hw/cxgb4/iw_cxgb4.h
index a11ed5c..e8b88a0 100644
--- a/drivers/infiniband/hw/cxgb4/iw_cxgb4.h
+++ b/drivers/infiniband/hw/cxgb4/iw_cxgb4.h
@@ -246,7 +246,7 @@ static inline int _insert_handle(struct c4iw_dev *rhp, struct idr *idr,
int newid;
do {
- if (!idr_pre_get(idr, GFP_KERNEL))
+ if (!idr_pre_get(idr, lock ? GFP_KERNEL : GFP_ATOMIC))
return -ENOMEM;
if (lock)
spin_lock_irq(&rhp->lock);
--
1.7.1
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply related
* [PATCH 03/10] cxgb4: DB Drop Recovery for RDMA and LLD queues.
From: Vipul Pandya @ 2012-05-18 9:59 UTC (permalink / raw)
To: linux-rdma-u79uwXL29TY76Z2rM5mHXA, netdev-u79uwXL29TY76Z2rM5mHXA
Cc: roland-BHEL68pLQRGGvPXPguhicg, davem-fT/PcQaiUtIeIZ0/mPfg9Q,
divy-ut6Up61K2wZBDgjK7y7TUQ, dm-ut6Up61K2wZBDgjK7y7TUQ,
kumaras-ut6Up61K2wZBDgjK7y7TUQ,
swise-7bPotxP6k4+P2YhJcF5u+vpXobYPEAuW, Vipul Pandya
In-Reply-To: <1337335173-3226-1-git-send-email-vipul-ut6Up61K2wZBDgjK7y7TUQ@public.gmane.org>
recover LLD EQs for DB drop interrupts. This includes adding a new
db_lock, a spin lock disabling BH too, used by the recovery thread and
the ring_tx_db() paths to allow db drop recovery.
cleaned up initial db avoidance code.
add read_eq_indices() - allows the LLD to use the pcie mw to efficiently
read hw eq contexts.
add cxgb4_sync_txq_pidx() - called by iw_cxgb4 to sync up the sw/hw pidx
value.
add flush_eq_cache() and cxgb4_flush_eq_cache(). This allows iw_cxgb4
to flush the sge eq context cache before beginning db drop recovery.
add module parameter, dbfoifo_int_thresh, to allow tuning the db
interrupt threshold value.
add dbfifo_int_thresh to cxgb4_lld_info so iw_cxgb4 knows the threshold.
add module parameter, dbfoifo_drain_delay, to allow tuning the amount
of time delay between DB FULL and EMPTY upcalls to iw_cxgb4.
Signed-off-by: Vipul Pandya <vipul-ut6Up61K2wZBDgjK7y7TUQ@public.gmane.org>
Signed-off-by: Steve Wise <swise-7bPotxP6k4+P2YhJcF5u+vpXobYPEAuW@public.gmane.org>
---
drivers/net/ethernet/chelsio/cxgb4/cxgb4.h | 16 ++
drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c | 214 +++++++++++++++++++----
drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h | 4 +
drivers/net/ethernet/chelsio/cxgb4/sge.c | 20 ++-
drivers/net/ethernet/chelsio/cxgb4/t4_regs.h | 53 ++++++
drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h | 15 ++
6 files changed, 280 insertions(+), 42 deletions(-)
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h
index 5f3c0a7..ec2dafe 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h
@@ -51,6 +51,8 @@
#define FW_VERSION_MINOR 1
#define FW_VERSION_MICRO 0
+#define CH_WARN(adap, fmt, ...) dev_warn(adap->pdev_dev, fmt, ## __VA_ARGS__)
+
enum {
MAX_NPORTS = 4, /* max # of ports */
SERNUM_LEN = 24, /* Serial # length */
@@ -64,6 +66,15 @@ enum {
MEM_MC
};
+enum {
+ MEMWIN0_APERTURE = 65536,
+ MEMWIN0_BASE = 0x30000,
+ MEMWIN1_APERTURE = 32768,
+ MEMWIN1_BASE = 0x28000,
+ MEMWIN2_APERTURE = 2048,
+ MEMWIN2_BASE = 0x1b800,
+};
+
enum dev_master {
MASTER_CANT,
MASTER_MAY,
@@ -403,6 +414,9 @@ struct sge_txq {
struct tx_sw_desc *sdesc; /* address of SW Tx descriptor ring */
struct sge_qstat *stat; /* queue status entry */
dma_addr_t phys_addr; /* physical address of the ring */
+ spinlock_t db_lock;
+ int db_disabled;
+ unsigned short db_pidx;
};
struct sge_eth_txq { /* state for an SGE Ethernet Tx queue */
@@ -475,6 +489,7 @@ struct adapter {
void __iomem *regs;
struct pci_dev *pdev;
struct device *pdev_dev;
+ unsigned int mbox;
unsigned int fn;
unsigned int flags;
@@ -607,6 +622,7 @@ irqreturn_t t4_sge_intr_msix(int irq, void *cookie);
void t4_sge_init(struct adapter *adap);
void t4_sge_start(struct adapter *adap);
void t4_sge_stop(struct adapter *adap);
+extern int dbfifo_int_thresh;
#define for_each_port(adapter, iter) \
for (iter = 0; iter < (adapter)->params.nports; ++iter)
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
index c243f93..e1f96fb 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
@@ -149,15 +149,6 @@ static unsigned int pfvfres_pmask(struct adapter *adapter,
#endif
enum {
- MEMWIN0_APERTURE = 65536,
- MEMWIN0_BASE = 0x30000,
- MEMWIN1_APERTURE = 32768,
- MEMWIN1_BASE = 0x28000,
- MEMWIN2_APERTURE = 2048,
- MEMWIN2_BASE = 0x1b800,
-};
-
-enum {
MAX_TXQ_ENTRIES = 16384,
MAX_CTRL_TXQ_ENTRIES = 1024,
MAX_RSPQ_ENTRIES = 16384,
@@ -371,6 +362,15 @@ static int set_addr_filters(const struct net_device *dev, bool sleep)
uhash | mhash, sleep);
}
+int dbfifo_int_thresh = 10; /* 10 == 640 entry threshold */
+module_param(dbfifo_int_thresh, int, 0644);
+MODULE_PARM_DESC(dbfifo_int_thresh, "doorbell fifo interrupt threshold");
+
+int dbfifo_drain_delay = 1000; /* usecs to sleep while draining the dbfifo */
+module_param(dbfifo_drain_delay, int, 0644);
+MODULE_PARM_DESC(dbfifo_drain_delay,
+ "usecs to sleep while draining the dbfifo");
+
/*
* Set Rx properties of a port, such as promiscruity, address filters, and MTU.
* If @mtu is -1 it is left unchanged.
@@ -389,6 +389,8 @@ static int set_rxmode(struct net_device *dev, int mtu, bool sleep_ok)
return ret;
}
+static struct workqueue_struct *workq;
+
/**
* link_start - enable a port
* @dev: the port to enable
@@ -2196,7 +2198,7 @@ static void cxgb4_queue_tid_release(struct tid_info *t, unsigned int chan,
adap->tid_release_head = (void **)((uintptr_t)p | chan);
if (!adap->tid_release_task_busy) {
adap->tid_release_task_busy = true;
- schedule_work(&adap->tid_release_task);
+ queue_work(workq, &adap->tid_release_task);
}
spin_unlock_bh(&adap->tid_release_lock);
}
@@ -2423,6 +2425,59 @@ void cxgb4_iscsi_init(struct net_device *dev, unsigned int tag_mask,
}
EXPORT_SYMBOL(cxgb4_iscsi_init);
+int cxgb4_flush_eq_cache(struct net_device *dev)
+{
+ struct adapter *adap = netdev2adap(dev);
+ int ret;
+
+ ret = t4_fwaddrspace_write(adap, adap->mbox,
+ 0xe1000000 + A_SGE_CTXT_CMD, 0x20000000);
+ return ret;
+}
+EXPORT_SYMBOL(cxgb4_flush_eq_cache);
+
+static int read_eq_indices(struct adapter *adap, u16 qid, u16 *pidx, u16 *cidx)
+{
+ u32 addr = t4_read_reg(adap, A_SGE_DBQ_CTXT_BADDR) + 24 * qid + 8;
+ __be64 indices;
+ int ret;
+
+ ret = t4_mem_win_read_len(adap, addr, (__be32 *)&indices, 8);
+ if (!ret) {
+ indices = be64_to_cpu(indices);
+ *cidx = (indices >> 25) & 0xffff;
+ *pidx = (indices >> 9) & 0xffff;
+ }
+ return ret;
+}
+
+int cxgb4_sync_txq_pidx(struct net_device *dev, u16 qid, u16 pidx,
+ u16 size)
+{
+ struct adapter *adap = netdev2adap(dev);
+ u16 hw_pidx, hw_cidx;
+ int ret;
+
+ ret = read_eq_indices(adap, qid, &hw_pidx, &hw_cidx);
+ if (ret)
+ goto out;
+
+ if (pidx != hw_pidx) {
+ u16 delta;
+
+ if (pidx >= hw_pidx)
+ delta = pidx - hw_pidx;
+ else
+ delta = size - hw_pidx + pidx;
+ wmb();
+ t4_write_reg(adap, MYPF_REG(A_SGE_PF_KDOORBELL),
+ V_QID(qid) | V_PIDX(delta));
+ }
+out:
+ return ret;
+}
+EXPORT_SYMBOL(cxgb4_sync_txq_pidx);
+
static struct pci_driver cxgb4_driver;
static void check_neigh_update(struct neighbour *neigh)
@@ -2456,6 +2511,95 @@ static struct notifier_block cxgb4_netevent_nb = {
.notifier_call = netevent_cb
};
+static void drain_db_fifo(struct adapter *adap, int usecs)
+{
+ u32 v;
+
+ do {
+ set_current_state(TASK_UNINTERRUPTIBLE);
+ schedule_timeout(usecs_to_jiffies(usecs));
+ v = t4_read_reg(adap, A_SGE_DBFIFO_STATUS);
+ if (G_LP_COUNT(v) == 0 && G_HP_COUNT(v) == 0)
+ break;
+ } while (1);
+}
+
+static void disable_txq_db(struct sge_txq *q)
+{
+ spin_lock_irq(&q->db_lock);
+ q->db_disabled = 1;
+ spin_unlock_irq(&q->db_lock);
+}
+
+static void enable_txq_db(struct sge_txq *q)
+{
+ spin_lock_irq(&q->db_lock);
+ q->db_disabled = 0;
+ spin_unlock_irq(&q->db_lock);
+}
+
+static void disable_dbs(struct adapter *adap)
+{
+ int i;
+
+ for_each_ethrxq(&adap->sge, i)
+ disable_txq_db(&adap->sge.ethtxq[i].q);
+ for_each_ofldrxq(&adap->sge, i)
+ disable_txq_db(&adap->sge.ofldtxq[i].q);
+ for_each_port(adap, i)
+ disable_txq_db(&adap->sge.ctrlq[i].q);
+}
+
+static void enable_dbs(struct adapter *adap)
+{
+ int i;
+
+ for_each_ethrxq(&adap->sge, i)
+ enable_txq_db(&adap->sge.ethtxq[i].q);
+ for_each_ofldrxq(&adap->sge, i)
+ enable_txq_db(&adap->sge.ofldtxq[i].q);
+ for_each_port(adap, i)
+ enable_txq_db(&adap->sge.ctrlq[i].q);
+}
+
+static void sync_txq_pidx(struct adapter *adap, struct sge_txq *q)
+{
+ u16 hw_pidx, hw_cidx;
+ int ret;
+
+ spin_lock_bh(&q->db_lock);
+ ret = read_eq_indices(adap, (u16)q->cntxt_id, &hw_pidx, &hw_cidx);
+ if (ret)
+ goto out;
+ if (q->db_pidx != hw_pidx) {
+ u16 delta;
+
+ if (q->db_pidx >= hw_pidx)
+ delta = q->db_pidx - hw_pidx;
+ else
+ delta = q->size - hw_pidx + q->db_pidx;
+ wmb();
+ t4_write_reg(adap, MYPF_REG(A_SGE_PF_KDOORBELL),
+ V_QID(q->cntxt_id) | V_PIDX(delta));
+ }
+out:
+ q->db_disabled = 0;
+ spin_unlock_bh(&q->db_lock);
+ if (ret)
+ CH_WARN(adap, "DB drop recovery failed.\n");
+}
+static void recover_all_queues(struct adapter *adap)
+{
+ int i;
+
+ for_each_ethrxq(&adap->sge, i)
+ sync_txq_pidx(adap, &adap->sge.ethtxq[i].q);
+ for_each_ofldrxq(&adap->sge, i)
+ sync_txq_pidx(adap, &adap->sge.ofldtxq[i].q);
+ for_each_port(adap, i)
+ sync_txq_pidx(adap, &adap->sge.ctrlq[i].q);
+}
+
static void notify_rdma_uld(struct adapter *adap, enum cxgb4_control cmd)
{
mutex_lock(&uld_mutex);
@@ -2468,55 +2612,41 @@ static void notify_rdma_uld(struct adapter *adap, enum cxgb4_control cmd)
static void process_db_full(struct work_struct *work)
{
struct adapter *adap;
- static int delay = 1000;
- u32 v;
adap = container_of(work, struct adapter, db_full_task);
-
- /* stop LLD queues */
-
notify_rdma_uld(adap, CXGB4_CONTROL_DB_FULL);
- do {
- set_current_state(TASK_UNINTERRUPTIBLE);
- schedule_timeout(usecs_to_jiffies(delay));
- v = t4_read_reg(adap, A_SGE_DBFIFO_STATUS);
- if (G_LP_COUNT(v) == 0 && G_HP_COUNT(v) == 0)
- break;
- } while (1);
+ drain_db_fifo(adap, dbfifo_drain_delay);
+ t4_set_reg_field(adap, A_SGE_INT_ENABLE3,
+ F_DBFIFO_HP_INT | F_DBFIFO_LP_INT,
+ F_DBFIFO_HP_INT | F_DBFIFO_LP_INT);
notify_rdma_uld(adap, CXGB4_CONTROL_DB_EMPTY);
-
-
- /*
- * The more we get db full interrupts, the more we'll delay
- * in re-enabling db rings on queues, capped off at 200ms.
- */
- delay = min(delay << 1, 200000);
-
- /* resume LLD queues */
}
static void process_db_drop(struct work_struct *work)
{
struct adapter *adap;
- adap = container_of(work, struct adapter, db_drop_task);
+ adap = container_of(work, struct adapter, db_drop_task);
- /*
- * sync the PIDX values in HW and SW for LLD queues.
- */
-
+ t4_set_reg_field(adap, A_SGE_DOORBELL_CONTROL, F_DROPPED_DB, 0);
+ disable_dbs(adap);
notify_rdma_uld(adap, CXGB4_CONTROL_DB_DROP);
+ drain_db_fifo(adap, 1);
+ recover_all_queues(adap);
+ enable_dbs(adap);
}
void t4_db_full(struct adapter *adap)
{
- schedule_work(&adap->db_full_task);
+ t4_set_reg_field(adap, A_SGE_INT_ENABLE3,
+ F_DBFIFO_HP_INT | F_DBFIFO_LP_INT, 0);
+ queue_work(workq, &adap->db_full_task);
}
void t4_db_dropped(struct adapter *adap)
{
- schedule_work(&adap->db_drop_task);
+ queue_work(workq, &adap->db_drop_task);
}
static void uld_attach(struct adapter *adap, unsigned int uld)
@@ -2552,6 +2682,7 @@ static void uld_attach(struct adapter *adap, unsigned int uld)
lli.gts_reg = adap->regs + MYPF_REG(SGE_PF_GTS);
lli.db_reg = adap->regs + MYPF_REG(SGE_PF_KDOORBELL);
lli.fw_vers = adap->params.fw_vers;
+ lli.dbfifo_int_thresh = dbfifo_int_thresh;
handle = ulds[uld].add(&lli);
if (IS_ERR(handle)) {
@@ -3668,6 +3799,7 @@ static int __devinit init_one(struct pci_dev *pdev,
adapter->pdev = pdev;
adapter->pdev_dev = &pdev->dev;
+ adapter->mbox = func;
adapter->fn = func;
adapter->msg_enable = dflt_msg_enable;
memset(adapter->chan_map, 0xff, sizeof(adapter->chan_map));
@@ -3865,6 +3997,10 @@ static int __init cxgb4_init_module(void)
{
int ret;
+ workq = create_singlethread_workqueue("cxgb4");
+ if (!workq)
+ return -ENOMEM;
+
/* Debugfs support is optional, just warn if this fails */
cxgb4_debugfs_root = debugfs_create_dir(KBUILD_MODNAME, NULL);
if (!cxgb4_debugfs_root)
@@ -3880,6 +4016,8 @@ static void __exit cxgb4_cleanup_module(void)
{
pci_unregister_driver(&cxgb4_driver);
debugfs_remove(cxgb4_debugfs_root); /* NULL ok */
+ flush_workqueue(workq);
+ destroy_workqueue(workq);
}
module_init(cxgb4_init_module);
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h
index 5cc2f27..d79980c 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h
@@ -218,6 +218,7 @@ struct cxgb4_lld_info {
unsigned short ucq_density; /* # of user CQs/page */
void __iomem *gts_reg; /* address of GTS register */
void __iomem *db_reg; /* address of kernel doorbell */
+ int dbfifo_int_thresh; /* doorbell fifo int threshold */
};
struct cxgb4_uld_info {
@@ -226,6 +227,7 @@ struct cxgb4_uld_info {
int (*rx_handler)(void *handle, const __be64 *rsp,
const struct pkt_gl *gl);
int (*state_change)(void *handle, enum cxgb4_state new_state);
+ int (*control)(void *handle, enum cxgb4_control control, ...);
};
int cxgb4_register_uld(enum cxgb4_uld type, const struct cxgb4_uld_info *p);
@@ -243,4 +245,6 @@ void cxgb4_iscsi_init(struct net_device *dev, unsigned int tag_mask,
const unsigned int *pgsz_order);
struct sk_buff *cxgb4_pktgl_to_skb(const struct pkt_gl *gl,
unsigned int skb_len, unsigned int pull_len);
+int cxgb4_sync_txq_pidx(struct net_device *dev, u16 qid, u16 pidx, u16 size);
+int cxgb4_flush_eq_cache(struct net_device *dev);
#endif /* !__CXGB4_OFLD_H */
diff --git a/drivers/net/ethernet/chelsio/cxgb4/sge.c b/drivers/net/ethernet/chelsio/cxgb4/sge.c
index 234c157..e111d97 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/sge.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/sge.c
@@ -767,8 +767,13 @@ static void write_sgl(const struct sk_buff *skb, struct sge_txq *q,
static inline void ring_tx_db(struct adapter *adap, struct sge_txq *q, int n)
{
wmb(); /* write descriptors before telling HW */
- t4_write_reg(adap, MYPF_REG(SGE_PF_KDOORBELL),
- QID(q->cntxt_id) | PIDX(n));
+ spin_lock(&q->db_lock);
+ if (!q->db_disabled) {
+ t4_write_reg(adap, MYPF_REG(A_SGE_PF_KDOORBELL),
+ V_QID(q->cntxt_id) | V_PIDX(n));
+ }
+ q->db_pidx = q->pidx;
+ spin_unlock(&q->db_lock);
}
/**
@@ -2081,6 +2086,7 @@ static void init_txq(struct adapter *adap, struct sge_txq *q, unsigned int id)
q->stops = q->restarts = 0;
q->stat = (void *)&q->desc[q->size];
q->cntxt_id = id;
+ spin_lock_init(&q->db_lock);
adap->sge.egr_map[id - adap->sge.egr_start] = q;
}
@@ -2415,9 +2421,15 @@ void t4_sge_init(struct adapter *adap)
RXPKTCPLMODE |
(STAT_LEN == 128 ? EGRSTATUSPAGESIZE : 0));
+ /*
+ * Set up to drop DOORBELL writes when the DOORBELL FIFO overflows
+ * and generate an interrupt when this occurs so we can recover.
+ */
t4_set_reg_field(adap, A_SGE_DBFIFO_STATUS,
- V_HP_INT_THRESH(5) | V_LP_INT_THRESH(5),
- V_HP_INT_THRESH(5) | V_LP_INT_THRESH(5));
+ V_HP_INT_THRESH(M_HP_INT_THRESH) |
+ V_LP_INT_THRESH(M_LP_INT_THRESH),
+ V_HP_INT_THRESH(dbfifo_int_thresh) |
+ V_LP_INT_THRESH(dbfifo_int_thresh));
t4_set_reg_field(adap, A_SGE_DOORBELL_CONTROL, F_ENABLE_DROP,
F_ENABLE_DROP);
diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_regs.h b/drivers/net/ethernet/chelsio/cxgb4/t4_regs.h
index 0adc5bc..111fc32 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/t4_regs.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/t4_regs.h
@@ -190,6 +190,59 @@
#define SGE_DEBUG_DATA_LOW 0x10d4
#define SGE_INGRESS_QUEUES_PER_PAGE_PF 0x10f4
+#define S_LP_INT_THRESH 12
+#define V_LP_INT_THRESH(x) ((x) << S_LP_INT_THRESH)
+#define S_HP_INT_THRESH 28
+#define V_HP_INT_THRESH(x) ((x) << S_HP_INT_THRESH)
+#define A_SGE_DBFIFO_STATUS 0x10a4
+
+#define S_ENABLE_DROP 13
+#define V_ENABLE_DROP(x) ((x) << S_ENABLE_DROP)
+#define F_ENABLE_DROP V_ENABLE_DROP(1U)
+#define A_SGE_DOORBELL_CONTROL 0x10a8
+
+#define A_SGE_CTXT_CMD 0x11fc
+#define A_SGE_DBQ_CTXT_BADDR 0x1084
+
+#define A_SGE_PF_KDOORBELL 0x0
+
+#define S_QID 15
+#define V_QID(x) ((x) << S_QID)
+
+#define S_PIDX 0
+#define V_PIDX(x) ((x) << S_PIDX)
+
+#define M_LP_COUNT 0x7ffU
+#define S_LP_COUNT 0
+#define G_LP_COUNT(x) (((x) >> S_LP_COUNT) & M_LP_COUNT)
+
+#define M_HP_COUNT 0x7ffU
+#define S_HP_COUNT 16
+#define G_HP_COUNT(x) (((x) >> S_HP_COUNT) & M_HP_COUNT)
+
+#define A_SGE_INT_ENABLE3 0x1040
+
+#define S_DBFIFO_HP_INT 8
+#define V_DBFIFO_HP_INT(x) ((x) << S_DBFIFO_HP_INT)
+#define F_DBFIFO_HP_INT V_DBFIFO_HP_INT(1U)
+
+#define S_DBFIFO_LP_INT 7
+#define V_DBFIFO_LP_INT(x) ((x) << S_DBFIFO_LP_INT)
+#define F_DBFIFO_LP_INT V_DBFIFO_LP_INT(1U)
+
+#define S_DROPPED_DB 0
+#define V_DROPPED_DB(x) ((x) << S_DROPPED_DB)
+#define F_DROPPED_DB V_DROPPED_DB(1U)
+
+#define S_ERR_DROPPED_DB 18
+#define V_ERR_DROPPED_DB(x) ((x) << S_ERR_DROPPED_DB)
+#define F_ERR_DROPPED_DB V_ERR_DROPPED_DB(1U)
+
+#define A_PCIE_MEM_ACCESS_OFFSET 0x306c
+
+#define M_HP_INT_THRESH 0xfU
+#define M_LP_INT_THRESH 0xfU
+
#define PCIE_PF_CLI 0x44
#define PCIE_INT_CAUSE 0x3004
#define UNXSPLCPLERR 0x20000000U
diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h b/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h
index edcfd7e..ad53f79 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h
@@ -1620,4 +1620,19 @@ struct fw_hdr {
#define FW_HDR_FW_VER_MINOR_GET(x) (((x) >> 16) & 0xff)
#define FW_HDR_FW_VER_MICRO_GET(x) (((x) >> 8) & 0xff)
#define FW_HDR_FW_VER_BUILD_GET(x) (((x) >> 0) & 0xff)
+
+#define S_FW_CMD_OP 24
+#define V_FW_CMD_OP(x) ((x) << S_FW_CMD_OP)
+
+#define S_FW_CMD_REQUEST 23
+#define V_FW_CMD_REQUEST(x) ((x) << S_FW_CMD_REQUEST)
+#define F_FW_CMD_REQUEST V_FW_CMD_REQUEST(1U)
+
+#define S_FW_CMD_WRITE 21
+#define V_FW_CMD_WRITE(x) ((x) << S_FW_CMD_WRITE)
+#define F_FW_CMD_WRITE V_FW_CMD_WRITE(1U)
+
+#define S_FW_LDST_CMD_ADDRSPACE 0
+#define V_FW_LDST_CMD_ADDRSPACE(x) ((x) << S_FW_LDST_CMD_ADDRSPACE)
+
#endif /* _T4FW_INTERFACE_H_ */
--
1.7.1
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply related
* [PATCH 02/10] cxgb4: Common platform specific changes for DB Drop Recovery
From: Vipul Pandya @ 2012-05-18 9:59 UTC (permalink / raw)
To: linux-rdma-u79uwXL29TY76Z2rM5mHXA, netdev-u79uwXL29TY76Z2rM5mHXA
Cc: roland-BHEL68pLQRGGvPXPguhicg, davem-fT/PcQaiUtIeIZ0/mPfg9Q,
divy-ut6Up61K2wZBDgjK7y7TUQ, dm-ut6Up61K2wZBDgjK7y7TUQ,
kumaras-ut6Up61K2wZBDgjK7y7TUQ,
swise-7bPotxP6k4+P2YhJcF5u+vpXobYPEAuW, Vipul Pandya
In-Reply-To: <1337335173-3226-1-git-send-email-vipul-ut6Up61K2wZBDgjK7y7TUQ@public.gmane.org>
Add platform-specific callback functions for interrupts. This is
needed to do a single read-clear of the CAUSE register and then call
out to platform specific functions for DB threshold interrupts and DB
drop interrupts.
Add t4_mem_win_read_len() - mem-window reads for arbitrary lengths.
This is used to read the CIDX/PIDX values from EC contexts during DB
drop recovery.
Add t4_fwaddrspace_write() - sends addrspace write cmds to the fw.
Needed to flush the sge eq context cache.
Signed-off-by: Vipul Pandya <vipul-ut6Up61K2wZBDgjK7y7TUQ@public.gmane.org>
Signed-off-by: Steve Wise <swise-7bPotxP6k4+P2YhJcF5u+vpXobYPEAuW@public.gmane.org>
---
drivers/net/ethernet/chelsio/cxgb4/cxgb4.h | 3 +
drivers/net/ethernet/chelsio/cxgb4/t4_hw.c | 69 +++++++++++++++++++++++----
2 files changed, 61 insertions(+), 11 deletions(-)
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h
index f91b259..5f3c0a7 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h
@@ -723,4 +723,7 @@ int t4_ofld_eq_free(struct adapter *adap, unsigned int mbox, unsigned int pf,
int t4_handle_fw_rpl(struct adapter *adap, const __be64 *rpl);
void t4_db_full(struct adapter *adapter);
void t4_db_dropped(struct adapter *adapter);
+int t4_mem_win_read_len(struct adapter *adap, u32 addr, __be32 *data, int len);
+int t4_fwaddrspace_write(struct adapter *adap, unsigned int mbox,
+ u32 addr, u32 val);
#endif /* __CXGB4_H__ */
diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c
index 13609bf..32e1dd5 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c
@@ -868,11 +868,14 @@ int t4_restart_aneg(struct adapter *adap, unsigned int mbox, unsigned int port)
return t4_wr_mbox(adap, mbox, &c, sizeof(c), NULL);
}
+typedef void (*int_handler_t)(struct adapter *adap);
+
struct intr_info {
unsigned int mask; /* bits to check in interrupt status */
const char *msg; /* message to print or NULL */
short stat_idx; /* stat counter to increment or -1 */
unsigned short fatal; /* whether the condition reported is fatal */
+ int_handler_t int_handler; /* platform-specific int handler */
};
/**
@@ -905,6 +908,8 @@ static int t4_handle_intr_status(struct adapter *adapter, unsigned int reg,
} else if (acts->msg && printk_ratelimit())
dev_warn(adapter->pdev_dev, "%s (0x%x)\n", acts->msg,
status & acts->mask);
+ if (acts->int_handler)
+ acts->int_handler(adapter);
mask |= acts->mask;
}
status &= mask;
@@ -1013,9 +1018,9 @@ static void sge_intr_handler(struct adapter *adapter)
{ ERR_INVALID_CIDX_INC,
"SGE GTS CIDX increment too large", -1, 0 },
{ ERR_CPL_OPCODE_0, "SGE received 0-length CPL", -1, 0 },
- { F_DBFIFO_LP_INT, NULL, -1, 0 },
- { F_DBFIFO_HP_INT, NULL, -1, 0 },
- { ERR_DROPPED_DB, "SGE doorbell dropped", -1, 0 },
+ { F_DBFIFO_LP_INT, NULL, -1, 0, t4_db_full },
+ { F_DBFIFO_HP_INT, NULL, -1, 0, t4_db_full },
+ { F_ERR_DROPPED_DB, NULL, -1, 0, t4_db_dropped },
{ ERR_DATA_CPL_ON_HIGH_QID1 | ERR_DATA_CPL_ON_HIGH_QID0,
"SGE IQID > 1023 received CPL for FL", -1, 0 },
{ ERR_BAD_DB_PIDX3, "SGE DBP 3 pidx increment too large", -1,
@@ -1036,20 +1041,14 @@ static void sge_intr_handler(struct adapter *adapter)
};
v = (u64)t4_read_reg(adapter, SGE_INT_CAUSE1) |
- ((u64)t4_read_reg(adapter, SGE_INT_CAUSE2) << 32);
+ ((u64)t4_read_reg(adapter, SGE_INT_CAUSE2) << 32);
if (v) {
dev_alert(adapter->pdev_dev, "SGE parity error (%#llx)\n",
- (unsigned long long)v);
+ (unsigned long long)v);
t4_write_reg(adapter, SGE_INT_CAUSE1, v);
t4_write_reg(adapter, SGE_INT_CAUSE2, v >> 32);
}
- err = t4_read_reg(adapter, A_SGE_INT_CAUSE3);
- if (err & (F_DBFIFO_HP_INT|F_DBFIFO_LP_INT))
- t4_db_full(adapter);
- if (err & F_ERR_DROPPED_DB)
- t4_db_dropped(adapter);
-
if (t4_handle_intr_status(adapter, SGE_INT_CAUSE3, sge_intr_info) ||
v != 0)
t4_fatal_err(adapter);
@@ -1995,6 +1994,54 @@ int t4_wol_pat_enable(struct adapter *adap, unsigned int port, unsigned int map,
(var).retval_len16 = htonl(FW_LEN16(var)); \
} while (0)
+int t4_fwaddrspace_write(struct adapter *adap, unsigned int mbox,
+ u32 addr, u32 val)
+{
+ struct fw_ldst_cmd c;
+
+ memset(&c, 0, sizeof(c));
+ c.op_to_addrspace = htonl(V_FW_CMD_OP(FW_LDST_CMD) | F_FW_CMD_REQUEST |
+ F_FW_CMD_WRITE |
+ V_FW_LDST_CMD_ADDRSPACE(FW_LDST_ADDRSPC_FIRMWARE));
+ c.cycles_to_len16 = htonl(FW_LEN16(c));
+ c.u.addrval.addr = htonl(addr);
+ c.u.addrval.val = htonl(val);
+
+ return t4_wr_mbox(adap, mbox, &c, sizeof(c), NULL);
+}
+
+/*
+ * t4_mem_win_read_len - read memory through PCIE memory window
+ * @adap: the adapter
+ * @addr: address of first byte requested aligned on 32b.
+ * @data: len bytes to hold the data read
+ * @len: amount of data to read from window. Must be <=
+ * MEMWIN0_APERATURE after adjusting for 16B alignment
+ * requirements of the the memory window.
+ *
+ * Read len bytes of data from MC starting at @addr.
+ */
+int t4_mem_win_read_len(struct adapter *adap, u32 addr, __be32 *data, int len)
+{
+ int i;
+ int off;
+
+ /*
+ * Align on a 16B boundary.
+ */
+ off = addr & 15;
+ if ((addr & 3) || (len + off) > MEMWIN0_APERTURE)
+ return -EINVAL;
+
+ t4_write_reg(adap, A_PCIE_MEM_ACCESS_OFFSET, addr & ~15);
+ t4_read_reg(adap, A_PCIE_MEM_ACCESS_OFFSET);
+
+ for (i = 0; i < len; i += 4)
+ *data++ = t4_read_reg(adap, (MEMWIN0_BASE + off + i));
+
+ return 0;
+}
+
/**
* t4_mdio_rd - read a PHY register through MDIO
* @adap: the adapter
--
1.7.1
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply related
* [PATCH 01/10] cxgb4: Detect DB FULL events and notify RDMA ULD.
From: Vipul Pandya @ 2012-05-18 9:59 UTC (permalink / raw)
To: linux-rdma-u79uwXL29TY76Z2rM5mHXA, netdev-u79uwXL29TY76Z2rM5mHXA
Cc: roland-BHEL68pLQRGGvPXPguhicg, davem-fT/PcQaiUtIeIZ0/mPfg9Q,
divy-ut6Up61K2wZBDgjK7y7TUQ, dm-ut6Up61K2wZBDgjK7y7TUQ,
kumaras-ut6Up61K2wZBDgjK7y7TUQ,
swise-7bPotxP6k4+P2YhJcF5u+vpXobYPEAuW, Vipul Pandya
In-Reply-To: <1337335173-3226-1-git-send-email-vipul-ut6Up61K2wZBDgjK7y7TUQ@public.gmane.org>
Signed-off-by: Vipul Pandya <vipul-ut6Up61K2wZBDgjK7y7TUQ@public.gmane.org>
Signed-off-by: Steve Wise <swise-7bPotxP6k4+P2YhJcF5u+vpXobYPEAuW@public.gmane.org>
---
drivers/net/ethernet/chelsio/cxgb4/cxgb4.h | 4 +
drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c | 77 +++++++++++++++++++++++
drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h | 7 ++
drivers/net/ethernet/chelsio/cxgb4/sge.c | 6 ++
drivers/net/ethernet/chelsio/cxgb4/t4_hw.c | 9 +++
5 files changed, 103 insertions(+), 0 deletions(-)
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h
index 0fe1885..f91b259 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h
@@ -504,6 +504,8 @@ struct adapter {
void **tid_release_head;
spinlock_t tid_release_lock;
struct work_struct tid_release_task;
+ struct work_struct db_full_task;
+ struct work_struct db_drop_task;
bool tid_release_task_busy;
struct dentry *debugfs_root;
@@ -719,4 +721,6 @@ int t4_ctrl_eq_free(struct adapter *adap, unsigned int mbox, unsigned int pf,
int t4_ofld_eq_free(struct adapter *adap, unsigned int mbox, unsigned int pf,
unsigned int vf, unsigned int eqid);
int t4_handle_fw_rpl(struct adapter *adap, const __be64 *rpl);
+void t4_db_full(struct adapter *adapter);
+void t4_db_dropped(struct adapter *adapter);
#endif /* __CXGB4_H__ */
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
index b126b98..c243f93 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
@@ -2366,6 +2366,16 @@ unsigned int cxgb4_port_chan(const struct net_device *dev)
}
EXPORT_SYMBOL(cxgb4_port_chan);
+unsigned int cxgb4_dbfifo_count(const struct net_device *dev, int lpfifo)
+{
+ struct adapter *adap = netdev2adap(dev);
+ u32 v;
+
+ v = t4_read_reg(adap, A_SGE_DBFIFO_STATUS);
+ return lpfifo ? G_LP_COUNT(v) : G_HP_COUNT(v);
+}
+EXPORT_SYMBOL(cxgb4_dbfifo_count);
+
/**
* cxgb4_port_viid - get the VI id of a port
* @dev: the net device for the port
@@ -2446,6 +2456,69 @@ static struct notifier_block cxgb4_netevent_nb = {
.notifier_call = netevent_cb
};
+static void notify_rdma_uld(struct adapter *adap, enum cxgb4_control cmd)
+{
+ mutex_lock(&uld_mutex);
+ if (adap->uld_handle[CXGB4_ULD_RDMA])
+ ulds[CXGB4_ULD_RDMA].control(adap->uld_handle[CXGB4_ULD_RDMA],
+ cmd);
+ mutex_unlock(&uld_mutex);
+}
+
+static void process_db_full(struct work_struct *work)
+{
+ struct adapter *adap;
+ static int delay = 1000;
+ u32 v;
+
+ adap = container_of(work, struct adapter, db_full_task);
+
+
+ /* stop LLD queues */
+
+ notify_rdma_uld(adap, CXGB4_CONTROL_DB_FULL);
+ do {
+ set_current_state(TASK_UNINTERRUPTIBLE);
+ schedule_timeout(usecs_to_jiffies(delay));
+ v = t4_read_reg(adap, A_SGE_DBFIFO_STATUS);
+ if (G_LP_COUNT(v) == 0 && G_HP_COUNT(v) == 0)
+ break;
+ } while (1);
+ notify_rdma_uld(adap, CXGB4_CONTROL_DB_EMPTY);
+
+
+ /*
+ * The more we get db full interrupts, the more we'll delay
+ * in re-enabling db rings on queues, capped off at 200ms.
+ */
+ delay = min(delay << 1, 200000);
+
+ /* resume LLD queues */
+}
+
+static void process_db_drop(struct work_struct *work)
+{
+ struct adapter *adap;
+ adap = container_of(work, struct adapter, db_drop_task);
+
+
+ /*
+ * sync the PIDX values in HW and SW for LLD queues.
+ */
+
+ notify_rdma_uld(adap, CXGB4_CONTROL_DB_DROP);
+}
+
+void t4_db_full(struct adapter *adap)
+{
+ schedule_work(&adap->db_full_task);
+}
+
+void t4_db_dropped(struct adapter *adap)
+{
+ schedule_work(&adap->db_drop_task);
+}
+
static void uld_attach(struct adapter *adap, unsigned int uld)
{
void *handle;
@@ -2649,6 +2722,8 @@ static void cxgb_down(struct adapter *adapter)
{
t4_intr_disable(adapter);
cancel_work_sync(&adapter->tid_release_task);
+ cancel_work_sync(&adapter->db_full_task);
+ cancel_work_sync(&adapter->db_drop_task);
adapter->tid_release_task_busy = false;
adapter->tid_release_head = NULL;
@@ -3601,6 +3676,8 @@ static int __devinit init_one(struct pci_dev *pdev,
spin_lock_init(&adapter->tid_release_lock);
INIT_WORK(&adapter->tid_release_task, process_tid_release_list);
+ INIT_WORK(&adapter->db_full_task, process_db_full);
+ INIT_WORK(&adapter->db_drop_task, process_db_drop);
err = t4_prep_adapter(adapter);
if (err)
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h
index b1d39b8..5cc2f27 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h
@@ -163,6 +163,12 @@ enum cxgb4_state {
CXGB4_STATE_DETACH
};
+enum cxgb4_control {
+ CXGB4_CONTROL_DB_FULL,
+ CXGB4_CONTROL_DB_EMPTY,
+ CXGB4_CONTROL_DB_DROP,
+};
+
struct pci_dev;
struct l2t_data;
struct net_device;
@@ -225,6 +231,7 @@ struct cxgb4_uld_info {
int cxgb4_register_uld(enum cxgb4_uld type, const struct cxgb4_uld_info *p);
int cxgb4_unregister_uld(enum cxgb4_uld type);
int cxgb4_ofld_send(struct net_device *dev, struct sk_buff *skb);
+unsigned int cxgb4_dbfifo_count(const struct net_device *dev, int lpfifo);
unsigned int cxgb4_port_chan(const struct net_device *dev);
unsigned int cxgb4_port_viid(const struct net_device *dev);
unsigned int cxgb4_port_idx(const struct net_device *dev);
diff --git a/drivers/net/ethernet/chelsio/cxgb4/sge.c b/drivers/net/ethernet/chelsio/cxgb4/sge.c
index 2dae795..234c157 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/sge.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/sge.c
@@ -2415,6 +2415,12 @@ void t4_sge_init(struct adapter *adap)
RXPKTCPLMODE |
(STAT_LEN == 128 ? EGRSTATUSPAGESIZE : 0));
+ t4_set_reg_field(adap, A_SGE_DBFIFO_STATUS,
+ V_HP_INT_THRESH(5) | V_LP_INT_THRESH(5),
+ V_HP_INT_THRESH(5) | V_LP_INT_THRESH(5));
+ t4_set_reg_field(adap, A_SGE_DOORBELL_CONTROL, F_ENABLE_DROP,
+ F_ENABLE_DROP);
+
for (i = v = 0; i < 32; i += 4)
v |= (PAGE_SHIFT - 10) << i;
t4_write_reg(adap, SGE_HOST_PAGE_SIZE, v);
diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c
index d1ec111..13609bf 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c
@@ -1013,6 +1013,8 @@ static void sge_intr_handler(struct adapter *adapter)
{ ERR_INVALID_CIDX_INC,
"SGE GTS CIDX increment too large", -1, 0 },
{ ERR_CPL_OPCODE_0, "SGE received 0-length CPL", -1, 0 },
+ { F_DBFIFO_LP_INT, NULL, -1, 0 },
+ { F_DBFIFO_HP_INT, NULL, -1, 0 },
{ ERR_DROPPED_DB, "SGE doorbell dropped", -1, 0 },
{ ERR_DATA_CPL_ON_HIGH_QID1 | ERR_DATA_CPL_ON_HIGH_QID0,
"SGE IQID > 1023 received CPL for FL", -1, 0 },
@@ -1042,6 +1044,12 @@ static void sge_intr_handler(struct adapter *adapter)
t4_write_reg(adapter, SGE_INT_CAUSE2, v >> 32);
}
+ err = t4_read_reg(adapter, A_SGE_INT_CAUSE3);
+ if (err & (F_DBFIFO_HP_INT|F_DBFIFO_LP_INT))
+ t4_db_full(adapter);
+ if (err & F_ERR_DROPPED_DB)
+ t4_db_dropped(adapter);
+
if (t4_handle_intr_status(adapter, SGE_INT_CAUSE3, sge_intr_info) ||
v != 0)
t4_fatal_err(adapter);
@@ -1513,6 +1521,7 @@ void t4_intr_enable(struct adapter *adapter)
ERR_BAD_DB_PIDX2 | ERR_BAD_DB_PIDX1 |
ERR_BAD_DB_PIDX0 | ERR_ING_CTXT_PRIO |
ERR_EGR_CTXT_PRIO | INGRESS_SIZE_ERR |
+ F_DBFIFO_HP_INT | F_DBFIFO_LP_INT |
EGRESS_SIZE_ERR);
t4_write_reg(adapter, MYPF_REG(PL_PF_INT_ENABLE), PF_INTR_MASK);
t4_set_reg_field(adapter, PL_INT_MAP0, 0, 1 << pf);
--
1.7.1
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply related
* Re: [V2 PATCH 9/9] vhost: zerocopy: poll vq in zerocopy callback
From: Jason Wang @ 2012-05-18 9:58 UTC (permalink / raw)
To: Shirley Ma
Cc: Michael S. Tsirkin, eric.dumazet, netdev, linux-kernel, ebiederm,
davem
In-Reply-To: <1337268862.10741.58.camel@oc3660625478.ibm.com>
On 05/17/2012 11:34 PM, Shirley Ma wrote:
> On Thu, 2012-05-17 at 10:50 +0800, Jason Wang wrote:
>> The problem is we may stop the tx queue when there no enough capacity
>> to
>> place packets, at this moment we depends on the tx interrupt to
>> re-enable the tx queue. So if we didn't poll the vhost during
>> callback,
>> guest may lose the tx interrupt to re-enable the tx queue which could
>> stall the whole tx queue.
> VHOST_MAX_PEND should handle the capacity.
>
> Hasn't the above situation been handled in handle_tx() code?:
> ...
> if (unlikely(num_pends> VHOST_MAX_PEND)) {
> tx_poll_start(net, sock);
> set_bit(SOCK_ASYNC_NOSPACE,&sock->flags);
> break;
> }
> ...
>
> Thanks
> Shirley
It may not help in because:
- tx polling depends on skb_orphan() which is often called by device
driver when it place the packet into the queue of the devices instead
of when the packets were sent. So it was too early for vhost to be
notified.
- it only works when the pending DMAs exceeds VHOST_MAX_PEND, it's
highly possible that guest needs to be notified when the pending packets
isn't so much.
So this piece of code may not help and could be removed and we need to
poll the virt-queue during zerocopy callback ( through it could be
further optimized but may not be easy).
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
^ permalink raw reply
* [PATCH] pktgen: fix module unload for good
From: Eric Dumazet @ 2012-05-18 9:52 UTC (permalink / raw)
To: David Miller; +Cc: netdev
From: Eric Dumazet <edumazet@google.com>
commit c57b5468406 (pktgen: fix crash at module unload) did a very poor
job with list primitives.
1) list_splice() arguments were in the wrong order
2) list_splice(list, head) has undefined behavior if head is not
initialized.
3) We should use the list_splice_init() variant to clear pktgen_threads
list.
Signed-off-by: Eric Dumazet <edumazet@google.com>
---
Sorry for the mess, I really dont understand how my machine did not
crash with c57b5468406 commit...
net/core/pktgen.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/net/core/pktgen.c b/net/core/pktgen.c
index 77a5998..b81369b 100644
--- a/net/core/pktgen.c
+++ b/net/core/pktgen.c
@@ -3755,13 +3755,13 @@ static void __exit pg_cleanup(void)
{
struct pktgen_thread *t;
struct list_head *q, *n;
- struct list_head list;
+ LIST_HEAD(list);
/* Stop all interfaces & threads */
pktgen_exiting = true;
mutex_lock(&pktgen_thread_lock);
- list_splice(&list, &pktgen_threads);
+ list_splice_init(&pktgen_threads, &list);
mutex_unlock(&pktgen_thread_lock);
list_for_each_safe(q, n, &list) {
^ permalink raw reply related
* [PATCH net-next v5] be2net: Fix to allow get/set of debug levels in the firmware.
From: Somnath Kotur @ 2012-05-18 8:59 UTC (permalink / raw)
To: netdev, bhutchings; +Cc: Somnath Kotur, Suresh Reddy
Patch re-spin.
Incorporated review comments by Ben Hutchings.
Signed-off-by: Suresh Reddy <suresh.reddy@emulex.com>
Signed-off-by: Somnath Kotur <somnath.kotur@emulex.com>
---
drivers/net/ethernet/emulex/benet/be.h | 3 +
drivers/net/ethernet/emulex/benet/be_cmds.c | 56 +++++++++++++++++
drivers/net/ethernet/emulex/benet/be_cmds.h | 57 +++++++++++++++++
drivers/net/ethernet/emulex/benet/be_ethtool.c | 77 ++++++++++++++++++++++++
drivers/net/ethernet/emulex/benet/be_main.c | 37 +++++++++++
5 files changed, 230 insertions(+), 0 deletions(-)
diff --git a/drivers/net/ethernet/emulex/benet/be.h b/drivers/net/ethernet/emulex/benet/be.h
index ecf1a81..ff4eb8f 100644
--- a/drivers/net/ethernet/emulex/benet/be.h
+++ b/drivers/net/ethernet/emulex/benet/be.h
@@ -421,6 +421,7 @@ struct be_adapter {
bool wol;
u32 max_pmac_cnt; /* Max secondary UC MACs programmable */
u32 uc_macs; /* Count of secondary UC MAC programmed */
+ u32 msg_enable;
};
#define be_physfn(adapter) (!adapter->virtfn)
@@ -603,4 +604,6 @@ extern void be_parse_stats(struct be_adapter *adapter);
extern int be_load_fw(struct be_adapter *adapter, u8 *func);
extern bool be_is_wol_supported(struct be_adapter *adapter);
extern bool be_pause_supported(struct be_adapter *adapter);
+extern u32 be_get_fw_log_level(struct be_adapter *adapter);
+
#endif /* BE_H */
diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.c b/drivers/net/ethernet/emulex/benet/be_cmds.c
index 43167e8..b24623c 100644
--- a/drivers/net/ethernet/emulex/benet/be_cmds.c
+++ b/drivers/net/ethernet/emulex/benet/be_cmds.c
@@ -2589,4 +2589,60 @@ err:
mutex_unlock(&adapter->mbox_lock);
pci_free_consistent(adapter->pdev, cmd.size, cmd.va, cmd.dma);
return status;
+
+}
+int be_cmd_get_ext_fat_capabilites(struct be_adapter *adapter,
+ struct be_dma_mem *cmd)
+{
+ struct be_mcc_wrb *wrb;
+ struct be_cmd_req_get_ext_fat_caps *req;
+ int status;
+
+ if (mutex_lock_interruptible(&adapter->mbox_lock))
+ return -1;
+
+ wrb = wrb_from_mbox(adapter);
+ if (!wrb) {
+ status = -EBUSY;
+ goto err;
+ }
+
+ req = cmd->va;
+ be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
+ OPCODE_COMMON_GET_EXT_FAT_CAPABILITES,
+ cmd->size, wrb, cmd);
+ req->parameter_type = cpu_to_le32(1);
+
+ status = be_mbox_notify_wait(adapter);
+err:
+ mutex_unlock(&adapter->mbox_lock);
+ return status;
+}
+
+int be_cmd_set_ext_fat_capabilites(struct be_adapter *adapter,
+ struct be_dma_mem *cmd,
+ struct be_fat_conf_params *configs)
+{
+ struct be_mcc_wrb *wrb;
+ struct be_cmd_req_set_ext_fat_caps *req;
+ int status;
+
+ spin_lock_bh(&adapter->mcc_lock);
+
+ wrb = wrb_from_mccq(adapter);
+ if (!wrb) {
+ status = -EBUSY;
+ goto err;
+ }
+
+ req = cmd->va;
+ memcpy(&req->set_params, configs, sizeof(struct be_fat_conf_params));
+ be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
+ OPCODE_COMMON_SET_EXT_FAT_CAPABILITES,
+ cmd->size, wrb, cmd);
+
+ status = be_mcc_notify_wait(adapter);
+err:
+ spin_unlock_bh(&adapter->mcc_lock);
+ return status;
}
diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.h b/drivers/net/ethernet/emulex/benet/be_cmds.h
index 944f031..0b1029b 100644
--- a/drivers/net/ethernet/emulex/benet/be_cmds.h
+++ b/drivers/net/ethernet/emulex/benet/be_cmds.h
@@ -189,6 +189,8 @@ struct be_mcc_mailbox {
#define OPCODE_COMMON_GET_PHY_DETAILS 102
#define OPCODE_COMMON_SET_DRIVER_FUNCTION_CAP 103
#define OPCODE_COMMON_GET_CNTL_ADDITIONAL_ATTRIBUTES 121
+#define OPCODE_COMMON_GET_EXT_FAT_CAPABILITES 125
+#define OPCODE_COMMON_SET_EXT_FAT_CAPABILITES 126
#define OPCODE_COMMON_GET_MAC_LIST 147
#define OPCODE_COMMON_SET_MAC_LIST 148
#define OPCODE_COMMON_GET_HSW_CONFIG 152
@@ -1602,6 +1604,56 @@ static inline void *be_erx_stats_from_cmd(struct be_adapter *adapter)
}
}
+
+/************** get fat capabilites *******************/
+#define MAX_MODULES 27
+#define MAX_MODES 4
+#define MODE_UART 0
+#define FW_LOG_LEVEL_DEFAULT 48
+#define FW_LOG_LEVEL_FATAL 64
+
+struct ext_fat_mode {
+ u8 mode;
+ u8 rsvd0;
+ u16 port_mask;
+ u32 dbg_lvl;
+ u64 fun_mask;
+} __packed;
+
+struct ext_fat_modules {
+ u8 modules_str[32];
+ u32 modules_id;
+ u32 num_modes;
+ struct ext_fat_mode trace_lvl[MAX_MODES];
+} __packed;
+
+struct be_fat_conf_params {
+ u32 max_log_entries;
+ u32 log_entry_size;
+ u8 log_type;
+ u8 max_log_funs;
+ u8 max_log_ports;
+ u8 rsvd0;
+ u32 supp_modes;
+ u32 num_modules;
+ struct ext_fat_modules module[MAX_MODULES];
+} __packed;
+
+struct be_cmd_req_get_ext_fat_caps {
+ struct be_cmd_req_hdr hdr;
+ u32 parameter_type;
+};
+
+struct be_cmd_resp_get_ext_fat_caps {
+ struct be_cmd_resp_hdr hdr;
+ struct be_fat_conf_params get_params;
+};
+
+struct be_cmd_req_set_ext_fat_caps {
+ struct be_cmd_req_hdr hdr;
+ struct be_fat_conf_params set_params;
+};
+
extern int be_pci_fnum_get(struct be_adapter *adapter);
extern int be_cmd_POST(struct be_adapter *adapter);
extern int be_cmd_mac_addr_query(struct be_adapter *adapter, u8 *mac_addr,
@@ -1707,4 +1759,9 @@ extern int be_cmd_set_hsw_config(struct be_adapter *adapter, u16 pvid,
extern int be_cmd_get_hsw_config(struct be_adapter *adapter, u16 *pvid,
u32 domain, u16 intf_id);
extern int be_cmd_get_acpi_wol_cap(struct be_adapter *adapter);
+extern int be_cmd_get_ext_fat_capabilites(struct be_adapter *adapter,
+ struct be_dma_mem *cmd);
+extern int be_cmd_set_ext_fat_capabilites(struct be_adapter *adapter,
+ struct be_dma_mem *cmd,
+ struct be_fat_conf_params *cfgs);
diff --git a/drivers/net/ethernet/emulex/benet/be_ethtool.c b/drivers/net/ethernet/emulex/benet/be_ethtool.c
index 747f68f..63e51d4 100644
--- a/drivers/net/ethernet/emulex/benet/be_ethtool.c
+++ b/drivers/net/ethernet/emulex/benet/be_ethtool.c
@@ -878,6 +878,81 @@ be_read_eeprom(struct net_device *netdev, struct ethtool_eeprom *eeprom,
return status;
}
+static u32 be_get_msg_level(struct net_device *netdev)
+{
+ struct be_adapter *adapter = netdev_priv(netdev);
+
+ if (lancer_chip(adapter)) {
+ dev_err(&adapter->pdev->dev, "Operation not supported\n");
+ return -EOPNOTSUPP;
+ }
+
+ return adapter->msg_enable;
+}
+
+static void be_set_fw_log_level(struct be_adapter *adapter, u32 level)
+{
+ struct be_dma_mem extfat_cmd;
+ struct be_fat_conf_params *cfgs;
+ int status;
+ int i, j;
+
+ memset(&extfat_cmd, 0, sizeof(struct be_dma_mem));
+ extfat_cmd.size = sizeof(struct be_cmd_resp_get_ext_fat_caps);
+ extfat_cmd.va = pci_alloc_consistent(adapter->pdev, extfat_cmd.size,
+ &extfat_cmd.dma);
+ if (!extfat_cmd.va) {
+ dev_err(&adapter->pdev->dev, "%s: Memory allocation failure\n",
+ __func__);
+ goto err;
+ }
+ status = be_cmd_get_ext_fat_capabilites(adapter, &extfat_cmd);
+ if (!status) {
+ cfgs = (struct be_fat_conf_params *)(extfat_cmd.va +
+ sizeof(struct be_cmd_resp_hdr));
+ for (i = 0; i < cfgs->num_modules; i++) {
+ for (j = 0; j < cfgs->module[i].num_modes; j++) {
+ if (cfgs->module[i].trace_lvl[j].mode ==
+ MODE_UART)
+ cfgs->module[i].trace_lvl[j].dbg_lvl =
+ cpu_to_le32(level);
+ }
+ }
+ status = be_cmd_set_ext_fat_capabilites(adapter, &extfat_cmd,
+ cfgs);
+ if (status)
+ dev_err(&adapter->pdev->dev,
+ "Message level set failed\n");
+ } else {
+ dev_err(&adapter->pdev->dev, "Message level get failed\n");
+ }
+
+ pci_free_consistent(adapter->pdev, extfat_cmd.size, extfat_cmd.va,
+ extfat_cmd.dma);
+err:
+ return;
+}
+
+static void be_set_msg_level(struct net_device *netdev, u32 level)
+{
+ struct be_adapter *adapter = netdev_priv(netdev);
+
+ if (lancer_chip(adapter)) {
+ dev_err(&adapter->pdev->dev, "Operation not supported\n");
+ return;
+ }
+
+ if (adapter->msg_enable == level)
+ return;
+
+ if ((level & NETIF_MSG_HW) != (adapter->msg_enable & NETIF_MSG_HW))
+ be_set_fw_log_level(adapter, level & NETIF_MSG_HW ?
+ FW_LOG_LEVEL_DEFAULT : FW_LOG_LEVEL_FATAL);
+ adapter->msg_enable = level;
+
+ return;
+}
+
const struct ethtool_ops be_ethtool_ops = {
.get_settings = be_get_settings,
.get_drvinfo = be_get_drvinfo,
@@ -893,6 +968,8 @@ const struct ethtool_ops be_ethtool_ops = {
.set_pauseparam = be_set_pauseparam,
.get_strings = be_get_stat_strings,
.set_phys_id = be_set_phys_id,
+ .get_msglevel = be_get_msg_level,
+ .set_msglevel = be_set_msg_level,
.get_sset_count = be_get_sset_count,
.get_ethtool_stats = be_get_ethtool_stats,
.get_regs_len = be_get_reg_len,
diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c
index a01f734..081c777 100644
--- a/drivers/net/ethernet/emulex/benet/be_main.c
+++ b/drivers/net/ethernet/emulex/benet/be_main.c
@@ -3403,9 +3403,43 @@ bool be_is_wol_supported(struct be_adapter *adapter)
!be_is_wol_excluded(adapter)) ? true : false;
}
+u32 be_get_fw_log_level(struct be_adapter *adapter)
+{
+ struct be_dma_mem extfat_cmd;
+ struct be_fat_conf_params *cfgs;
+ int status;
+ u32 level = 0;
+ int j;
+
+ memset(&extfat_cmd, 0, sizeof(struct be_dma_mem));
+ extfat_cmd.size = sizeof(struct be_cmd_resp_get_ext_fat_caps);
+ extfat_cmd.va = pci_alloc_consistent(adapter->pdev, extfat_cmd.size,
+ &extfat_cmd.dma);
+
+ if (!extfat_cmd.va) {
+ dev_err(&adapter->pdev->dev, "%s: Memory allocation failure\n",
+ __func__);
+ goto err;
+ }
+
+ status = be_cmd_get_ext_fat_capabilites(adapter, &extfat_cmd);
+ if (!status) {
+ cfgs = (struct be_fat_conf_params *)(extfat_cmd.va +
+ sizeof(struct be_cmd_resp_hdr));
+ for (j = 0; j < cfgs->module[0].num_modes; j++) {
+ if (cfgs->module[0].trace_lvl[j].mode == MODE_UART)
+ level = cfgs->module[0].trace_lvl[j].dbg_lvl;
+ }
+ }
+ pci_free_consistent(adapter->pdev, extfat_cmd.size, extfat_cmd.va,
+ extfat_cmd.dma);
+err:
+ return level;
+}
static int be_get_initial_config(struct be_adapter *adapter)
{
int status;
+ u32 level;
status = be_cmd_query_fw_cfg(adapter, &adapter->port_num,
&adapter->function_mode, &adapter->function_caps);
@@ -3443,6 +3477,9 @@ static int be_get_initial_config(struct be_adapter *adapter)
if (be_is_wol_supported(adapter))
adapter->wol = true;
+ level = be_get_fw_log_level(adapter);
+ adapter->msg_enable = level <= FW_LOG_LEVEL_DEFAULT ? NETIF_MSG_HW : 0;
+
return 0;
}
--
1.5.6.1
^ permalink raw reply related
* [v2 PATCH] netfilter: xt_HMARK: fix endian bugs and warnings
From: Hans Schillstrom @ 2012-05-18 8:35 UTC (permalink / raw)
To: pablo, kaber, jengelh, netfilter-devel, netdev, dan.carpenter
Cc: hans.schillstrom, Hans Schillstrom
A mix of u32 and __be32 causes endian warning.
The hash value produced is now the same for big and little endian machines.
i.e. a mix of Big and Little endian in a cluster is now possible.
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Hans Schillstrom <hans@schillstrom.com>
---
include/linux/netfilter/xt_HMARK.h | 5 +++
net/netfilter/xt_HMARK.c | 69 ++++++++++++++++++++----------------
2 files changed, 43 insertions(+), 31 deletions(-)
diff --git a/include/linux/netfilter/xt_HMARK.h b/include/linux/netfilter/xt_HMARK.h
index abb1650..826fc58 100644
--- a/include/linux/netfilter/xt_HMARK.h
+++ b/include/linux/netfilter/xt_HMARK.h
@@ -27,7 +27,12 @@ union hmark_ports {
__u16 src;
__u16 dst;
} p16;
+ struct {
+ __be16 src;
+ __be16 dst;
+ } b16;
__u32 v32;
+ __be32 b32;
};
struct xt_hmark_info {
diff --git a/net/netfilter/xt_HMARK.c b/net/netfilter/xt_HMARK.c
index 0a96a43..5119666 100644
--- a/net/netfilter/xt_HMARK.c
+++ b/net/netfilter/xt_HMARK.c
@@ -32,13 +32,13 @@ MODULE_ALIAS("ipt_HMARK");
MODULE_ALIAS("ip6t_HMARK");
struct hmark_tuple {
- u32 src;
- u32 dst;
+ __be32 src;
+ __be32 dst;
union hmark_ports uports;
uint8_t proto;
};
-static inline u32 hmark_addr6_mask(const __u32 *addr32, const __u32 *mask)
+static inline __be32 hmark_addr6_mask(const __be32 *addr32, const __be32 *mask)
{
return (addr32[0] & mask[0]) ^
(addr32[1] & mask[1]) ^
@@ -46,8 +46,8 @@ static inline u32 hmark_addr6_mask(const __u32 *addr32, const __u32 *mask)
(addr32[3] & mask[3]);
}
-static inline u32
-hmark_addr_mask(int l3num, const __u32 *addr32, const __u32 *mask)
+static inline __be32
+hmark_addr_mask(int l3num, const __be32 *addr32, const __be32 *mask)
{
switch (l3num) {
case AF_INET:
@@ -58,6 +58,22 @@ hmark_addr_mask(int l3num, const __u32 *addr32, const __u32 *mask)
return 0;
}
+static inline void hmark_swap_ports(union hmark_ports *uports,
+ const struct xt_hmark_info *info)
+{
+ union hmark_ports hp;
+ u16 src,dst;
+
+ hp.b32 = (uports->b32 & info->port_mask.b32) | info->port_set.b32;
+ src = ntohs(hp.b16.src);
+ dst = ntohs(hp.b16.dst);
+
+ if (dst > src)
+ uports->v32 = (dst << 16) | src;
+ else
+ uports->v32 = (src << 16) | dst;
+}
+
static int
hmark_ct_set_htuple(const struct sk_buff *skb, struct hmark_tuple *t,
const struct xt_hmark_info *info)
@@ -74,22 +90,19 @@ hmark_ct_set_htuple(const struct sk_buff *skb, struct hmark_tuple *t,
otuple = &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple;
rtuple = &ct->tuplehash[IP_CT_DIR_REPLY].tuple;
- t->src = hmark_addr_mask(otuple->src.l3num, otuple->src.u3.all,
- info->src_mask.all);
- t->dst = hmark_addr_mask(otuple->src.l3num, rtuple->src.u3.all,
- info->dst_mask.all);
+ t->src = hmark_addr_mask(otuple->src.l3num, otuple->src.u3.ip6,
+ info->src_mask.ip6);
+ t->dst = hmark_addr_mask(otuple->src.l3num, rtuple->src.u3.ip6,
+ info->dst_mask.ip6);
if (info->flags & XT_HMARK_FLAG(XT_HMARK_METHOD_L3))
return 0;
t->proto = nf_ct_protonum(ct);
if (t->proto != IPPROTO_ICMP) {
- t->uports.p16.src = otuple->src.u.all;
- t->uports.p16.dst = rtuple->src.u.all;
- t->uports.v32 = (t->uports.v32 & info->port_mask.v32) |
- info->port_set.v32;
- if (t->uports.p16.dst < t->uports.p16.src)
- swap(t->uports.p16.dst, t->uports.p16.src);
+ t->uports.b16.src = otuple->src.u.all;
+ t->uports.b16.dst = rtuple->src.u.all;
+ hmark_swap_ports(&t->uports, info);
}
return 0;
@@ -102,11 +115,13 @@ static inline u32
hmark_hash(struct hmark_tuple *t, const struct xt_hmark_info *info)
{
u32 hash;
+ u32 src = ntohl(t->src);
+ u32 dst = ntohl(t->dst);
- if (t->dst < t->src)
- swap(t->src, t->dst);
+ if (dst < src)
+ swap(src, dst);
- hash = jhash_3words(t->src, t->dst, t->uports.v32, info->hashrnd);
+ hash = jhash_3words(src, dst, t->uports.v32, info->hashrnd);
hash = hash ^ (t->proto & info->proto_mask);
return (((u64)hash * info->hmodulus) >> 32) + info->hoffset;
@@ -125,12 +140,7 @@ hmark_set_tuple_ports(const struct sk_buff *skb, unsigned int nhoff,
nhoff += protoff;
if (skb_copy_bits(skb, nhoff, &t->uports, sizeof(t->uports)) < 0)
return;
-
- t->uports.v32 = (t->uports.v32 & info->port_mask.v32) |
- info->port_set.v32;
-
- if (t->uports.p16.dst < t->uports.p16.src)
- swap(t->uports.p16.dst, t->uports.p16.src);
+ hmark_swap_ports(&t->uports, info);
}
#if IS_ENABLED(CONFIG_IP6_NF_IPTABLES)
@@ -178,8 +188,8 @@ hmark_pkt_set_htuple_ipv6(const struct sk_buff *skb, struct hmark_tuple *t,
return -1;
}
noicmp:
- t->src = hmark_addr6_mask(ip6->saddr.s6_addr32, info->src_mask.all);
- t->dst = hmark_addr6_mask(ip6->daddr.s6_addr32, info->dst_mask.all);
+ t->src = hmark_addr6_mask(ip6->saddr.s6_addr32, info->src_mask.ip6);
+ t->dst = hmark_addr6_mask(ip6->daddr.s6_addr32, info->dst_mask.ip6);
if (info->flags & XT_HMARK_FLAG(XT_HMARK_METHOD_L3))
return 0;
@@ -255,11 +265,8 @@ hmark_pkt_set_htuple_ipv4(const struct sk_buff *skb, struct hmark_tuple *t,
}
}
- t->src = (__force u32) ip->saddr;
- t->dst = (__force u32) ip->daddr;
-
- t->src &= info->src_mask.ip;
- t->dst &= info->dst_mask.ip;
+ t->src = ip->saddr & info->src_mask.ip;
+ t->dst = ip->daddr & info->dst_mask.ip;
if (info->flags & XT_HMARK_FLAG(XT_HMARK_METHOD_L3))
return 0;
--
1.7.2.3
^ permalink raw reply related
* Re: [PATCH net-next] ipv6: bool conversions phase1
From: David Miller @ 2012-05-18 6:24 UTC (permalink / raw)
To: eric.dumazet; +Cc: netdev
In-Reply-To: <1337321651.3403.92.camel@edumazet-glaptop>
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Fri, 18 May 2012 08:14:11 +0200
> From: Eric Dumazet <edumazet@google.com>
>
> ipv6_opt_accepted() returns a bool, and can use const pointers
>
> ipv6_addr_equal(), ipv6_addr_any(), ipv6_addr_loopback(),
> ipv6_addr_orchid() return a bool.
>
> Signed-off-by: Eric Dumazet <edumazet@google.com>
Applied, thanks Eric.
^ permalink raw reply
* [PATCH net-next] ipv6: bool conversions phase1
From: Eric Dumazet @ 2012-05-18 6:14 UTC (permalink / raw)
To: David Miller; +Cc: netdev
From: Eric Dumazet <edumazet@google.com>
ipv6_opt_accepted() returns a bool, and can use const pointers
ipv6_addr_equal(), ipv6_addr_any(), ipv6_addr_loopback(),
ipv6_addr_orchid() return a bool.
Signed-off-by: Eric Dumazet <edumazet@google.com>
---
include/net/ipv6.h | 30 +++++++++++++++---------------
net/ipv6/af_inet6.c | 10 +++++-----
net/ipv6/ip6_tunnel.c | 2 +-
3 files changed, 21 insertions(+), 21 deletions(-)
diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index 5f65e7c..1402139 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -263,7 +263,7 @@ extern struct ipv6_txoptions * ipv6_renew_options(struct sock *sk, struct ipv6_t
struct ipv6_txoptions *ipv6_fixup_options(struct ipv6_txoptions *opt_space,
struct ipv6_txoptions *opt);
-extern int ipv6_opt_accepted(struct sock *sk, struct sk_buff *skb);
+extern bool ipv6_opt_accepted(const struct sock *sk, const struct sk_buff *skb);
int ip6_frag_nqueues(struct net *net);
int ip6_frag_mem(struct net *net);
@@ -332,8 +332,8 @@ static inline void ipv6_addr_set(struct in6_addr *addr,
addr->s6_addr32[3] = w4;
}
-static inline int ipv6_addr_equal(const struct in6_addr *a1,
- const struct in6_addr *a2)
+static inline bool ipv6_addr_equal(const struct in6_addr *a1,
+ const struct in6_addr *a2)
{
return ((a1->s6_addr32[0] ^ a2->s6_addr32[0]) |
(a1->s6_addr32[1] ^ a2->s6_addr32[1]) |
@@ -341,27 +341,27 @@ static inline int ipv6_addr_equal(const struct in6_addr *a1,
(a1->s6_addr32[3] ^ a2->s6_addr32[3])) == 0;
}
-static inline int __ipv6_prefix_equal(const __be32 *a1, const __be32 *a2,
- unsigned int prefixlen)
+static inline bool __ipv6_prefix_equal(const __be32 *a1, const __be32 *a2,
+ unsigned int prefixlen)
{
unsigned int pdw, pbi;
/* check complete u32 in prefix */
pdw = prefixlen >> 5;
if (pdw && memcmp(a1, a2, pdw << 2))
- return 0;
+ return false;
/* check incomplete u32 in prefix */
pbi = prefixlen & 0x1f;
if (pbi && ((a1[pdw] ^ a2[pdw]) & htonl((0xffffffff) << (32 - pbi))))
- return 0;
+ return false;
- return 1;
+ return true;
}
-static inline int ipv6_prefix_equal(const struct in6_addr *a1,
- const struct in6_addr *a2,
- unsigned int prefixlen)
+static inline bool ipv6_prefix_equal(const struct in6_addr *a1,
+ const struct in6_addr *a2,
+ unsigned int prefixlen)
{
return __ipv6_prefix_equal(a1->s6_addr32, a2->s6_addr32,
prefixlen);
@@ -389,19 +389,19 @@ struct ip6_create_arg {
void ip6_frag_init(struct inet_frag_queue *q, void *a);
bool ip6_frag_match(struct inet_frag_queue *q, void *a);
-static inline int ipv6_addr_any(const struct in6_addr *a)
+static inline bool ipv6_addr_any(const struct in6_addr *a)
{
return (a->s6_addr32[0] | a->s6_addr32[1] |
a->s6_addr32[2] | a->s6_addr32[3]) == 0;
}
-static inline int ipv6_addr_loopback(const struct in6_addr *a)
+static inline bool ipv6_addr_loopback(const struct in6_addr *a)
{
return (a->s6_addr32[0] | a->s6_addr32[1] |
a->s6_addr32[2] | (a->s6_addr32[3] ^ htonl(1))) == 0;
}
-static inline int ipv6_addr_v4mapped(const struct in6_addr *a)
+static inline bool ipv6_addr_v4mapped(const struct in6_addr *a)
{
return (a->s6_addr32[0] | a->s6_addr32[1] |
(a->s6_addr32[2] ^ htonl(0x0000ffff))) == 0;
@@ -411,7 +411,7 @@ static inline int ipv6_addr_v4mapped(const struct in6_addr *a)
* Check for a RFC 4843 ORCHID address
* (Overlay Routable Cryptographic Hash Identifiers)
*/
-static inline int ipv6_addr_orchid(const struct in6_addr *a)
+static inline bool ipv6_addr_orchid(const struct in6_addr *a)
{
return (a->s6_addr32[0] & htonl(0xfffffff0)) == htonl(0x20010010);
}
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 138d498..e22e6d8 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -678,10 +678,10 @@ int inet6_sk_rebuild_header(struct sock *sk)
}
EXPORT_SYMBOL_GPL(inet6_sk_rebuild_header);
-int ipv6_opt_accepted(struct sock *sk, struct sk_buff *skb)
+bool ipv6_opt_accepted(const struct sock *sk, const struct sk_buff *skb)
{
- struct ipv6_pinfo *np = inet6_sk(sk);
- struct inet6_skb_parm *opt = IP6CB(skb);
+ const struct ipv6_pinfo *np = inet6_sk(sk);
+ const struct inet6_skb_parm *opt = IP6CB(skb);
if (np->rxopt.all) {
if ((opt->hop && (np->rxopt.bits.hopopts ||
@@ -693,9 +693,9 @@ int ipv6_opt_accepted(struct sock *sk, struct sk_buff *skb)
np->rxopt.bits.osrcrt)) ||
((opt->dst1 || opt->dst0) &&
(np->rxopt.bits.dstopts || np->rxopt.bits.odstopts)))
- return 1;
+ return true;
}
- return 0;
+ return false;
}
EXPORT_SYMBOL_GPL(ipv6_opt_accepted);
diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c
index e65c560..c9015fa 100644
--- a/net/ipv6/ip6_tunnel.c
+++ b/net/ipv6/ip6_tunnel.c
@@ -818,7 +818,7 @@ static void init_tel_txopt(struct ipv6_tel_txoption *opt, __u8 encap_limit)
* 0 else
**/
-static inline int
+static inline bool
ip6_tnl_addr_conflict(const struct ip6_tnl *t, const struct ipv6hdr *hdr)
{
return ipv6_addr_equal(&t->parms.raddr, &hdr->saddr);
^ permalink raw reply related
* Re: [PATCH] STA2X11 CAN: CAN driver for the STA2X11 board
From: Wolfgang Grandegger @ 2012-05-18 6:00 UTC (permalink / raw)
To: Federico Vaga
Cc: Marc Kleine-Budde, linux-can, netdev, linux-kernel,
Giancarlo Asnaghi, Alan Cox
In-Reply-To: <1337288364-17572-1-git-send-email-federico.vaga@gmail.com>
On 05/17/2012 10:59 PM, Federico Vaga wrote:
> Signed-off-by: Federico Vaga <federico.vaga@gmail.com>
> Acked-by: Giancarlo Asnaghi <giancarlo.asnaghi@st.com>
> Cc: Alan Cox <alan@linux.intel.com>
> ---
> drivers/net/can/Kconfig | 11 +
> drivers/net/can/Makefile | 1 +
> drivers/net/can/sta2x11_can.c | 1085 +++++++++++++++++++++++++++++++++++++++++
> 3 files changed, 1097 insertions(+), 0 deletions(-)
> create mode 100644 drivers/net/can/sta2x11_can.c
Thanks for your contribution. At a first glance, this driver looks
similar to the pch_can and the c_can driver. It seems that a C_CAN based
controller is used on that board as well. If that's true, it should be
handled by the C_CAN driver. To get ride of the obsolete pch_can driver,
I sent some time ago the patch "[RFC/PATCH] c_can: add driver for the
PCH CAN controller":
http://marc.info/?t=132991563600003&r=1&w=4
I could serve as base of a generic c_can_pci driver.
Wolfgang.
^ permalink raw reply
* Re: [RFC] econet: remove ancient bug ridden protocol
From: David Miller @ 2012-05-18 5:41 UTC (permalink / raw)
To: remi; +Cc: shemminger, netdev
In-Reply-To: <201205180831.47820.remi@remlab.net>
From: "Rémi Denis-Courmont" <remi@remlab.net>
Date: Fri, 18 May 2012 08:31:47 +0300
> Shouldn't this patch remove ec_ptr from struct net_device?
I took care of it with a follow-on patch.
^ permalink raw reply
* Re: [PATCH net-next] ip_frag: struct inet_frags match() method returns a bool
From: David Miller @ 2012-05-18 5:40 UTC (permalink / raw)
To: eric.dumazet; +Cc: netdev
In-Reply-To: <1337313433.3403.73.camel@edumazet-glaptop>
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Fri, 18 May 2012 05:57:13 +0200
> From: Eric Dumazet <edumazet@google.com>
>
> - match() method returns a boolean
> - return (A && B && C && D) -> return A && B && C && D
> - fix indentation
>
> Signed-off-by: Eric Dumazet <edumazet@google.com>
Applied, thanks Eric.
^ permalink raw reply
* [PATCH] net: Remove netdevice ec_ptr, no longer used.
From: David Miller @ 2012-05-18 5:40 UTC (permalink / raw)
To: netdev
ECONET is gone, thus this can be deleted as well.
Signed-off-by: David S. Miller <davem@davemloft.net>
---
include/linux/netdevice.h | 1 -
1 file changed, 1 deletion(-)
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index b0f6f22..9c710bd 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1155,7 +1155,6 @@ struct net_device {
struct in_device __rcu *ip_ptr; /* IPv4 specific data */
struct dn_dev __rcu *dn_ptr; /* DECnet specific data */
struct inet6_dev __rcu *ip6_ptr; /* IPv6 specific data */
- void *ec_ptr; /* Econet specific data */
void *ax25_ptr; /* AX.25 specific data */
struct wireless_dev *ieee80211_ptr; /* IEEE 802.11 specific data,
assign before registering */
--
1.7.10.1
^ permalink raw reply related
* Re: [RFC] econet: remove ancient bug ridden protocol
From: David Miller @ 2012-05-18 5:35 UTC (permalink / raw)
To: mjg59
Cc: shemminger, paul.gortmaker, arnd, john.calixto, xemul,
linux-kernel, netdev
In-Reply-To: <20120518042301.GA723@srcf.ucam.org>
From: Matthew Garrett <mjg59@srcf.ucam.org>
Date: Fri, 18 May 2012 05:23:01 +0100
> On Thu, May 17, 2012 at 08:59:51PM -0700, Stephen Hemminger wrote:
>
>> The ancient Econet protocol should go. Most of the bug fixes in recent
>> years have been fixing security vulnerabilities. The hardware hasn't
>> been made since 1981, it is only interesting as an archeological curiosity.
>
> Pedantry - the hardware was introduced in 1981, it was still being made
> into at least the 90s. Still probably time for it to die, though.
Patch applied with this pedantic adjustment :-)
^ permalink raw reply
* Re: [RFC] econet: remove ancient bug ridden protocol
From: Rémi Denis-Courmont @ 2012-05-18 5:31 UTC (permalink / raw)
To: Stephen Hemminger, netdev
In-Reply-To: <20120517205951.462d336f@nehalam.linuxnetplumber.net>
Hello,
Shouldn't this patch remove ec_ptr from struct net_device?
--
Rémi Denis-Courmont
http://www.remlab.info
^ permalink raw reply
* Re: [PATCH net-next] net: netdev_alloc_skb() use build_skb()
From: Eric Dumazet @ 2012-05-18 4:41 UTC (permalink / raw)
To: David Miller; +Cc: w, netdev
In-Reply-To: <20120517.155318.1557653193070030300.davem@davemloft.net>
On Thu, 2012-05-17 at 15:53 -0400, David Miller wrote:
> Applied, we can sort out any fallout very easily before 3.5 is released.
>
> Awesome work Eric.
Thanks David
I'll send a followup patch to fix issues :
- Must be IRQ safe (non NAPI drivers can use it)
- Must not leak the frag is build_skb() fails to allocate sk_buff
- Factorize code so that dev_alloc_skb()/__dev_alloc_skb() are a wrapper
around __netdev_alloc_skb()
^ permalink raw reply
* Re: [RFC] econet: remove ancient bug ridden protocol
From: Matthew Garrett @ 2012-05-18 4:23 UTC (permalink / raw)
To: Stephen Hemminger
Cc: David Miller, Paul Gortmaker, Arnd Bergmann, John Calixto,
Pavel Emelyanov, linux-kernel, netdev
In-Reply-To: <20120517205951.462d336f@nehalam.linuxnetplumber.net>
On Thu, May 17, 2012 at 08:59:51PM -0700, Stephen Hemminger wrote:
> The ancient Econet protocol should go. Most of the bug fixes in recent
> years have been fixing security vulnerabilities. The hardware hasn't
> been made since 1981, it is only interesting as an archeological curiosity.
Pedantry - the hardware was introduced in 1981, it was still being made
into at least the 90s. Still probably time for it to die, though.
--
Matthew Garrett | mjg59@srcf.ucam.org
^ permalink raw reply
* [RFC] econet: remove ancient bug ridden protocol
From: Stephen Hemminger @ 2012-05-18 3:59 UTC (permalink / raw)
To: David Miller, Paul Gortmaker
Cc: Arnd Bergmann, John Calixto, Pavel Emelyanov, linux-kernel,
netdev
More spring cleaning!
The ancient Econet protocol should go. Most of the bug fixes in recent
years have been fixing security vulnerabilities. The hardware hasn't
been made since 1981, it is only interesting as an archeological curiosity.
For the truly curious, or insomniac, go read up on it.
http://en.wikipedia.org/wiki/Econet
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
--- a/net/Kconfig 2012-05-17 14:27:38.898050909 -0700
+++ b/net/Kconfig 2012-05-17 20:51:42.199027402 -0700
@@ -207,7 +207,6 @@ source "net/ipx/Kconfig"
source "drivers/net/appletalk/Kconfig"
source "net/x25/Kconfig"
source "net/lapb/Kconfig"
-source "net/econet/Kconfig"
source "net/wanrouter/Kconfig"
source "net/phonet/Kconfig"
source "net/ieee802154/Kconfig"
--- a/net/econet/Kconfig 2012-02-13 09:23:59.142512298 -0800
+++ /dev/null 1970-01-01 00:00:00.000000000 +0000
@@ -1,36 +0,0 @@
-#
-# Acorn Econet/AUN protocols
-#
-
-config ECONET
- tristate "Acorn Econet/AUN protocols (EXPERIMENTAL)"
- depends on EXPERIMENTAL && INET
- ---help---
- Econet is a fairly old and slow networking protocol mainly used by
- Acorn computers to access file and print servers. It uses native
- Econet network cards. AUN is an implementation of the higher level
- parts of Econet that runs over ordinary Ethernet connections, on
- top of the UDP packet protocol, which in turn runs on top of the
- Internet protocol IP.
-
- If you say Y here, you can choose with the next two options whether
- to send Econet/AUN traffic over a UDP Ethernet connection or over
- a native Econet network card.
-
- To compile this driver as a module, choose M here: the module
- will be called econet.
-
-config ECONET_AUNUDP
- bool "AUN over UDP"
- depends on ECONET
- help
- Say Y here if you want to send Econet/AUN traffic over a UDP
- connection (UDP is a packet based protocol that runs on top of the
- Internet protocol IP) using an ordinary Ethernet network card.
-
-config ECONET_NATIVE
- bool "Native Econet"
- depends on ECONET
- help
- Say Y here if you have a native Econet network card installed in
- your computer.
--- a/net/econet/Makefile 2012-02-13 09:23:59.142512298 -0800
+++ /dev/null 1970-01-01 00:00:00.000000000 +0000
@@ -1,7 +0,0 @@
-#
-# Makefile for Econet support code.
-#
-
-obj-$(CONFIG_ECONET) += econet.o
-
-econet-y := af_econet.o
--- a/net/econet/af_econet.c 2012-04-25 09:29:00.115222896 -0700
+++ /dev/null 1970-01-01 00:00:00.000000000 +0000
@@ -1,1172 +0,0 @@
-/*
- * An implementation of the Acorn Econet and AUN protocols.
- * Philip Blundell <philb@gnu.org>
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License
- * as published by the Free Software Foundation; either version
- * 2 of the License, or (at your option) any later version.
- *
- */
-
-#define pr_fmt(fmt) fmt
-
-#include <linux/module.h>
-
-#include <linux/types.h>
-#include <linux/kernel.h>
-#include <linux/string.h>
-#include <linux/mm.h>
-#include <linux/socket.h>
-#include <linux/sockios.h>
-#include <linux/in.h>
-#include <linux/errno.h>
-#include <linux/interrupt.h>
-#include <linux/if_ether.h>
-#include <linux/netdevice.h>
-#include <linux/inetdevice.h>
-#include <linux/route.h>
-#include <linux/inet.h>
-#include <linux/etherdevice.h>
-#include <linux/if_arp.h>
-#include <linux/wireless.h>
-#include <linux/skbuff.h>
-#include <linux/udp.h>
-#include <linux/slab.h>
-#include <linux/vmalloc.h>
-#include <net/sock.h>
-#include <net/inet_common.h>
-#include <linux/stat.h>
-#include <linux/init.h>
-#include <linux/if_ec.h>
-#include <net/udp.h>
-#include <net/ip.h>
-#include <linux/spinlock.h>
-#include <linux/rcupdate.h>
-#include <linux/bitops.h>
-#include <linux/mutex.h>
-
-#include <linux/uaccess.h>
-
-static const struct proto_ops econet_ops;
-static struct hlist_head econet_sklist;
-static DEFINE_SPINLOCK(econet_lock);
-static DEFINE_MUTEX(econet_mutex);
-
-/* Since there are only 256 possible network numbers (or fewer, depends
- how you count) it makes sense to use a simple lookup table. */
-static struct net_device *net2dev_map[256];
-
-#define EC_PORT_IP 0xd2
-
-#ifdef CONFIG_ECONET_AUNUDP
-static DEFINE_SPINLOCK(aun_queue_lock);
-static struct socket *udpsock;
-#define AUN_PORT 0x8000
-
-struct aunhdr {
- unsigned char code; /* AUN magic protocol byte */
- unsigned char port;
- unsigned char cb;
- unsigned char pad;
- unsigned long handle;
-};
-
-static unsigned long aun_seq;
-
-/* Queue of packets waiting to be transmitted. */
-static struct sk_buff_head aun_queue;
-static struct timer_list ab_cleanup_timer;
-
-#endif /* CONFIG_ECONET_AUNUDP */
-
-/* Per-packet information */
-struct ec_cb {
- struct sockaddr_ec sec;
- unsigned long cookie; /* Supplied by user. */
-#ifdef CONFIG_ECONET_AUNUDP
- int done;
- unsigned long seq; /* Sequencing */
- unsigned long timeout; /* Timeout */
- unsigned long start; /* jiffies */
-#endif
-#ifdef CONFIG_ECONET_NATIVE
- void (*sent)(struct sk_buff *, int result);
-#endif
-};
-
-static void econet_remove_socket(struct hlist_head *list, struct sock *sk)
-{
- spin_lock_bh(&econet_lock);
- sk_del_node_init(sk);
- spin_unlock_bh(&econet_lock);
-}
-
-static void econet_insert_socket(struct hlist_head *list, struct sock *sk)
-{
- spin_lock_bh(&econet_lock);
- sk_add_node(sk, list);
- spin_unlock_bh(&econet_lock);
-}
-
-/*
- * Pull a packet from our receive queue and hand it to the user.
- * If necessary we block.
- */
-
-static int econet_recvmsg(struct kiocb *iocb, struct socket *sock,
- struct msghdr *msg, size_t len, int flags)
-{
- struct sock *sk = sock->sk;
- struct sk_buff *skb;
- size_t copied;
- int err;
-
- msg->msg_namelen = sizeof(struct sockaddr_ec);
-
- mutex_lock(&econet_mutex);
-
- /*
- * Call the generic datagram receiver. This handles all sorts
- * of horrible races and re-entrancy so we can forget about it
- * in the protocol layers.
- *
- * Now it will return ENETDOWN, if device have just gone down,
- * but then it will block.
- */
-
- skb = skb_recv_datagram(sk, flags, flags & MSG_DONTWAIT, &err);
-
- /*
- * An error occurred so return it. Because skb_recv_datagram()
- * handles the blocking we don't see and worry about blocking
- * retries.
- */
-
- if (skb == NULL)
- goto out;
-
- /*
- * You lose any data beyond the buffer you gave. If it worries a
- * user program they can ask the device for its MTU anyway.
- */
-
- copied = skb->len;
- if (copied > len) {
- copied = len;
- msg->msg_flags |= MSG_TRUNC;
- }
-
- /* We can't use skb_copy_datagram here */
- err = memcpy_toiovec(msg->msg_iov, skb->data, copied);
- if (err)
- goto out_free;
- sk->sk_stamp = skb->tstamp;
-
- if (msg->msg_name)
- memcpy(msg->msg_name, skb->cb, msg->msg_namelen);
-
- /*
- * Free or return the buffer as appropriate. Again this
- * hides all the races and re-entrancy issues from us.
- */
- err = copied;
-
-out_free:
- skb_free_datagram(sk, skb);
-out:
- mutex_unlock(&econet_mutex);
- return err;
-}
-
-/*
- * Bind an Econet socket.
- */
-
-static int econet_bind(struct socket *sock, struct sockaddr *uaddr,
- int addr_len)
-{
- struct sockaddr_ec *sec = (struct sockaddr_ec *)uaddr;
- struct sock *sk;
- struct econet_sock *eo;
-
- /*
- * Check legality
- */
-
- if (addr_len < sizeof(struct sockaddr_ec) ||
- sec->sec_family != AF_ECONET)
- return -EINVAL;
-
- mutex_lock(&econet_mutex);
-
- sk = sock->sk;
- eo = ec_sk(sk);
-
- eo->cb = sec->cb;
- eo->port = sec->port;
- eo->station = sec->addr.station;
- eo->net = sec->addr.net;
-
- mutex_unlock(&econet_mutex);
-
- return 0;
-}
-
-#if defined(CONFIG_ECONET_AUNUDP) || defined(CONFIG_ECONET_NATIVE)
-/*
- * Queue a transmit result for the user to be told about.
- */
-
-static void tx_result(struct sock *sk, unsigned long cookie, int result)
-{
- struct sk_buff *skb = alloc_skb(0, GFP_ATOMIC);
- struct ec_cb *eb;
- struct sockaddr_ec *sec;
-
- if (skb == NULL) {
- pr_debug("econet: memory squeeze, transmit result dropped\n");
- return;
- }
-
- eb = (struct ec_cb *)&skb->cb;
- sec = (struct sockaddr_ec *)&eb->sec;
- memset(sec, 0, sizeof(struct sockaddr_ec));
- sec->cookie = cookie;
- sec->type = ECTYPE_TRANSMIT_STATUS | result;
- sec->sec_family = AF_ECONET;
-
- if (sock_queue_rcv_skb(sk, skb) < 0)
- kfree_skb(skb);
-}
-#endif
-
-#ifdef CONFIG_ECONET_NATIVE
-/*
- * Called by the Econet hardware driver when a packet transmit
- * has completed. Tell the user.
- */
-
-static void ec_tx_done(struct sk_buff *skb, int result)
-{
- struct ec_cb *eb = (struct ec_cb *)&skb->cb;
- tx_result(skb->sk, eb->cookie, result);
-}
-#endif
-
-/*
- * Send a packet. We have to work out which device it's going out on
- * and hence whether to use real Econet or the UDP emulation.
- */
-
-static int econet_sendmsg(struct kiocb *iocb, struct socket *sock,
- struct msghdr *msg, size_t len)
-{
- struct sockaddr_ec *saddr = (struct sockaddr_ec *)msg->msg_name;
- struct net_device *dev;
- struct ec_addr addr;
- int err;
- unsigned char port, cb;
-#if defined(CONFIG_ECONET_AUNUDP) || defined(CONFIG_ECONET_NATIVE)
- struct sock *sk = sock->sk;
- struct sk_buff *skb;
- struct ec_cb *eb;
-#endif
-#ifdef CONFIG_ECONET_AUNUDP
- struct msghdr udpmsg;
- struct iovec iov[2];
- struct aunhdr ah;
- struct sockaddr_in udpdest;
- __kernel_size_t size;
- mm_segment_t oldfs;
- char *userbuf;
-#endif
-
- /*
- * Check the flags.
- */
-
- if (msg->msg_flags & ~(MSG_DONTWAIT|MSG_CMSG_COMPAT))
- return -EINVAL;
-
- /*
- * Get and verify the address.
- */
-
- mutex_lock(&econet_mutex);
-
- if (saddr == NULL || msg->msg_namelen < sizeof(struct sockaddr_ec)) {
- mutex_unlock(&econet_mutex);
- return -EINVAL;
- }
- addr.station = saddr->addr.station;
- addr.net = saddr->addr.net;
- port = saddr->port;
- cb = saddr->cb;
-
- /* Look for a device with the right network number. */
- dev = net2dev_map[addr.net];
-
- /* If not directly reachable, use some default */
- if (dev == NULL) {
- dev = net2dev_map[0];
- /* No interfaces at all? */
- if (dev == NULL) {
- mutex_unlock(&econet_mutex);
- return -ENETDOWN;
- }
- }
-
- if (dev->type == ARPHRD_ECONET) {
- /* Real hardware Econet. We're not worthy etc. */
-#ifdef CONFIG_ECONET_NATIVE
- unsigned short proto = 0;
- int hlen, tlen;
- int res;
-
- if (len + 15 > dev->mtu) {
- mutex_unlock(&econet_mutex);
- return -EMSGSIZE;
- }
-
- dev_hold(dev);
-
- hlen = LL_RESERVED_SPACE(dev);
- tlen = dev->needed_tailroom;
- skb = sock_alloc_send_skb(sk, len + hlen + tlen,
- msg->msg_flags & MSG_DONTWAIT, &err);
- if (skb == NULL)
- goto out_unlock;
-
- skb_reserve(skb, hlen);
- skb_reset_network_header(skb);
-
- eb = (struct ec_cb *)&skb->cb;
-
- eb->cookie = saddr->cookie;
- eb->sec = *saddr;
- eb->sent = ec_tx_done;
-
- err = -EINVAL;
- res = dev_hard_header(skb, dev, ntohs(proto), &addr, NULL, len);
- if (res < 0)
- goto out_free;
- if (res > 0) {
- struct ec_framehdr *fh;
- /* Poke in our control byte and
- port number. Hack, hack. */
- fh = (struct ec_framehdr *)skb->data;
- fh->cb = cb;
- fh->port = port;
- if (sock->type != SOCK_DGRAM) {
- skb_reset_tail_pointer(skb);
- skb->len = 0;
- }
- }
-
- /* Copy the data. Returns -EFAULT on error */
- err = memcpy_fromiovec(skb_put(skb, len), msg->msg_iov, len);
- skb->protocol = proto;
- skb->dev = dev;
- skb->priority = sk->sk_priority;
- if (err)
- goto out_free;
-
- err = -ENETDOWN;
- if (!(dev->flags & IFF_UP))
- goto out_free;
-
- /*
- * Now send it
- */
-
- dev_queue_xmit(skb);
- dev_put(dev);
- mutex_unlock(&econet_mutex);
- return len;
-
-out_free:
- kfree_skb(skb);
-out_unlock:
- if (dev)
- dev_put(dev);
-#else
- err = -EPROTOTYPE;
-#endif
- mutex_unlock(&econet_mutex);
-
- return err;
- }
-
-#ifdef CONFIG_ECONET_AUNUDP
- /* AUN virtual Econet. */
-
- if (udpsock == NULL) {
- mutex_unlock(&econet_mutex);
- return -ENETDOWN; /* No socket - can't send */
- }
-
- if (len > 32768) {
- err = -E2BIG;
- goto error;
- }
-
- /* Make up a UDP datagram and hand it off to some higher intellect. */
-
- memset(&udpdest, 0, sizeof(udpdest));
- udpdest.sin_family = AF_INET;
- udpdest.sin_port = htons(AUN_PORT);
-
- /* At the moment we use the stupid Acorn scheme of Econet address
- y.x maps to IP a.b.c.x. This should be replaced with something
- more flexible and more aware of subnet masks. */
- {
- struct in_device *idev;
- unsigned long network = 0;
-
- rcu_read_lock();
- idev = __in_dev_get_rcu(dev);
- if (idev) {
- if (idev->ifa_list)
- network = ntohl(idev->ifa_list->ifa_address) &
- 0xffffff00; /* !!! */
- }
- rcu_read_unlock();
- udpdest.sin_addr.s_addr = htonl(network | addr.station);
- }
-
- memset(&ah, 0, sizeof(ah));
- ah.port = port;
- ah.cb = cb & 0x7f;
- ah.code = 2; /* magic */
-
- /* tack our header on the front of the iovec */
- size = sizeof(struct aunhdr);
- iov[0].iov_base = (void *)&ah;
- iov[0].iov_len = size;
-
- userbuf = vmalloc(len);
- if (userbuf == NULL) {
- err = -ENOMEM;
- goto error;
- }
-
- iov[1].iov_base = userbuf;
- iov[1].iov_len = len;
- err = memcpy_fromiovec(userbuf, msg->msg_iov, len);
- if (err)
- goto error_free_buf;
-
- /* Get a skbuff (no data, just holds our cb information) */
- skb = sock_alloc_send_skb(sk, 0, msg->msg_flags & MSG_DONTWAIT, &err);
- if (skb == NULL)
- goto error_free_buf;
-
- eb = (struct ec_cb *)&skb->cb;
-
- eb->cookie = saddr->cookie;
- eb->timeout = 5 * HZ;
- eb->start = jiffies;
- ah.handle = aun_seq;
- eb->seq = (aun_seq++);
- eb->sec = *saddr;
-
- skb_queue_tail(&aun_queue, skb);
-
- udpmsg.msg_name = (void *)&udpdest;
- udpmsg.msg_namelen = sizeof(udpdest);
- udpmsg.msg_iov = &iov[0];
- udpmsg.msg_iovlen = 2;
- udpmsg.msg_control = NULL;
- udpmsg.msg_controllen = 0;
- udpmsg.msg_flags = 0;
-
- oldfs = get_fs();
- set_fs(KERNEL_DS); /* More privs :-) */
- err = sock_sendmsg(udpsock, &udpmsg, size);
- set_fs(oldfs);
-
-error_free_buf:
- vfree(userbuf);
-error:
-#else
- err = -EPROTOTYPE;
-#endif
- mutex_unlock(&econet_mutex);
-
- return err;
-}
-
-/*
- * Look up the address of a socket.
- */
-
-static int econet_getname(struct socket *sock, struct sockaddr *uaddr,
- int *uaddr_len, int peer)
-{
- struct sock *sk;
- struct econet_sock *eo;
- struct sockaddr_ec *sec = (struct sockaddr_ec *)uaddr;
-
- if (peer)
- return -EOPNOTSUPP;
-
- memset(sec, 0, sizeof(*sec));
- mutex_lock(&econet_mutex);
-
- sk = sock->sk;
- eo = ec_sk(sk);
-
- sec->sec_family = AF_ECONET;
- sec->port = eo->port;
- sec->addr.station = eo->station;
- sec->addr.net = eo->net;
-
- mutex_unlock(&econet_mutex);
-
- *uaddr_len = sizeof(*sec);
- return 0;
-}
-
-static void econet_destroy_timer(unsigned long data)
-{
- struct sock *sk = (struct sock *)data;
-
- if (!sk_has_allocations(sk)) {
- sk_free(sk);
- return;
- }
-
- sk->sk_timer.expires = jiffies + 10 * HZ;
- add_timer(&sk->sk_timer);
- pr_debug("econet: socket destroy delayed\n");
-}
-
-/*
- * Close an econet socket.
- */
-
-static int econet_release(struct socket *sock)
-{
- struct sock *sk;
-
- mutex_lock(&econet_mutex);
-
- sk = sock->sk;
- if (!sk)
- goto out_unlock;
-
- econet_remove_socket(&econet_sklist, sk);
-
- /*
- * Now the socket is dead. No more input will appear.
- */
-
- sk->sk_state_change(sk); /* It is useless. Just for sanity. */
-
- sock_orphan(sk);
-
- /* Purge queues */
-
- skb_queue_purge(&sk->sk_receive_queue);
-
- if (sk_has_allocations(sk)) {
- sk->sk_timer.data = (unsigned long)sk;
- sk->sk_timer.expires = jiffies + HZ;
- sk->sk_timer.function = econet_destroy_timer;
- add_timer(&sk->sk_timer);
-
- goto out_unlock;
- }
-
- sk_free(sk);
-
-out_unlock:
- mutex_unlock(&econet_mutex);
- return 0;
-}
-
-static struct proto econet_proto = {
- .name = "ECONET",
- .owner = THIS_MODULE,
- .obj_size = sizeof(struct econet_sock),
-};
-
-/*
- * Create an Econet socket
- */
-
-static int econet_create(struct net *net, struct socket *sock, int protocol,
- int kern)
-{
- struct sock *sk;
- struct econet_sock *eo;
- int err;
-
- if (!net_eq(net, &init_net))
- return -EAFNOSUPPORT;
-
- /* Econet only provides datagram services. */
- if (sock->type != SOCK_DGRAM)
- return -ESOCKTNOSUPPORT;
-
- sock->state = SS_UNCONNECTED;
-
- err = -ENOBUFS;
- sk = sk_alloc(net, PF_ECONET, GFP_KERNEL, &econet_proto);
- if (sk == NULL)
- goto out;
-
- sk->sk_reuse = SK_CAN_REUSE;
- sock->ops = &econet_ops;
- sock_init_data(sock, sk);
-
- eo = ec_sk(sk);
- sock_reset_flag(sk, SOCK_ZAPPED);
- sk->sk_family = PF_ECONET;
- eo->num = protocol;
-
- econet_insert_socket(&econet_sklist, sk);
- return 0;
-out:
- return err;
-}
-
-/*
- * Handle Econet specific ioctls
- */
-
-static int ec_dev_ioctl(struct socket *sock, unsigned int cmd, void __user *arg)
-{
- struct ifreq ifr;
- struct ec_device *edev;
- struct net_device *dev;
- struct sockaddr_ec *sec;
- int err;
-
- /*
- * Fetch the caller's info block into kernel space
- */
-
- if (copy_from_user(&ifr, arg, sizeof(struct ifreq)))
- return -EFAULT;
-
- dev = dev_get_by_name(&init_net, ifr.ifr_name);
- if (dev == NULL)
- return -ENODEV;
-
- sec = (struct sockaddr_ec *)&ifr.ifr_addr;
-
- mutex_lock(&econet_mutex);
-
- err = 0;
- switch (cmd) {
- case SIOCSIFADDR:
- if (!capable(CAP_NET_ADMIN)) {
- err = -EPERM;
- break;
- }
-
- edev = dev->ec_ptr;
- if (edev == NULL) {
- /* Magic up a new one. */
- edev = kzalloc(sizeof(struct ec_device), GFP_KERNEL);
- if (edev == NULL) {
- err = -ENOMEM;
- break;
- }
- dev->ec_ptr = edev;
- } else
- net2dev_map[edev->net] = NULL;
- edev->station = sec->addr.station;
- edev->net = sec->addr.net;
- net2dev_map[sec->addr.net] = dev;
- if (!net2dev_map[0])
- net2dev_map[0] = dev;
- break;
-
- case SIOCGIFADDR:
- edev = dev->ec_ptr;
- if (edev == NULL) {
- err = -ENODEV;
- break;
- }
- memset(sec, 0, sizeof(struct sockaddr_ec));
- sec->addr.station = edev->station;
- sec->addr.net = edev->net;
- sec->sec_family = AF_ECONET;
- dev_put(dev);
- if (copy_to_user(arg, &ifr, sizeof(struct ifreq)))
- err = -EFAULT;
- break;
-
- default:
- err = -EINVAL;
- break;
- }
-
- mutex_unlock(&econet_mutex);
-
- dev_put(dev);
-
- return err;
-}
-
-/*
- * Handle generic ioctls
- */
-
-static int econet_ioctl(struct socket *sock, unsigned int cmd,
- unsigned long arg)
-{
- struct sock *sk = sock->sk;
- void __user *argp = (void __user *)arg;
-
- switch (cmd) {
- case SIOCGSTAMP:
- return sock_get_timestamp(sk, argp);
-
- case SIOCGSTAMPNS:
- return sock_get_timestampns(sk, argp);
-
- case SIOCSIFADDR:
- case SIOCGIFADDR:
- return ec_dev_ioctl(sock, cmd, argp);
-
- }
-
- return -ENOIOCTLCMD;
-}
-
-static const struct net_proto_family econet_family_ops = {
- .family = PF_ECONET,
- .create = econet_create,
- .owner = THIS_MODULE,
-};
-
-static const struct proto_ops econet_ops = {
- .family = PF_ECONET,
- .owner = THIS_MODULE,
- .release = econet_release,
- .bind = econet_bind,
- .connect = sock_no_connect,
- .socketpair = sock_no_socketpair,
- .accept = sock_no_accept,
- .getname = econet_getname,
- .poll = datagram_poll,
- .ioctl = econet_ioctl,
- .listen = sock_no_listen,
- .shutdown = sock_no_shutdown,
- .setsockopt = sock_no_setsockopt,
- .getsockopt = sock_no_getsockopt,
- .sendmsg = econet_sendmsg,
- .recvmsg = econet_recvmsg,
- .mmap = sock_no_mmap,
- .sendpage = sock_no_sendpage,
-};
-
-#if defined(CONFIG_ECONET_AUNUDP) || defined(CONFIG_ECONET_NATIVE)
-/*
- * Find the listening socket, if any, for the given data.
- */
-
-static struct sock *ec_listening_socket(unsigned char port, unsigned char
- station, unsigned char net)
-{
- struct sock *sk;
- struct hlist_node *node;
-
- spin_lock(&econet_lock);
- sk_for_each(sk, node, &econet_sklist) {
- struct econet_sock *opt = ec_sk(sk);
- if ((opt->port == port || opt->port == 0) &&
- (opt->station == station || opt->station == 0) &&
- (opt->net == net || opt->net == 0)) {
- sock_hold(sk);
- goto found;
- }
- }
- sk = NULL;
-found:
- spin_unlock(&econet_lock);
- return sk;
-}
-
-/*
- * Queue a received packet for a socket.
- */
-
-static int ec_queue_packet(struct sock *sk, struct sk_buff *skb,
- unsigned char stn, unsigned char net,
- unsigned char cb, unsigned char port)
-{
- struct ec_cb *eb = (struct ec_cb *)&skb->cb;
- struct sockaddr_ec *sec = (struct sockaddr_ec *)&eb->sec;
-
- memset(sec, 0, sizeof(struct sockaddr_ec));
- sec->sec_family = AF_ECONET;
- sec->type = ECTYPE_PACKET_RECEIVED;
- sec->port = port;
- sec->cb = cb;
- sec->addr.net = net;
- sec->addr.station = stn;
-
- return sock_queue_rcv_skb(sk, skb);
-}
-#endif
-
-#ifdef CONFIG_ECONET_AUNUDP
-/*
- * Send an AUN protocol response.
- */
-
-static void aun_send_response(__u32 addr, unsigned long seq, int code, int cb)
-{
- struct sockaddr_in sin = {
- .sin_family = AF_INET,
- .sin_port = htons(AUN_PORT),
- .sin_addr = {.s_addr = addr}
- };
- struct aunhdr ah = {.code = code, .cb = cb, .handle = seq};
- struct kvec iov = {.iov_base = (void *)&ah, .iov_len = sizeof(ah)};
- struct msghdr udpmsg;
-
- udpmsg.msg_name = (void *)&sin;
- udpmsg.msg_namelen = sizeof(sin);
- udpmsg.msg_control = NULL;
- udpmsg.msg_controllen = 0;
- udpmsg.msg_flags = 0;
-
- kernel_sendmsg(udpsock, &udpmsg, &iov, 1, sizeof(ah));
-}
-
-
-/*
- * Handle incoming AUN packets. Work out if anybody wants them,
- * and send positive or negative acknowledgements as appropriate.
- */
-
-static void aun_incoming(struct sk_buff *skb, struct aunhdr *ah, size_t len)
-{
- struct iphdr *ip = ip_hdr(skb);
- unsigned char stn = ntohl(ip->saddr) & 0xff;
- struct dst_entry *dst = skb_dst(skb);
- struct ec_device *edev = NULL;
- struct sock *sk = NULL;
- struct sk_buff *newskb;
-
- if (dst)
- edev = dst->dev->ec_ptr;
-
- if (!edev)
- goto bad;
-
- sk = ec_listening_socket(ah->port, stn, edev->net);
- if (sk == NULL)
- goto bad; /* Nobody wants it */
-
- newskb = alloc_skb((len - sizeof(struct aunhdr) + 15) & ~15,
- GFP_ATOMIC);
- if (newskb == NULL) {
- pr_debug("AUN: memory squeeze, dropping packet\n");
- /* Send nack and hope sender tries again */
- goto bad;
- }
-
- memcpy(skb_put(newskb, len - sizeof(struct aunhdr)), (void *)(ah + 1),
- len - sizeof(struct aunhdr));
-
- if (ec_queue_packet(sk, newskb, stn, edev->net, ah->cb, ah->port)) {
- /* Socket is bankrupt. */
- kfree_skb(newskb);
- goto bad;
- }
-
- aun_send_response(ip->saddr, ah->handle, 3, 0);
- sock_put(sk);
- return;
-
-bad:
- aun_send_response(ip->saddr, ah->handle, 4, 0);
- if (sk)
- sock_put(sk);
-}
-
-/*
- * Handle incoming AUN transmit acknowledgements. If the sequence
- * number matches something in our backlog then kill it and tell
- * the user. If the remote took too long to reply then we may have
- * dropped the packet already.
- */
-
-static void aun_tx_ack(unsigned long seq, int result)
-{
- struct sk_buff *skb;
- unsigned long flags;
- struct ec_cb *eb;
-
- spin_lock_irqsave(&aun_queue_lock, flags);
- skb_queue_walk(&aun_queue, skb) {
- eb = (struct ec_cb *)&skb->cb;
- if (eb->seq == seq)
- goto foundit;
- }
- spin_unlock_irqrestore(&aun_queue_lock, flags);
- pr_debug("AUN: unknown sequence %ld\n", seq);
- return;
-
-foundit:
- tx_result(skb->sk, eb->cookie, result);
- skb_unlink(skb, &aun_queue);
- spin_unlock_irqrestore(&aun_queue_lock, flags);
- kfree_skb(skb);
-}
-
-/*
- * Deal with received AUN frames - sort out what type of thing it is
- * and hand it to the right function.
- */
-
-static void aun_data_available(struct sock *sk, int slen)
-{
- int err;
- struct sk_buff *skb;
- unsigned char *data;
- struct aunhdr *ah;
- size_t len;
-
- while ((skb = skb_recv_datagram(sk, 0, 1, &err)) == NULL) {
- if (err == -EAGAIN) {
- pr_err("AUN: no data available?!\n");
- return;
- }
- pr_debug("AUN: recvfrom() error %d\n", -err);
- }
-
- data = skb_transport_header(skb) + sizeof(struct udphdr);
- ah = (struct aunhdr *)data;
- len = skb->len - sizeof(struct udphdr);
-
- switch (ah->code) {
- case 2:
- aun_incoming(skb, ah, len);
- break;
- case 3:
- aun_tx_ack(ah->handle, ECTYPE_TRANSMIT_OK);
- break;
- case 4:
- aun_tx_ack(ah->handle, ECTYPE_TRANSMIT_NOT_LISTENING);
- break;
- default:
- pr_debug("AUN: unknown packet type: %d\n", data[0]);
- }
-
- skb_free_datagram(sk, skb);
-}
-
-/*
- * Called by the timer to manage the AUN transmit queue. If a packet
- * was sent to a dead or nonexistent host then we will never get an
- * acknowledgement back. After a few seconds we need to spot this and
- * drop the packet.
- */
-
-static void ab_cleanup(unsigned long h)
-{
- struct sk_buff *skb, *n;
- unsigned long flags;
-
- spin_lock_irqsave(&aun_queue_lock, flags);
- skb_queue_walk_safe(&aun_queue, skb, n) {
- struct ec_cb *eb = (struct ec_cb *)&skb->cb;
- if ((jiffies - eb->start) > eb->timeout) {
- tx_result(skb->sk, eb->cookie,
- ECTYPE_TRANSMIT_NOT_PRESENT);
- skb_unlink(skb, &aun_queue);
- kfree_skb(skb);
- }
- }
- spin_unlock_irqrestore(&aun_queue_lock, flags);
-
- mod_timer(&ab_cleanup_timer, jiffies + (HZ * 2));
-}
-
-static int __init aun_udp_initialise(void)
-{
- int error;
- struct sockaddr_in sin;
-
- skb_queue_head_init(&aun_queue);
- setup_timer(&ab_cleanup_timer, ab_cleanup, 0);
- ab_cleanup_timer.expires = jiffies + (HZ * 2);
- add_timer(&ab_cleanup_timer);
-
- memset(&sin, 0, sizeof(sin));
- sin.sin_port = htons(AUN_PORT);
-
- /* We can count ourselves lucky Acorn machines are too dim to
- speak IPv6. :-) */
- error = sock_create_kern(PF_INET, SOCK_DGRAM, 0, &udpsock);
- if (error < 0) {
- pr_err("AUN: socket error %d\n", -error);
- return error;
- }
-
- udpsock->sk->sk_reuse = SK_CAN_REUSE;
- udpsock->sk->sk_allocation = GFP_ATOMIC; /* we're going to call it
- from interrupts */
-
- error = udpsock->ops->bind(udpsock, (struct sockaddr *)&sin,
- sizeof(sin));
- if (error < 0) {
- pr_err("AUN: bind error %d\n", -error);
- goto release;
- }
-
- udpsock->sk->sk_data_ready = aun_data_available;
-
- return 0;
-
-release:
- sock_release(udpsock);
- udpsock = NULL;
- return error;
-}
-#endif
-
-#ifdef CONFIG_ECONET_NATIVE
-
-/*
- * Receive an Econet frame from a device.
- */
-
-static int econet_rcv(struct sk_buff *skb, struct net_device *dev,
- struct packet_type *pt, struct net_device *orig_dev)
-{
- struct ec_framehdr *hdr;
- struct sock *sk = NULL;
- struct ec_device *edev = dev->ec_ptr;
-
- if (!net_eq(dev_net(dev), &init_net))
- goto drop;
-
- if (skb->pkt_type == PACKET_OTHERHOST)
- goto drop;
-
- if (!edev)
- goto drop;
-
- skb = skb_share_check(skb, GFP_ATOMIC);
- if (skb == NULL)
- return NET_RX_DROP;
-
- if (!pskb_may_pull(skb, sizeof(struct ec_framehdr)))
- goto drop;
-
- hdr = (struct ec_framehdr *)skb->data;
-
- /* First check for encapsulated IP */
- if (hdr->port == EC_PORT_IP) {
- skb->protocol = htons(ETH_P_IP);
- skb_pull(skb, sizeof(struct ec_framehdr));
- netif_rx(skb);
- return NET_RX_SUCCESS;
- }
-
- sk = ec_listening_socket(hdr->port, hdr->src_stn, hdr->src_net);
- if (!sk)
- goto drop;
-
- if (ec_queue_packet(sk, skb, edev->net, hdr->src_stn, hdr->cb,
- hdr->port))
- goto drop;
- sock_put(sk);
- return NET_RX_SUCCESS;
-
-drop:
- if (sk)
- sock_put(sk);
- kfree_skb(skb);
- return NET_RX_DROP;
-}
-
-static struct packet_type econet_packet_type __read_mostly = {
- .type = cpu_to_be16(ETH_P_ECONET),
- .func = econet_rcv,
-};
-
-static void econet_hw_initialise(void)
-{
- dev_add_pack(&econet_packet_type);
-}
-
-#endif
-
-static int econet_notifier(struct notifier_block *this, unsigned long msg,
- void *data)
-{
- struct net_device *dev = data;
- struct ec_device *edev;
-
- if (!net_eq(dev_net(dev), &init_net))
- return NOTIFY_DONE;
-
- switch (msg) {
- case NETDEV_UNREGISTER:
- /* A device has gone down - kill any data we hold for it. */
- edev = dev->ec_ptr;
- if (edev) {
- if (net2dev_map[0] == dev)
- net2dev_map[0] = NULL;
- net2dev_map[edev->net] = NULL;
- kfree(edev);
- dev->ec_ptr = NULL;
- }
- break;
- }
-
- return NOTIFY_DONE;
-}
-
-static struct notifier_block econet_netdev_notifier = {
- .notifier_call = econet_notifier,
-};
-
-static void __exit econet_proto_exit(void)
-{
-#ifdef CONFIG_ECONET_AUNUDP
- del_timer(&ab_cleanup_timer);
- if (udpsock)
- sock_release(udpsock);
-#endif
- unregister_netdevice_notifier(&econet_netdev_notifier);
-#ifdef CONFIG_ECONET_NATIVE
- dev_remove_pack(&econet_packet_type);
-#endif
- sock_unregister(econet_family_ops.family);
- proto_unregister(&econet_proto);
-}
-
-static int __init econet_proto_init(void)
-{
- int err = proto_register(&econet_proto, 0);
-
- if (err != 0)
- goto out;
- sock_register(&econet_family_ops);
-#ifdef CONFIG_ECONET_AUNUDP
- aun_udp_initialise();
-#endif
-#ifdef CONFIG_ECONET_NATIVE
- econet_hw_initialise();
-#endif
- register_netdevice_notifier(&econet_netdev_notifier);
-out:
- return err;
-}
-
-module_init(econet_proto_init);
-module_exit(econet_proto_exit);
-
-MODULE_LICENSE("GPL");
-MODULE_ALIAS_NETPROTO(PF_ECONET);
--- a/include/linux/Kbuild 2012-05-17 07:28:39.911045825 -0700
+++ b/include/linux/Kbuild 2012-05-17 20:53:29.972167302 -0700
@@ -167,7 +167,6 @@ header-y += if_arp.h
header-y += if_bonding.h
header-y += if_bridge.h
header-y += if_cablemodem.h
-header-y += if_ec.h
header-y += if_eql.h
header-y += if_ether.h
header-y += if_fc.h
--- a/include/linux/if_ec.h 2012-02-13 09:23:58.694507006 -0800
+++ /dev/null 1970-01-01 00:00:00.000000000 +0000
@@ -1,68 +0,0 @@
-/* Definitions for Econet sockets. */
-
-#ifndef __LINUX_IF_EC
-#define __LINUX_IF_EC
-
-/* User visible stuff. Glibc provides its own but libc5 folk will use these */
-
-struct ec_addr {
- unsigned char station; /* Station number. */
- unsigned char net; /* Network number. */
-};
-
-struct sockaddr_ec {
- unsigned short sec_family;
- unsigned char port; /* Port number. */
- unsigned char cb; /* Control/flag byte. */
- unsigned char type; /* Type of message. */
- struct ec_addr addr;
- unsigned long cookie;
-};
-
-#define ECTYPE_PACKET_RECEIVED 0 /* Packet received */
-#define ECTYPE_TRANSMIT_STATUS 0x10 /* Transmit completed,
- low nibble holds status */
-
-#define ECTYPE_TRANSMIT_OK 1
-#define ECTYPE_TRANSMIT_NOT_LISTENING 2
-#define ECTYPE_TRANSMIT_NET_ERROR 3
-#define ECTYPE_TRANSMIT_NO_CLOCK 4
-#define ECTYPE_TRANSMIT_LINE_JAMMED 5
-#define ECTYPE_TRANSMIT_NOT_PRESENT 6
-
-#ifdef __KERNEL__
-
-#define EC_HLEN 6
-
-/* This is what an Econet frame looks like on the wire. */
-struct ec_framehdr {
- unsigned char dst_stn;
- unsigned char dst_net;
- unsigned char src_stn;
- unsigned char src_net;
- unsigned char cb;
- unsigned char port;
-};
-
-struct econet_sock {
- /* struct sock has to be the first member of econet_sock */
- struct sock sk;
- unsigned char cb;
- unsigned char port;
- unsigned char station;
- unsigned char net;
- unsigned short num;
-};
-
-static inline struct econet_sock *ec_sk(const struct sock *sk)
-{
- return (struct econet_sock *)sk;
-}
-
-struct ec_device {
- unsigned char station, net; /* Econet protocol address */
-};
-
-#endif
-
-#endif
--- a/net/Makefile 2012-05-17 07:28:39.927046007 -0700
+++ b/net/Makefile 2012-05-17 20:53:08.867944023 -0700
@@ -40,7 +40,6 @@ obj-$(CONFIG_AF_RXRPC) += rxrpc/
obj-$(CONFIG_ATM) += atm/
obj-$(CONFIG_L2TP) += l2tp/
obj-$(CONFIG_DECNET) += decnet/
-obj-$(CONFIG_ECONET) += econet/
obj-$(CONFIG_PHONET) += phonet/
ifneq ($(CONFIG_VLAN_8021Q),)
obj-y += 8021q/
^ permalink raw reply
* [PATCH net-next] ip_frag: struct inet_frags match() method returns a bool
From: Eric Dumazet @ 2012-05-18 3:57 UTC (permalink / raw)
To: David Miller; +Cc: netdev
From: Eric Dumazet <edumazet@google.com>
- match() method returns a boolean
- return (A && B && C && D) -> return A && B && C && D
- fix indentation
Signed-off-by: Eric Dumazet <edumazet@google.com>
---
include/net/inet_frag.h | 3 +--
include/net/ipv6.h | 2 +-
net/ipv4/ip_fragment.c | 10 +++++-----
net/ipv6/reassembly.c | 9 +++++----
4 files changed, 12 insertions(+), 12 deletions(-)
diff --git a/include/net/inet_frag.h b/include/net/inet_frag.h
index 16ff29a..2431cf8 100644
--- a/include/net/inet_frag.h
+++ b/include/net/inet_frag.h
@@ -46,8 +46,7 @@ struct inet_frags {
void *arg);
void (*destructor)(struct inet_frag_queue *);
void (*skb_free)(struct sk_buff *);
- int (*match)(struct inet_frag_queue *q,
- void *arg);
+ bool (*match)(struct inet_frag_queue *q, void *arg);
void (*frag_expire)(unsigned long data);
};
diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index 4332e9a..5f65e7c 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -387,7 +387,7 @@ struct ip6_create_arg {
};
void ip6_frag_init(struct inet_frag_queue *q, void *a);
-int ip6_frag_match(struct inet_frag_queue *q, void *a);
+bool ip6_frag_match(struct inet_frag_queue *q, void *a);
static inline int ipv6_addr_any(const struct in6_addr *a)
{
diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c
index 9f9bd13..695b27f 100644
--- a/net/ipv4/ip_fragment.c
+++ b/net/ipv4/ip_fragment.c
@@ -148,17 +148,17 @@ static unsigned int ip4_hashfn(struct inet_frag_queue *q)
return ipqhashfn(ipq->id, ipq->saddr, ipq->daddr, ipq->protocol);
}
-static int ip4_frag_match(struct inet_frag_queue *q, void *a)
+static bool ip4_frag_match(struct inet_frag_queue *q, void *a)
{
struct ipq *qp;
struct ip4_create_arg *arg = a;
qp = container_of(q, struct ipq, q);
return qp->id == arg->iph->id &&
- qp->saddr == arg->iph->saddr &&
- qp->daddr == arg->iph->daddr &&
- qp->protocol == arg->iph->protocol &&
- qp->user == arg->user;
+ qp->saddr == arg->iph->saddr &&
+ qp->daddr == arg->iph->daddr &&
+ qp->protocol == arg->iph->protocol &&
+ qp->user == arg->user;
}
/* Memory Tracking Functions. */
diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c
index f1b86fd..5d32dfa 100644
--- a/net/ipv6/reassembly.c
+++ b/net/ipv6/reassembly.c
@@ -134,15 +134,16 @@ static unsigned int ip6_hashfn(struct inet_frag_queue *q)
return inet6_hash_frag(fq->id, &fq->saddr, &fq->daddr, ip6_frags.rnd);
}
-int ip6_frag_match(struct inet_frag_queue *q, void *a)
+bool ip6_frag_match(struct inet_frag_queue *q, void *a)
{
struct frag_queue *fq;
struct ip6_create_arg *arg = a;
fq = container_of(q, struct frag_queue, q);
- return (fq->id == arg->id && fq->user == arg->user &&
- ipv6_addr_equal(&fq->saddr, arg->src) &&
- ipv6_addr_equal(&fq->daddr, arg->dst));
+ return fq->id == arg->id &&
+ fq->user == arg->user &&
+ ipv6_addr_equal(&fq->saddr, arg->src) &&
+ ipv6_addr_equal(&fq->daddr, arg->dst);
}
EXPORT_SYMBOL(ip6_frag_match);
^ permalink raw reply related
* Re: [PATCH net-next v4] be2net: Fix to allow get/set of debug levels in the firmware.
From: David Miller @ 2012-05-18 2:45 UTC (permalink / raw)
To: Somnath.Kotur; +Cc: netdev, bhutchings, Suresh.Reddy
In-Reply-To: <80fb6a42-aa9a-4288-aa96-c6aa5972bc03@exht1.ad.emulex.com>
From: <Somnath.Kotur@Emulex.Com>
Date: Thu, 17 May 2012 18:47:50 -0700
> At the moment our driver does not have any extensive logging
> facility, but yes going forward we will use this infrastructure to
> gate all our driver's logging and will send separate patch(es) for
> that in future.
>
> Does that sound good for now? A respin for this current patch
Yes.
^ permalink raw reply
* [PATCH v5 6/6] net: sh_eth: use NAPI
From: Shimoda, Yoshihiro @ 2012-05-18 1:58 UTC (permalink / raw)
To: netdev; +Cc: SH-Linux
This patch modifies the driver to use NAPI.
Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
---
about v5:
- fix two racing sh_eth_txfree() in sh_eth_start_xmit() and sh_eth_poll()
drivers/net/ethernet/renesas/sh_eth.c | 109 +++++++++++++++++++++------------
drivers/net/ethernet/renesas/sh_eth.h | 3 +
2 files changed, 73 insertions(+), 39 deletions(-)
diff --git a/drivers/net/ethernet/renesas/sh_eth.c b/drivers/net/ethernet/renesas/sh_eth.c
index c64a31c..1dc9b6e 100644
--- a/drivers/net/ethernet/renesas/sh_eth.c
+++ b/drivers/net/ethernet/renesas/sh_eth.c
@@ -1035,7 +1035,7 @@ static int sh_eth_txfree(struct net_device *ndev)
}
/* Packet receive function */
-static int sh_eth_rx(struct net_device *ndev)
+static int sh_eth_rx(struct net_device *ndev, int *work, int budget)
{
struct sh_eth_private *mdp = netdev_priv(ndev);
struct sh_eth_rxdesc *rxdesc;
@@ -1047,7 +1047,8 @@ static int sh_eth_rx(struct net_device *ndev)
u32 desc_status;
rxdesc = &mdp->rx_ring[entry];
- while (!(rxdesc->status & cpu_to_edmac(mdp, RD_RACT))) {
+ while (!(rxdesc->status & cpu_to_edmac(mdp, RD_RACT)) &&
+ *work < budget) {
desc_status = edmac_to_cpu(mdp, rxdesc->status);
pkt_len = rxdesc->frame_length;
@@ -1087,13 +1088,17 @@ static int sh_eth_rx(struct net_device *ndev)
skb_reserve(skb, NET_IP_ALIGN);
skb_put(skb, pkt_len);
skb->protocol = eth_type_trans(skb, ndev);
- netif_rx(skb);
- ndev->stats.rx_packets++;
- ndev->stats.rx_bytes += pkt_len;
+ if (netif_receive_skb(skb) == NET_RX_DROP) {
+ ndev->stats.rx_dropped++;
+ } else {
+ ndev->stats.rx_packets++;
+ ndev->stats.rx_bytes += pkt_len;
+ }
}
rxdesc->status |= cpu_to_edmac(mdp, RD_RACT);
entry = (++mdp->cur_rx) % mdp->num_rx_ring;
rxdesc = &mdp->rx_ring[entry];
+ (*work)++;
}
/* Refill the Rx ring buffers. */
@@ -1125,7 +1130,7 @@ static int sh_eth_rx(struct net_device *ndev)
/* Restart Rx engine if stopped. */
/* If we don't need to check status, don't. -KDU */
- if (!(sh_eth_read(ndev, EDRRR) & EDRRR_R)) {
+ if (*work < budget && !(sh_eth_read(ndev, EDRRR) & EDRRR_R)) {
/* fix the values for the next receiving */
mdp->cur_rx = mdp->dirty_rx = (sh_eth_read(ndev, RDFAR) -
sh_eth_read(ndev, RDLAR)) >> 4;
@@ -1281,38 +1286,61 @@ static irqreturn_t sh_eth_interrupt(int irq, void *netdev)
/* Get interrpt stat */
intr_status = sh_eth_read(ndev, EESR);
- /* Clear interrupt */
if (intr_status & (EESR_FRC | EESR_RMAF | EESR_RRF |
EESR_RTLF | EESR_RTSF | EESR_PRE | EESR_CERF |
cd->tx_check | cd->eesr_err_check)) {
- sh_eth_write(ndev, intr_status, EESR);
+ if (napi_schedule_prep(&mdp->napi)) {
+ /* Disable interrupts of the channel */
+ sh_eth_write(ndev, 0, EESIPR);
+ __napi_schedule(&mdp->napi);
+ }
ret = IRQ_HANDLED;
- } else
- goto other_irq;
-
- if (intr_status & (EESR_FRC | /* Frame recv*/
- EESR_RMAF | /* Multi cast address recv*/
- EESR_RRF | /* Bit frame recv */
- EESR_RTLF | /* Long frame recv*/
- EESR_RTSF | /* short frame recv */
- EESR_PRE | /* PHY-LSI recv error */
- EESR_CERF)){ /* recv frame CRC error */
- sh_eth_rx(ndev);
}
- /* Tx Check */
- if (intr_status & cd->tx_check) {
- sh_eth_txfree(ndev);
- netif_wake_queue(ndev);
+ spin_unlock(&mdp->lock);
+
+ return ret;
+}
+
+static int sh_eth_poll(struct napi_struct *napi, int budget)
+{
+ struct sh_eth_private *mdp = container_of(napi, struct sh_eth_private,
+ napi);
+ struct net_device *ndev = mdp->ndev;
+ struct sh_eth_cpu_data *cd = mdp->cd;
+ int work_done = 0, txfree_num;
+ u32 intr_status = sh_eth_read(ndev, EESR);
+
+ /* Clear interrupt flags */
+ sh_eth_write(ndev, intr_status, EESR);
+
+ /* check txdesc */
+ txfree_num = sh_eth_txfree(ndev);
+ if (txfree_num) {
+ netif_tx_lock(ndev);
+ if (netif_queue_stopped(ndev))
+ netif_wake_queue(ndev);
+ netif_tx_unlock(ndev);
}
+ /* check rxdesc */
+ sh_eth_rx(ndev, &work_done, budget);
+
+ /* check error flags */
if (intr_status & cd->eesr_err_check)
sh_eth_error(ndev, intr_status);
-other_irq:
- spin_unlock(&mdp->lock);
+ /* get current interrupt flags */
+ intr_status = sh_eth_read(ndev, EESR);
- return ret;
+ /* check whether this driver should call napi_complete() */
+ if (work_done < budget) {
+ napi_complete(napi);
+ /* Enable all interrupts */
+ sh_eth_write(ndev, cd->eesipr_value, EESIPR);
+ }
+
+ return work_done;
}
/* PHY state control function */
@@ -1545,6 +1573,7 @@ static int sh_eth_set_ringparam(struct net_device *ndev,
/* Stop the chip's Tx and Rx processes. */
sh_eth_write(ndev, 0, EDTRR);
sh_eth_write(ndev, 0, EDRRR);
+ napi_disable(&mdp->napi);
synchronize_irq(ndev->irq);
}
@@ -1569,6 +1598,7 @@ static int sh_eth_set_ringparam(struct net_device *ndev,
}
if (netif_running(ndev)) {
+ napi_enable(&mdp->napi);
sh_eth_write(ndev, mdp->cd->eesipr_value, EESIPR);
/* Setting the Rx mode will start the Rx process. */
sh_eth_write(ndev, EDRRR_R, EDRRR);
@@ -1600,6 +1630,8 @@ static int sh_eth_open(struct net_device *ndev)
pm_runtime_get_sync(&mdp->pdev->dev);
+ napi_enable(&mdp->napi);
+
ret = request_irq(ndev->irq, sh_eth_interrupt,
#if defined(CONFIG_CPU_SUBTYPE_SH7763) || \
defined(CONFIG_CPU_SUBTYPE_SH7764) || \
@@ -1678,19 +1710,6 @@ static int sh_eth_start_xmit(struct sk_buff *skb, struct net_device *ndev)
struct sh_eth_private *mdp = netdev_priv(ndev);
struct sh_eth_txdesc *txdesc;
u32 entry;
- unsigned long flags;
-
- spin_lock_irqsave(&mdp->lock, flags);
- if ((mdp->cur_tx - mdp->dirty_tx) >= (mdp->num_tx_ring - 4)) {
- if (!sh_eth_txfree(ndev)) {
- if (netif_msg_tx_queued(mdp))
- dev_warn(&ndev->dev, "TxFD exhausted.\n");
- netif_stop_queue(ndev);
- spin_unlock_irqrestore(&mdp->lock, flags);
- return NETDEV_TX_BUSY;
- }
- }
- spin_unlock_irqrestore(&mdp->lock, flags);
entry = mdp->cur_tx % mdp->num_tx_ring;
mdp->tx_skbuff[entry] = skb;
@@ -1716,6 +1735,13 @@ static int sh_eth_start_xmit(struct sk_buff *skb, struct net_device *ndev)
if (!(sh_eth_read(ndev, EDTRR) & sh_eth_get_edtrr_trns(mdp)))
sh_eth_write(ndev, sh_eth_get_edtrr_trns(mdp), EDTRR);
+ if ((mdp->cur_tx - mdp->dirty_tx) >= (mdp->num_tx_ring - 4)) {
+ if (netif_msg_tx_queued(mdp)) {
+ dev_warn(&ndev->dev, "TxFD exhausted.\n");
+ netif_stop_queue(ndev);
+ }
+ }
+
return NETDEV_TX_OK;
}
@@ -1739,6 +1765,8 @@ static int sh_eth_close(struct net_device *ndev)
phy_disconnect(mdp->phydev);
}
+ napi_disable(&mdp->napi);
+
free_irq(ndev->irq, ndev);
/* Free all the skbuffs in the Rx queue. */
@@ -2368,6 +2396,9 @@ static int sh_eth_drv_probe(struct platform_device *pdev)
#endif
sh_eth_set_default_cpu_data(mdp->cd);
+ mdp->ndev = ndev;
+ netif_napi_add(ndev, &mdp->napi, sh_eth_poll, SH_ETH_NAPI_WEIGHT);
+
/* set function */
ndev->netdev_ops = &sh_eth_netdev_ops;
SET_ETHTOOL_OPS(ndev, &sh_eth_ethtool_ops);
diff --git a/drivers/net/ethernet/renesas/sh_eth.h b/drivers/net/ethernet/renesas/sh_eth.h
index f1dbc27..93dad7b 100644
--- a/drivers/net/ethernet/renesas/sh_eth.h
+++ b/drivers/net/ethernet/renesas/sh_eth.h
@@ -35,6 +35,7 @@
#define PKT_BUF_SZ 1538
#define SH_ETH_TSU_TIMEOUT_MS 500
#define SH_ETH_TSU_CAM_ENTRIES 32
+#define SH_ETH_NAPI_WEIGHT 32
enum {
/* E-DMAC registers */
@@ -728,6 +729,8 @@ struct sh_eth_private {
int duplex;
int port; /* for TSU */
int vlan_num_ids; /* for VLAN tag filter */
+ struct napi_struct napi;
+ struct net_device *ndev;
unsigned no_ether_link:1;
unsigned ether_link_active_low:1;
--
1.7.1
^ permalink raw reply related
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox