2009-01-26 17:47:06

by Divy Le ray

Subject: [PATCH 2.6.30 2/2] iw_cxgb3 - handle chip reset notifications

From: Divy Le Ray <[email protected]>

Freeze activity when notified that the underlying chip
is getting reset on a EEH event or fatal error.

Signed-off-by: Divy Le Ray <[email protected]>
---

drivers/infiniband/hw/cxgb3/cxio_hal.c | 3 +++
drivers/infiniband/hw/cxgb3/cxio_hal.h | 2 ++
drivers/infiniband/hw/cxgb3/iwch.c | 15 ++++++++++++++-
3 files changed, 19 insertions(+), 1 deletions(-)

diff --git a/drivers/infiniband/hw/cxgb3/cxio_hal.c b/drivers/infiniband/hw/cxgb3/cxio_hal.c
index 4dcf08b..11efd35 100644
--- a/drivers/infiniband/hw/cxgb3/cxio_hal.c
+++ b/drivers/infiniband/hw/cxgb3/cxio_hal.c
@@ -701,6 +701,9 @@ static int __cxio_tpt_op(struct cxio_rdev *rdev_p, u32 reset_tpt_entry,
u32 stag_idx;
u32 wptr;

+ if (rdev_p->flags)
+ return -EIO;
+
stag_state = stag_state > 0;
stag_idx = (*stag) >> 8;

diff --git a/drivers/infiniband/hw/cxgb3/cxio_hal.h b/drivers/infiniband/hw/cxgb3/cxio_hal.h
index 656fe47..9ed65b0 100644
--- a/drivers/infiniband/hw/cxgb3/cxio_hal.h
+++ b/drivers/infiniband/hw/cxgb3/cxio_hal.h
@@ -108,6 +108,8 @@ struct cxio_rdev {
struct gen_pool *pbl_pool;
struct gen_pool *rqt_pool;
struct list_head entry;
+ u32 flags;
+#define CXIO_ERROR_FATAL 1
};

static inline int cxio_num_stags(struct cxio_rdev *rdev_p)
diff --git a/drivers/infiniband/hw/cxgb3/iwch.c b/drivers/infiniband/hw/cxgb3/iwch.c
index 4489c89..37a4fc2 100644
--- a/drivers/infiniband/hw/cxgb3/iwch.c
+++ b/drivers/infiniband/hw/cxgb3/iwch.c
@@ -51,13 +51,15 @@ cxgb3_cpl_handler_func t3c_handlers[NUM_CPL_CMDS];

static void open_rnic_dev(struct t3cdev *);
static void close_rnic_dev(struct t3cdev *);
+static void iwch_err_handler(struct t3cdev *, u32, u32);

struct cxgb3_client t3c_client = {
.name = "iw_cxgb3",
.add = open_rnic_dev,
.remove = close_rnic_dev,
.handlers = t3c_handlers,
- .redirect = iwch_ep_redirect
+ .redirect = iwch_ep_redirect,
+ .err_handler = iwch_err_handler
};

static LIST_HEAD(dev_list);
@@ -160,6 +162,17 @@ static void close_rnic_dev(struct t3cdev *tdev)
mutex_unlock(&dev_mutex);
}

+static void iwch_err_handler(struct t3cdev *tdev, u32 status, u32 error)
+{
+ struct cxio_rdev *rdev = tdev->ulp;
+
+ if (status == OFFLOAD_STATUS_DOWN)
+ rdev->flags = CXIO_ERROR_FATAL;
+
+ return;
+
+}
+
static int __init iwch_init_module(void)
{
int err;

2009-01-27 06:23:08

by David Miller

Subject: Re: [PATCH 2.6.30 2/2] iw_cxgb3 - handle chip reset notifications

From: Divy Le Ray <[email protected]>
Date: Mon, 26 Jan 2009 09:46:11 -0800

> Freeze activity when notified that the underlying chip
> is getting reset on a EEH event or fatal error.
>
> Signed-off-by: Divy Le Ray <[email protected]>

Applied.