* [PATCH 01/17] cxgb3: handle global variables properly
From: Christoph Hellwig @ 2016-10-13 19:56 UTC (permalink / raw)
To: linux-rdma-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <1476388576-22863-1-git-send-email-hch-jcswGhMUV9g@public.gmane.org>
We'll need to define them in one compilation unit and use an extern
declaration in the headers.
Signed-off-by: Christoph Hellwig <hch-jcswGhMUV9g@public.gmane.org>
---
providers/cxgb3/cxio_wr.h | 6 +++---
providers/cxgb3/iwch.c | 4 ++++
2 files changed, 7 insertions(+), 3 deletions(-)
diff --git a/providers/cxgb3/cxio_wr.h b/providers/cxgb3/cxio_wr.h
index e154795..d56c529 100644
--- a/providers/cxgb3/cxio_wr.h
+++ b/providers/cxgb3/cxio_wr.h
@@ -633,9 +633,9 @@ static inline unsigned t3_cq_depth(struct t3_cq *cq)
return (1UL<<cq->size_log2);
}
-unsigned long iwch_page_size;
-unsigned long iwch_page_shift;
-unsigned long iwch_page_mask;
+extern unsigned long iwch_page_size;
+extern unsigned long iwch_page_shift;
+extern unsigned long iwch_page_mask;
#define PAGE_ALIGN(x) (((x) + iwch_page_mask) & ~iwch_page_mask)
diff --git a/providers/cxgb3/iwch.c b/providers/cxgb3/iwch.c
index aa9b108..804726c 100644
--- a/providers/cxgb3/iwch.c
+++ b/providers/cxgb3/iwch.c
@@ -105,6 +105,10 @@ static struct ibv_context_ops iwch_ctx_ops = {
.req_notify_cq = iwch_arm_cq,
};
+unsigned long iwch_page_size;
+unsigned long iwch_page_shift;
+unsigned long iwch_page_mask;
+
static struct ibv_context *iwch_alloc_context(struct ibv_device *ibdev,
int cmd_fd)
{
--
2.1.4
--
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 02/17] cxgb3: mark symbols static where possible
From: Christoph Hellwig @ 2016-10-13 19:56 UTC (permalink / raw)
To: linux-rdma-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <1476388576-22863-1-git-send-email-hch-jcswGhMUV9g@public.gmane.org>
Signed-off-by: Christoph Hellwig <hch-jcswGhMUV9g@public.gmane.org>
---
providers/cxgb3/cq.c | 2 +-
providers/cxgb3/iwch.c | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/providers/cxgb3/cq.c b/providers/cxgb3/cq.c
index 1510d1e..a6158ce 100644
--- a/providers/cxgb3/cq.c
+++ b/providers/cxgb3/cq.c
@@ -255,7 +255,7 @@ skip_cqe:
* -EAGAIN caller must try again
* any other -errno fatal error
*/
-int iwch_poll_cq_one(struct iwch_device *rhp, struct iwch_cq *chp,
+static int iwch_poll_cq_one(struct iwch_device *rhp, struct iwch_cq *chp,
struct ibv_wc *wc)
{
struct iwch_qp *qhp = NULL;
diff --git a/providers/cxgb3/iwch.c b/providers/cxgb3/iwch.c
index 804726c..a03bced 100644
--- a/providers/cxgb3/iwch.c
+++ b/providers/cxgb3/iwch.c
@@ -61,7 +61,7 @@
.device = PCI_DEVICE_ID_CHELSIO_##d, \
.type = CHELSIO_##t }
-struct {
+static struct {
unsigned vendor;
unsigned device;
enum iwch_hca_type type;
--
2.1.4
--
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 03/17] hfi1verbs: mark symbols static where possible
From: Christoph Hellwig @ 2016-10-13 19:56 UTC (permalink / raw)
To: linux-rdma-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <1476388576-22863-1-git-send-email-hch-jcswGhMUV9g@public.gmane.org>
Signed-off-by: Christoph Hellwig <hch-jcswGhMUV9g@public.gmane.org>
---
providers/hfi1verbs/hfiverbs.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/providers/hfi1verbs/hfiverbs.c b/providers/hfi1verbs/hfiverbs.c
index e28afc7..7605347 100644
--- a/providers/hfi1verbs/hfiverbs.c
+++ b/providers/hfi1verbs/hfiverbs.c
@@ -81,7 +81,7 @@
{ .vendor = PCI_VENDOR_ID_##v, \
.device = PCI_DEVICE_ID_HFI_##d }
-struct {
+static struct {
unsigned vendor;
unsigned device;
} hca_table[] = {
--
2.1.4
--
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 04/17] ipathverbs: mark symbols static where possible
From: Christoph Hellwig @ 2016-10-13 19:56 UTC (permalink / raw)
To: linux-rdma-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <1476388576-22863-1-git-send-email-hch-jcswGhMUV9g@public.gmane.org>
Signed-off-by: Christoph Hellwig <hch-jcswGhMUV9g@public.gmane.org>
---
providers/ipathverbs/ipathverbs.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/providers/ipathverbs/ipathverbs.c b/providers/ipathverbs/ipathverbs.c
index 4b945f6..e953864 100644
--- a/providers/ipathverbs/ipathverbs.c
+++ b/providers/ipathverbs/ipathverbs.c
@@ -77,7 +77,7 @@
{ .vendor = PCI_VENDOR_ID_##v, \
.device = PCI_DEVICE_ID_INFINIPATH_##d }
-struct {
+static struct {
unsigned vendor;
unsigned device;
} hca_table[] = {
--
2.1.4
--
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 05/17] i40iw: mark symbols static where possible
From: Christoph Hellwig @ 2016-10-13 19:56 UTC (permalink / raw)
To: linux-rdma-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <1476388576-22863-1-git-send-email-hch-jcswGhMUV9g@public.gmane.org>
Signed-off-by: Christoph Hellwig <hch-jcswGhMUV9g@public.gmane.org>
---
providers/i40iw/i40iw_umain.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/providers/i40iw/i40iw_umain.c b/providers/i40iw/i40iw_umain.c
index 1756e65..0540ee7 100644
--- a/providers/i40iw/i40iw_umain.c
+++ b/providers/i40iw/i40iw_umain.c
@@ -57,7 +57,7 @@ unsigned int i40iw_debug_level;
.device = d, \
.type = INTEL_ ## t }
-struct {
+static struct {
unsigned int vendor;
unsigned int device;
enum i40iw_uhca_type type;
--
2.1.4
--
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 06/17] i40iw: remove unused variable i40iw_debug_level
From: Christoph Hellwig @ 2016-10-13 19:56 UTC (permalink / raw)
To: linux-rdma-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <1476388576-22863-1-git-send-email-hch-jcswGhMUV9g@public.gmane.org>
Signed-off-by: Christoph Hellwig <hch-jcswGhMUV9g@public.gmane.org>
---
providers/i40iw/i40iw_umain.c | 2 --
1 file changed, 2 deletions(-)
diff --git a/providers/i40iw/i40iw_umain.c b/providers/i40iw/i40iw_umain.c
index 0540ee7..8d7b655 100644
--- a/providers/i40iw/i40iw_umain.c
+++ b/providers/i40iw/i40iw_umain.c
@@ -46,8 +46,6 @@
#include "i40iw_umain.h"
#include "i40iw-abi.h"
-unsigned int i40iw_debug_level;
-
#include <sys/types.h>
#include <sys/stat.h>
#include <fcntl.h>
--
2.1.4
--
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: [PATCH rdma-next 00/10] Hardware tag matching support
From: Jason Gunthorpe @ 2016-10-13 19:56 UTC (permalink / raw)
To: Hefty, Sean
Cc: Christoph Hellwig, Doug Ledford, Leon Romanovsky,
linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
In-Reply-To: <1828884A29C6694DAF28B7E6B8A82373AB095647-P5GAC/sN6hkd3b2yrw5b5LfspsVTdybXVpNB7YpNyf8@public.gmane.org>
On Thu, Oct 13, 2016 at 07:43:25PM +0000, Hefty, Sean wrote:
> > What makes OFVWG more qualified to review an ABI than linux-rdma?
>
> We've had this discussion on the list before. The issue is that
> verbs isn't just a software API. It defines a hardware
> implementation. The consensus was that some 'standards' body needed
> to be able to review the proposed changes and its impact. The OFVWG
> was selected as the best available option. If we want to say no,
> and let the linux-rdma maintainer dictate the hardware interface
> exported by the kernel, then, yes, that's an option.
A bigger objection I have is that much of this new stuff is not
adaquately documented.
Tag matching is a great example. It looks like this introduces many new
headers into the on-wire protocol. Where are these headers documented?
Where is the spec for the on-wire format? Where is the implementation
for rxe?
Basically, is there enough public information that another vendor
could implement this wire protocol? (if not, why are we even
*contemplating* merging something like this???)
'Hopefully' the answer is that this information is all hidden away in
the IBTA... Which opens up an entire new line of problems reviewing
something like this without seeing the agreed spec..
The basic problem with linux-rdma is that it is not a reasonable place
to review a new on-wire protocol amendment to the IBA. Nor is ofvwg
for that matter.
Jason
--
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
* [PATCH 07/17] mlx4: mark symbols static where possible
From: Christoph Hellwig @ 2016-10-13 19:56 UTC (permalink / raw)
To: linux-rdma-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <1476388576-22863-1-git-send-email-hch-jcswGhMUV9g@public.gmane.org>
Signed-off-by: Christoph Hellwig <hch-jcswGhMUV9g@public.gmane.org>
---
providers/mlx4/mlx4.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/providers/mlx4/mlx4.c b/providers/mlx4/mlx4.c
index 326ead4..b59c202 100644
--- a/providers/mlx4/mlx4.c
+++ b/providers/mlx4/mlx4.c
@@ -51,7 +51,7 @@
{ .vendor = PCI_VENDOR_ID_##v, \
.device = d }
-struct {
+static struct {
unsigned vendor;
unsigned device;
} hca_table[] = {
--
2.1.4
--
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 08/17] mthca: mark symbols static where possible
From: Christoph Hellwig @ 2016-10-13 19:56 UTC (permalink / raw)
To: linux-rdma-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <1476388576-22863-1-git-send-email-hch-jcswGhMUV9g@public.gmane.org>
Signed-off-by: Christoph Hellwig <hch-jcswGhMUV9g@public.gmane.org>
---
providers/mthca/mthca.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/providers/mthca/mthca.c b/providers/mthca/mthca.c
index d5660ce..4103a3a 100644
--- a/providers/mthca/mthca.c
+++ b/providers/mthca/mthca.c
@@ -77,7 +77,7 @@
.device = PCI_DEVICE_ID_MELLANOX_##d, \
.type = MTHCA_##t }
-struct {
+static struct {
unsigned vendor;
unsigned device;
enum mthca_hca_type type;
--
2.1.4
--
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 09/17] rxe: mark symbols static where possible
From: Christoph Hellwig @ 2016-10-13 19:56 UTC (permalink / raw)
To: linux-rdma-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <1476388576-22863-1-git-send-email-hch-jcswGhMUV9g@public.gmane.org>
Signed-off-by: Christoph Hellwig <hch-jcswGhMUV9g@public.gmane.org>
---
providers/rxe/rxe.c | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/providers/rxe/rxe.c b/providers/rxe/rxe.c
index 8f8cc91..5aa77ea 100644
--- a/providers/rxe/rxe.c
+++ b/providers/rxe/rxe.c
@@ -305,7 +305,7 @@ static struct ibv_srq *rxe_create_srq(struct ibv_pd *pd,
return &srq->ibv_srq;
}
-int rxe_modify_srq(struct ibv_srq *ibsrq,
+static int rxe_modify_srq(struct ibv_srq *ibsrq,
struct ibv_srq_attr *attr, int attr_mask)
{
struct rxe_srq *srq = to_rsrq(ibsrq);
@@ -553,7 +553,7 @@ static int validate_send_wr(struct rxe_wq *sq, struct ibv_send_wr *ibwr,
return 0;
}
-void convert_send_wr(struct rxe_send_wr *kwr, struct ibv_send_wr *uwr)
+static void convert_send_wr(struct rxe_send_wr *kwr, struct ibv_send_wr *uwr)
{
memset(kwr, 0, sizeof(*kwr));
@@ -593,7 +593,7 @@ void convert_send_wr(struct rxe_send_wr *kwr, struct ibv_send_wr *uwr)
}
}
-int init_send_wqe(struct rxe_qp *qp, struct rxe_wq *sq,
+static int init_send_wqe(struct rxe_qp *qp, struct rxe_wq *sq,
struct ibv_send_wr *ibwr, unsigned int length,
struct rxe_send_wqe *wqe)
{
@@ -668,7 +668,7 @@ static int post_one_send(struct rxe_qp *qp, struct rxe_wq *sq,
}
/* send a null post send as a doorbell */
-int post_send_db(struct ibv_qp *ibqp)
+static int post_send_db(struct ibv_qp *ibqp)
{
struct ibv_post_send cmd;
struct ibv_post_send_resp resp;
--
2.1.4
--
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 10/17] ocrdma: mark symbols static where possible
From: Christoph Hellwig @ 2016-10-13 19:56 UTC (permalink / raw)
To: linux-rdma-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <1476388576-22863-1-git-send-email-hch-jcswGhMUV9g@public.gmane.org>
Signed-off-by: Christoph Hellwig <hch-jcswGhMUV9g@public.gmane.org>
---
providers/ocrdma/ocrdma_main.c | 2 +-
providers/ocrdma/ocrdma_verbs.c | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/providers/ocrdma/ocrdma_main.c b/providers/ocrdma/ocrdma_main.c
index f3830d9..39da8f0 100644
--- a/providers/ocrdma/ocrdma_main.c
+++ b/providers/ocrdma/ocrdma_main.c
@@ -59,7 +59,7 @@
{ .vendor = PCI_VENDOR_ID_##v, \
.device = PCI_DEVICE_ID_EMULEX_##d }
-struct {
+static struct {
unsigned vendor;
unsigned device;
} ucna_table[] = {
diff --git a/providers/ocrdma/ocrdma_verbs.c b/providers/ocrdma/ocrdma_verbs.c
index 163bf23..c5dd74b 100644
--- a/providers/ocrdma/ocrdma_verbs.c
+++ b/providers/ocrdma/ocrdma_verbs.c
@@ -608,7 +608,7 @@ mbx_err:
return NULL;
}
-enum ocrdma_qp_state get_ocrdma_qp_state(enum ibv_qp_state qps)
+static enum ocrdma_qp_state get_ocrdma_qp_state(enum ibv_qp_state qps)
{
switch (qps) {
case IBV_QPS_RESET:
--
2.1.4
--
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 11/17] ocrdma: use NULL instead of 0 for NULL pointers
From: Christoph Hellwig @ 2016-10-13 19:56 UTC (permalink / raw)
To: linux-rdma-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <1476388576-22863-1-git-send-email-hch-jcswGhMUV9g@public.gmane.org>
Signed-off-by: Christoph Hellwig <hch-jcswGhMUV9g@public.gmane.org>
---
providers/ocrdma/ocrdma_verbs.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/providers/ocrdma/ocrdma_verbs.c b/providers/ocrdma/ocrdma_verbs.c
index c5dd74b..7e066dc 100644
--- a/providers/ocrdma/ocrdma_verbs.c
+++ b/providers/ocrdma/ocrdma_verbs.c
@@ -455,7 +455,7 @@ int ocrdma_destroy_srq(struct ibv_srq *ibsrq)
free(srq->rqe_wr_id_tbl);
if (srq->db_va) {
munmap((void *)srq->db_va, srq->db_size);
- srq->db_va = 0;
+ srq->db_va = NULL;
}
if (srq->rq.va) {
munmap(srq->rq.va, srq->rq.len);
@@ -591,7 +591,7 @@ struct ibv_qp *ocrdma_create_qp(struct ibv_pd *pd,
} else {
if (qp->dpp_cq) {
ocrdma_destroy_cq(&qp->dpp_cq->ibv_cq);
- qp->dpp_cq = 0;
+ qp->dpp_cq = NULL;
}
}
qp->state = OCRDMA_QPS_RST;
@@ -1789,7 +1789,7 @@ static void ocrdma_update_free_srq_cqe(struct ibv_wc *ibwc,
struct ocrdma_cqe *cqe,
struct ocrdma_qp *qp)
{
- struct ocrdma_srq *srq = 0;
+ struct ocrdma_srq *srq = NULL;
uint32_t wqe_idx;
srq = get_ocrdma_srq(qp->ibv_qp.srq);
--
2.1.4
--
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 12/17] nes: move external declaration to nes_umain.h
From: Christoph Hellwig @ 2016-10-13 19:56 UTC (permalink / raw)
To: linux-rdma-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <1476388576-22863-1-git-send-email-hch-jcswGhMUV9g@public.gmane.org>
Signed-off-by: Christoph Hellwig <hch-jcswGhMUV9g@public.gmane.org>
---
providers/nes/nes_umain.h | 2 ++
providers/nes/nes_uverbs.c | 2 --
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/providers/nes/nes_umain.h b/providers/nes/nes_umain.h
index f3cc348..093956a 100644
--- a/providers/nes/nes_umain.h
+++ b/providers/nes/nes_umain.h
@@ -400,4 +400,6 @@ static inline uint32_t le32_to_cpu(uint32_t x)
return le32toh(x);
}
+extern long int page_size;
+
#endif /* nes_umain_H */
diff --git a/providers/nes/nes_uverbs.c b/providers/nes/nes_uverbs.c
index aa9c2b9..287e773 100644
--- a/providers/nes/nes_uverbs.c
+++ b/providers/nes/nes_uverbs.c
@@ -49,8 +49,6 @@
#include "nes_umain.h"
#include "nes-abi.h"
-extern long int page_size;
-
#define STATIC static
#define INLINE inline
--
2.1.4
--
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 13/17] nes: mark symbols static where possible
From: Christoph Hellwig @ 2016-10-13 19:56 UTC (permalink / raw)
To: linux-rdma-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <1476388576-22863-1-git-send-email-hch-jcswGhMUV9g@public.gmane.org>
Signed-off-by: Christoph Hellwig <hch-jcswGhMUV9g@public.gmane.org>
---
providers/nes/nes_umain.c | 2 +-
providers/nes/nes_uverbs.c | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/providers/nes/nes_umain.c b/providers/nes/nes_umain.c
index b4e0bc1..4b19dd7 100644
--- a/providers/nes/nes_umain.c
+++ b/providers/nes/nes_umain.c
@@ -60,7 +60,7 @@ long int page_size;
.device = d, \
.type = NETEFFECT_##t }
-struct {
+static struct {
unsigned vendor;
unsigned device;
enum nes_uhca_type type;
diff --git a/providers/nes/nes_uverbs.c b/providers/nes/nes_uverbs.c
index 287e773..f404ffd 100644
--- a/providers/nes/nes_uverbs.c
+++ b/providers/nes/nes_uverbs.c
@@ -759,7 +759,7 @@ int nes_upoll_cq_no_db_read(struct ibv_cq *cq, int num_entries, struct ibv_wc *e
/**
* nes_arm_cq
*/
-void nes_arm_cq(struct nes_ucq *nesucq, struct nes_uvcontext *nesvctx, int sol)
+static void nes_arm_cq(struct nes_ucq *nesucq, struct nes_uvcontext *nesvctx, int sol)
{
uint32_t cq_arm;
--
2.1.4
--
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 14/17] nes: use NULL instead of 0 for NULL pointers
From: Christoph Hellwig @ 2016-10-13 19:56 UTC (permalink / raw)
To: linux-rdma-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <1476388576-22863-1-git-send-email-hch-jcswGhMUV9g@public.gmane.org>
Signed-off-by: Christoph Hellwig <hch-jcswGhMUV9g@public.gmane.org>
---
providers/nes/nes_uverbs.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/providers/nes/nes_uverbs.c b/providers/nes/nes_uverbs.c
index f404ffd..7e7bb90 100644
--- a/providers/nes/nes_uverbs.c
+++ b/providers/nes/nes_uverbs.c
@@ -324,7 +324,7 @@ int nes_ima_upoll_cq(struct ibv_cq *cq, int num_entries, struct ibv_wc *entry)
uint32_t head;
uint32_t cq_size;
- volatile struct nes_hw_nic_cqe *cqe = 0;
+ volatile struct nes_hw_nic_cqe *cqe = NULL;
volatile struct nes_hw_nic_cqe *cqes;
struct nes_uqp *nesuqp = nesucq->udqp;
--
2.1.4
--
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 15/17] mlx5: mark symbols static where possible
From: Christoph Hellwig @ 2016-10-13 19:56 UTC (permalink / raw)
To: linux-rdma-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <1476388576-22863-1-git-send-email-hch-jcswGhMUV9g@public.gmane.org>
Signed-off-by: Christoph Hellwig <hch-jcswGhMUV9g@public.gmane.org>
---
providers/mlx5/mlx5.c | 2 +-
providers/mlx5/verbs.c | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/providers/mlx5/mlx5.c b/providers/mlx5/mlx5.c
index f24e504..4b05dc4 100644
--- a/providers/mlx5/mlx5.c
+++ b/providers/mlx5/mlx5.c
@@ -62,7 +62,7 @@
{ .vendor = PCI_VENDOR_ID_##v, \
.device = d }
-struct {
+static struct {
unsigned vendor;
unsigned device;
} hca_table[] = {
diff --git a/providers/mlx5/verbs.c b/providers/mlx5/verbs.c
index ef85a63..e288ebf 100644
--- a/providers/mlx5/verbs.c
+++ b/providers/mlx5/verbs.c
@@ -1184,7 +1184,7 @@ enum {
IBV_QP_INIT_ATTR_RX_HASH),
};
-struct ibv_qp *create_qp(struct ibv_context *context,
+static struct ibv_qp *create_qp(struct ibv_context *context,
struct ibv_qp_init_attr_ex *attr)
{
struct mlx5_create_qp cmd;
--
2.1.4
--
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 16/17] mlx5: use NULL instead of 0 for NULL pointers
From: Christoph Hellwig @ 2016-10-13 19:56 UTC (permalink / raw)
To: linux-rdma-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <1476388576-22863-1-git-send-email-hch-jcswGhMUV9g@public.gmane.org>
Signed-off-by: Christoph Hellwig <hch-jcswGhMUV9g@public.gmane.org>
---
providers/mlx5/mlx5.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/providers/mlx5/mlx5.c b/providers/mlx5/mlx5.c
index 4b05dc4..747e242 100644
--- a/providers/mlx5/mlx5.c
+++ b/providers/mlx5/mlx5.c
@@ -310,7 +310,7 @@ static void mlx5_local_cpu_set(struct ibv_device *ibdev, cpu_set_t *cpu_set)
p ++;
}
- word = strtoul(p, 0, 16);
+ word = strtoul(p, NULL, 16);
for (k = 0; word; ++k, word >>= 1)
if (word & 1)
--
2.1.4
--
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 17/17] mlx5: use proper ANSI-C prototypes for functions without arguments
From: Christoph Hellwig @ 2016-10-13 19:56 UTC (permalink / raw)
To: linux-rdma-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <1476388576-22863-1-git-send-email-hch-jcswGhMUV9g@public.gmane.org>
Signed-off-by: Christoph Hellwig <hch-jcswGhMUV9g@public.gmane.org>
---
providers/mlx5/cq.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/providers/mlx5/cq.c b/providers/mlx5/cq.c
index 2a53c89..ecfd6eb 100644
--- a/providers/mlx5/cq.c
+++ b/providers/mlx5/cq.c
@@ -384,7 +384,7 @@ static enum ibv_wc_status mlx5_handle_error_cqe(struct mlx5_err_cqe *cqe)
}
#if defined(__x86_64__) || defined (__i386__)
-static inline unsigned long get_cycles()
+static inline unsigned long get_cycles(void)
{
uint32_t low, high;
uint64_t val;
@@ -394,7 +394,7 @@ static inline unsigned long get_cycles()
return val;
}
-static void mlx5_stall_poll_cq()
+static void mlx5_stall_poll_cq(void)
{
int i;
--
2.1.4
--
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: [PATCH 01/32] libibumad: include umad_str.h in umad_str.c
From: Hal Rosenstock @ 2016-10-13 19:58 UTC (permalink / raw)
To: Christoph Hellwig, linux-rdma-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <1476381095-20041-2-git-send-email-hch-jcswGhMUV9g@public.gmane.org>
On 10/13/2016 1:51 PM, Christoph Hellwig wrote:
> So that we have the prototypes available in the implementation file
> and do get proper type checking for them.
>
> Signed-off-by: Christoph Hellwig <hch-jcswGhMUV9g@public.gmane.org>
Reviewed-by: Hal Rosenstock <hal-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
--
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
* Re: [PATCH 02/32] libibumad: create sysfs.h instead of using extern declarations in umad.c
From: Hal Rosenstock @ 2016-10-13 19:58 UTC (permalink / raw)
To: Christoph Hellwig, linux-rdma-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <1476381095-20041-3-git-send-email-hch-jcswGhMUV9g@public.gmane.org>
On 10/13/2016 1:51 PM, Christoph Hellwig wrote:
> So that we have the prototypes available in the implementation file
> and do get proper type checking for them.
>
> Signed-off-by: Christoph Hellwig <hch-jcswGhMUV9g@public.gmane.org>
Reviewed-by: Hal Rosenstock <hal-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
> ---
> libibumad/sysfs.c | 1 +
> libibumad/sysfs.h | 10 ++++++++++
> libibumad/umad.c | 7 +------
> 3 files changed, 12 insertions(+), 6 deletions(-)
> create mode 100644 libibumad/sysfs.h
>
> diff --git a/libibumad/sysfs.c b/libibumad/sysfs.c
> index 011e411..83f89b4 100644
> --- a/libibumad/sysfs.c
> +++ b/libibumad/sysfs.c
> @@ -44,6 +44,7 @@
> #include <byteswap.h>
> #include <netinet/in.h>
> #include <infiniband/arch.h>
> +#include "sysfs.h"
>
> static int ret_code(void)
> {
> diff --git a/libibumad/sysfs.h b/libibumad/sysfs.h
> new file mode 100644
> index 0000000..7f8cbfe
> --- /dev/null
> +++ b/libibumad/sysfs.h
> @@ -0,0 +1,10 @@
> +#ifndef _UMAD_SYSFS_H
> +#define _UMAD_SYSFS_H
What about copyright for this new file ?
> +
> +extern int sys_read_string(const char *dir_name, const char *file_name, char *str, int len);
> +extern int sys_read_guid(const char *dir_name, const char *file_name, uint64_t * net_guid);
> +extern int sys_read_gid(const char *dir_name, const char *file_name, uint8_t * gid);
> +extern int sys_read_uint64(const char *dir_name, const char *file_name, uint64_t * u);
> +extern int sys_read_uint(const char *dir_name, const char *file_name, unsigned *u);
> +
> +#endif /* _UMAD_SYSFS_H */
> diff --git a/libibumad/umad.c b/libibumad/umad.c
> index 0c969f1..8860b99 100644
> --- a/libibumad/umad.c
> +++ b/libibumad/umad.c
> @@ -53,6 +53,7 @@
> #define IB_OPENIB_OUI (0x001405)
>
> #include <valgrind/memcheck.h>
> +#include "sysfs.h"
>
> typedef struct ib_user_mad_reg_req {
> uint32_t id;
> @@ -77,12 +78,6 @@ struct ib_user_mad_reg_req2 {
> uint8_t reserved[3];
> };
>
> -extern int sys_read_string(const char *dir_name, const char *file_name, char *str, int len);
> -extern int sys_read_guid(const char *dir_name, const char *file_name, uint64_t * net_guid);
> -extern int sys_read_gid(const char *dir_name, const char *file_name, uint8_t * gid);
> -extern int sys_read_uint64(const char *dir_name, const char *file_name, uint64_t * u);
> -extern int sys_read_uint(const char *dir_name, const char *file_name, unsigned *u);
> -
> #define IBWARN(fmt, args...) fprintf(stderr, "ibwarn: [%d] %s: " fmt "\n", getpid(), __func__, ## args)
>
> #define TRACE if (umaddebug) IBWARN
>
--
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
* Re: [PATCH 04/32] libibumad: mark symbols as static where possible
From: Hal Rosenstock @ 2016-10-13 19:58 UTC (permalink / raw)
To: Christoph Hellwig, linux-rdma-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <1476381095-20041-5-git-send-email-hch-jcswGhMUV9g@public.gmane.org>
On 10/13/2016 1:51 PM, Christoph Hellwig wrote:
> Signed-off-by: Christoph Hellwig <hch-jcswGhMUV9g@public.gmane.org>
Reviewed-by: Hal Rosenstock <hal-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
--
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
* Re: [PATCH 03/32] libibumad: use NULL instead of 0 for NULL pointers
From: Hal Rosenstock @ 2016-10-13 19:58 UTC (permalink / raw)
To: Christoph Hellwig, linux-rdma-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <1476381095-20041-4-git-send-email-hch-jcswGhMUV9g@public.gmane.org>
On 10/13/2016 1:51 PM, Christoph Hellwig wrote:
> Signed-off-by: Christoph Hellwig <hch-jcswGhMUV9g@public.gmane.org>
Reviewed-by: Hal Rosenstock <hal-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
--
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
* Re: [PATCH rdma-next 00/10] Hardware tag matching support
From: Christoph Hellwig @ 2016-10-13 20:02 UTC (permalink / raw)
To: Jason Gunthorpe
Cc: Hefty, Sean, Christoph Hellwig, Doug Ledford, Leon Romanovsky,
linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
In-Reply-To: <20161013195605.GA8077-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org>
On Thu, Oct 13, 2016 at 01:56:05PM -0600, Jason Gunthorpe wrote:
> A bigger objection I have is that much of this new stuff is not
> adaquately documented.
>
> Tag matching is a great example. It looks like this introduces many new
> headers into the on-wire protocol. Where are these headers documented?
> Where is the spec for the on-wire format? Where is the implementation
> for rxe?
Agreed on that part. For pon the wire changes we need the hardware
standards organization to act. IBTA for IB and RoCE and IETF for iWarp,
although not even having an active WG for iWarp there is a major problem
at the moment.
--
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
* Re: [PATCH rdma-next 00/10] Hardware tag matching support
From: Doug Ledford @ 2016-10-13 20:30 UTC (permalink / raw)
To: Christoph Hellwig, Jason Gunthorpe
Cc: Hefty, Sean, Leon Romanovsky,
linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
In-Reply-To: <20161013200208.GA8998-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>
[-- Attachment #1.1: Type: text/plain, Size: 1449 bytes --]
On 10/13/2016 4:02 PM, Christoph Hellwig wrote:
> On Thu, Oct 13, 2016 at 01:56:05PM -0600, Jason Gunthorpe wrote:
>> A bigger objection I have is that much of this new stuff is not
>> adaquately documented.
>>
>> Tag matching is a great example. It looks like this introduces many new
>> headers into the on-wire protocol. Where are these headers documented?
>> Where is the spec for the on-wire format? Where is the implementation
>> for rxe?
>
> Agreed on that part. For pon the wire changes we need the hardware
> standards organization to act. IBTA for IB and RoCE and IETF for iWarp,
> although not even having an active WG for iWarp there is a major problem
> at the moment.
>
I think there are some fundamental questions here about what this is.
As I understand it (and Mellanox can chime in if I'm wrong), tag
matching already exists on the wire, it's just implemented above the
level of the link layer transmissions. This enables a hardware
accelerator for something that is already done at an upper layer, so the
implementation is mostly defined by the defacto existing use in upper
layer applications, not by the IBTA. I believe this could roughly be
considered analogous to a network driver enabling a layer3 or layer4
accelerator. It's not a layer1 thing, so the IBTA is not appropriate to
review it.
--
Doug Ledford <dledford-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
GPG Key ID: 0E572FDD
[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 884 bytes --]
^ permalink raw reply
* Re: [PATCH rdma-next 00/10] Hardware tag matching support
From: Jason Gunthorpe @ 2016-10-13 20:47 UTC (permalink / raw)
To: Doug Ledford
Cc: Christoph Hellwig, Hefty, Sean, Leon Romanovsky,
linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
In-Reply-To: <a3495685-69c7-26a7-ba97-9761848535a8-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
On Thu, Oct 13, 2016 at 04:30:10PM -0400, Doug Ledford wrote:
> On 10/13/2016 4:02 PM, Christoph Hellwig wrote:
> > On Thu, Oct 13, 2016 at 01:56:05PM -0600, Jason Gunthorpe wrote:
> >> A bigger objection I have is that much of this new stuff is not
> >> adaquately documented.
> >>
> >> Tag matching is a great example. It looks like this introduces many new
> >> headers into the on-wire protocol. Where are these headers documented?
> >> Where is the spec for the on-wire format? Where is the implementation
> >> for rxe?
> >
> > Agreed on that part. For pon the wire changes we need the hardware
> > standards organization to act. IBTA for IB and RoCE and IETF for iWarp,
> > although not even having an active WG for iWarp there is a major problem
> > at the moment.
> >
>
> I think there are some fundamental questions here about what this is.
> As I understand it (and Mellanox can chime in if I'm wrong), tag
> matching already exists on the wire, it's just implemented above the
> level of the link layer transmissions. This enables a hardware
> accelerator for something that is already done at an upper layer, so the
> implementation is mostly defined by the defacto existing use in upper
> layer applications, not by the IBTA. I believe this could roughly be
> considered analogous to a network driver enabling a layer3 or layer4
> accelerator. It's not a layer1 thing, so the IBTA is not appropriate to
> review it.
Maybe, but that doesn't change my original point. This is clearly not
defined well enough in the patches alone for any other vendor to
implement it.
If we can't even reach a basic threshold of understanding it is
pointless to expect other vendors to chime in on the design.
I'm starting to wonder if we should insist on an implementation of all
this new stuff in rxe before accepting. At least then we have code as
documentation ...
Jason
--
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
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