Re: [PATCH] IB/ipoib: Expose ioctl command to retrieve SGID of a given socket

On 06/01/2016 13:03, Yuval Shaia wrote: > On Thu, Dec 31, 2015 at 05:34:03PM +0200, Haggai Eran wrote: >>> + sock = sockfd_lookup(fd, ); >>> + if (IS_ERR_OR_NULL(sock)) >>> + return -EINVAL; >>> + >>> + inetsock = inet_sk(so

Re: SoftRoCE V1

On 29/12/2015 18:01, Wenda Ni wrote: > Hi experts, > > We have several Mellanox RoCE V1 NIC cards, and would like to try > communicating with SoftRoCE V1. > > We are using branch rxe-3.0 from https://github.com/SoftRoCE/rxe-dev > according to the Soft-RoCE README Rev 1.0 issued from Mellanox

Re: [PATCH 2/6] IB/uapi: expose uverbs send WR flags

+ IB_UVERBS_SEND_END = (1 << 5), Do you think we need a check in ib_uverbs_post_send to see only user flags are passed? I think it would be safer. Other than that, Reviewed-by: Haggai Eran <hagg...@mellanox.com> -- To unsubscribe from this list

Re: [PATCH 6/6] IB/uapi: expose device capability flags

On 24/12/2015 16:39, Christoph Hellwig wrote: > Expose the device capability flags which can be queried through uverbs in > the uapi headers. > > Signed-off-by: Christoph Hellwig > --- > include/rdma/ib_verbs.h | 94 > +++- >

Re: [PATCH] IB/ipoib: Expose ioctl command to retrieve SGID of a given socket

On 31/12/2015 16:41, Yuval Shaia wrote: > To support security applications, that need to filter out connections based > on SGID, an ioctl command to retrieve SGID of a given socket is added. Could you elaborate on the security applications? How do you see this ioctl being used? > diff --git

Re: [PATCH 10/10] IB: remove the unused usecnt field from struct ib_mr

On 20/12/2015 11:31, Sagi Grimberg wrote: > >>> diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h >>> index 284916d..e45776e 100644 >>> --- a/include/rdma/ib_verbs.h >>> +++ b/include/rdma/ib_verbs.h >>> @@ -1306,7 +1306,6 @@ struct ib_mr { >>> u64 iova; >>> u32

Re: [PATCH for-next V2 2/5] IB/core: Add ib_is_udata_cleared

On 15/12/2015 20:30, Matan Barak wrote: > Extending core and vendor verb commands require us to check that the > unknown part of the user's given command is all zeros. > Adding ib_is_udata_cleared in order to do so. > > Signed-off-by: Matan Barak <mat...@mellanox.com> Rev

Re: [PATCH for-next V1 2/5] IB/core: Add ib_is_udata_cleared

On 14/12/2015 18:31, Matan Barak wrote: > I'm not sure regarding the string.c location, as it deals with user > buffers, but in order not to > be dependent on this, I'll change this code to the following. > > static inline bool ib_is_udata_cleared(struct ib_udata *udata, >

Re: [PATCH for-next V1 2/5] IB/core: Add ib_is_udata_cleared

On 10/12/2015 19:29, Matan Barak wrote: > On Thu, Dec 10, 2015 at 5:20 PM, Haggai Eran <hagg...@mellanox.com> wrote: >> On 10/12/2015 16:59, Matan Barak wrote: >>> On Mon, Dec 7, 2015 at 3:18 PM, Haggai Eran <hagg...@mellanox.com> wrote: >>>>

Re: [PATCH 04/37] IB/rdmavt: Add ib core device attributes to rvt driver params list

On 07/12/2015 22:43, Dennis Dalessandro wrote: > struct rvt_dev_info { > + /* > + * Prior to calling for registration the driver will be responsible for > + * allocating space for this structure. The driver will also need to > + * allocate space for any private device or per

Re: [PATCH 04/37] IB/rdmavt: Add ib core device attributes to rvt driver params list

On 07/12/2015 22:43, Dennis Dalessandro wrote: > + /* > + * Drivers will need to support a number of notifications to rvt in > + * accordance with certain events. This structure should contain a mask > + * of the supported events. Such events that the rvt may need to know > +

Re: [PATCH 01/37] IB/rdmavt: Create module framework and handle driver registration

On 07/12/2015 22:43, Dennis Dalessandro wrote: > +struct rvt_dev_info { > + struct ib_device ibdev; > + int (*port_callback)(struct ib_device *, u8, struct kobject *); > + > + /* > + * TODO: > + * need to reflect module parameters that may vary by dev > + */ > +};

Re: [PATCH 02/37] IB/rdmavt: Consolidate dma ops in rdmavt.

A few nits: On 07/12/2015 22:43, Dennis Dalessandro wrote: > +static int rvt_map_sg(struct ib_device *dev, struct scatterlist *sgl, > + int nents, enum dma_data_direction direction) > +{ > + struct scatterlist *sg; > + u64 addr; > + int i; > + int ret = nents; >

Re: [PATCH 04/37] IB/rdmavt: Add ib core device attributes to rvt driver params list

On 10/12/2015 15:25, Dennis Dalessandro wrote: > On Thu, Dec 10, 2015 at 02:26:11PM +0200, Haggai Eran wrote: >> On 07/12/2015 22:43, Dennis Dalessandro wrote: >>> +/* >>> + * Drivers will need to support a number of notifications to rvt in >>> +

Re: [PATCH for-next V1 2/5] IB/core: Add ib_is_udata_cleared

On 10/12/2015 16:59, Matan Barak wrote: > On Mon, Dec 7, 2015 at 3:18 PM, Haggai Eran <hagg...@mellanox.com> wrote: >> On 12/03/2015 05:44 PM, Matan Barak wrote: >>> Extending core and vendor verb commands require us to check that the >>> unknown part of the

Re: [PATCH for-next V2 07/11] IB/core: Validate route in ib_init_ah_from_wc and ib_init_ah_from_path

On 12/03/2015 03:47 PM, Matan Barak wrote: > +static int addr_resolve_neigh(struct dst_entry *dst, > + const struct sockaddr *dst_in, > + struct rdma_dev_addr *addr) > +{ > + if (dst->dev->flags & IFF_LOOPBACK) { > + int ret; > +

Re: [PATCH libibverbs] init.c: increase sysfs read buffer size to 16

from ibv_devinfo when > reading usNIC devices. > > This commit therefore increases the buffer size to 16, which is long > enough to read the usNIC node_type value. Reviewed-by: Haggai Eran <hagg...@mellanox.com> -- To unsubscribe from this list: send the line "unsubscribe l

Re: [PATCH libmlx5 V1 6/6] Add always_inline check

On 12/03/2015 06:02 PM, Matan Barak wrote: > Always inline isn't supported by every compiler. Adding it to > configure.ac in order to support it only when possible. > Inline other poll_one data path functions in order to eliminate > "ifs". > > Signed-off-by: Matan Barak >

Re: [PATCH for-next V1 2/5] IB/core: Add ib_is_udata_cleared

On 12/03/2015 05:44 PM, Matan Barak wrote: > Extending core and vendor verb commands require us to check that the > unknown part of the user's given command is all zeros. > Adding ib_is_udata_cleared in order to do so. > Why not copy the data into kernel space and run memchr_inv() on it? -- To

Re: [PATCH] infiniband:core:Add needed error path in cm_init_av_by_path

On Friday, December 4, 2015 8:02 PM, Nicholas Krause <xerofo...@gmail.com> wrote: > To: dledf...@redhat.com > Cc: sean.he...@intel.com; hal.rosenst...@gmail.com; Haggai Eran; > jguntho...@obsidianresearch.com; Matan Barak; yun.w...@profitbricks.com; > ted.h@oracle.com

Re: [PATCH v1 10/10] IB/iser: Support the remote invalidation exception

On 24/11/2015 18:23, Sagi Grimberg wrote: > From: Jenny Derzhavetz > > Declare that we support remote invalidation in case we are: > 1. using Fastreg method > 2. always registering memory. > > Detect the invalidated rkey from the work completion info so we > won't

Re: [PATCH 3/3] IB/core: constify mmu_notifier_ops structures

On 30/11/2015 00:02, Julia Lawall wrote: > This mmu_notifier_ops structure is never modified, so declare it as > const, like the other mmu_notifier_ops structures. > > Done with the help of Coccinelle. > > Signed-off-by: Julia Lawall <julia.law...@lip6.fr> Review

Re: [PATCH] IB/cma: Add a missing rcu_read_unlock()

On 20/11/2015 21:04, Bart Van Assche wrote: > Ensure that validate_ipv4_net_dev() calls rcu_read_unlock() if > fib_lookup() fails. Detected by sparse. Compile-tested only. > > Fixes: "IB/cma: Validate routing of incoming requests" (commit f887f2ac87c2). > Cc: Haggai E

Re: [PATCH ib-next 2/3] IB/core: IB/core: Allow legacy verbs through extended interfaces

On 08/11/2015 17:04, Matan Barak wrote: >> @@ -704,6 +719,10 @@ static ssize_t ib_uverbs_write(struct file *filp, const >> char __user *buf, >> > } >> > >> > command = hdr.command & IB_USER_VERBS_CMD_COMMAND_MASK; >> > + if (verify_command_mask(ib_dev, command)) { >> > +

Re: RFC rdma cgroup

On 03/11/2015 21:11, Parav Pandit wrote: > So it looks like below, > #cat rdma.resources.verbs.list > Output: > mlx4_0 uctx ah pd cq mr mw srq qp flow > mlx4_1 uctx ah pd cq mr mw srq qp flow rss_wq What happens if you set a limit of rss_wq to mlx4_0 in this example? Would it fail? I think it

Re: RFC rdma cgroup

On 29/10/2015 20:46, Parav Pandit wrote: > On Thu, Oct 29, 2015 at 8:27 PM, Haggai Eran <hagg...@mellanox.com> wrote: >> On 28/10/2015 10:29, Parav Pandit wrote: >>> 3. Resources are not defined by the RDMA cgroup. Resources are defined >>> by RDMA/IB subsystem an

Re: RFC rdma cgroup

On 28/10/2015 10:29, Parav Pandit wrote: > 3. Resources are not defined by the RDMA cgroup. Resources are defined > by RDMA/IB subsystem and optionally by HCA vendor device drivers. > Rationale: This allows rdma cgroup to remain constant while RDMA/IB > subsystem can evolve without the need of

Re: [PATCH v6 0/4] Add network namespace support in the RDMA-CM

On 22/10/2015 06:37, Doug Ledford wrote: > On 08/27/2015 10:29 AM, Haggai Eran wrote: >> Hi, >> >> Now that the code for demuxing requests is inside rdma_cm, here are the >> patches >> to add InfiniBand network namespace again. >> >> Changes from

[PATCH v7 1/4] IB/addr: Pass network namespace as a parameter

upport is added on more levels. Signed-off-by: Haggai Eran <hagg...@mellanox.com> Signed-off-by: Yotam Kenneth <yota...@mellanox.com> Signed-off-by: Shachar Raindel <rain...@mellanox.com> Signed-off-by: Guy Shapiro <gu...@mellanox.com> --- drivers/infiniband/core/addr.c |

[PATCH v7 0/4] Add network namespace support in the RDMA-CM

as a parameter IB/cma: Add support for network namespaces IB/ucma: Take the network namespace from the process Haggai Eran (1): IB/cma: Separate port allocation to network namespaces drivers/infiniband/core/addr.c | 17 +-- drivers/infiniband/core/cma.c

[PATCH v7 3/4] IB/cma: Add support for network namespaces

alling rdma_destroy_id. In order to preserve the current behavior init_net is passed when calling from other modules. Signed-off-by: Guy Shapiro <gu...@mellanox.com> Signed-off-by: Haggai Eran <hagg...@mellanox.com> Signed-off-by: Yotam Kenneth <yota...@mellanox.com> Signed-off-b

[PATCH v7 4/4] IB/ucma: Take the network namespace from the process

From: Guy Shapiro <gu...@mellanox.com> Add support for network namespaces from user space. This is done by passing the network namespace of the process instead of init_net. Signed-off-by: Haggai Eran <hagg...@mellanox.com> Signed-off-by: Yotam Kenneth <yota...@mellanox.

Re: [PATCH v3 for-next 1/7] IB/core: Extend ib_uverbs_create_qp

drivers/infiniband/core/uverbs.h | 1 + > drivers/infiniband/core/uverbs_cmd.c | 254 > +- > drivers/infiniband/core/uverbs_main.c | 1 + > include/uapi/rdma/ib_user_verbs.h | 26 > 4 files changed, 217 insertions(+), 65 deletions(-) &

Re: [PATCH v2 for-next 1/7] IB/core: Extend ib_uverbs_create_qp

On 21/10/2015 16:46, eran ben elisha wrote: >>> +ssize_t ib_uverbs_create_qp(struct ib_uverbs_file *file, >>> >> + struct ib_device *ib_dev, >>> >> + const char __user *buf, int in_len, >>> >> + int out_len) >>> >> +{ >>> >> +

Re: [PATCH] IB/cma: Use inner P_Key to determine netdev

On 20/10/2015 19:44, Jason Gunthorpe wrote: > On Tue, Oct 20, 2015 at 09:45:27AM +0300, Haggai Eran wrote: >> On 19/10/2015 21:19, Jason Gunthorpe wrote: >>> On Mon, Oct 19, 2015 at 09:09:25PM +0300, Haggai Eran wrote: >>>> When discussing the patches to demux i

Re: [PATCH v2 for-next 1/7] IB/core: Extend ib_uverbs_create_qp

On 21/10/2015 12:53, Sagi Grimberg wrote: > On 10/15/2015 2:44 PM, Eran Ben Elisha wrote: >> ib_uverbs_ex_create_qp follows the extension verbs >> mechanism. New features (for example, QP creation flags >> field which is added in a downstream patch) could used >> via user-space libraries without

Re: [PATCH v2 for-next 1/7] IB/core: Extend ib_uverbs_create_qp

On 15/10/2015 14:44, Eran Ben Elisha wrote: > ib_uverbs_ex_create_qp follows the extension verbs > mechanism. New features (for example, QP creation flags > field which is added in a downstream patch) could used > via user-space libraries without breaking the ABI. > > Signed-off-by: Eran Ben

Re: [PATCH v2 for-next 2/7] IB/core: Allow setting create flags in QP init attribute

ret = -EINVAL; > goto err_put; > } > FWIW Reviewed-by: Haggai Eran <hagg...@mellanox.com> -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html

[PATCH v1] IB/cma: Use inner P_Key to determine netdev

during the GSI QP initialization. Change the rdma_cm code to look at the P_Key value that is part of the packet payload as a workaround. Once the drivers are fixed this patch can be reverted. Fixes: 4c21b5bcef73 ("IB/cma: Add net_dev and private data checks to RDMA CM") Signed-off-by: H

Re: [PATCH v1] IB/cma: Use inner P_Key to determine netdev

On 20/10/2015 10:20, Doug Ledford wrote: > On 10/20/2015 02:53 AM, Haggai Eran wrote: >> When discussing the patches to demux ids in rdma_cm instead of ib_cm, it >> was decided that it is best to use the P_Key value in the packet headers. >> However, the mlx5 and ipath drivers

Re: [PATCH] IB/cma: Use inner P_Key to determine netdev

On 19/10/2015 21:19, Jason Gunthorpe wrote: > On Mon, Oct 19, 2015 at 09:09:25PM +0300, Haggai Eran wrote: >> When discussing the patches to demux ids in rdma_cm instead of ib_cm, it >> was decided that it is best to use the P_Key value in the packet headers >> [1].

Re: [PATCH 0/6] IB/mad: Support devices taking pkey_index from the GSI QP

On 10/14/2015 08:54 PM, Jason Gunthorpe wrote: > What this series is doing (src QPN != 1) is absolutely not complainant > with the spec. Hal convinced me that the spec implicitly requires that. Originally I thought that it was allowed, based this paragraph from Section 13.5.1 (MAD Interfaces): >

[PATCH] IB/cma: Use inner P_Key to determine netdev

of the packet payload as a workaround. Once the drivers are fixed this patch can be reverted. Fixes: 4c21b5bcef73 ("IB/cma: Add net_dev and private data checks to RDMA CM") Signed-off-by: Haggai Eran <hagg...@mellanox.com> --- drivers/infiniband/core/cma.c | 4 ++-- 1 file chang

[PATCH 0/6] IB/mad: Support devices taking pkey_index from the GSI QP

ltiple QPs on the same ib_mad_qp_info struct, and patch 6 adds a P_Key change event handler to update the table when the SM changes the available P_Keys. [1] mlx5: Fix incorrect wc pkey_index assignment for GSI messages http://www.spinics.net/lists/linux-rdma/msg28374.html Regards, Haggai Haggai Eran (6)

[PATCH 2/6] IB/mad: Add QP parameters to ib_mad_qp_info

In preparation for having an array of QPs in each ib_mad_qp_info, add the qp_type and qp_num parameters to the ib_mad_qp_info struct so that clients won't need to access the QPs themselves for this information. Signed-off-by: Haggai Eran <hagg...@mellanox.com> --- drivers/infiniband/core

[PATCH 3/6] IB/core: Add capability bit to tell whether per-WR P_Key change in GSI is supported

<infinip...@intel.com> Cc: Eli Cohen <e...@mellanox.com> Signed-off-by: Haggai Eran <hagg...@mellanox.com> --- drivers/infiniband/hw/mlx5/main.c| 1 + drivers/staging/rdma/ipath/ipath_verbs.c | 1 + include/rdma/ib_verbs.h | 2 ++ 3 files changed, 4 inserti

[PATCH 5/6] IB/mad: Create multiple QPs for supporting different P_Keys

-by: Haggai Eran <hagg...@mellanox.com> --- drivers/infiniband/core/mad.c | 404 +++-- drivers/infiniband/core/mad_priv.h | 4 +- 2 files changed, 299 insertions(+), 109 deletions(-) diff --git a/drivers/infiniband/core/mad.c b/drivers/infiniband/core/mad.c

[PATCH 1/6] IB/mad: Use helpers to get ib_device and ib_pd from ib_mad_agent

ib_mad_agent currently exposes an ib_qp and an ib_device unnecessarily. Replace these fields with a single ib_pd, and use helper functions to get the device and pd instead of accessing the fields directly. Signed-off-by: Haggai Eran <hagg...@mellanox.com> --- drivers/infiniband/core/a

[PATCH 6/6] IB/mad: P_Key change event handler

be added later on. Signed-off-by: Haggai Eran <hagg...@mellanox.com> --- drivers/infiniband/core/mad.c | 51 -- drivers/infiniband/core/mad_priv.h | 2 ++ 2 files changed, 51 insertions(+), 2 deletions(-) diff --git a/drivers/infiniband/core/m

[PATCH 4/6] IB/mad: Use a SRQ for receiving GMPs

As a preparation for supporting multiple transmission QPs for each GSI QP, add a SRQ that will be used for all the receive buffers of these QPs. Signed-off-by: Haggai Eran <hagg...@mellanox.com> --- drivers/infiniband/core/mad.c | 58 ++ d

[PATCH] [RESEND] IB/cma: Accept connection without a valid netdev on RoCE

the netdev enforcement for each incoming connections when the link layer is RoCE. Fixes: 4c21b5bcef73 ("IB/cma: Add net_dev and private data checks to RDMA CM") Reported-by: Kamal Heib <kam...@mellanox.com> Signed-off-by: Haggai Eran <hagg...@mellanox.com> --- I accid

Re: [PATCH 3.2] IB/qib: Change lkey table allocation to support more MRs

On 02/10/2015 16:10, Marciniszyn, Mike wrote: >>> The lkey table is allocated with with a get_user_pages() with an >> Don't you mean __get_free_pages? >> > > I was a nit in the original upstream commit. > > I don’t think it is a big deal since the patch context clarifies. I agree it's not a big

Re: [PATCH 3.2] IB/qib: Change lkey table allocation to support more MRs

On 29/09/2015 17:51, Mike Marciniszyn wrote: > The lkey table is allocated with with a get_user_pages() with an Don't you mean __get_free_pages? Regards, Haggai > order based on a number of index bits from a module parameter. > > The underlying kernel code cannot allocate that many contiguous

Re: shrink struct ib_send_wr V4

On 13/09/2015 18:13, Christoph Hellwig wrote: > This series shrinks the WR size by splitting out the different WR > types. > > Patch number one is too large for the mailinglist, so if you didn't > get it grab it here: > > >

RoCE RDMA CM handling after the demux patches

Hi, I noticed that the "Demux IB CM requests in the rdma_cm module" patches, while intended to only apply to InfiniBand connections, accidentally also affect RoCE connections. The patches used a query to IPoIB for the netdev associated with a request, and RoCE devices therefore fail to answer

[PATCH] IB/cma: Potential NULL dereference in cma_id_from_event

If the lookup of a listening ID failed for an AF_IB request, the code would try to call dev_put() on a NULL net_dev. Fixes: be688195bd08 ("IB/cma: Fix net_dev reference leak with failed requests") Reported-by: Dan Carpenter <dan.carpen...@oracle.com> Signed-off-by: H

Re: [PATCH 0/7] devcg: device cgroup extension for rdma resource

On 15/09/2015 06:45, Jason Gunthorpe wrote: > No, I'm saying the resource pool is *well defined* and *fixed* by each > hardware. > > The only question is how do we expose the N resource limits, the list > of which is totally vendor specific. I don't see why you say the limits are vendor

Re: [PATCH 5/7] devcg: device cgroup's extension for RDMA resource.

On 07/09/2015 23:38, Parav Pandit wrote: > +void devcgroup_rdma_uncharge_resource(struct ib_ucontext *ucontext, > + enum devcgroup_rdma_rt type, int num) > +{ > + struct dev_cgroup *dev_cg, *p; > + struct task_struct *ctx_task; > + > + if (!num) > +

Re: [PATCH 6/7] devcg: Added support to use RDMA device cgroup.

On 07/09/2015 23:38, Parav Pandit wrote: > +static void init_ucontext_lists(struct ib_ucontext *ucontext) > +{ > + INIT_LIST_HEAD(>pd_list); > + INIT_LIST_HEAD(>mr_list); > + INIT_LIST_HEAD(>mw_list); > + INIT_LIST_HEAD(>cq_list); > + INIT_LIST_HEAD(>qp_list); > +

Re: [PATCH 5/7] devcg: device cgroup's extension for RDMA resource.

On 07/09/2015 23:38, Parav Pandit wrote: > +/* RDMA resources from device cgroup perspective */ > +enum devcgroup_rdma_rt { > + DEVCG_RDMA_RES_TYPE_UCTX, > + DEVCG_RDMA_RES_TYPE_CQ, > + DEVCG_RDMA_RES_TYPE_PD, > + DEVCG_RDMA_RES_TYPE_AH, > + DEVCG_RDMA_RES_TYPE_MR, > +

Re: [PATCH 4/7] devcg: Added rdma resource tracker object per task

On 08/09/2015 10:04, Parav Pandit wrote: > On Tue, Sep 8, 2015 at 11:18 AM, Haggai Eran <hagg...@mellanox.com> wrote: >> On 07/09/2015 23:38, Parav Pandit wrote: >>> @@ -2676,7 +2686,7 @@ static inline int thread_group_empty(struct >>> task_struct *p) >

Re: [PATCH 0/7] devcg: device cgroup extension for rdma resource

On 07/09/2015 23:38, Parav Pandit wrote: > Currently user space applications can easily take away all the rdma > device specific resources such as AH, CQ, QP, MR etc. Due to which other > applications in other cgroup or kernel space ULPs may not even get chance > to allocate any rdma resources. >

Re: [PATCH 6/7] devcg: Added support to use RDMA device cgroup.

On 08/09/2015 13:22, Parav Pandit wrote: > On Tue, Sep 8, 2015 at 2:10 PM, Haggai Eran <hagg...@mellanox.com> wrote: >> On 07/09/2015 23:38, Parav Pandit wrote: >>> +static void init_ucontext_lists(struct ib_ucontext *ucontext) >>> +{ >>> + INIT_LIS

Re: [PATCH 5/7] devcg: device cgroup's extension for RDMA resource.

On 08/09/2015 13:50, Parav Pandit wrote: > On Tue, Sep 8, 2015 at 2:06 PM, Haggai Eran <hagg...@mellanox.com> wrote: >> On 07/09/2015 23:38, Parav Pandit wrote: >>> +void devcgroup_rdma_uncharge_resource(struct ib_ucontext *ucontext, >>> +

Re: [PATCH 5/7] devcg: device cgroup's extension for RDMA resource.

On 08/09/2015 13:18, Parav Pandit wrote: >> > >>> >> + * RDMA resource limits are hierarchical, so the highest configured >>> >> limit of >>> >> + * the hierarchy is enforced. Allowing resource limit configuration to >>> >> default >>> >> + * cgroup allows fair share to kernel space ULPs as

Re: [PATCH 3/7] devcg: Added infrastructure for rdma device cgroup.

On 07/09/2015 23:38, Parav Pandit wrote: > diff --git a/include/linux/device_cgroup.h b/include/linux/device_cgroup.h > index 8b64221..cdbdd60 100644 > --- a/include/linux/device_cgroup.h > +++ b/include/linux/device_cgroup.h > @@ -1,6 +1,57 @@ > +#ifndef _DEVICE_CGROUP > +#define _DEVICE_CGROUP >

Re: [PATCH 4/7] devcg: Added rdma resource tracker object per task

On 07/09/2015 23:38, Parav Pandit wrote: > @@ -2676,7 +2686,7 @@ static inline int thread_group_empty(struct task_struct > *p) > * Protects ->fs, ->files, ->mm, ->group_info, ->comm, keyring > * subscriptions and synchronises with wait4(). Also used in procfs. Also > * pins the final

[PATCH] libmlx5: Implement query device extended verb

Simply pass the extended query device verb to back to libiverbs, in order to support it. Also share some code with the legacy query device verb. Signed-off-by: Haggai Eran <hagg...@mellanox.com> --- src/mlx5.c | 1 + src/mlx5.h | 3 +++ src/verbs.

[PATCH v1 1/3] Add support for extended query device capabilities

From: Eli Cohen <e...@mellanox.com> Add the verb ibv_query_device_ex which is extensible and allows following commits to add new features to define additional properties. Cc: Moshe Lazer <mos...@mellanox.com> Signed-off-by: Eli Cohen <e...@mellanox.com> Signed-off-by:

[PATCH v1 0/3] libibverbs: On-demand paging support

a const pointer instead of a by-value struct. * check that the application has enough space for ODP capabilities in the provided ibv_device_attr_ex struct. Eli Cohen (1): Add support for extended query device capabilities Haggai Eran (1): Add on-demand paging support Majd Dibbiny (1

[PATCH v1 3/3] libibverbs/examples: Support odp in rc_pingpong

From: Majd Dibbiny <m...@mellanox.com> Signed-off-by: Majd Dibbiny <m...@mellanox.com> Signed-off-by: Haggai Eran <hagg...@mellanox.com> --- examples/rc_pingpong.c | 31 +-- 1 file changed, 29 insertions(+), 2 deletions(-) diff --git a/exampl

[PATCH v1 2/3] Add on-demand paging support

the IBV_ACCESS_ON_DEMAND access flag to allow registration of on-demand paging enabled memory regions. Signed-off-by: Shachar Raindel <rain...@mellanox.com> Signed-off-by: Majd Dibbiny <m...@mellanox.com> Signed-off-by: Haggai Eran <hagg...@mellanox.com> --- examples/devinfo

Re: [PATCH 2/3] Add on-demand paging support

On 02/09/2015 22:17, Sagi Grimberg wrote: > On 8/27/2015 6:22 PM, Haggai Eran wrote: >> On-demand paging feature allows registering memory regions without >> pinning >> their pages. Unfortunately the feature doesn't work together will all >> transports and all

[PATCH] IB/mlx5: avoid destroying a NULL mr in reg_user_mr error flow

emand paging by adding support for MMU notifiers") Cc: Eli Cohen <e...@mellanox.com> Signed-off-by: Haggai Eran <hagg...@mellanox.com> --- A quick resend without spaces between the Fixes: line and the rest of the signatures, per Or Gerlitz's request. drivers/infiniband/hw/mlx5/mr.

[PATCH] IB/mlx5: avoid destroying a NULL mr in reg_user_mr error flow

emand paging by adding support for MMU notifiers") Cc: Eli Cohen <e...@mellanox.com> Signed-off-by: Haggai Eran <hagg...@mellanox.com> --- drivers/infiniband/hw/mlx5/mr.c | 12 1 file changed, 12 deletions(-) diff --git a/drivers/infiniband/hw/mlx5/mr.c b/drivers/i

Re: [PATCH v4 09/14] IB/cm: Expose BTH P_Key in CM and SIDR request events

On 30/08/2015 21:23, Sagi Grimberg wrote: > > Looks like for some reason cm_get_bth_pkey got pkey_index of 0x > instead of 0 (working on the default pkey 0x at entry 0). It looks like the mlx5 driver doesn't interpret the completion format correctly. It takes a field defined in the

[PATCH] IB/cma: Fix net_dev reference leak with failed requests

When no matching listening ID is found for a given request, the net_dev that was used to find the request isn't released. Fixes: 20c36836ecad (IB/cma: Use found net_dev for passive connections) Signed-off-by: Haggai Eran hagg...@mellanox.com --- drivers/infiniband/core/cma.c | 4 1 file

[PATCH v6 1/4] IB/addr: Pass network namespace as a parameter

is added on more levels. Signed-off-by: Haggai Eran hagg...@mellanox.com Signed-off-by: Yotam Kenneth yota...@mellanox.com Signed-off-by: Shachar Raindel rain...@mellanox.com Signed-off-by: Guy Shapiro gu...@mellanox.com --- drivers/infiniband/core/addr.c | 17 + drivers/infiniband

[PATCH v6 2/4] IB/cma: Separate port allocation to network namespaces

for the new port-space API. Signed-off-by: Haggai Eran hagg...@mellanox.com Signed-off-by: Yotam Kenneth yota...@mellanox.com Signed-off-by: Shachar Raindel rain...@mellanox.com Signed-off-by: Guy Shapiro gu...@mellanox.com --- drivers/infiniband/core/cma.c | 94

[PATCH v6 4/4] IB/ucma: Take the network namespace from the process

From: Guy Shapiro gu...@mellanox.com Add support for network namespaces from user space. This is done by passing the network namespace of the process instead of init_net. Signed-off-by: Haggai Eran hagg...@mellanox.com Signed-off-by: Yotam Kenneth yota...@mellanox.com Signed-off-by: Shachar

[PATCH v6 0/4] Add network namespace support in the RDMA-CM

for network namespaces IB/ucma: Take the network namespace from the process Haggai Eran (1): IB/cma: Separate port allocation to network namespaces drivers/infiniband/core/addr.c | 17 +-- drivers/infiniband/core/cma.c | 129 +++-- drivers

[PATCH v6 3/4] IB/cma: Add support for network namespaces

rdma_destroy_id. In order to preserve the current behavior init_net is passed when calling from other modules. Signed-off-by: Guy Shapiro gu...@mellanox.com Signed-off-by: Haggai Eran hagg...@mellanox.com Signed-off-by: Yotam Kenneth yota...@mellanox.com Signed-off-by: Shachar Raindel rain...@mellanox.com

[PATCH 2/3] Add on-demand paging support

the IBV_ACCESS_ON_DEMAND access flag to allow registration of on-demand paging enabled memory regions. Signed-off-by: Shachar Raindel rain...@mellanox.com Signed-off-by: Majd Dibbiny m...@mellanox.com Signed-off-by: Haggai Eran hagg...@mellanox.com --- examples/devinfo.c| 51

[PATCH 0/3] libibverbs: On-demand paging support

): Add support for extended query device capabilities Haggai Eran (1): Add on-demand paging support Majd Dibbiny (1): libibverbs/examples: Support odp in rc_pingpong Makefile.am | 3 +- examples/devinfo.c| 67 -- examples/rc_pingpong.c

[PATCH 1/3] Add support for extended query device capabilities

From: Eli Cohen e...@mellanox.com Add the verb ibv_query_device_ex which is extensible and allows following commits to add new features to define additional properties. Signed-off-by: Eli Cohen e...@mellanox.com Signed-off-by: Haggai Eran hagg...@mellanox.com --- Makefile.am

Re: [PATCH for-next 00/10] Add RoCE support to the mlx5 driver

On 25/08/2015 16:06, Achiad Shochat wrote: On 8/25/2015 3:32 PM, Tom Talpey wrote: But I don't understand how it will work per-GID. What if the target node is RoCEv2 and on another subnet? How will it discover the remote's capability and establish the right protocol? How does the initiator

Re: [PATCH v9 0/4] Sending kernel pathrecord query to user cache server

On 24/08/2015 17:32, Wan, Kaike wrote: On Fri, Aug 14, 2015 at 08:52:05AM -0400, kaike@intel.com wrote: Some tests with namespace have been performed: 1. An unprivileged user cannot bind to the RDMA_NL_GROUP_LS multicast group; 2. An unprivileged user cannot create a new

Re: [PATCH v9 0/4] Sending kernel pathrecord query to user cache server

On 24/08/2015 17:32, Wan, Kaike wrote: On Fri, Aug 14, 2015 at 08:52:05AM -0400, kaike@intel.com wrote: Some tests with namespace have been performed: 1. An unprivileged user cannot bind to the RDMA_NL_GROUP_LS multicast group; 2. An unprivileged user cannot create a new

Re: [PATCH v9 0/4] Sending kernel pathrecord query to user cache server

On 24/08/2015 17:32, Wan, Kaike wrote: On Fri, Aug 14, 2015 at 08:52:05AM -0400, kaike@intel.com wrote: Some tests with namespace have been performed: 1. An unprivileged user cannot bind to the RDMA_NL_GROUP_LS multicast group; 2. An unprivileged user cannot create a new

Re: [PATCH 1/3] IB/uverbs: reject invalid or unknown opcodes

On 24/08/2015 09:55, Christoph Hellwig wrote: On Mon, Aug 24, 2015 at 09:52:14AM +0300, Haggai Eran wrote: Okay. Maybe you can just add a case for IB_WR_SEND in this patch to avoid hurting bisectability. I've done this already, just waiting for more feedback before resending: http

Re: [PATCH 1/3] IB/uverbs: reject invalid or unknown opcodes

On 22/08/2015 11:25, Christoph Hellwig wrote: On Sat, Aug 22, 2015 at 06:38:47AM +, Haggai Eran wrote: It looks like the default case in the non-UD branch is currently used to handle plain IB_WR_SEND operations, so the patch would cause these to return an error. Indeed. It's handled

Re: [PATCH v9 0/4] Sending kernel pathrecord query to user cache server

On Saturday, August 22, 2015 2:07 AM, Jason Gunthorpe jguntho...@obsidianresearch.com wrote: On Fri, Aug 14, 2015 at 08:52:05AM -0400, kaike@intel.com wrote: Some tests with namespace have been performed: 1. An unprivileged user cannot bind to the RDMA_NL_GROUP_LS multicast group;

Re: [PATCH 1/3] IB/uverbs: reject invalid or unknown opcodes

On Thursday, August 20, 2015 11:52 AM, linux-rdma-ow...@vger.kernel.org linux-rdma-ow...@vger.kernel.org on behalf of Sagi Grimberg sa...@dev.mellanox.co.il wrote: On 8/19/2015 7:37 PM, Christoph Hellwig wrote: We have many WR opcodes that are only supported in kernel space and/or require

Re: [PATCH] IB/sa: Restrict SA Netlink to admin users

On 07/08/2015 00:08, ira.we...@intel.com wrote: @@ -754,6 +764,12 @@ static int ib_nl_handle_resolve_resp(struct sk_buff *skb, int found = 0; int ret; + if (!ns_capable(sock_net(skb-sk)-user_ns, CAP_NET_ADMIN)) { + pr_warn_ratelimited(SA netlink: invalid perm

Re: [PATCH for-next 1/2] IB/core: Add support for RX/TX checksum offload capabilities report

On Wednesday, August 5, 2015 8:16 PM, Jason Gunthorpe jguntho...@obsidianresearch.com wrote: On Wed, Aug 05, 2015 at 06:34:26PM +0300, Amir Vadai wrote: struct ib_uverbs_ex_query_device { __u32 comp_mask; + __u32 csum_caps; __u32 reserved; }; Uh no. This is the struct

Re: [PATCH for-next 1/2] IB/core: Add support for RX/TX checksum offload capabilities report

On 08/06/2015 02:18 PM, Parav Pandit wrote: On Thu, Aug 6, 2015 at 4:30 PM, Haggai Eran hagg...@mellanox.com mailto:hagg...@mellanox.com wrote: On Wednesday, August 5, 2015 8:16 PM, Jason Gunthorpe jguntho...@obsidianresearch.com mailto:jguntho...@obsidianresearch.com wrote

Re: [PATCH v2 01/12] IB/core: Guarantee that a local_dma_lkey is available

On 31/07/2015 02:22, Jason Gunthorpe wrote: int ib_dealloc_pd(struct ib_pd *pd) { + if (pd-local_mr) { + if (ib_dereg_mr(pd-local_mr)) + return -EBUSY; + pd-local_mr = NULL; + } + It looks like ib_uverbs_alloc_pd calls

[PATCH v4 12/14] IB/cma: Use found net_dev for passive connections

When receiving a new connection in cma_req_handler, we actually already know the net_dev that is used for the connection's creation. Instead of calling cma_translate_addr to resolve the new connection id's source address, just use the net_dev that was found. Signed-off-by: Haggai Eran hagg

[PATCH v4 03/14] IB/core: Find the network device matching connection parameters

ipvlan-like configurations for RDMA CM with IPoIB. To resolve the device in these cases the code will also take the IP address as an additional input. Reviewed-by: Jason Gunthorpe jguntho...@obsidianresearch.com Signed-off-by: Haggai Eran hagg...@mellanox.com Signed-off-by: Yotam Kenneth yota

[PATCH v4 13/14] IB/cma: Share ib_cm_ids between rdma_cm_ids

. Signed-off-by: Haggai Eran hagg...@mellanox.com --- drivers/infiniband/core/cma.c | 59 +++ 1 file changed, 4 insertions(+), 55 deletions(-) diff --git a/drivers/infiniband/core/cma.c b/drivers/infiniband/core/cma.c index c1cd47eab149..1f26bff5f780 100644

  1   2   3   4   >