29 Aug, 2017
25 commits
-
There are currently 3 spots in the qib and hfi1 driver that have
knowledge of the internal QP hash list that should only be in
scope to rdmavt QP code.Add an iterator API for processing all QPs to hide the
nature of the RCU hashlist.The API consists of:
- rvt_qp_iter_init()
* For iterating QPs one at a time for seq_file semantics
- rvt_qp_iter_next()
* For iterating QPs one at a time for seq_file semantics
- rvt_qp_iter()
* For iterating all QPsThe first two are used for things like seq_file prints.
The last is for code that just needs to iterate all QPs
in the system.Reviewed-by: Dennis Dalessandro
Signed-off-by: Mike Marciniszyn
Signed-off-by: Dennis Dalessandro
Signed-off-by: Doug Ledford -
The qp_stats print will soon be moving to rdmavt, so use the proper
accessor to get the ring size rather than a driver supplied constant.Fixes: Commit ff8d836efe06 ("IB/hfi1: Add receiving queue info to qp_stats")
Reviewed-by: Kaike Wan
Signed-off-by: Mike Marciniszyn
Signed-off-by: Dennis Dalessandro
Signed-off-by: Doug Ledford -
Replace 'strcpy' with 'strncpy' to restrict the number
of bytes copied to the buffer.Reviewed-by: Michael J. Ruhl
Signed-off-by: Kamenee Arumugam
Signed-off-by: Dennis Dalessandro
Signed-off-by: Doug Ledford -
The hfi1_cdbg() macro can be instantiated in the hot path even when it
is not in use. This shows up on perf profiles.Rework the macros (for SDMA and MMU), to use the trace interface directly
to eliminate this performance hit.Reviewed-by: Mike Marciniszyn
Signed-off-by: Michael J. Ruhl
Signed-off-by: Dennis Dalessandro
Signed-off-by: Doug Ledford -
Currently, QSFP information is not queried
in cases where loopback was set up and QSFP module is
present.Acquire QSFP information in case of loopback.
Reviewed-by: Dennis Dalessandro
Signed-off-by: Jan Sokolowski
Signed-off-by: Dennis Dalessandro
Signed-off-by: Doug Ledford -
Make some structures const as they are only used during a copy
operation.Signed-off-by: Bhumika Goyal
Signed-off-by: Doug Ledford -
vm_operations_struct are not supposed to change at runtime.
vm_area_struct structure working with const vm_operations_struct.
So mark the non-const vm_operations_struct structs as const.Signed-off-by: Arvind Yadav
Signed-off-by: Doug Ledford -
call to memset to assign 0 value immediately after allocating
memory with kzalloc is unnecesaary as kzalloc allocates the memory
filled with 0 value.Signed-off-by: Himanshu Jha
Reviewed-by: Yuval Shaia
Signed-off-by: Doug Ledford -
usnic_uiom_get_dev_list() can return ERR_PTR(-ENOMEM) so we should check
for that.Fixes: e3cf00d0a87f ("IB/usnic: Add Cisco VIC low-level hardware driver")
Signed-off-by: Dan Carpenter
Reviewed-by: Yuval Shaia
Signed-off-by: Doug Ledford -
These fields allow for debugging send engine processing.
Reviewed-by: Kaike Wan
Signed-off-by: Mike Marciniszyn
Signed-off-by: Dennis Dalessandro
Signed-off-by: Doug Ledford -
The rvt_ack_entry pointed to by s_tail_ack_queue provides important
info about the request that has just been processed or is being processed
on the responder side of a RC connection. This patch adds this info to
the qp_stats to assist debugging.Reviewed-by: Mike Marciniszyn
Signed-off-by: Kaike Wan
Signed-off-by: Dennis Dalessandro
Signed-off-by: Doug Ledford -
Fix a tab alignment issue present in pr_err_ratelimited
error message.Reviewed-by: Michael J. Ruhl
Signed-off-by: Kamenee Arumugam
Signed-off-by: Dennis Dalessandro
Signed-off-by: Doug Ledford -
Clean up user_sdma.c by moving the structure and MACRO definitions into
the header file user_sdma.hReviewed-by: Dennis Dalessandro
Signed-off-by: Harish Chegondi
Signed-off-by: Dennis Dalessandro
Signed-off-by: Doug Ledford -
Clean up user_exp_rcv.c file by moving structure definitions into header
file user_exp_rcv.h. Since these structure definitions depend on the
structure definitions in mmu_rb.h, move #include "mmu_rb.h" above
the include "user_exp_rcv.h" or include of header files that include
user_exp_rcv.hReviewed-by: Dennis Dalessandro
Signed-off-by: Harish Chegondi
Signed-off-by: Dennis Dalessandro
Signed-off-by: Doug Ledford -
num_user_pages() function has been defined in both user_exp_rcv.c file
and user_sdma.c file. Move the function definition to a header file so
there is only one definition in the source repo.Reviewed-by: Dennis Dalessandro
Signed-off-by: Harish Chegondi
Signed-off-by: Dennis Dalessandro
Signed-off-by: Doug Ledford -
In pin_vector_pages() function, if there is any error while pinning the
pages or while adding a pinned buffer to the cache, the bail out code
needs to unpin any pinned pages that are not in the cache and adjust the
n_locked counter that counts the total pages pinned. The current bail
out code doesn't seem to be doing it right in two cases:1. Before pinning required pages for a buffer, the SDMA pinned buffer
cache is searched to see if the virtual address range that needs to be
pinned is already pinned. If there isn't a hit in the cache, a new node
is created for the buffer and is added to the cache after the buffer is
pinned. If adding the new node to the cache fails, the n_locked count is
decremented properly but the pinned pages are not freed. This commit
fixes this issue.2. If there is a hit in the SDMA cache, but the cached buffer doesn't
have enough pages to cover the entire address range that needs to be
pinned, the node for the cached buffer is extracted from the cache,
remaining pages needed are pinned and added to the node. The node is
finally added back into the cache. If there is an error pinning the
extra pages, the bail out code frees all the pages in the node but the
n_locked count is not being decremented by the no of pages in the node
that are freed. This commit fixes this issue.This commit fixes the above two issues by creating a new function that
frees the pages in a node and decrements the n_locked count by the
number of pages freed.Reviewed-by: Dennis Dalessandro
Signed-off-by: Harish Chegondi
Signed-off-by: Dennis Dalessandro
Signed-off-by: Doug Ledford -
Clean up pin_vector_pages() function by moving page pinning related code
to a separate function since it really stands on its own.Reviewed-by: Dennis Dalessandro
Signed-off-by: Harish Chegondi
Signed-off-by: Dennis Dalessandro
Signed-off-by: Doug Ledford -
user_sdma_send_pkts() function is unnecessarily long. Clean it up by
moving some of its code into separate functions.Reviewed-by: Dennis Dalessandro
Signed-off-by: Harish Chegondi
Signed-off-by: Dennis Dalessandro
Signed-off-by: Doug Ledford -
Clean up hfi1_user_exp_rcv_setup function by moving page pinning and
unpinning related code to separate functions. In order to reduce the
number of parameters passed between functions, a new data structure
struct tid_user_buf is defined and used.Reviewed-by: Dennis Dalessandro
Signed-off-by: Harish Chegondi
Signed-off-by: Dennis Dalessandro
Signed-off-by: Doug Ledford -
Performance analysis shows that the cache callback function
sdma_kmem_cache_ctor contributes to 1/2 of the kmem_cache_allocs
time.Since all of the fields in the allocated data structure are initialized
in the code path, remove the _ctor function.Reviewed-by: Mike Marciniszyn
Signed-off-by: Michael J. Ruhl
Signed-off-by: Dennis Dalessandro
Signed-off-by: Doug Ledford -
Ratelimit error prints from sdma_interrupt function
that could swarm dmesg otherwise.Reviewed-by: Dennis Dalessandro
Signed-off-by: Grzegorz Morys
Signed-off-by: Dennis Dalessandro
Signed-off-by: Doug Ledford -
Added checking on index value of array 'guids' in qib_ruc.c.
Pass in corrrect size of array for memset operation in qib_mad.c.Reviewed-by: Michael J. Ruhl
Signed-off-by: Kamenee Arumugam
Signed-off-by: Dennis Dalessandro
Signed-off-by: Doug Ledford -
Remove all the memory allocation implemented for boardname and
directly assign the defined string literal.Reviewed-by: Michael J. Ruhl
Signed-off-by: Kamenee Arumugam
Signed-off-by: Dennis Dalessandro
Signed-off-by: Doug Ledford -
Section 9.7.7.2.5 of the 1.3 IBTA spec clearly says that receive
credits should never apply to RDMA write.qib and hfi1 were doing that. The following situation will result
in a QP hang:
- A prior SEND or RDMA_WRITE with immmediate consumed the last
credit for a QP using RC receive buffer credits
- The prior op is acked so there are no more acks
- The peer ULP fails to post receive for some reason
- An RDMA write sees that the credits are exhausted and waits
- The peer ULP posts receive buffers
- The ULP posts a send or RDMA write that will be hungThe fix is to avoid the credit test for the RDMA write operation.
Cc:
Reviewed-by: Kaike Wan
Signed-off-by: Mike Marciniszyn
Signed-off-by: Dennis Dalessandro
Signed-off-by: Doug Ledford -
hfi1 and qib were converted in previous patches, do the same for rdmavt.
Signed-off-by: Mike Marciniszyn
Signed-off-by: Dennis Dalessandro
Signed-off-by: Doug Ledford
25 Aug, 2017
15 commits
-
Signed-off-by: Doug Ledford
-
Expose enhanced multi packet WQE capability to user space through
query_device by uhw.Signed-off-by: Bodong Wang
Reviewed-by: Daniel Jurgens
Signed-off-by: Leon Romanovsky
Signed-off-by: Doug Ledford -
Set the field to allow posting multi packet send WQEs if hardware
supports this feature. This doesn't mean the send WQEs will be for
multi packet unless the send WQE was prepared according to multi
packet send WQE format.User space shall use flag MLX5_IB_ALLOW_MPW to check if hardware
supports MPW and allows MPW in SQ context.Signed-off-by: Bodong Wang
Reviewed-by: Daniel Jurgens
Signed-off-by: Leon Romanovsky
Signed-off-by: Doug Ledford -
Set underlay QPN as part of flow rule when it's applicable.
There is one root flow table in the NIC RX namespace and all the
underlay QPs steer the traffic to this flow table.
In order to prevent QP to get traffic which is not target to its
underlay QP, we need to set the underlay QP number as part of
the steering matching.Note:
When multicast traffic is sent the QPN filtering is done by the firmware
as some early step. Adding the QPN match on the flow table entry is
wrong as by that time the target QPN holds the multicast address (e.g.
FF(s)) and it won't match.Signed-off-by: Yishai Hadas
Signed-off-by: Leon Romanovsky
Signed-off-by: Doug Ledford -
Fix a bug where MR registration fails when mlx5_ib_cont_pages
indicates that the MR can be mapped using 2GB pages (page_shift == 31).Fixes: e126ba97dba9 ("mlx5: Add driver for Mellanox Connect-IB adapters")
Signed-off-by: Ilya Lesokhin
Signed-off-by: Leon Romanovsky
Signed-off-by: Doug Ledford -
In clean_mr error path the 'mr' should be freed.
Fixes: e126ba97dba9 ('mlx5: Add driver for Mellanox Connect-IB adapters')
Signed-off-by: Kamal Heib
Signed-off-by: Doug Ledford -
mlx5 compatible devices have two ways of populating the MTT
table of an MKEY: using a FW command and using a UMR WQE.A UMR is much faster, so it should be used whenever possible.
Unfortunately the code today uses UMR only if the MKEY was allocated
from the MR cache.Fix the code to use UMR even for MKEYs that were allocated using
a FW command.Signed-off-by: Ilya Lesokhin
Signed-off-by: Leon Romanovsky
Signed-off-by: Doug Ledford -
This patch is the first step in decoupling UMR usage and
allocation from the MR cache. The only functional change
in this patch is to enables UMR for MRs created with
reg_create.This change fixes a bug where ODP memory regions that
were not allocated from the MR cache did not have UMR
enabled.Signed-off-by: Ilya Lesokhin
Signed-off-by: Leon Romanovsky
Signed-off-by: Doug Ledford -
Software parsing (SWP) is a feature that can be used to instruct the
device to stop using its internal parser and to parse packets on the
transmit path according to offsets set for each packets.Through this feature, the device allows the handling of checksum and
LSO by the hardware according to the location of IP and TCP/UDP
headers.Enable SW parsing on Raw Ethernet send queue by default if firmware
supports it and report these capabilities to user space.Signed-off-by: Noa Osherovich
Reviewed-by: Maor Gottlieb
Signed-off-by: Leon Romanovsky
Signed-off-by: Doug Ledford -
None of the calls to i40iw_netdev_vlan_ipv6 are using mac so let's
remove it from func's args-list.Signed-off-by: Yuval Shaia
Signed-off-by: Doug Ledford -
Trivial fix to spelling mistake in DP_ERR error message
Signed-off-by: Colin Ian King
Reviewed-by: Leon Romanovsky
Reviewed-by: Ram Amrani
Signed-off-by: Doug Ledford -
IB CM calls ib_modify_port() irrespective of link layer. If the
failure is returned, the mad agent gets unregistered for those
devices. Recently, modify_port() hook was removed from some of the
low level drivers as it was always returning success. This breaks
rdma connection establishment over those devices.
For ethernet devices, Qkey violation and port capabilities are not
applicable. So returning success for RoCE when modify_port hook is
is not implemented.Cc: Leon Romanovsky
Signed-off-by: Selvin Xavier
Reviewed-by: Leon Romanovsky
Signed-off-by: Doug Ledford -
Added support for two device caps - max_sge_rd, max_fast_reg_page_list_len
and the IP_BASED_GIDS port cap flag.Reviewed-by: Jorgen Hansen
Reviewed-by: Bryan Tan
Reviewed-by: Aditya Sarwade
Signed-off-by: Adit Ranadive
Reviewed-by: Leon Romanovsky
Signed-off-by: Doug Ledford -
The driver version is bumped for compatibility purposes. Also, send correct
GID type during register to device. Added compatibility check macros for
the device.Reviewed-by: Jorgen Hansen
Reviewed-by: Aditya Sarwade
Signed-off-by: Bryan Tan
Signed-off-by: Adit Ranadive
Reviewed-by: Leon Romanovsky
Reviewed-by: Yuval Shaia
Signed-off-by: Doug Ledford -
Adds support for ioctl callback in the RDMA netdevs to allow
supporting functions not handled by the generic interface code.Signed-off-by: Feras Daoud
Signed-off-by: Eitan Rabin
Signed-off-by: Leon Romanovsky
Reviewed-by: Yuval Shaia
Signed-off-by: Doug Ledford
