driver inclusion
category: Bugfix
bugzilla: https://gitee.com/openeuler/kernel/issues/I61F1Q

--------------------------------------------------------------

Running roce v1 business on fpga, the following error occurs:
hns3 0000:35:00.0 hns_0: local work queue 0x2 catast error,
sub_event type is: 4

This is because the sl transmitted by the roce v1 service
driver after set dscp is incorrect, which makes the sl of
db inconsistent with the sl of qpc, resulting in an sl error
on the hardware.

Fixes: 11ef2ec6 ("RDMA/hns: Support DSCP of userspace")
Signed-off-by: NYixing Liu <liuyixing1@huawei.com>
Reviewed-by: NYangyang Li <liyangyang20@huawei.com>
Reviewed-by: NYue Haibing <yuehaibing@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>

driver inclusion
category: Bugfix
bugzilla: https://gitee.com/openeuler/kernel/issues/I61FED

-----------------------------------------------------------

When it is roce v1, if the traffic_class value exceeds 63,
the following error will appear:
modify qp to 2 state failed(22)
Failed to create AH

This is because the driver intercepts the over-spec value
in set dscp, and there is no need to obtain dscp for roce v1,
so the driver does not intercept v1.

Fixes: 11ef2ec6 ("RDMA/hns: Support DSCP of userspace")
Signed-off-by: NYixing Liu <liuyixing1@huawei.com>
Reviewed-by: NYangyang Li <liyangyang20@huawei.com>
Reviewed-by: NYue Haibing <yuehaibing@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>

driver inclusion
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I5WKYW

-----------------------------------------------------------------------

The user mode driver needs to use the mac type to sense whether
it is in ROH mode, so it is necessary to add the mac type to
the user mode driver.
Signed-off-by: NGuofeng Yue <yueguofeng@hisilicon.com>
Signed-off-by: NKe Chen <chenke54@huawei.com>
Reviewed-by: NYangyang Li <liyangyang20@huawei.com>

driver inclusion
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I5WKYW

-----------------------------------------------------------------------

Support RDMA_CM in ROH mode
Signed-off-by: NKe Chen <chenke54@huawei.com>
Reviewed-by: NYangyang Li <liyangyang20@huawei.com>

driver inclusion
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I5WKYW

-----------------------------------------------------------------------

ROH MAC do not support ROCEv1 and ROCEv2@IPv6.
ROCE driver adapts to these limiations.
Signed-off-by: NLang Cheng <chenglang@huawei.com>
Signed-off-by: NKe Chen <chenke54@huawei.com>
Reviewed-by: NYangyang Li <liyangyang20@huawei.com>

driver inclusion
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I5WKYW

-----------------------------------------------------------------------

Add ROH device ID.
Signed-off-by: NLang Cheng <chenglang@huawei.com>
Signed-off-by: NKe Chen <chenke54@huawei.com>
Reviewed-by: NYangyang Li <liyangyang20@huawei.com>

driver inclusion
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I5Z6L8

----------------------------------------------------------

Support hns roce bonding
Signed-off-by: NJunxian Huang <huangjunxian6@hisilicon.com>
Signed-off-by: NChunZhi Hu <huchunzhi@huawei.com>
Reviewed-by: NYangyang Li <liyangyang20@huawei.com>

driver inclusion
category: bugfix
bugzilla: https://gitee.com/openeuler/kernel/issues/I5YZ0Q

----------------------------------------------------------

When the driver does not have an ex_cmd flag, related ioctl()
or syscall() will fail to execute. for example, if the modify_qp
flag is missing, the following error will appear when running
perftest:
Failed to modify QP to INIT, ret=95
Failed to modify QP to INIT
Failed to modify QP 24 to RTR
Failed to modify QP 25 to RTR
Unable to Connect the HCA's through the link
Unable to Connect the HCA's through the link

So add query_device、create_cq、modify_qp and create_qp flag.

Fixes: df065107 ("RDMA/hns: Enable modify_cq for uverbs.")
Signed-off-by: NYixing Liu <liuyixing1@huawei.com>
Reviewed-by: NYangyang Li <liyangyang20@huawei.com>
Reviewed-by: NYue Haibing <yuehaibing@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>

driver inclusion
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I5Z2DS

----------------------------------------------------------

When the netdev port status changes, the roce driver sends a port down
event by parsing the netdev event dispatched by IB_CORE, which takes about
a few hundred milliseconds.
But,it is not fast enough for ULP sometimes.

The HNS NIC driver can directly notify the ROCE driver send port
event via callback function, this takes only a few milliseconds.
This patch implements this callback function.
Signed-off-by: NLang Cheng <chenglang@huawei.com>
Signed-off-by: NChengchang Tang <tangchengchang@huawei.com>
Reviewed-by: NYangyang Li <liyangyang20@huawei.com>
Reviewed-by: NYue Haibing <yuehaibing@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>

driver inclusion
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I5Z2DS

----------------------------------------------------------

Driver can notify ulp with IB event when net link down/up.
Signed-off-by: NLang Cheng <chenglang@huawei.com>
Signed-off-by: NChengchang Tang <tangchengchang@huawei.com>
Reviewed-by: NYangyang Li <liyangyang20@huawei.com>
Reviewed-by: NYue Haibing <yuehaibing@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>

driver inclusion
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I5Z2HJ

----------------------------------------------------------

Currently, we have dumped some fields in the QP/CQ/MR resource. This
information is not enough. It is very inconvenient to continue to
expand on the current field, and it will also introduce some trouble
to parse this raw data.

This patch dump whole resource in raw to avoid the above problems.
Signed-off-by: NChengchang Tang <tangchengchang@huawei.com>
Reviewed-by: NYangyang Li <liyangyang20@huawei.com>
Reviewed-by: NYue Haibing <yuehaibing@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>

driver inclusion
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I5XYHV

----------------------------------------------------------

This patch add support to config dscp map to tc by
nic interface get_dscp_prio. Driver will convert
mapping relationship from dscp-prio to dscp-tc.

The dependency patch is as follows:
7a1313b3 ("net: hns3: add support config dscp map to tc")
Signed-off-by: NYixing Liu <liuyixing1@huawei.com>
Reviewed-by: NYangyang Li <liyangyang20@huawei.com>
Reviewed-by: NYue Haibing <yuehaibing@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>

driver inclusion
category: bugfix
bugzilla: https://gitee.com/openeuler/kernel/issues/I5Y79T

------------------------------------------------------------------------

An invalid roce page size cap must include the system page size.
Signed-off-by: NChengchang Tang <tangchengchang@huawei.com>
Reviewed-by: NYangyang Li <liyangyang20@huawei.com>
Reviewed-by: NYueHaibing <yuehaibing@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>

driver inclusion
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I5Y79T

-------------------------------------------------------------------

Currently, when creating a MR, there is a certain probability of
causing a hardware exception when addressing at level 0 with a huge
page. This is because the PA address is now aligned to 4K instead of
the actual page size set by the driver.

When the MTR of multiple regions is addressed at level 0 with a huge
page, multiple regions will share the same huge page. This huge page
will be divided into multiple 4K pages, and these pages will be
allocated to different regions. At this point, all the regions of this
MTR only require 4k alignment instead of the actual pagesize. This can
help reduce memory consumption.

But when there is only one region, the MTR with level-0 addressing can
directly use this huge page. At this point the hardware needs this page
to be aligned to the actual page size, not 4K.

This patch set page size to 4k only when the MTR with level 0 addressing
has mutliple regions.

Fixes: 0e0ab04b ("RDMA/hns: Refactor the MTR creation flow")
Signed-off-by: NChengchang Tang <tangchengchang@huawei.com>
Reviewed-by: NYangyang Li <liyangyang20@huawei.com>
Reviewed-by: NYueHaibing <yuehaibing@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>

driver inclusion
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I5Y79T

-------------------------------------------------------------------

In the current implementation, we configure a fixed addressing level
for the MTR. But in fact, the necessary addressing level is related
to the page size and the size of the memory.

This patch calculates the addressing level according to the page size
and the size of the memory, and uses the addressing level to configure
the MTR.

It only works for PBL now.
Signed-off-by: NChengchang Tang <tangchengchang@huawei.com>
Reviewed-by: NYangyang Li <liyangyang20@huawei.com>
Reviewed-by: NYueHaibing <yuehaibing@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>

driver inclusion
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I5Y79T

-------------------------------------------------------------------

In the current implementation, we use a fixed page size to
configure the MTR, which is not flexible enough and is not
conducive to the performance of the HW.

Now the best page size will be calculated and used to configure
the MTR. It only works for PBL now.
Signed-off-by: NChengchang Tang <tangchengchang@huawei.com>
Reviewed-by: NYangyang Li <liyangyang20@huawei.com>
Reviewed-by: NYueHaibing <yuehaibing@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>

driver inclusion
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I5Y79T

-------------------------------------------------------------------

MTR memory allocation do not depend on allocation of MTT. And MTT
allocation can be adjusted according to MTR, so make MTR allocation
before MTT allocation.
Signed-off-by: NChengchang Tang <tangchengchang@huawei.com>
Reviewed-by: NYangyang Li <liyangyang20@huawei.com>
Reviewed-by: NYueHaibing <yuehaibing@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>

driver inclusion
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I5Y79T

-------------------------------------------------------------------

page_shift and page_cnt is only used in mtr_map_bufs(). And these
parameter could be calculated indepedently.

Strip the computation of page_shift and page_cnt from mtr_init_buf_cfg(),
reducing the number of parameters of it. This help reducing coupling
between mtr_inif_buf_cfg() and mtr_map_bufs().

And the parameter validation in mtr_init_buf_cfg has also been abstracted
into a separate function and placed at the beginning of the function.
Signed-off-by: NChengchang Tang <tangchengchang@huawei.com>
Reviewed-by: NYangyang Li <liyangyang20@huawei.com>
Reviewed-by: NYueHaibing <yuehaibing@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>

mainline inclusion
from mailine v6.0-rc5
commit 55af9d49
category: bugfix
bugzilla: https://gitee.com/openeuler/kernel/issues/I5Y79T
CVE: NA

Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=55af9d498556

------------------------------------------------------------------------

The supported page size for hns is (4K, 128M), not (4K, 2G).

Fixes: cfc85f3e ("RDMA/hns: Add profile support for hip08 driver")
Link: https://lore.kernel.org/r/20220829105021.1427804-2-liangwenpeng@huawei.comSigned-off-by: NChengchang Tang <tangchengchang@huawei.com>
Signed-off-by: NWenpeng Liang <liangwenpeng@huawei.com>
Signed-off-by: NLeon Romanovsky <leon@kernel.org>
Reviewed-by: NYangyang Li <liyangyang20@huawei.com>
Reviewed-by: NYueHaibing <yuehaibing@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>

driver inclusion
category: bugfix
bugzilla: https://gitee.com/openeuler/kernel/issues/I5Y79T

-------------------------------------------------------------------

Now, The address of the first two pages in the MR will be searched.
And an exception will occur when there is only one page in this MR.

This patch fix the number of page to search.

Fixes: 9b2cf76c ("RDMA/hns: Optimize PBL buffer allocation process")
Signed-off-by: NChengchang Tang <tangchengchang@huawei.com>
Reviewed-by: NYangyang Li <liyangyang20@huawei.com>
Reviewed-by: NYueHaibing <yuehaibing@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>

driver inclusion
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I5X0C3

---------------------------------------------------------------------

Log return value of clear_hem() to help diagnose.
Signed-off-by: NChengchang Tang <tangchengchang@huawei.com>
Reviewed-by: NYangyang Li <liyangyang20@huawei.com>
Reviewed-by: NHaoyue Xu <xuhaoyue1@hisilicon.com>
Reviewed-by: NYueHaibing <yuehaibing@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>

driver inclusion
category: bugfix
bugzilla: https://gitee.com/openeuler/kernel/issues/I5X0C3

---------------------------------------------------------------------

The return value of set_hem has been fixed to ENODEV, which will
lead a diagnostic information missing.

Fixes: 9a443537 ("IB/hns: Add driver files for hns RoCE driver")
Signed-off-by: NChengchang Tang <tangchengchang@huawei.com>
Reviewed-by: NYangyang Li <liyangyang20@huawei.com>
Reviewed-by: NHaoyue Xu <xuhaoyue1@hisilicon.com>
Reviewed-by: NYueHaibing <yuehaibing@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>

mainline inclusion
from mainline-v5.15-rc1
commit 4ad81814
category: bugfix
bugzilla: https://gitee.com/openeuler/kernel/issues/I5X0C3
CVE: NA

Reference: https://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git/commit/?id=4ad8181426df92976feee5fbc55236293d069b37

---------------------------------------------------------------------

Due to the discrete nature of the HIP08 timer unit, a requester might
finish the timeout period sooner, in elapsed real time, than its responder
does, even when both sides share the identical RNR timeout length included
in the RNR Nak packet and the responder indeed starts the timing prior to
the requester. Furthermore, if a 'providential' resend packet arrived
before the responder's timeout period expired, the responder is certainly
entitled to drop the packet silently in the light of IB protocol.

To address this problem, our team made good use of certain hardware facts:

1) The timing resolution regards the transmission arrangements is 1
   microsecond, e.g. if cq_period field is set to 3, it would be
   interpreted as 3 microsecond by hardware

2) A QPC field shall inform the hardware how many timing unit (ticks)
   constitutes a full microsecond, which, by default, is 1000

3) It takes 14ns for the processor to handle a packet in the buffer, so
   the RNR timeout length of 10ns would ensure our processing mechanism is
   disabled during the entire timeout period and the packet won't be
   dropped silently

To achieve (3), we permanently set the QPC field mentioned in (2) to zero
which nominally indicates every time tick is equivalent to a microsecond
in wall-clock time; now, a RNR timeout period at face value of 10 would
only last 10 ticks, which is 10ns in wall-clock time.

It's worth noting that we adapt the driver by magnifying certain
configuration parameters(cq_period, eq_period and ack_timeout)by 1000
given the user assumes the configuring timing unit to be microseconds.

Also, this particular improvisation is only deployed on HIP08 since other
hardware has already solved this issue.

Fixes: cfc85f3e ("RDMA/hns: Add profile support for hip08 driver")
Link: https://lore.kernel.org/r/20211209140655.49493-1-liangwenpeng@huawei.comSigned-off-by: NYangyang Li <liyangyang20@huawei.com>
Signed-off-by: NWenpeng Liang <liangwenpeng@huawei.com>
Signed-off-by: NJason Gunthorpe <jgg@nvidia.com>
Reviewed-by: NYangyang Li <liyangyang20@huawei.com>
Reviewed-by: NHaoyue Xu <xuhaoyue1@hisilicon.com>
Reviewed-by: NYueHaibing <yuehaibing@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>

mainline inclusion
from mainline-v5.14-rc1
commit 7b0006db
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I5X0C3
CVE: NA

Reference: https://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git/commit/?id=7b0006db6800da4c05883584befa46502d85dede

---------------------------------------------------------------------

The base address table is allocated by dma allocator, and the size is
always aligned to PAGE_SIZE. If a fixed size is used to allocate the
table, the number of base address entries stored in the table will be
smaller than that can actually stored.

Link: https://lore.kernel.org/r/1621589395-2435-2-git-send-email-liweihang@huawei.comSigned-off-by: NXi Wang <wangxi11@huawei.com>
Signed-off-by: NWeihang Li <liweihang@huawei.com>
Signed-off-by: NJason Gunthorpe <jgg@nvidia.com>
Reviewed-by: NYangyang Li <liyangyang20@huawei.com>
Reviewed-by: NHaoyue Xu <xuhaoyue1@hisilicon.com>
Reviewed-by: NYueHaibing <yuehaibing@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>

driver inclusion
category: cleanup
bugzilla: https://gitee.com/openeuler/kernel/issues/I5X0C3

-----------------------------------------------------------

This reverts commit f9d4c848.

This patch is synchronized from the linux mainline, but some important
code is lost during synchronization. This results in a large code
difference between openEuler and the linux mainline.
It is a better option to revert the bad patch and resync a full patch
from the linux mainline. The resynchronized patch is the second patch
in this patchlist.
Signed-off-by: NChengchang Tang <tangchengchang@huawei.com>
Reviewed-by: NYangyang Li <liyangyang20@huawei.com>
Reviewed-by: NHaoyue Xu <xuhaoyue1@hisilicon.com>
Reviewed-by: NYueHaibing <yuehaibing@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>

driver inclusion
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I5USIG

----------------------------------------------------------

The kernel space does not support the rq inline feature anymore,
so remove the code associated with rq inline.
Signed-off-by: NLuoyouming <luoyouming@huawei.com>
Reviewed-by: NYangyang Li <liyangyang20@huawei.com>
Reviewed-by: NHaoyue Xu <xuhaoyue1@hisilicon.com>
Reviewed-by: NYueHaibing <yuehaibing@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>

driver inclusion
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I5USIG

----------------------------------------------------------

Enable the CQEIE field and configure the CQEIS field of QPC.
And add compatibility handling.
Signed-off-by: NLuoyouming <luoyouming@huawei.com>
Reviewed-by: NYangyang Li <liyangyang20@huawei.com>
Reviewed-by: NHaoyue Xu <xuhaoyue1@hisilicon.com>
Reviewed-by: NYueHaibing <yuehaibing@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>

driver inclusion
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I5USIG

----------------------------------------------------------

The rq inline makes some changes as follows, Firstly, it is only
used in user space. Secondly, it should notify hardware in QP RTR
status. Thirdly, Add compatibility processing between different
user space and kernel space. Change the HNS_ROCE_CAP_FLAG_RQ_INLINE
to a new bit to prevent old kernel spaces / spaced from enabling
rq inline.
Signed-off-by: NLuoyouming <luoyouming@huawei.com>
Reviewed-by: NYangyang Li <liyangyang20@huawei.com>
Reviewed-by: NHaoyue Xu <xuhaoyue1@hisilicon.com>
Reviewed-by: NYueHaibing <yuehaibing@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>

driver inclusion
category: bugfix
bugzilla: https://gitee.com/openeuler/kernel/issues/I5USIG

----------------------------------------------------------

Currently, the driver only uses max_send_sge to initialize sge num
when creating_qp. So, in the sq inline scenario, the driver may not
has enough sge to send data. For example, if max_send_sge is 16 and
max_inline_data is 1024, the driver needs 1024/16=64 sge to send data.
Therefore, the calculation method of sge num is modified to take the
maximum value of max_send_sge and max_inline_data/16 to solve this
problem.

Fixes:05201e01("RDMA/hns: Refactor process of setting extended sge")
Fixes:30b70788("RDMA/hns: Support inline data in extented sge space for RC")
Signed-off-by: NLuoyouming <luoyouming@huawei.com>
Reviewed-by: NYangyang Li <liyangyang20@huawei.com>
Reviewed-by: NHaoyue Xu <xuhaoyue1@hisilicon.com>
Reviewed-by: NYueHaibing <yuehaibing@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>

driver inclusion
category: cleanup
bugzilla: https://gitee.com/openeuler/kernel/issues/I5USIG

-----------------------------------------------------------

This reverts commit 34542349.

Fixes: 34542349 ("RDMA/hns: Fix the problem of sge nums")
Signed-off-by: NLuoyouming <luoyouming@huawei.com>
Reviewed-by: NYangyang Li <liyangyang20@huawei.com>
Reviewed-by: NHaoyue Xu <xuhaoyue1@hisilicon.com>
Reviewed-by: NYueHaibing <yuehaibing@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>

driver inclusion
category: cleanup
bugzilla: https://gitee.com/openeuler/kernel/issues/I5USIG

-----------------------------------------------------------

This reverts commit d0b40fc5.

Fixes: d0b40fc5 ("RDMA/hns: Remove enable rq inline in kernel and add compatibility handling")
Signed-off-by: NLuoyouming <luoyouming@huawei.com>
Reviewed-by: NYangyang Li <liyangyang20@huawei.com>
Reviewed-by: NHaoyue Xu <xuhaoyue1@hisilicon.com>
Reviewed-by: NYueHaibing <yuehaibing@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>

driver inclusion
category: cleanup
bugzilla: https://gitee.com/openeuler/kernel/issues/I5USIG

-----------------------------------------------------------

This reverts commit 5a18385d.

Fixes: 5a18385d ("RDMA/hns: Support cqe inline in user space")
Signed-off-by: NLuoyouming <luoyouming@huawei.com>
Reviewed-by: NYangyang Li <liyangyang20@huawei.com>
Reviewed-by: NHaoyue Xu <xuhaoyue1@hisilicon.com>
Reviewed-by: NYueHaibing <yuehaibing@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>

driver inclusion
category: bugfix
bugzilla: https://gitee.com/openeuler/kernel/issues/I5XJ01

----------------------------------------------------------------------

Add default algorithm processing.

Fixes: f91696f2 ("RDMA/hns: Support congestion control type selection according to the FW")
Signed-off-by: Nzhengfeng luo <luozhengfeng@h-partners.com>
Reviewed-by: NYangyang Li <liyangyang20@huawei.com>
Reviewed-by: NHaoyue Xu <xuhaoyue1@hisilicon.com>
Reviewed-by: NYue Haibing <yuehaibing@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>

driver inclusion
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I5USIG

----------------------------------------------------------

Enable the CQEIE field and configure the CQEIS field of QPC.
And add compatibility handling.
Signed-off-by: NLuoyouming <luoyouming@huawei.com>
Reviewed-by: NYangyang Li <liyangyang20@huawei.com>
Reviewed-by: NYueHaibing <yuehaibing@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>

driver inclusion
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I5USIG

----------------------------------------------------------

The rq inline only use in userspace, and it should enable in RTR
status. Added compatibility processing between different user space
and kernel space.
Signed-off-by: NLuoyouming <luoyouming@huawei.com>
Reviewed-by: NYangyang Li <liyangyang20@huawei.com>
Reviewed-by: NYueHaibing <yuehaibing@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>

driver inclusion
category: bugfix
bugzilla: https://gitee.com/openeuler/kernel/issues/I5USIG

----------------------------------------------------------

Modify the sge num calculation algorithm when creating qp. The value of the
sge-num takes the maximum value calculated by the parameter max_send_sge or
parameter max_inline_data set by the user.In the sq inline mode, the
following 2 cases will use extended sge:
    1)UD:     payload bytes more than 8
    2)RC/XRC: payload bytes more than 32

Fixes:05201e01("RDMA/hns: Refactor process of setting extended sge")
Fixes:30b70788("RDMA/hns: Support inline data in extented sge space for RC")
Signed-off-by: NLuoyouming <luoyouming@huawei.com>
Reviewed-by: NYangyang Li <liyangyang20@huawei.com>
Reviewed-by: NYueHaibing <yuehaibing@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>

driver inclusion
category: bugfix
bugzilla: https://gitee.com/openeuler/kernel/issues/I5USIG

-----------------------------------------------------------

The max_gs is the sum of extended sge and standard sge. In function
fill_ext_sge_inl_data, max_gs does not subtract the number of extended
sges, but is directly used to calculate the size of extended sges.
Signed-off-by: NLuoyouming <luoyouming@huawei.com>
Reviewed-by: NYangyang Li <liyangyang20@huawei.com>
Reviewed-by: NYueHaibing <yuehaibing@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>

driver inclusion
category: cleanup
bugzilla: https://gitee.com/openeuler/kernel/issues/I5USIG

-----------------------------------------------------------

The sge size is known to be constant, so it's unnecessary to use sizeof to
calculate.

Fixes: 30b70788 ("RDMA/hns: Support inline data in extented sge space for RC")
Signed-off-by: NLuoyouming <luoyouming@huawei.com>
Reviewed-by: NYangyang Li <liyangyang20@huawei.com>
Reviewed-by: NYueHaibing <yuehaibing@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>

mainline inclusion
from mainline-for-next
commit 3d67e7e2
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I5TAQ5
CVE: NA

Reference: https://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git/commit/?id=3d67e7e236ad

----------------------------------------------------------------------

The MR raw restrack attributes come from the queue context maintained by
the ROCEE.

For example:

$ rdma res show mr dev hns_0 mrn 6 -dd -jp -r
[ {
        "ifindex": 4,
        "ifname": "hns_0",
        "data": [ 1,0,0,0,2,0,0,0,0,3,0,0,0,0,2,0,0,0,0,0,32,0,0,0,2,0,0,0,
		  2,0,0,0,0,0,0,0 ]
    } ]

Link: https://lore.kernel.org/r/20220822104455.2311053-8-liangwenpeng@huawei.comSigned-off-by: NWenpeng Liang <liangwenpeng@huawei.com>
Signed-off-by: NLeon Romanovsky <leon@kernel.org>
Signed-off-by: NZhengfeng Luo <luozhengfeng@h-partners.com>
Reviewed-by: NYangyang Li <liyangyang20@huawei.com>
Reviewed-by: NYueHaibing <yuehaibing@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>

mainline inclusion
from mainline-for-next
commit dc9981ef
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I5TAQ5
CVE: NA

Reference: https://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git/commit/?id=dc9981ef17c6

----------------------------------------------------------------------

The MR restrack attributes come from the queue information maintained by
the driver.

For example:

$ rdma res show mr dev hns_0 mrn 6 -dd -jp
[ {
        "ifindex": 4,
        "ifname": "hns_0",
        "mrn": 6,
        "rkey": "300",
        "lkey": "300",
        "mrlen": 131072,
        "pdn": 8,
        "pid": 1524,
        "comm": "ib_send_bw"
    },
    "drv_pbl_hop_num": 2,
    "drv_ba_pg_shift": 14,
    "drv_buf_pg_shift": 12
}

Link: https://lore.kernel.org/r/20220822104455.2311053-7-liangwenpeng@huawei.comSigned-off-by: NWenpeng Liang <liangwenpeng@huawei.com>
Signed-off-by: NLeon Romanovsky <leon@kernel.org>
Signed-off-by: NZhengfeng Luo <luozhengfeng@h-partners.com>
Reviewed-by: NYangyang Li <liyangyang20@huawei.com>
Reviewed-by: NYueHaibing <yuehaibing@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>