driver inclusion
category: bugfix
bugzilla: NA
CVE: NA

This patch adds more information for reset DFX. Also, adds some
cleanups to reset info, move reset_fail_cnt into struct
hclge_rst_stats, and modifies some print formats.

Feature or Bugfix:Bugfix
Signed-off-by: NHuazhong Tan <tanhuazhong@huawei.com>
Reviewed-by: Nshenjian <shenjian15@huawei.com>
Reviewed-by: Nlinyunsheng <linyunsheng@huawei.com>
Reviewed-by: NYang Yingliang <yangyingliang@huawei.com>
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: NA
CVE: NA

This patch adds support for spoof check configuration for VFs.
When it is enabled, "spoof checking" is done for both mac address
and VLAN. For each VF, the HW ensures that the source MAC address
(or VLAN) of every outgoing packet exists in the MAC-list (or
VLAN-list) configured for RX filtering for that VF. If not,
the packet is dropped.

Feature or Bugfix:Bugfix
Signed-off-by: Nshenjian (K) <shenjian15@huawei.com>
Reviewed-by: Nlipeng <lipeng321@huawei.com>
Reviewed-by: NYunsheng Lin <linyunsheng@huawei.com>
Reviewed-by: NYang Yingliang <yangyingliang@huawei.com>
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: NA
CVE: NA

According to user manual, there are several independent  parameters
for switch parameter configure command. If they are configured
in two thread concurrently, we may encounter RMW issue.

Now this command is refreshed, adding a new parameter "mask", then
we can tell firmware only to change the specified parameter without
affecting others.

This patch refactors the switch parameter configure function to
support the new command.

Feature or Bugfix:Bugfix
Signed-off-by: Nshenjian (K) <shenjian15@huawei.com>
Reviewed-by: Nlipeng <lipeng321@huawei.com>
Reviewed-by: NYunsheng Lin <linyunsheng@huawei.com>
Reviewed-by: NYang Yingliang <yangyingliang@huawei.com>
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: NA
CVE: NA

This patch adds the function of setting VF status independently. The
options are auto, enable, and disable. Even if the PF is down, the
communication between VFs will be normal if the VFs are set to enable.
The commands are as follows:

'ip link set <pf> vf <vf_id> state <auto|enable|disable>'
set the vf status

'ip link show'
view the setting status

Feature or Bugfix:Bugfix
Signed-off-by: NYufeng Mo <moyufeng@huawei.com>
Reviewed-by: Nlipeng <lipeng321@huawei.com>
Reviewed-by: NYunsheng Lin <linyunsheng@huawei.com>
Reviewed-by: NYang Yingliang <yangyingliang@huawei.com>
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: NA
CVE: NA

Currently, the reset interrupt will be cleared firstly, so when
reset fails, if interrupt status register has reset interrupt,
it means there is a new coming reset.

Fixes: 764a6254ce5c ("net: hns3: clear reset interrupt status in hclge_irq_handle()")

Feature or Bugfix:Bugfix
Signed-off-by: NHuazhong Tan <tanhuazhong@huawei.com>
Reviewed-by: Nlipeng <lipeng321@huawei.com>
Reviewed-by: NYunsheng Lin <linyunsheng@huawei.com>
Reviewed-by: NYang Yingliang <yangyingliang@huawei.com>
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: NA
CVE: NA

Currently we use hclge_task_schedule() to handle link change
event when receive a link change msg. It's neccessary to remove
the HCLGE_STATE_LINK_CHANGE_REPORT_EN checking in the function
hclge_ae_stop(), because it set the HCLGE_STATE_DOWN bit at the
beginning, which will stop the hdev->service_task being scheduled.

Feature or Bugfix:Bugfix
Signed-off-by: Nshenjian (K) <shenjian15@huawei.com>
Reviewed-by: Nlipeng <lipeng321@huawei.com>
Reviewed-by: NYunsheng Lin <linyunsheng@huawei.com>
Reviewed-by: NYang Yingliang <yangyingliang@huawei.com>
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: NA
CVE: NA

Async two macro definition with M7.

Also adds a hardware error check to hclge_handle_mpf_ras_error()
which has been removed incorrectly.

Fixes: 7cc5f1c13564 ("net: hns3: add support for handling IMP error")

Feature or Bugfix:Bugfix
Signed-off-by: NHuazhong Tan <tanhuazhong@huawei.com>
Reviewed-by: Nlinyunsheng <linyunsheng@huawei.com>
Reviewed-by: NYang Yingliang <yangyingliang@huawei.com>
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: NA
CVE: NA

In some case, hns3_nic_net_up() may fail. But the firmware is
unware of this. It may still send link change message to PF,
PF needs to drop it, otherwise it may cause panic when net
stack is up and device is down.

Feature or Bugfix:Bugfix
Signed-off-by: Nshenjian (K) <shenjian15@huawei.com>
Reviewed-by: Nlipeng <lipeng321@huawei.com>
Reviewed-by: NYunsheng Lin <linyunsheng@huawei.com>
Reviewed-by: NYang Yingliang <yangyingliang@huawei.com>
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: NA
CVE: NA

Using reviewbot_c to check the code within hns3 module,
we found some mismatch to the rule set by tools.
Now we fix all the mismatch, including Comments format,
Alignment, The devil figure, etc.

Feature or Bugfix:Bugfix
Signed-off-by: Nliaoguojia <liaoguojia@huawei.com>
Reviewed-by: Nlipeng <lipeng321@huawei.com>
Reviewed-by: NYang Yingliang <yangyingliang@huawei.com>
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: NA
CVE: NA

When IMP goes errors, the hardware reports a RAS to the driver,
the driver record this kind of error. Then a IMP reset will happen,
the driver checks the reason and takes the corresponding action
when doing IMP reset.

So this patch adds imp_err_state field to the struct hclge_dev
to record the error type, and handle_imp_error ops to handle it.

Feature or Bugfix: Bugfix
Signed-off-by: NWeihang Li <liweihang@hisilicon.com>
Signed-off-by: NHuazhong Tan <tanhuazhong@huawei.com>
Reviewed-by: Nlipeng <lipeng321@huawei.com>
Reviewed-by: NYang Yingliang <yangyingliang@huawei.com>
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: NA
CVE: NA

Use delayed work instead of using timers to trigger the
hclge_serive.

Simplify the code with one less middle function and in order
to support misc irq affinity.

Feature or Bugfix:Bugfix
Signed-off-by: NYunsheng Lin <linyunsheng@huawei.com>
Reviewed-by: Nlipeng <lipeng321@huawei.com>
Reviewed-by: NYang Yingliang <yangyingliang@huawei.com>
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: NA
CVE: NA

This patch fixes a repeat code warning in two functions,
and wrap the same codes into a function.

Feature or Bugfix:Bugfix
Signed-off-by: Nhuangguangbin (A) <huangguangbin2@huawei.com>
Reviewed-by: Nlipeng <lipeng321@huawei.com>
Reviewed-by: NYang Yingliang <yangyingliang@huawei.com>
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: NA
CVE: NA

Now we can use ethtool -d command to dump some registers. However,
these registers information is not enough to find out where the problem is.

This patch add dfx registers information after original registers
when use ethtool -d commmand to dump registers.

Feature or Bugfix:Bugfix
Signed-off-by: Nhuangguangbin (A) <huangguangbin2@huawei.com>
Reviewed-by: Nlipeng <lipeng321@huawei.com>
Reviewed-by: NYang Yingliang <yangyingliang@huawei.com>
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: NA
CVE: NA

Previously, PF update link status per seconed. For some scenario,
such as storage, it requires link down event being reported
more quickly. To solve it, firmware pushes the link change event
to PF with cmdq message, and driver updates the link status
directly.
Signed-off-by: Nshenjian (K) <shenjian15@huawei.com>
Reviewed-by: Nlipeng <lipeng321@huawei.com>
Reviewed-by: NYang Yingliang <yangyingliang@huawei.com>
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: NA
CVE: NA

Two previous patch "net: hns3: delay setting of reset level for hw errors
until slot_reset is called" and "RDMA/hns: modification for the changed
set_default_reset_request interface" is hard to upstream to community
because they belong to two different repositories. We modify patch of
hns to avoid changing the set_default_reset_request(), at the same time,
we change back related code in roce.
Signed-off-by: Nliweihang <liweihang@huawei.com>
Reviewed-by: Nlipeng <lipeng321@huawei.com>
Reviewed-by: NYang Yingliang <yangyingliang@huawei.com>
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: NA
CVE: NA

The misc interrupt is used to schedule the reset and mailbox
subtask, and a 1 sec timer is used to schedule the service
subtask, which does periodic work.

This patch sets the above three subtask's affinity using the
misc interrupt' affinity.

Also this patch setup a affinity notify for misc interrupt to
allow user to change the above three subtask's affinity.

Feature or Bugfix:Bugfix
Signed-off-by: NYunsheng Lin <linyunsheng@huawei.com>
Reviewed-by: Nlipeng <lipeng321@huawei.com>
Reviewed-by: NYang Yingliang <yangyingliang@huawei.com>
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: NA
CVE: NA

When loading or unloading module, it should wait for the reset task
done before it un-initializes the client, otherwise the reset task
may cause a NULL pointer reference.

Fixes: 31218a981f90 ("net: hns3: fix race conditions between reset and module loading & unloading")

Feature or Bugfix:Bugfix
Signed-off-by: NHuazhong Tan <tanhuazhong@huawei.com>
Reviewed-by: Nlipeng <lipeng321@huawei.com>
Reviewed-by: NYang Yingliang <yangyingliang@huawei.com>
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: NA
CVE: NA

When PF ROCE client's init_instance() succeeds, it means this client
has been registered successfully, so we use HCLGE_STATE_ROCE_REGISTERED
to indicate that. And before calling PF ROCE client's uninit_instance(),
we clear this state.

So any operation of the ROCE client from HCLGE is not allowed if this
state is not set.

Feature or Bugfix:Bugfix
Signed-off-by: NHuazhong Tan <tanhuazhong@huawei.com>
Reviewed-by: Nlinyunsheng <linyunsheng@huawei.com>
Reviewed-by: NYang Yingliang <yangyingliang@huawei.com>
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: NA
CVE: NA

Currently, firmware helps manage the unicast mac vlan table
space for each PF. PF just needs to tell firmware its wanted
space when initializing, and unnecessary to free it when
unintializing.

For adding and removing mac address may be called in set_rx_mode(),
so we should use spin_lock to protect shared_umv_size instead
of mutex.

This patch removes the umv space free and changes the lock type.

Feature or Bugfix:Bugfix
Signed-off-by: Nshenjian (K) <shenjian15@huawei.com>
Reviewed-by: Nlipeng <lipeng321@huawei.com>
Reviewed-by: NYang Yingliang <yangyingliang@huawei.com>
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: NA
CVE: NA

Previously, the user fec mode may be lost after reset, the firmware
always uses the default fec mode, which may cause link down if the
link partner is using the usr fec mod. This patch fixes it.
Signed-off-by: Nshenjian (K) <shenjian15@huawei.com>
Reviewed-by: Nlipeng <lipeng321@huawei.com>
Reviewed-by: NYang Yingliang <yangyingliang@huawei.com>
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: NA
CVE: NA

This patch optimizs some variable types to remove the static check
warning.

Feature or Bugfix:Bugfix
Signed-off-by: Nmoyufeng <moyufeng@huawei.com>
Reviewed-by: Nlipeng <lipeng321@huawei.com>
Reviewed-by: NXie XiuQi <xiexiuqi@huawei.com>
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: NA
CVE: NA

This patch optimizs some variable types to remove the static check
warning.

Feature or Bugfix:Bugfix
Signed-off-by: Nmoyufeng <moyufeng@huawei.com>
Reviewed-by: Nlipeng <lipeng321@huawei.com>
Reviewed-by: NXie XiuQi <xiexiuqi@huawei.com>
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: NA
CVE: NA

When hw is resetting, firmware is unable to handle commands
from driver. So if remove vlan device from stack at this time,
it will fail to remove the vlan id from from hw vlan filter,
then the vlan filter status is unsynced with stack.

This patch fixes it by recording the vlan id delete failed,
and remove them again when reset complete.
Signed-off-by: Nshenjian (K) <shenjian15@huawei.com>
Reviewed-by: Nlipeng <lipeng321@huawei.com>
Reviewed-by: NHanjun Guo <guohanjun@huawei.com>
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: NA
CVE: NA

The next-net has fixed it, so do not need upstreaming.
Signed-off-by: NYunsheng Lin <linyunsheng@huawei.com>
Reviewed-by: Nlipeng <lipeng321@huawei.com>
Reviewed-by: NHanjun Guo <guohanjun@huawei.com>
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: NA
CVE: NA

Vf vlan table can only supports no more than 256 vlans. When user
adds too many vlans, the vf vlan table will be full, and firmware
will close the vf vlan table for the function. When vf vlan table
is full, and user keeps adding new vlans, it's unnecssary to
configure the vf vlan table, because it will always fail, and print
warning message. The extreme case is adding 4k vlans, and do reset,
it will take much time to restore these vlans, which may cause vf
reset fail by timeout.
Signed-off-by: Nshenjian (K) <shenjian15@huawei.com>
Reviewed-by: Nlipeng <lipeng321@huawei.com>
Reviewed-by: NHanjun Guo <guohanjun@huawei.com>
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: NA
CVE: NA

Mac tunnel interruptions are different from other type of RAS and MSI-X
errors, because some bits, such as OVF/LR/RF will occur during link up
and down.
HNS3 drivers should clear status of all mac tnl interruption bits but
shouldn't print any message that would mislead the users.
In case that link down and re-up in a short time because of some reasons,
we record when they occurred, and users can query them by debugfs.
Signed-off-by: Nliweihang <liweihang@huawei.com>
Reviewed-by: Nlipeng <lipeng321@huawei.com>
Reviewed-by: NHanjun Guo <guohanjun@huawei.com>
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: NA
CVE: NA

This patch adds supports for fec configuration and autoneg.
Signed-off-by: Nshenjian (K) <shenjian15@huawei.com>
Reviewed-by: Nlipeng <lipeng321@huawei.com>
Reviewed-by: NHanjun Guo <guohanjun@huawei.com>
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: NA
CVE: NA

Move HCLGE_SET_DEFAULT_RESET_REQUEST to hclge_err.h
Remove keyword static from hclge_get_reset_level()
Signed-off-by: Nliweihang <liweihang@huawei.com>
Signed-off-by: Ntianjinchuan <tianjinchuan1@huawei.com>
Reviewed-by: Nlipeng <lipeng321@huawei.com>
Reviewed-by: NXie XiuQi <xiexiuqi@huawei.com>
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: NA
CVE: NA

{topost} net: hns3: stop mailbox handling when command queue need re-init
{topost} net: hns3: modify HNS3_NIC_STATE_INITED flag in
	 	    hns3_reset_notify_uninit_enet
{topost} net: hns3: enlarge reset event time interval
{topost} net: hns3: add error handler for initializing command queue
{topost} net: hns3: prevent change mtu when resetting
{topost} net: hns3: check return value of reset TQP in hclge_ae_stop()
{topost} net: hns3: fix improper error handling for
		    hns3_reset_notify_init_enet
{topost} net: hns3: assign ring_data to NULL after free in
		    hns3_get_ring_config
{topost} net: hns3: check resetting status in hns3_get_stats()
Signed-off-by: NHuazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: Ntanhuazhong <tanhuazhong@huawei.com>
Reviewed-by: Nlipeng <lipeng321@huawei.com>
Reviewed-by: NXie XiuQi <xiexiuqi@huawei.com>
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: NA
CVE: NA

Optimize the waiting time for obtaining tqp reset completion status.
Signed-off-by: NLiuzhongzhu <liuzhongzhu@huawei.com>
Reviewed-by: Nlipeng <lipeng321@huawei.com>
Reviewed-by: NXie XiuQi <xiexiuqi@huawei.com>
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: Feature
bugzilla: NA
CVE: NA

This patch adds statistics info for PF reset info,
also, provides a debugfs command to dump these statistics.
Signed-off-by: NHuazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: Ntanhuazhong <tanhuazhong@huawei.com>
Reviewed-by: Nlipeng <lipeng321@huawei.com>
Reviewed-by: NXie XiuQi <xiexiuqi@huawei.com>
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: feature
bugzilla: NA
CVE: NA

This patch fixes comments by code review.
Signed-off-by: Nshenjian (K) <shenjian15@huawei.com>
Reviewed-by: Nlipeng <lipeng321@huawei.com>
Reviewed-by: NXie XiuQi <xiexiuqi@huawei.com>
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: feature
bugzilla: NA
CVE: NA

This patch adds aRFS support for hns3 driver. Use the existing
filter management functions to insert TCPv4/UDPv4/TCPv6/UDPv6
flow director filters.

The aRFS rule is added automatically by stack, and user can
also add flow director rule with ethtool command. But the two.
modes can not work simultaneously.
Signed-off-by: Nshenjian (K) <shenjian15@huawei.com>
Reviewed-by: Nlipeng <lipeng321@huawei.com>
Reviewed-by: NXie XiuQi <xiexiuqi@huawei.com>
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: feature

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

Based on add763cbef9424c6ea624dce6d6d2d51048cf9da
("net: hns3: Reduce resources use in kdump kernel")
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: feature

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

Based on dde21a4eedd17955f173f055e2d702dadb1e70ea
("performance optimizations")
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

Sync hns3 driver from drivers team.
Based on 5db38ff635972569325bbf95db88fa3558f27ea7
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

The offset of msix number for roce is different between different
revision id. We should get it from firmware, instead of a fix value.
This patch refines the msix allocation, make it compatible.
Signed-off-by: NJian Shen <shenjian15@huawei.com>
Signed-off-by: NPeng Li <lipeng321@huawei.com>
Signed-off-by: NSalil Mehta <salil.mehta@huawei.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add the SPDX identifiers to HNS3 PF driver.
Signed-off-by: NJian Shen <shenjian15@huawei.com>
Signed-off-by: NPeng Li <lipeng321@huawei.com>
Signed-off-by: NSalil Mehta <salil.mehta@huawei.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The struct hclge_desc_cb and hclge_desc_cb are never used in
anywhere. This patch removes them.
Signed-off-by: NJian Shen <shenjian15@huawei.com>
Signed-off-by: NPeng Li <lipeng321@huawei.com>
Signed-off-by: NSalil Mehta <salil.mehta@huawei.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Use BIT() and GENMASK() to convert the bit mask, modify
the inconsistent ones, and remove useless ones.
Signed-off-by: NJian Shen <shenjian15@huawei.com>
Signed-off-by: NPeng Li <lipeng321@huawei.com>
Signed-off-by: NSalil Mehta <salil.mehta@huawei.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>