hulk inclusion
category: feature
feature: TIPC
bugzilla: NA
CVE: NA
Link: https://gitee.com/openeuler/kernel/issues/I1TDS3



The Transparent Inter Process Communication (TIPC) protocol is
specially designed for intra cluster communication. This protocol
originates from Ericsson where it has been used in carrier grade
cluster applications for many years.

Signed-off-by: Xie XiuQi <xiexiuqi@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: NA
CVE: NA

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

This patch is used to update driver version to 1.9.38.6.

Signed-off-by: Shengzui You <youshengzui@huawei.com>
Reviewed-by: Weiwei Deng <dengweiwei@huawei.com>
Reviewed-by: Zhaohui Zhong <zhongzhaohui@huawei.com>
Reviewed-by: Junxin Chen <chenjunxin1@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: NA
CVE: NA

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

This patch adds support of dumping MAC umv counter in debugfs,
which will be helpful for debugging.

Signed-off-by: Guojia Liao <liaoguojia@huawei.com>
Reviewed-by: Peng Li <lipeng321@huawei.com>
Reviewed-by: Weiwei Deng <dengweiwei@huawei.com>
Reviewed-by: Zhaohui Zhong <zhongzhaohui@huawei.com>
Reviewed-by: Junxin Chen <chenjunxin1@huawei.com>
Signed-off-by: Shengzui You <youshengzui@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: NA
CVE: NA

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

We consider that the uniqueness of the MAC address configuration should
be guaranteed by the user. So the restriction on the configuration of
duplicate MAC addresses is removed in patch "remove the limitation of
MAC address duplicate configuration".

What happens if a duplicate MAC address is still configured?

If the MAC address A is configured to vport A and then vport B. The MAC
address of vport A in the hardware becomes invalid. If the address of
vport A is changed to MAC address B, the driver needs to delete the MAC
address A of vport A. Due to the MAC address A of vport A has become
invalid in the hardware entry, so "-ENOENT" is returned. In this case, the
"used_umv_size" value recorded in driver is not updated. As a result, the
MAC entry status of the software is inconsistent with that of the hardware.

Therefore, the driver updates the umv size even if the MAC entry cannot be
found. Ensure that the software and hardware status is consistent.

Signed-off-by: Guojia Liao <liaoguojia@huawei.com>
Reviewed-by: Peng Li <lipeng321@huawei.com>
Reviewed-by: Weiwei Deng <dengweiwei@huawei.com>
Reviewed-by: Zhaohui Zhong <zhongzhaohui@huawei.com>
Reviewed-by: Junxin Chen <chenjunxin1@huawei.com>
Signed-off-by: Shengzui You <youshengzui@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: 4472

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

The out_size in nictool is passed in by the user mode. If out_size is
not checked, it will cause too much memory in the kernel space to be
copied to the user space.

Signed-off-by: Chiqijun <chiqijun@huawei.com>
Reviewed-by: Zengweiliang <zengweiliang.zengweiliang@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: 4472

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

When the user mode tool sets ETS, the input value tc id exceeds the
maximum value of 7, which will cause the array to be accessed
out of bounds. Driver check the legality of tc id.

Signed-off-by: Chiqijun <chiqijun@huawei.com>
Reviewed-by: Zengweiliang <zengweiliang.zengweiliang@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: 4472

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

Rename camelCase used in nictool.

Signed-off-by: Chiqijun <chiqijun@huawei.com>
Reviewed-by: Zengweiliang <zengweiliang.zengweiliang@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: 4472

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

Fix alignment and code style.

Signed-off-by: Chiqijun <chiqijun@huawei.com>
Reviewed-by: Zengweiliang <zengweiliang.zengweiliang@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: 4472

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

Delete unused heartbeat enhancement feature.

Signed-off-by: Chiqijun <chiqijun@huawei.com>
Reviewed-by: Zengweiliang <zengweiliang.zengweiliang@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: 4472

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

Delete the unused chip fault handling process.

Signed-off-by: Chiqijun <chiqijun@huawei.com>
Reviewed-by: Zengweiliang <zengweiliang.zengweiliang@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: 4472

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

Delete unused microcode back pressure feature.

Signed-off-by: Chiqijun <chiqijun@huawei.com>
Reviewed-by: Zengweiliang <zengweiliang.zengweiliang@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: 4472

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

Fix misspelled word and wrong print format.

Signed-off-by: Chiqijun <chiqijun@huawei.com>
Reviewed-by: Zengweiliang <zengweiliang.zengweiliang@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: 4472

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

update hinic version to 2.3.2.15

Signed-off-by: Chiqijun <chiqijun@huawei.com>
Reviewed-by: Zengweiliang <zengweiliang.zengweiliang@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: 4472

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

When the module parameter poll_weight is set to 2^32-1, driver cannot
receive packets. We limit the maximum value of poll weight with
queue depth(4K).

Signed-off-by: Chiqijun <chiqijun@huawei.com>
Reviewed-by: Zengweiliang <zengweiliang.zengweiliang@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: 4472

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

PFC and Pause of 1822 cannot be enabled at the same time, so if PFC is
turned on, pause is not allowed.

Signed-off-by: Chiqijun <chiqijun@huawei.com>
Reviewed-by: Zengweiliang <zengweiliang.zengweiliang@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: 4472

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

event_flag in struct hinic_msg_pf_to_mgmt marks the sending status of
the message, add lock for mgmt channel event_flag in
hinic_force_complete_all.

Signed-off-by: Chiqijun <chiqijun@huawei.com>
Reviewed-by: Zengweiliang <zengweiliang.zengweiliang@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

driver inclusion
category: bugfix
bugzilla: 4472

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

Use u16 for cons_idx and delta in struct hinic_rxq, or we'll meet a
overflow when a large number of packets are received.

Signed-off-by: Chiqijun <chiqijun@huawei.com>
Reviewed-by: Zengweiliang <zengweiliang.zengweiliang@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

stable inclusion
from linux-4.19.132
commit 2b3faab744b04de198aee4ed8dd74fca7a2a8306
CVE: CVE-2020-24394

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

commit 22cf8419 upstream.

The server is failing to apply the umask when creating new objects on
filesystems without ACL support.

To reproduce this, you need to use NFSv4.2 and a client and server
recent enough to support umask, and you need to export a filesystem that
lacks ACL support (for example, ext4 with the "noacl" mount option).

Filesystems with ACL support are expected to take care of the umask
themselves (usually by calling posix_acl_create).

For filesystems without ACL support, this is up to the caller of
vfs_create(), vfs_mknod(), or vfs_mkdir().

Reported-by: Elliott Mitchell <ehem+debian@m5p.com>
Reported-by: Salvatore Bonaccorso <carnil@debian.org>
Tested-by: Salvatore Bonaccorso <carnil@debian.org>
Fixes: 47057abd ("nfsd: add support for the umask attribute")
Cc: stable@vger.kernel.org
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

ascend inclusion
category: feature
bugzilla: NA
CVE: NA

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

There are too many ascend features enable flag, all of them is
used for all ascend soc till now, so use a new enable flag
to enable all of them for ascend platform by default, it would
clean and simplify the bootargs.

Also clean some code warning.

v2: modify the wrong config name.

v3: modify the wrong include head file.

Signed-off-by: Ding Tianhong <dingtianhong@huawei.com>
Reviewed-by: Zefan Li <lizefan@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

hulk inclusion
category: bugfix
bugzilla: NA
CVE: NA

Enable ARM_SBSA_WATCHDOG_PANIC_NOTIFIER in hulk_defconfig.

Signed-off-by: Wang Wensheng <wangwensheng4@huawei.com>
Reviewed-by: Xiongfeng Wang <wangxiongfeng2@huawei.com>
Reviewed-by: Ding Tianhong <dingtianhong@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

hulk inclusion
category: feature
bugzilla: NA
CVE: NA

When the kernel panic we reset the timeout to pretimeout, then after
pretimeout seconds the sbsa_gwdt reset the system.

The pretimeout to be set is configured via the watchdog-core, but the
meaning is different from that of the watchdog-core. The pretimeout here
defines the new timeout for the sbsa_gwdt after panic while that of the
watchdog-core is meaningful where the WDT would raise an interrupt
before timeout and it defines the length of that period. That period of
sbsa_gwdt cannot be changed separately so the redefination here doesn't
make trouble.

The pretimeout here follow the same limit of that of watchdog-core that
the pretimeout shall be smaller than timeout, since we prefer that the
sbsa_gwdt would reset the system sooner on panic.

We add a new kconfig ARM_SBSA_WATCHDOG_PANIC_NOTIFIER to control the
feature. Set action to 2 to enable the notifier.

Signed-off-by: Wang Wensheng <wangwensheng4@huawei.com>
Reviewed-by: Xiongfeng Wang <wangxiongfeng2@huawei.com>
Reviewed-by: Ding Tianhong <dingtianhong@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

ascend inclusion
category: feature
bugzilla: NA
CVE: NA

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

Normally, pages can not be allocated from CDM node(explicit allocation
requests from kernel or user process MPOL_BIND policy which has CDM
nodes).

This situation may happen. Within the limit bytes range of the memcg,
the CDM nodes have a large amount of free memory, and other nodes
have no available free memory. Then, the kernel or user process can not
get required memory resources normally.

For example:
  size of CMD : A mbytes
  size of non CMD : B mbytes
  limit mbytes of memcg : C mbytes. A,B < C < (A + B).
If app(user app and OS service app) used up non CMD memory, but a large
amount of CDM memory is available. Since OS service app can't get pages
from CDM node, the allocating of pages should be failed. This is not
what we expect. We hope that the memcg can be used to restrict the
memory used by some user apps to ensure that certain memory is available
for system services.

Therefore, the CDM memory does not need to be charged in the memcg. The
total size of CDM is already a limit.

Signed-off-by: Zhou Guanghui <zhouguanghui1@huawei.com>
Reviewed-by: Ding Tianhong <dingtianhong@huawei.com>
Reviewed-by: Kefeng Wang <wangkefeng.wang@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

ascend inclusion
category: feature
bugzilla: NA
CVE: NA

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

Add Message Base SPI optional property for hisilicon

Signed-off-by: Zhou Guanghui <zhouguanghui1@huawei.com>
Reviewed-by: Ding Tianhong <dingtianhong@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

ascend inclusion
category: feature
bugzilla: NA
CVE: NA

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

The current SMMU driver supports SPI WireInterrupt and Message Based
SPI. However, some hisilicon chips use the Messaged Based SPI. Therefore,
a special attribute is added to indentify this situation. Add an option
"hisicion,message-based-spi" and addr of GICD_SETSPI to dts node of SMMU,
like this:
    hisicion,message-based-spi
    iommu-spi-base=<0x10 0x9000040> //Addr of GICD_SETSPI: 0x1009000040

Signed-off-by: Zhou Guanghui <zhouguanghui1@huawei.com>
Reviewed-by: Ding Tianhong <dingtianhong@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

hulk inclusion
category: bugfix
bugzilla: NA
CVE: CVE-2019-16089

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

nla_nest_start may fail and return NULL. The check is inserted, and
errno is selected based on other call sites within the same source code.
Update: removed extra new line.
v3 Update: added release reply, thanks to Michal Kubecek for pointing
out.

Signed-off-by: Navid Emamdoost <navid.emamdoost@gmail.com>
Reviewed-by: Michal Kubecek <mkubecek@suse.cz>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Reviewed-by: Jason Yan <yanaijie@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

ascend inclusion
category: feature
Bugzilla: N/A
CVE: N/A

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

Signed-off-by: Xu Qiang <xuqiang36@huawei.com>
Reviewed-by: Ding Tianhong <dingtianhong@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

ascend inclusion
category: feature
bugzilla: NA
CVE: NA

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

For Ascend platform, other NON-OS managed GICRs need be initialized
in OS.

Signed-off-by: Yang Yingliang <yaingyingliang@huawei.com>
Signed-off-by: Xu Qiang <xuqiang36@huawei.com>
Signed-off-by: Lijun Fang <fanglijun3@huawei.com>
Reviewed-by: Hanjun Guo <guohanjun@huawei.com>
Signed-off-by: Xu Qiang <xuqiang36@huawei.com>
Reviewed-by: Ding Tianhong <dingtianhong@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

ascend inclusion
category: feature
bugzilla: NA
CVE: NA

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

Enable charge migrate hugepages for hulk defconfig.

Signed-off-by: Zhou Guanghui <zhouguanghui1@huawei.com>
Reviewed-by: Ding Tianhong <dingtianhong@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

ascend inclusion
category: feature
Bugzilla: N/A
CVE: N/A

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

When the driver gets huge pages by alloc_huge_page_node, it attempts
to apply for migrate hugepages after the reserved memory hugepages
are used up. We expect that the migrated hugepages that are applied
for can be charged in memcg to limit the memory usage.

__GFP_ACOUNT flag is added to gfp mask before we allocate migrage
hugepages. Then, if memcg is set by memalloc_use_memcg(), the
allocated migrate hugepages will be charged to this memcg.

Signed-off-by: Zhou Guanghui <zhouguanghui1@huawei.com>
Reviewed-by: Ding Tianhong <dingtianhong@huawei.com>
Reviewed-by: Kefeng Wang <wangkefeng.wang@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

ascend inclusion
category: bugfix
Bugzilla: N/A
CVE: N/A

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

Signed-off-by: Xu Qiang <xuqiang36@huawei.com>
Reviewed-by: Hanjun Guo <guohanjun@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

serial: amba-pl011: Fix serial port discard interrupt when interrupt signal line of serial port is connected to mbigen.

ascend inclusion
category: bugfix
Bugzilla: N/A
CVE: N/A

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

Hisi when designing ascend chip, connect the serial port interrupt
signal lines to mbigen equipment, mbigen write GICD_SETSPI_NSR
register trigger the SPI interrupt. This can result in serial
port drop interrupts.

Signed-off-by: Xu Qiang <xuqiang36@huawei.com>
Reviewed-by: Hanjun Guo <guohanjun@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

ascend inclusion
category: bugfix
bugzilla: NA
CVE: NA

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

Fix a mistake check for commit bd6c06e0917d
("iommu: introduce device fault report API")

Fixes: bd6c06e0917d ("iommu: introduce device fault report API")
Signed-off-by: Bixuan Cui <cuibixuan@huawei.com>
Reviewed-by: Hanjun Guo <guohanjun@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

ascend inclusion
category: bugfix
Bugzilla: N/A
CVE: N/A

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

Signed-off-by: Xu Qiang <xuqiang36@huawei.com>
Acked-by: Hanjun Guo <guohanjun@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

ascend inclusion
category: feature
bugzilla: NA
CVE: NA

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

Enable the high priority of iopf workqueue for hulk default config.

Signed-off-by: Zhou Guanghui <zhouguanghui1@huawei.com>
Reviewed-by: Hanjun Guo <guohanjun@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

ascend inclusion
category: feature
bugzilla: NA
CVE: NA

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

The iopf workqueue which means IO Page Fault workqueue is used for device
page fault.

During the processing of the iopf work, the handle_mm_fault() is invoked
to alloc page and create a mapping. It is expected that the CPU that
allocates memory and the device that triggers the fault are affinity to
the same NUMA node to ensure that the memory allocated is local access to
CPU and device. This will also improve the performance.

Signed-off-by: Zhou Guanghui <zhouguanghui1@huawei.com>
Reviewed-by: Hanjun Guo <guohanjun@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

ascend inclusion
category: Bugfix
bugzilla: NA
CVE: NA

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

System cann't use the cdm nodes memory, but it can mmap all nodes huge
pages, so it will cause Bus error when mmap succeed but the huge pages
were not enough.

When set the cdmmask, users will transfer the numa id by mmap flag to
map the specific numa node hugepages, if there was not enough hugepages
on this node, return -ENOMEM.

v2: Fix compile error when disable CONFIG_COHERENT_DEVICE

Signed-off-by: Fang Lijun <fanglijun3@huawei.com>
Reviewed-by: Ding Tianhong <dingtianhong@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

ascend inclusion
category: feature
bugzilla: NA
CVE: NA

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

Enable the ascend oom control features for hulk default config.

Signed-off-by: Weilong Chen <chenweilong@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

ascend inclusion
category: feature
bugzilla: NA
CVE: NA

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

Support disable oom-killer, and report oom events to bbox
vm.enable_oom_killer:
	0: disable oom killer
	1: enable oom killer (default,compatible with mainline)
	2: disable oom killer and panic on oom

Signed-off-by: Weilong Chen <chenweilong@huawei.com>
Reviewed-by: Kefeng Wang <wangkefeng.wang@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

ascend inclusion
category: feature
bugzilla: NA
CVE: NA

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

Add alloc and release memory functions in svm. And the physical address
of the memory is within 4GB.

For example:
    /* alloc */
    fd = open("dev/svm0",);
    mmap(0, ALLOC_SIZE,, MAP_PA32BIT, fd, 0);

    /* free */
    ioctl(fd, SVM_IOCTL_RELEASE_PHYS32,);
    close(fd);

Signed-off-by: Bixuan Cui <cuibixuan@huawei.com>
Reviewed-by: Zefan Li <lizefan@huawei.com>
Reviewed-by: Hanjun Guo <guohanjun@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

ascend inclusion
category: feature
bugzilla: NA
CVE: NA

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

Export cpu_suspend/cpu_resume/psci_ops for lowpower driver.

Signed-off-by: Bixuan Cui <cuibixuan@huawei.com>
Reviewed-by: Hanjun Guo <guohanjun@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>