Merge Pull Request from: @ci-robot 
 
PR sync from: Ruan Jinjie <ruanjinjie@huawei.com>
https://mailweb.openeuler.org/hyperkitty/list/kernel@openeuler.org/message/EHN6ENC7DMIVST6HGFD53IQMXC7J6VJF/ 
arm64: Add framework to turn an IPI as NMI

Sumit Garg (7):
  arm64: Add framework to turn IPI as NMI
  irqchip/gic-v3: Enable support for SGIs to act as NMIs
  arm64: smp: Assign and setup an IPI as NMI
  nmi: backtrace: Allow runtime arch specific override
  arm64: ipi_nmi: Add support for NMI backtrace
  kgdb: Expose default CPUs roundup fallback mechanism
  arm64: kgdb: Roundup cpus using IPI as NMI

Xiongfeng Wang (1):
  arm64: ipi_nmi: fix compile error when CONFIG_KGDB is disabled


-- 
2.34.1
 
https://gitee.com/openeuler/kernel/issues/I7R4EN 
 
Link:https://gitee.com/openeuler/kernel/pulls/1665 

Reviewed-by: Wei Li <liwei391@huawei.com> 
Reviewed-by: Xiongfeng Wang <wangxiongfeng2@huawei.com> 
Signed-off-by: Wei Li <liwei391@huawei.com> 

Merge Pull Request from: @xingmz 
 
This config is based on riscv64-config.all.final of
https://gitee.com/openeuler-risc-v/kernel.git
openEuler-22.03-LTS-SP2:5.10.0-153.1.0.80, generated by 'make olddefconfig',
merged configs tested in kernel v6.1.19,
and cleanup compiler's indicators. 
 
Link:https://gitee.com/openeuler/kernel/pulls/1511 

Reviewed-by: Zhang Xuzhou <zhangxuzhou4@huawei.com> 
Signed-off-by: Zheng Zengkai <zhengzengkai@huawei.com> 
Acked-by: Xie XiuQi <xiexiuqi@huawei.com> 

hulk inclusion
category: bugfix
bugzilla: https://gitee.com/openeuler/kernel/issues/I7R4EN
CVE: NA

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

Fix the following compile error.

arch/arm64/kernel/ipi_nmi.c: In function ‘ipi_nmi_handler’:
arch/arm64/kernel/ipi_nmi.c:54:7: error: implicit declaration of function ‘kgdb_nmicallback’ [-Werror=implicit-function-declaration]
  if (!kgdb_nmicallback(cpu, get_irq_regs()))
       ^~~~~~~~~~~~~~~~
Signed-off-by: NXiongfeng Wang <wangxiongfeng2@huawei.com>
Reviewed-by: NXie XiuQi <xiexiuqi@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>
Signed-off-by: NRuan Jinjie <ruanjinjie@huawei.com>

maillist inclusion
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I7R4EN
CVE: NA
Reference: https://www.spinics.net/lists/arm-kernel/msg851005.html

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

arm64 platforms with GICv3 or later supports pseudo NMIs which can be
leveraged to roundup CPUs which are stuck in hard lockup state with
interrupts disabled that wouldn't be possible with a normal IPI.

So instead switch to roundup CPUs using IPI turned as NMI. And in
case a particular arm64 platform doesn't supports pseudo NMIs,
it will switch back to default kgdb CPUs roundup mechanism.
Signed-off-by: NSumit Garg <sumit.garg@linaro.org>
Signed-off-by: NWei Li <liwei391@huawei.com>
Reviewed-by: NXie XiuQi <xiexiuqi@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>
Signed-off-by: NRuan Jinjie <ruanjinjie@huawei.com>

maillist inclusion
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I7R4EN
CVE: NA
Reference: https://www.spinics.net/lists/arm-kernel/msg851005.html

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

Add a new API kgdb_smp_call_nmi_hook() to expose default CPUs roundup
mechanism to a particular archichecture as a runtime fallback if it
detects to not support NMI roundup.

Currently such an architecture example is arm64 supporting pseudo NMIs
feature which is only available on platforms which have support for GICv3
or later version.
Signed-off-by: NSumit Garg <sumit.garg@linaro.org>
Signed-off-by: NWei Li <liwei391@huawei.com>
Reviewed-by: NXie XiuQi <xiexiuqi@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>
Signed-off-by: NRuan Jinjie <ruanjinjie@huawei.com>

maillist inclusion
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I7R4EN
CVE: NA
Reference: https://www.spinics.net/lists/arm-kernel/msg851005.html

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

Enable NMI backtrace support on arm64 using IPI turned as an NMI
leveraging pseudo NMIs support. It is now possible for users to get a
backtrace of a CPU stuck in hard-lockup using magic SYSRQ.
Signed-off-by: NSumit Garg <sumit.garg@linaro.org>
Signed-off-by: NWei Li <liwei391@huawei.com>
Reviewed-by: NXie XiuQi <xiexiuqi@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>
Signed-off-by: NRuan Jinjie <ruanjinjie@huawei.com>

maillist inclusion
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I7R4EN
CVE: NA
Reference: https://www.spinics.net/lists/arm-kernel/msg851005.html

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

Add a boolean return to arch_trigger_cpumask_backtrace() to support a
use-case where a particular architecture detects at runtime if it supports
NMI backtrace or it would like to fallback to default implementation using
SMP cross-calls.

Currently such an architecture example is arm64 supporting pseudo NMIs
feature which is only available on platforms which have support for GICv3
or later version.
Signed-off-by: NSumit Garg <sumit.garg@linaro.org>
Signed-off-by: NWei Li <liwei391@huawei.com>
Reviewed-by: NXie XiuQi <xiexiuqi@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>
Signed-off-by: NRuan Jinjie <ruanjinjie@huawei.com>

maillist inclusion
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I7R4EN
CVE: NA
Reference: https://www.spinics.net/lists/arm-kernel/msg851005.html

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

Assign an unused IPI which can be turned as NMI using ipi_nmi framework.
Also, invoke corresponding dynamic IPI setup/teardown APIs.
Signed-off-by: NSumit Garg <sumit.garg@linaro.org>
Signed-off-by: NWei Li <liwei391@huawei.com>
Reviewed-by: NXie XiuQi <xiexiuqi@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>
Signed-off-by: NRuan Jinjie <ruanjinjie@huawei.com>

maillist inclusion
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I7R4EN
CVE: NA
Reference: https://www.spinics.net/lists/arm-kernel/msg851005.html

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

Add support to handle SGIs as pseudo NMIs. As SGIs or IPIs default to a
special flow handler: handle_percpu_devid_fasteoi_ipi(), so skip NMI
handler update in case of SGIs.

Also, enable NMI support prior to gic_smp_init() as allocation of SGIs
as IRQs/NMIs happen as part of this routine.
Signed-off-by: NSumit Garg <sumit.garg@linaro.org>
Signed-off-by: NWei Li <liwei391@huawei.com>
Reviewed-by: NXie XiuQi <xiexiuqi@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>
Signed-off-by: NRuan Jinjie <ruanjinjie@huawei.com>

maillist inclusion
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I7R4EN
CVE: NA
Reference: https://www.spinics.net/lists/arm-kernel/msg851005.html

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

Introduce framework to turn an IPI as NMI using pseudo NMIs. The main
motivation for this feature is to have an IPI that can be leveraged to
invoke NMI functions on other CPUs.

And current prospective users are NMI backtrace and KGDB CPUs round-up
whose support is added via future patches.
Signed-off-by: NSumit Garg <sumit.garg@linaro.org>
Signed-off-by: NWei Li <liwei391@huawei.com>
Reviewed-by: NXie XiuQi <xiexiuqi@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>
Signed-off-by: NRuan Jinjie <ruanjinjie@huawei.com>

riscv inclusion
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I7NK25

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

This config is based on riscv64-config.all.final of
https://gitee.com/openeuler-risc-v/kernel.git
openEuler-22.03-LTS-SP2:5.10.0-153.1.0.80
generated by 'make olddefconfig', merged configs tested in kernel v6.1.19,
and cleanup compiler's indicators.
Signed-off-by: NMingzheng Xing <xingmingzheng@iscas.ac.cn>

Merge Pull Request from: @fanhuimeng 
 
 
 
Link:https://gitee.com/openeuler/kernel/pulls/1407 

Reviewed-by: Xie XiuQi <xiexiuqi@huawei.com> 
Reviewed-by: Wei Li <liwei391@huawei.com> 
Signed-off-by: Wei Li <liwei391@huawei.com> 

Merge Pull Request from: @kwb0523 
 
add gid field in sock;
add bpf_get_sockops_uid_gid and bpf_sk_original_addr helper function;
original issue：https://gitee.com/openeuler/kernel/issues/I545NW 
 
Link:https://gitee.com/openeuler/kernel/pulls/1443 

Reviewed-by: Wei Li <liwei391@huawei.com> 
Signed-off-by: Wei Li <liwei391@huawei.com> 

Merge Pull Request from: @cherry530 
 
openeuler inclusion
category: feature
bugzilla:  https://gitee.com/openeuler/kernel/issues/I7LM8U

Reference:https://gitee.com/src-openeuler/kernel/pulls/933

----------------------------------------------------
issue causes：
Module lz4_compress.o will not be compiled into the kernel
unless option CONFIG_LZ4_COMPRESS setted to 'y' or 'm'.
When kernel does not provide the lz4_compress module,
kmod-kvdo will not find the missing symbols LZ4_compress_default.

Solution：
Add CONFIG_LZ4_COMPRESS and CONFIG_DRM_GUD option for
x86_64 architecture.

 
 
Link:https://gitee.com/openeuler/kernel/pulls/1430 

Reviewed-by: Liu Chao <liuchao173@huawei.com> 
Reviewed-by: Wei Li <liwei391@huawei.com> 
Signed-off-by: Wei Li <liwei391@huawei.com> 

hulk inclusion
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I7LTRR
CVE: NA

Reference: https://gitee.com/openeuler/kernel/commit/97aeb284efece2a8af5bb424d4e980905927f7bb

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

Add new optname(BPF_SO_ORIGINAL_DST 800, BPF_SO_REPLY_SRC 801)
to get origdst/reply src for bpf progs.
Now only support IPv4.
Signed-off-by: NWang Yufen <wangyufen@huawei.com>
Signed-off-by: NLiu Jian <liujian56@huawei.com>
Signed-off-by: NJofDiamonds <kwb0523@163.com>
Reviewed-by: Nwuchangye <wuchangye@huawei.com>

hulk inclusion
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I7LTRR
CVE: NA

Reference: https://gitee.com/openeuler/kernel/commit/9d4b4a05ae00d7e5b2f8a33fdbdf974df182ccb7

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

Add the function for bpf sock_ops hook to get sock's uid and gid.
Signed-off-by: NLiu Jian <liujian56@huawei.com>
Conflicts:
	include/uapi/linux/bpf.h
	net/core/filter.c
	tools/include/uapi/linux/bpf.h
Signed-off-by: NJofDiamonds <kwb0523@163.com>
Reviewed-by: Nwuchangye <wuchangye@huawei.com>

hulk inclusion
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I7LTRR
CVE: NA

Reference: https://gitee.com/openeuler/kernel/commit/f6740a11189620e5fd5ec0642c41b00f71b01689

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

UID and GID are requested as filters for socketmap, but we can only get
UID from sock structure. This patch adds GID field to struct sock as UID.
Signed-off-by: NLu Wei <luwei32@huawei.com>
Signed-off-by: NLiu Jian <liujian56@huawei.com>
Conflicts:
	include/net/sock.h
	net/core/sock.c
Signed-off-by: NJofDiamonds <kwb0523@163.com>
Reviewed-by: Nwuchangye <wuchangye@huawei.com>

x86_64 architecture.

openeuler inclusion
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I7LM8U
Reference: https://gitee.com/src-openeuler/kernel/pulls/933

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

issue causes：
Module lz4_compress.o will not be compiled into the kernel
unless option CONFIG_LZ4_COMPRESS setted to 'y' or 'm'.
When kernel does not provide the lz4_compress module,
kmod-kvdo will not find the missing symbols LZ4_compress_default.

Solution：
Add CONFIG_LZ4_COMPRESS and CONFIG_DRM_GUD option for
x86_64 architecture.
Signed-off-by: Nwu-leilei <wu18740459704@163.com>
Signed-off-by: Ncherry530 <707078654@qq.com>

Merge Pull Request from: @zhang-mingyi66 
 
modify sockops.

change connect function to implement defer connect. 
 
Link:https://gitee.com/openeuler/kernel/pulls/1414 

Reviewed-by: Wei Li <liwei391@huawei.com> 
Signed-off-by: Wei Li <liwei391@huawei.com> 

hulk inclusion
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I7LE1H

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

Since commit 2585cd62 ("bpf: Only reply field should be writeable"),
sockops is not allowd to modify the replylong field except replylong[0].
The reason is that the replylong[1] to replylong[3] field is not used
at that time.

But in actual use, we can call `BPF_CGROUP_RUN_PROG_SOCK_OPS` in the
kernel modules and expect sockops to return some useful data.

The design comment about bpf_sock_ops::replylong in
include/uapi/linux/bpf.h is described as follows:

```
  struct bpf_sock_ops {
        __u32 op;
        union {
                __u32 args[4];          /* Optionally passed to bpf program */
                __u32 reply;            /* Returned by bpf program          */
                __u32 replylong[4];     /* Optioznally returned by bpf prog  */
        };
  ...
```

It seems to contradict the purpose for which the field was originally
designed. Let's remove this restriction.

Fixes: 2585cd62 ("bpf: Only reply field should be writeable")
Signed-off-by: Nzhang-mingyi66 <zhangmingyi5@huawei.com>

hulk inclusion
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I7LE1H

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

A trace point is added to the connection process. Theebpf program
can be mounted to modify the return value of the function. This is
mandatory for delaying the establishment of an ebpf link. After
the connection is complete, a message is returned immediately and
no unnecessary operation is performed.
Signed-off-by: Nzhang-mingyi66 <zhangmingyi5@huawei.com>

hulk inclusion
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I7LE1H

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

The bpf_defer_connect bit is added for inet_sock to indicate
whether the current socket is changed to the bpf program to
delay link establishment.
Signed-off-by: Nzhang-mingyi66 <zhangmingyi5@huawei.com>

sockets during setopt

hulk inclusion
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I7LE1H

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

Currently, the ebpf program can distinguish sockets according to
the address accessed by the client, and use the ULP framework to
modify the matched sockets to delay link establishment.
Signed-off-by: Nzhang-mingyi66 <zhangmingyi5@huawei.com>

the sockops call

hulk inclusion
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I7LE1H

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

Currently, a permission status code is required to identify
that the access to the sockops is from the delayed link establishment
scenario. Therefore, "BPF_SOCK_OPS_TCP_DEFFER_CONNECT_CB"
Signed-off-by: Nzhang-mingyi66 <zhangmingyi5@huawei.com>

kylin inclusion
category: bugfix
bugzilla: https://gitee.com/openeuler/kernel/issues/I7KYTH
CVE: NA

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

While mounting the filesystem named 'gfs2' with the command such as
"mount /dev/sda10 /mnt",
we get:
        "mount: /mnt: unknown filesystem type 'gfs2'"

Fix this by requiring "CONFIG_GFS2_FS=m" instead of
"# CONFIG_GFS2_FS is not set" in the loongson3_defconfig configuration file
After this commit, this issue is now resolved.
Tested on loongarch 3C5000.
Signed-off-by: Nmengfanhui <mengfanhui@kylinos.cn>

Merge Pull Request from: @ci-robot 
 
PR sync from: Wei Li <liwei391@huawei.com>
https://mailweb.openeuler.org/hyperkitty/list/kernel@openeuler.org/message/R3ZWBSDS6K7GJ6Q63W4KJVEU2C4YJXKR/ 
 
 
Link:https://gitee.com/openeuler/kernel/pulls/1375 

Reviewed-by: Xie XiuQi <xiexiuqi@huawei.com> 
Signed-off-by: Zheng Zengkai <zhengzengkai@huawei.com> 

hulk inclusion
category: bugfix
bugzilla: https://gitee.com/openeuler/kernel/issues/I7KBJI

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

Custom 'bugzilla' and 'Reference' tags can also be URL, update the check
of link tags.
Signed-off-by: NWei Li <liwei391@huawei.com>

Merge Pull Request from: @stkid 
 
Add initial openeuler_defconfig of arm64 and x86_64, inheritted from
openEuler-22.03-LTS-SP2. 
 
Link:https://gitee.com/openeuler/kernel/pulls/1360 

Reviewed-by: Xie XiuQi <xiexiuqi@huawei.com> 
Signed-off-by: Xie XiuQi <xiexiuqi@huawei.com> 

hulk inclusion
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I7JK23

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

This config is based on x86_64 openeuler_defconfig of
openEuler-22.03-LTS-SP2:5.10.0-153.17.0, generated by 'make olddefconfig'
and cleanup compiler's indicators.
Signed-off-by: NWei Li <liwei391@huawei.com>

hulk inclusion
category: feature
bugzilla: https://gitee.com/openeuler/kernel/issues/I7JK23

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

This config is based on arm64 openeuler_defconfig of
openEuler-22.03-LTS-SP2:5.10.0-153.17.0, generated by 'make olddefconfig'
and cleanup compiler's indicators.
Signed-off-by: NWei Li <liwei391@huawei.com>

Pull i2c fixes from Wolfram Sang:
 "Nothing fancy. Two driver and one DT binding fix"

* tag 'i2c-for-6.4-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux:
  i2c: imx-lpi2c: fix type char overflow issue when calculating the clock cycle
  i2c: qup: Add missing unwind goto in qup_i2c_probe()
  dt-bindings: i2c: opencores: Add missing type for "regstep"

Pull perf fixes from Borislav Petkov:

 - Drop the __weak attribute from a function prototype as it otherwise
   leads to the function getting replaced by a dummy stub

 - Fix the umask value setup of the frontend event as former is
   different on two Intel cores

* tag 'perf_urgent_for_v6.4' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  perf/x86/intel: Fix the FRONTEND encoding on GNR and MTL
  perf/core: Drop __weak attribute from arch_perf_update_userpage() prototype

Pull objtool fix from Borislav Petkov:

 - Add a ORC format hash to vmlinux and modules in order for other tools
   which use it, to detect changes to it and adapt accordingly

* tag 'objtool_urgent_for_v6.4' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/unwind/orc: Add ELF section with ORC version identifier

Pull x86 fixes from Borislav Petkov:

 - Do not use set_pgd() when updating the KASLR trampoline pgd entry
   because that updates the user PGD too on KPTI builds, resulting in
   memory corruption

 - Prevent a panic in the IO-APIC setup code due to conflicting command
   line parameters

* tag 'x86_urgent_for_v6.4' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/apic: Fix kernel panic when booting with intremap=off and x2apic_phys
  x86/mm: Avoid using set_pgd() outside of real PGD pages

Pull drm fixes from Dave Airlie:
 "Very quiet last week, just two misc fixes, one dp-mst and one qaic:

  qaic:
   - dma-buf import fix

  dp-mst:
   - fix NULL ptr deref"

[ It turns out it was a quiet week because Alex Deucher hadn't sent in
  his pending AMD changes. So they are coming next    - Linus ]

* tag 'drm-fixes-2023-06-23' of git://anongit.freedesktop.org/drm/drm:
  drm: use mgr->dev in drm_dbg_kms in drm_dp_add_payload_part2
  accel/qaic: Call DRM helper function to destroy prime GEM

Pull ARM SoC fixes from Arnd Bergmann:
 "The final bug fixes for Qualcomm and Rockchips came in, all of them
  for devicetree files:

   - Devices on Qualcomm SC7180/SC7280 that are cache coherent are now
     marked so correctly to fix a regression after a change in kernel
     behavior

   - Rockchips has a few minor changes for correctness of regulator and
     cache properties, as well as fixes for incorrect behavior of the
     RK3568 PCI controller and reset pins on two boards"

* tag 'arm-fixes-6.4-3' of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc:
  arm64: dts: qcom: sc7280: Mark SCM as dma-coherent for chrome devices
  arm64: dts: qcom: sc7180: Mark SCM as dma-coherent for trogdor
  arm64: dts: qcom: sc7180: Mark SCM as dma-coherent for IDP
  dt-bindings: firmware: qcom,scm: Document that SCM can be dma-coherent
  arm64: dts: rockchip: Fix rk356x PCIe register and range mappings
  arm64: dts: rockchip: fix button reset pin for nanopi r5c
  arm64: dts: rockchip: fix nEXTRST on SOQuartz
  arm64: dts: rockchip: add missing cache properties
  arm64: dts: rockchip: fix USB regulator on ROCK64

Pull btrfs fix from David Sterba:
 "Unfortunately the recent u32 overflow fix was not complete, there was
  one conversion left, assertion not triggered by my tests but caught by
  Qu's fstests case.

  The "cleanup for later" has been promoted to a proper fix and wraps
  all uses of the stripe left shift so the diffstat has grown but leaves
  no potentially problematic uses.

  We should have done it that way before, sorry"

* tag 'for-6.4-rc7-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:
  btrfs: fix remaining u32 overflows when left shifting stripe_nr

Pull block fix from Jens Axboe:
 "It's apparently the week of 'fixup something from last week', because
  the same is true for this block pull request.

  Fix up a lock grab that needs to be IRQ saving, rather than just IRQ
  disabling, in the block cgroup code"

* tag 'block-6.4-2023-06-23' of git://git.kernel.dk/linux:
  block: make sure local irq is disabled when calling __blkcg_rstat_flush

Pull iommu fix from Joerg Roedel:

 - Fix potential memory leak in AMD IOMMU domain allocation path

* tag 'iommu-fix-v6.4-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu:
  iommu/amd: Fix possible memory leak of 'domain'