ASoC: qcom: cleanup and fix dependency of QCOM_COMMON

[ Upstream commit 8d89cf6ff229ff31cd4f73f5b3928564b81fc41e ]

SND_SOC_QCOM_COMMON depends on SOUNDWIRE for some symbols but this
is not explicitly specified using Kconfig depends. On the other hand
SND_SOC_QCOM_COMMON is also directly selected by the sound card
Kconfigs, this could result in various combinations and some symbols
ending up in modules and soundcard that uses those symbols as in-build
driver.

Fix these issues by explicitly specifying the dependencies of
SND_SOC_QCOM_COMMON and also use imply a to select SND_SOC_QCOM_COMMON
so that the symbol is selected based on its dependencies.

Also remove dummy stubs in common.c around CONFIG_SOUNDWIRE

Fixes: 3bd975f3ae0a ("ASoC: qcom: sm8250: move some code to common")
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Link: https://lore.kernel.org/r/20221124140351.407506-1-srinivas.kandagatla@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

ASoC: qcom: Add checks for devm_kcalloc

[ Upstream commit 1bf5ee979076ceb121ee51c95197d890b1cee7f4 ]

As the devm_kcalloc may return NULL, the return value needs to be checked
to avoid NULL poineter dereference.

Fixes: 24caf8d9eb10 ("ASoC: qcom: lpass-sc7180: Add platform driver for lpass audio")
Signed-off-by: Yuan Can <yuancan@huawei.com>
Link: https://lore.kernel.org/r/20221124140510.63468-1-yuancan@huawei.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

drbd: destroy workqueue when drbd device was freed

[ Upstream commit 8692814b77ca4228a99da8a005de0acf40af6132 ]

A submitter workqueue is dynamically allocated by init_submitter()
called by drbd_create_device(), we should destroy it when this
device is not needed or destroyed.

Fixes: 113fef9e20e0 ("drbd: prepare to queue write requests on a submit worker")
Signed-off-by: Wang ShaoBo <bobo.shaobowang@huawei.com>
Link: https://lore.kernel.org/r/20221124015817.2729789-3-bobo.shaobowang@huawei.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>

drbd: remove call to memset before free device/resource/connection

[ Upstream commit 6e7b854e4c1b02dba00760dfa79d8dbf6cce561e ]

This revert c2258ffc56f2 ("drbd: poison free'd device, resource and
connection structs"), add memset is odd here for debugging, there are
some methods to accurately show what happened, such as kdump.

Signed-off-by: Wang ShaoBo <bobo.shaobowang@huawei.com>
Link: https://lore.kernel.org/r/20221124015817.2729789-2-bobo.shaobowang@huawei.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Stable-dep-of: 8692814b77ca ("drbd: destroy workqueue when drbd device was freed")
Signed-off-by: Sasha Levin <sashal@kernel.org>

mtd: maps: pxa2xx-flash: fix memory leak in probe

[ Upstream commit 2399401feee27c639addc5b7e6ba519d3ca341bf ]

Free 'info' upon remapping error to avoid a memory leak.

Fixes: e644f7d62894 ("[MTD] MAPS: Merge Lubbock and Mainstone drivers into common PXA2xx driver")
Signed-off-by: Zheng Yongjun <zhengyongjun3@huawei.com>
[<miquel.raynal@bootlin.com>: Reword the commit log]
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/linux-mtd/20221119073307.22929-1-zhengyongjun3@huawei.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

mtd: core: Fix refcount error in del_mtd_device()

[ Upstream commit 56570bdad5e31c5c538cd6efff5c4510256e1bb4 ]

del_mtd_device() will call of_node_put() to mtd_get_of_node(mtd), which
is mtd->dev.of_node. However, memset(&mtd->dev, 0) is called before
of_node_put(). As the result, of_node_put() won't do anything in
del_mtd_device(), and causes the refcount leak.

del_mtd_device()
    memset(&mtd->dev, 0, sizeof(mtd->dev) # clear mtd->dev
    of_node_put()
        mtd_get_of_node(mtd) # mtd->dev is cleared, can't locate of_node
                             # of_node_put(NULL) won't do anything

Fix the error by caching the pointer of the device_node.

OF: ERROR: memory leak, expected refcount 1 instead of 2,
of_node_get()/of_node_put() unbalanced - destroy cset entry: attach
overlay node /spi/spi-sram@0
CPU: 3 PID: 275 Comm: python3 Tainted: G N 6.1.0-rc3+ #54
    0d8a1edddf51f172ff5226989a7565c6313b08e2
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS
rel-1.15.0-0-g2dd4b9b3f840-prebuilt.qemu.org 04/01/2014
Call Trace:
<TASK>
    dump_stack_lvl+0x67/0x83
    kobject_get+0x155/0x160
    of_node_get+0x1f/0x30
    of_fwnode_get+0x43/0x70
    fwnode_handle_get+0x54/0x80
    fwnode_get_nth_parent+0xc9/0xe0
    fwnode_full_name_string+0x3f/0xa0
    device_node_string+0x30f/0x750
    pointer+0x598/0x7a0
    vsnprintf+0x62d/0x9b0
    ...
    cfs_overlay_release+0x30/0x90
    config_item_release+0xbe/0x1a0
    config_item_put+0x5e/0x80
    configfs_rmdir+0x3bd/0x540
    vfs_rmdir+0x18c/0x320
    do_rmdir+0x198/0x330
    __x64_sys_rmdir+0x2c/0x40
    do_syscall_64+0x37/0x90
    entry_SYSCALL_64_after_hwframe+0x63/0xcd

Fixes: 00596576a051 ("mtd: core: clear out unregistered devices a bit more")
Signed-off-by: Shang XiaoJing <shangxiaojing@huawei.com>
[<miquel.raynal@bootlin.com>: Light reword of the commit log]
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/linux-mtd/20221119063915.11108-1-shangxiaojing@huawei.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

clk: microchip: check for null return of devm_kzalloc()

[ Upstream commit e2e6a217a84d09785848a82599729c9a41566e3a ]

Because of the possible failure of devm_kzalloc(), name might be NULL and
will cause null pointer dereference later.

Therefore, it might be better to check it and directly return -ENOMEM.

Fixes: d39fb172760e ("clk: microchip: add PolarFire SoC fabric clock support")
Signed-off-by: Hui Tang <tanghui20@huawei.com>
Reviewed-by: Conor Dooley <conor.dooley@microchip.com>
[claudiu.beznea: s/refrence/reference/, s/possilble/possible]
Signed-off-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Link: https://lore.kernel.org/r/20221119054858.178629-1-tanghui20@huawei.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

bonding: fix link recovery in mode 2 when updelay is nonzero

[ Upstream commit f8a65ab2f3ff7410921ebbf0dc55453102c33c56 ]

Before this change when a bond in mode 2 lost link, all of its slaves
lost link, the bonding device would never recover even after the
expiration of updelay. This change removes the updelay when the bond
currently has no usable links. Conforming to bonding.txt section 13.1
paragraph 4.

Fixes: 41f891004063 ("bonding: ignore updelay param when there is no active slave")
Signed-off-by: Jonathan Toppins <jtoppins@redhat.com>
Acked-by: Jay Vosburgh <jay.vosburgh@canonical.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

selftests/bpf: Mount debugfs in setns_by_fd

[ Upstream commit 8ac88eece8009428e2577c345080a458e4507e2f ]

Jiri reports broken test_progs after recent commit 68f8e3d4b916
("selftests/bpf: Make sure zero-len skbs aren't redirectable").
Apparently we don't remount debugfs when we switch back networking namespace.
Let's explicitly mount /sys/kernel/debug.

0: https://lore.kernel.org/bpf/63b85917-a2ea-8e35-620c-808560910819@meta.com/T/#ma66ca9c92e99eee0a25e40f422489b26ee0171c1

Fixes: a30338840fa5 ("selftests/bpf: Move open_netns() and close_netns() into network_helpers.c")
Reported-by: Jiri Olsa <olsajiri@gmail.com>
Signed-off-by: Stanislav Fomichev <sdf@google.com>
Link: https://lore.kernel.org/r/20221123200829.2226254-1-sdf@google.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

selftests/bpf: Make sure zero-len skbs aren't redirectable

[ Upstream commit 68f8e3d4b916531ea3bb8b83e35138cf78f2fce5 ]

LWT_XMIT to test L3 case, TC to test L2 case.

v2:
- s/veth_ifindex/ipip_ifindex/ in two places (Martin)
- add comment about which condition triggers the rejection (Martin)

Signed-off-by: Stanislav Fomichev <sdf@google.com>
Link: https://lore.kernel.org/r/20221121180340.1983627-2-sdf@google.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Stable-dep-of: 8ac88eece800 ("selftests/bpf: Mount debugfs in setns_by_fd")
Signed-off-by: Sasha Levin <sashal@kernel.org>

drm/i915/guc: make default_lists const data

[ Upstream commit dfa5e6ef3ccefff9fa8a70d9f5fa6ef6244aa312 ]

The default_lists array should be in rodata.

Fixes: dce2bd542337 ("drm/i915/guc: Add Gen9 registers for GuC error state capture.")
Cc: Alan Previn <alan.previn.teres.alexis@intel.com>
Cc: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20221122141616.3469214-1-jani.nikula@intel.com
(cherry picked from commit 8b7f7a9b10b704ba7d73199ff0f01354e0bad7a5)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

drm/amdgpu: fix pci device refcount leak

[ Upstream commit b85e285e3d6352b02947fc1b72303673dfacb0aa ]

As comment of pci_get_domain_bus_and_slot() says, it returns
a pci device with refcount increment, when finish using it,
the caller must decrement the reference count by calling
pci_dev_put().

So before returning from amdgpu_device_resume|suspend_display_audio(),
pci_dev_put() is called to avoid refcount leak.

Fixes: 3f12acc8d6d4 ("drm/amdgpu: put the audio codec into suspend state before gpu reset V3")
Reviewed-by: Evan Quan <evan.quan@amd.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

clk: rockchip: Fix memory leak in rockchip_clk_register_pll()

[ Upstream commit 739a6a6bbdb793bd57938cb24aa5a6df89983546 ]

If clk_register() fails, @pll->rate_table may have allocated memory by
kmemdup(), so it needs to be freed, otherwise will cause memory leak
issue, this patch fixes it.

Fixes: 90c590254051 ("clk: rockchip: add clock type for pll clocks and pll used on rk3066")
Signed-off-by: Xiu Jianfeng <xiujianfeng@huawei.com>
Link: https://lore.kernel.org/r/20221123091201.199819-1-xiujianfeng@huawei.com
Signed-off-by: Heiko Stuebner <heiko@sntech.de>
Signed-off-by: Sasha Levin <sashal@kernel.org>

regulator: core: use kfree_const() to free space conditionally

[ Upstream commit dc8d006d15b623c1d80b90b45d6dcb6e890dad09 ]

Use kfree_const() to free supply_name conditionally in create_regulator()
as supply_name may be allocated from kmalloc() or directly from .rodata
section.

Fixes: 87fe29b61f95 ("regulator: push allocations in create_regulator() outside of lock")
Signed-off-by: Wang ShaoBo <bobo.shaobowang@huawei.com>
Link: https://lore.kernel.org/r/20221123034616.3609537-1-bobo.shaobowang@huawei.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

ALSA: seq: fix undefined behavior in bit shift for SNDRV_SEQ_FILTER_USE_EVENT

[ Upstream commit cf59e1e4c79bf741905484cdb13c130b53576a16 ]

Shifting signed 32-bit value by 31 bits is undefined, so changing
significant bit to unsigned. The UBSAN warning calltrace like below:

UBSAN: shift-out-of-bounds in sound/core/seq/seq_clientmgr.c:509:22
left shift of 1 by 31 places cannot be represented in type 'int'
...
Call Trace:
 <TASK>
 dump_stack_lvl+0x8d/0xcf
 ubsan_epilogue+0xa/0x44
 __ubsan_handle_shift_out_of_bounds+0x1e7/0x208
 snd_seq_deliver_single_event.constprop.21+0x191/0x2f0
 snd_seq_deliver_event+0x1a2/0x350
 snd_seq_kernel_client_dispatch+0x8b/0xb0
 snd_seq_client_notify_subscription+0x72/0xa0
 snd_seq_ioctl_subscribe_port+0x128/0x160
 snd_seq_kernel_client_ctl+0xce/0xf0
 snd_seq_oss_create_client+0x109/0x15b
 alsa_seq_oss_init+0x11c/0x1aa
 do_one_initcall+0x80/0x440
 kernel_init_freeable+0x370/0x3c3
 kernel_init+0x1b/0x190
 ret_from_fork+0x1f/0x30
 </TASK>

Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Signed-off-by: Baisong Zhong <zhongbaisong@huawei.com>
Link: https://lore.kernel.org/r/20221121111630.3119259-1-zhongbaisong@huawei.com
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Sasha Levin <sashal@kernel.org>

ALSA: pcm: fix undefined behavior in bit shift for SNDRV_PCM_RATE_KNOT

[ Upstream commit b5172e62458f8e6ff359e5f096044a488db90ac5 ]

Shifting signed 32-bit value by 31 bits is undefined, so changing
significant bit to unsigned. The UBSAN warning calltrace like below:

UBSAN: shift-out-of-bounds in sound/core/pcm_native.c:2676:21
left shift of 1 by 31 places cannot be represented in type 'int'
...
Call Trace:
 <TASK>
 dump_stack_lvl+0x8d/0xcf
 ubsan_epilogue+0xa/0x44
 __ubsan_handle_shift_out_of_bounds+0x1e7/0x208
 snd_pcm_open_substream+0x9f0/0xa90
 snd_pcm_oss_open.part.26+0x313/0x670
 snd_pcm_oss_open+0x30/0x40
 soundcore_open+0x18b/0x2e0
 chrdev_open+0xe2/0x270
 do_dentry_open+0x2f7/0x620
 path_openat+0xd66/0xe70
 do_filp_open+0xe3/0x170
 do_sys_openat2+0x357/0x4a0
 do_sys_open+0x87/0xd0
 do_syscall_64+0x34/0x80

Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Signed-off-by: Baisong Zhong <zhongbaisong@huawei.com>
Link: https://lore.kernel.org/r/20221121110044.3115686-1-zhongbaisong@huawei.com
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Sasha Levin <sashal@kernel.org>

ASoC: Intel: avs: Lock substream before snd_pcm_stop()

[ Upstream commit c30c8f9d51ec24b36e2c65a6307a5c8cbc5a0ebc ]

snd_pcm_stop() shall be called with stream lock held to prevent any
races between nonatomic streaming operations.

Fixes: 2f1f570cd730 ("ASoC: Intel: avs: Coredump and recovery flow")
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://lore.kernel.org/r/20221116115550.1100398-2-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

ASoC: Intel: Skylake: Fix Kconfig dependency

[ Upstream commit e5d4d2b23aed20a7815d1b500dbcd50af1da0023 ]

Commit e4746d94d00c ("ASoC: Intel: Skylake: Introduce HDA codec init and
exit routines") introduced HDA codec init routine which depends on SND_HDA.
Select SND_SOC_HDAC_HDA unconditionally to fix following compile error:
ERROR: modpost: "snd_hda_codec_device_init" [sound/soc/intel/skylake/snd-soc-skl.ko] undefined!

Fixes: e4746d94d00c ("ASoC: Intel: Skylake: Introduce HDA codec init and exit routines")
Reviewed-by: Junxiao Chang <junxiao.chang@intel.com>
Suggested-by: Cezary Rojewski <cezary.rojewski@intel.com>
Signed-off-by: Lili Li <lili.li@intel.com>
Reviewed-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://lore.kernel.org/r/20221121104742.1007486-1-lili.li@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

wifi: rtw89: fix physts IE page check

[ Upstream commit 9e2f177de1bfb7d891bf38140bda54831ecef30d ]

The index RTW89_PHYSTS_BITMAP_NUM is not a valid physts IE page.
So, fix the check condition.

Fixes: eb4e52b3f38d ("rtw89: fix incorrect channel info during scan")
Signed-off-by: Zong-Zhe Yang <kevin_yang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20221118042322.26794-1-pkshih@realtek.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

pinctrl: k210: call of_node_put()

[ Upstream commit a8acc11643082a706de86a19f1f824712d971984 ]

Since for_each_available_child_of_node() will increase the refcount of
node, we need to call of_node_put() manually when breaking out of the
iteration.

Fixes: d4c34d09ab03 ("pinctrl: Add RISC-V Canaan Kendryte K210 FPIOA driver")
Signed-off-by: ZhangPeng <zhangpeng362@huawei.com>
Reviewed-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Link: https://lore.kernel.org/r/20221122075853.2496680-1-zhangpeng362@huawei.com
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

clk: imx: imxrt1050: fix IMXRT1050_CLK_LCDIF_APB offsets

[ Upstream commit 3095c02f95e537c553e0b30948c2f6c7cbed87ee ]

Fix IMXRT1050_CLK_LCDIF_APB offsets.

Fixes: 7154b046d8f3 ("clk: imx: Add initial support for i.MXRT1050 clock driver")
Cc: Jesse Taube <mr.bossman075@gmail.com>
Signed-off-by: Giulio Benetti <giulio.benetti@benettiengineering.com>
Reviewed-by: Abel Vesa <abel.vesa@linaro.org>
Signed-off-by: Abel Vesa <abel.vesa@linaro.org>
Link: https://lore.kernel.org/r/20221117181014.851505-1-giulio.benetti@benettiengineering.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

HID: hid-sensor-custom: set fixed size for custom attributes

[ Upstream commit 9d013910df22de91333a0acc81d1dbb115bd76f6 ]

This is no bugfix (so no Fixes: tag is necessary) as it is
taken care of in hid_sensor_custom_add_attributes().

The motivation for this patch is that:
hid_sensor_custom_field.attr_name and
hid_sensor_custom_field.attrs
has the size of HID_CUSTOM_TOTAL_ATTRS and used in same context.

We compare against HID_CUSTOM_TOTAL_ATTRS when
looping through hid_custom_attrs.

We will silent the smatch error:
hid_sensor_custom_add_attributes() error: buffer overflow
'hid_custom_attrs' 8 <= 10

Signed-off-by: Marcus Folkesson <marcus.folkesson@gmail.com>
Acked-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
Signed-off-by: Sasha Levin <sashal@kernel.org>

bpf: Move skb->len == 0 checks into __bpf_redirect

[ Upstream commit 114039b342014680911c35bd6b72624180fd669a ]

To avoid potentially breaking existing users.

Both mac/no-mac cases have to be amended; mac_header >= network_header
is not enough (verified with a new test, see next patch).

Fixes: fd1894224407 ("bpf: Don't redirect packets with invalid pkt_len")
Signed-off-by: Stanislav Fomichev <sdf@google.com>
Link: https://lore.kernel.org/r/20221121180340.1983627-1-sdf@google.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

clk: imx93: correct enet clock

[ Upstream commit 4be5d91b9433f1dc76de485e240ca6aaa2d19f65 ]

Per update Reference Mannual, correct the enet clock parent to
wakeup_axi_root.

Fixes: 24defbe194b6 ("clk: imx: add i.MX93 clk")
Reviewed-by: Ye Li <ye.li@nxp.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
Reviewed-by: Abel Vesa <abel.vesa@linaro.org>
Signed-off-by: Abel Vesa <abel.vesa@linaro.org>
Link: https://lore.kernel.org/r/20221028095211.2598312-3-peng.fan@oss.nxp.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

clk: imx93: unmap anatop base in error handling path

[ Upstream commit bda7b7f396f94d8df89ecacc88f2826908e8762c ]

The anatop base is not unmapped during error handling path, fix it.

Fixes: 24defbe194b6 ("clk: imx: add i.MX93 clk")
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
Reviewed-by: Abel Vesa <abel.vesa@linaro.org>
Signed-off-by: Abel Vesa <abel.vesa@linaro.org>
Link: https://lore.kernel.org/r/20221028095211.2598312-2-peng.fan@oss.nxp.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

HID: i2c: let RMI devices decide what constitutes wakeup event

[ Upstream commit 9984fbf55b9bd998b4ff66395cbb118020c1effa ]

HID-RMI is special in the sense that it does not carry HID events
directly, but rather uses HID protocol as a wrapper/transport for RMI
protocol.  Therefore we should not assume that all data coming from the
device via interrupt is associated with user activity and report wakeup
event indiscriminately, but rather let HID-RMI do that when appropriate.

HID-RMI devices tag responses to the commands issued by the host as
RMI_READ_DATA_REPORT_ID whereas motion and other input events from the
device are tagged as RMI_ATTN_REPORT_ID. Change hid-rmi to report wakeup
events when receiving the latter packets. This allows ChromeOS to
accurately identify wakeup source and make correct decision on the mode
of the resume the system should take ("dark" where the display stays off
vs normal one).

Fixes: d951ae1ce803 ("HID: i2c-hid: Report wakeup events")
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
Signed-off-by: Sasha Levin <sashal@kernel.org>

bpf: Pin the start cgroup in cgroup_iter_seq_init()

[ Upstream commit 1a5160d4d8fe63ba4964cfff4a85831b6af75f2d ]

bpf_iter_attach_cgroup() has already acquired an extra reference for the
start cgroup, but the reference may be released if the iterator link fd
is closed after the creation of iterator fd, and it may lead to
user-after-free problem when reading the iterator fd.

An alternative fix is pinning iterator link when opening iterator,
but it will make iterator link being still visible after the close of
iterator link fd and the behavior is different with other link types, so
just fixing it by acquiring another reference for the start cgroup.

Fixes: d4ccaf58a847 ("bpf: Introduce cgroup iter")
Signed-off-by: Hou Tao <houtao1@huawei.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20221121073440.1828292-2-houtao@huaweicloud.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

clk: imx93: correct the flexspi1 clock setting

[ Upstream commit 62dfdbcc16e767b91ed35d4fc0428c86d4688505 ]

Correct IMX93_CLK_FLEXSPI1_GATE CCGR setting. Otherwise the flexspi
always can't be assigned to a parent clock when dump the clock tree.

Fixes: 24defbe194b6 ("clk: imx: add i.MX93 clk")
Reviewed-by: Peng Fan <peng.fan@nxp.com>
Signed-off-by: Haibo Chen <haibo.chen@nxp.com>
Reviewed-by: Abel Vesa <abel.vesa@linaro.org>
Signed-off-by: Abel Vesa <abel.vesa@linaro.org>
Link: https://lore.kernel.org/r/1666589199-1199-1-git-send-email-haibo.chen@nxp.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

mtd: spi-nor: Fix the number of bytes for the dummy cycles

[ Upstream commit fdc20370d93e8c6d2f448a539d08c2c064af7694 ]

The number of bytes used by spi_nor_spimem_check_readop() may be
incorrect for the dummy cycles. Since nor->read_dummy is not initialized
before spi_nor_spimem_adjust_hwcaps().

We use both mode and wait state clock cycles instead of nor->read_dummy.

Fixes: 0e30f47232ab ("mtd: spi-nor: add support for DTR protocol")
Co-developed-by: Bayi Cheng <bayi.cheng@mediatek.com>
Signed-off-by: Bayi Cheng <bayi.cheng@mediatek.com>
Signed-off-by: Allen-KH Cheng <allen-kh.cheng@mediatek.com>
Signed-off-by: Tudor Ambarus <tudor.ambarus@microchip.com>
Tested-by: Dhruva Gole <d-gole@ti.com>
Tested-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Pratyush Yadav <pratyush@kernel.org>
Link: https://lore.kernel.org/r/20221031124633.13189-1-allen-kh.cheng@mediatek.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

mtd: spi-nor: hide jedec_id sysfs attribute if not present

[ Upstream commit 7d388551b6888f3725e6c957f472526b35161a5b ]

Some non-jedec compliant flashes (like the Everspin flashes) don't have
an ID at all. Hide the attribute in this case.

Fixes: 36ac02286265 ("mtd: spi-nor: add initial sysfs support")
Signed-off-by: Michael Walle <michael@walle.cc>
Signed-off-by: Tudor Ambarus <tudor.ambarus@microchip.com>
Reviewed-by: Takahiro Kuwano <Takahiro.Kuwano@infineon.com>
Link: https://lore.kernel.org/r/20220810220654.1297699-2-michael@walle.cc
Signed-off-by: Sasha Levin <sashal@kernel.org>

net: Return errno in sk->sk_prot->get_port().

[ Upstream commit 7a7160edf1bfde25422262fb26851cef65f695d3 ]

We assume the correct errno is -EADDRINUSE when sk->sk_prot->get_port()
fails, so some ->get_port() functions return just 1 on failure and the
callers return -EADDRINUSE instead.

However, mptcp_get_port() can return -EINVAL.  Let's not ignore the error.

Note the only exception is inet_autobind(), all of whose callers return
-EAGAIN instead.

Fixes: cec37a6e41aa ("mptcp: Handle MP_CAPABLE options for outgoing connections")
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>

udp: Clean up some functions.

[ Upstream commit 919dfa0b20ae56060dce0436eb710717f8987d18 ]

This patch adds no functional change and cleans up some functions
that the following patches touch around so that we make them tidy
and easy to review/revert.  The change is mainly to keep reverse
christmas tree order.

Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Stable-dep-of: 7a7160edf1bf ("net: Return errno in sk->sk_prot->get_port().")
Signed-off-by: Sasha Levin <sashal@kernel.org>

net: ethernet: mtk_eth_soc: fix RSTCTRL_PPE{0,1} definitions

[ Upstream commit ef8c373bd91df3cf70596497da0955d218961ead ]

Fix RSTCTRL_PPE0 and RSTCTRL_PPE1 register mask definitions for
MTK_NETSYS_V2.
Remove duplicated definitions.

Fixes: 160d3a9b1929 ("net: ethernet: mtk_eth_soc: introduce MTK_NETSYS_V2 support")
Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>

media: videobuf-dma-contig: use dma_mmap_coherent

[ Upstream commit b3dc3f8e49577840dc8ac8a365c5b3da4edb10b8 ]

dma_alloc_coherent does not return a physical address, but a DMA address,
which might be remapped or have an offset.  Passing the DMA address to
vm_iomap_memory is thus broken.

Use the proper dma_mmap_coherent helper instead, and stop passing
__GFP_COMP to dma_alloc_coherent, as the memory management inside the
DMA allocator is hidden from the callers and does not require it.

With this the gfp_t argument to __videobuf_dc_alloc can be removed and
hard coded to GFP_KERNEL.

Fixes: a8f3c203e19b ("[media] videobuf-dma-contig: add cache support")
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Sasha Levin <sashal@kernel.org>

media: amphion: Fix error handling in vpu_driver_init()

[ Upstream commit a95cc6d11aae16a7b2d043b073a40de81bbea689 ]

A problem about modprobe amphion-vpu failed is triggered with the
following log given:

 [ 2208.634841] Error: Driver 'amphion-vpu' is already registered, aborting...
 modprobe: ERROR: could not insert 'amphion_vpu': Device or resource busy

The reason is that vpu_driver_init() returns vpu_core_driver_init()
directly without checking its return value, if vpu_core_driver_init()
failed, it returns without unregister amphion_vpu_driver, resulting the
amphion-vpu can never be installed later.
A simple call graph is shown as below:

 vpu_driver_init()
   platform_driver_register() # register amphion_vpu_driver
   vpu_core_driver_init()
     platform_driver_register()
       driver_register()
         bus_add_driver()
           dev = kzalloc(...) # OOM happened
   # return without unregister amphion_vpu_driver

Fix by unregister amphion_vpu_driver when vpu_core_driver_init() returns
error.

Fixes: b50a64fc54af ("media: amphion: add amphion vpu device driver")
Signed-off-by: Yuan Can <yuancan@huawei.com>
Reviewed-by: ming_qian <ming.qian@nxp.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Sasha Levin <sashal@kernel.org>

media: platform: exynos4-is: Fix error handling in fimc_md_init()

[ Upstream commit b434422c45282a0573d8123239abc41fa72665d4 ]

A problem about modprobe s5p_fimc failed is triggered with the
following log given:

 [  272.075275] Error: Driver 'exynos4-fimc' is already registered, aborting...
 modprobe: ERROR: could not insert 's5p_fimc': Device or resource busy

The reason is that fimc_md_init() returns platform_driver_register()
directly without checking its return value, if platform_driver_register()
failed, it returns without unregister fimc_driver, resulting the
s5p_fimc can never be installed later.
A simple call graph is shown as below:

 fimc_md_init()
   fimc_register_driver() # register fimc_driver
   platform_driver_register()
     platform_driver_register()
       driver_register()
         bus_add_driver()
           dev = kzalloc(...) # OOM happened
   # return without unregister fimc_driver

Fix by unregister fimc_driver when platform_driver_register() returns
error.

Fixes: d3953223b090 ("[media] s5p-fimc: Add the media device driver")
Signed-off-by: Yuan Can <yuancan@huawei.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Sasha Levin <sashal@kernel.org>

media: solo6x10: fix possible memory leak in solo_sysfs_init()

[ Upstream commit 7f5866dd96d95b74e439f6ee17b8abd8195179fb ]

If device_register() returns error in solo_sysfs_init(), the
name allocated by dev_set_name() need be freed. As comment of
device_register() says, it should use put_device() to give up
the reference in the error path. So fix this by calling
put_device(), then the name can be freed in kobject_cleanup().

Fixes: dcae5dacbce5 ("[media] solo6x10: sync to latest code from Bluecherry's git repo")
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Sasha Levin <sashal@kernel.org>

media: vidtv: Fix use-after-free in vidtv_bridge_dvb_init()

[ Upstream commit ba8d9405935097e296bcf7a942c3a01df0edb865 ]

KASAN reports a use-after-free:
BUG: KASAN: use-after-free in dvb_dmxdev_release+0x4d5/0x5d0 [dvb_core]
Call Trace:
 ...
 dvb_dmxdev_release+0x4d5/0x5d0 [dvb_core]
 vidtv_bridge_probe+0x7bf/0xa40 [dvb_vidtv_bridge]
 platform_probe+0xb6/0x170
 ...
Allocated by task 1238:
 ...
 dvb_register_device+0x1a7/0xa70 [dvb_core]
 dvb_dmxdev_init+0x2af/0x4a0 [dvb_core]
 vidtv_bridge_probe+0x766/0xa40 [dvb_vidtv_bridge]
 ...
Freed by task 1238:
 dvb_register_device+0x6d2/0xa70 [dvb_core]
 dvb_dmxdev_init+0x2af/0x4a0 [dvb_core]
 vidtv_bridge_probe+0x766/0xa40 [dvb_vidtv_bridge]
 ...

It is because the error handling in vidtv_bridge_dvb_init() is wrong.

First, vidtv_bridge_dmx(dev)_init() will clean themselves when fail, but
goto fail_dmx(_dev): calls release functions again, which causes
use-after-free.

Also, in fail_fe, fail_tuner_probe and fail_demod_probe, j = i will cause
out-of-bound when i finished its loop (i == NUM_FE). And the loop
releasing is wrong, although now NUM_FE is 1 so it won't cause problem.

Fix this by correctly releasing everything.

Fixes: f90cf6079bf6 ("media: vidtv: add a bridge driver")
Signed-off-by: Chen Zhongjin <chenzhongjin@huawei.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Sasha Levin <sashal@kernel.org>

media: amphion: apply vb2_queue_error instead of setting manually

[ Upstream commit 9d175a81e28f260916a0a13f457dd8b940eafb4e ]

vb2_queue_error is help to set the error of vb2_queue,
don't need to set it manually

Fixes: 3cd084519c6f ("media: amphion: add vpu v4l2 m2m support")
Signed-off-by: Ming Qian <ming.qian@nxp.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Sasha Levin <sashal@kernel.org>

media: amphion: add lock around vdec_g_fmt

[ Upstream commit 8480dd5fb3c82b5887d456b3fbe4201d99231814 ]

the capture format may be changed when
sequence header is parsed,
it may be read and write in the same time,
add lock around vdec_g_fmt to synchronize it

Fixes: 6de8d628df6e ("media: amphion: add v4l2 m2m vpu decoder stateful driver")
Signed-off-by: Ming Qian <ming.qian@nxp.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Sasha Levin <sashal@kernel.org>

net: ethernet: mtk_eth_soc: do not overwrite mtu configuration running reset routine

[ Upstream commit b677d6c7a695dad1b02d2e0e428c39b3b344f270 ]

Restore user configured MTU running mtk_hw_init() during tx timeout routine
since it will be overwritten after a hw reset.

Reported-by: Felix Fietkau <nbd@nbd.name>
Fixes: 9ea4d311509f ("net: ethernet: mediatek: add the whole ethernet reset into the reset process")
Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>

ASoC: amd: acp: Fix possible UAF in acp_dma_open

[ Upstream commit 3420fdb8ae99f0a08d78d2b80f42a71971cf478d ]

Smatch report warning as follows:

sound/soc/amd/acp/acp-platform.c:199 acp_dma_open() warn:
  '&stream->list' not removed from list

If snd_pcm_hw_constraint_integer() fails in acp_dma_open(),
stream will be freed, but stream->list will not be removed from
adata->stream_list, then list traversal may cause UAF.

Fix by adding the newly allocated stream to the list once it's fully
initialised.

Fixes: 7929985cfe36 ("ASoC: amd: acp: Initialize list to store acp_stream during pcm_open")
Signed-off-by: Gaosheng Cui <cuigaosheng1@huawei.com>
Link: https://lore.kernel.org/r/20221118030056.3135960-1-cuigaosheng1@huawei.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

Input: elants_i2c - properly handle the reset GPIO when power is off

[ Upstream commit a85fbd6498441694475716a4d5c65f9d3e073faf ]

As can be seen in elants_i2c_power_off(), we want the reset GPIO
asserted when power is off. The reset GPIO is active low so we need
the reset line logic low when power is off to avoid leakage.

We have a problem, though, at probe time. At probe time we haven't
powered the regulators on yet but we have:

  devm_gpiod_get(&client->dev, "reset", GPIOD_OUT_LOW);

While that _looks_ right, it turns out that it's not. The
GPIOD_OUT_LOW doesn't mean to init the GPIO to low. It means init the
GPIO to "not asserted". Since this is an active low GPIO that inits it
to be high.

Let's fix this to properly init the GPIO. Now after both probe and
power off the state of the GPIO is consistent (it's "asserted" or
level low).

Once we fix this, we can see that at power on time we no longer to
assert the reset GPIO as the first thing. The reset GPIO is _always_
asserted before powering on. Let's fix powering on to account for
this.

Fixes: afe10358e47a ("Input: elants_i2c - wire up regulator support")
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20221117123805.1.I9959ac561dd6e1e8e1ce7085e4de6167b27c574f@changeid
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

mtd: lpddr2_nvm: Fix possible null-ptr-deref

[ Upstream commit 6bdd45d795adf9e73b38ced5e7f750cd199499ff ]

It will cause null-ptr-deref when resource_size(add_range) invoked,
if platform_get_resource() returns NULL.

Fixes: 96ba9dd65788 ("mtd: lpddr: add driver for LPDDR2-NVM PCM memories")
Signed-off-by: Hui Tang <tanghui20@huawei.com>
Acked-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/linux-mtd/20221114090240.244172-1-tanghui20@huawei.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

drm/msm/a6xx: Fix speed-bin detection vs probe-defer

[ Upstream commit f6d1918794ef92b4e26b80c3d40365347b76b1fd ]

If we get an error (other than -ENOENT) we need to propagate that up the
stack.  Otherwise if the nvmem driver hasn't probed yet, we'll end up
end up claiming that we support all the OPPs which is not likely to be
true (and on some generations impossible to be true, ie. if there are
conflicting OPPs).

v2: Update commit msg, gc unused label, etc
v3: Add previously missing \n's

Fixes: fe7952c629da ("drm/msm: Add speed-bin support to a618 gpu")
Signed-off-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Reviewed-by: Akhil P Oommen <quic_akhilpo@quicinc.com>
Patchwork: https://patchwork.freedesktop.org/patch/511690/
Link: https://lore.kernel.org/r/20221115154637.1613968-1-robdclark@gmail.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

wifi: ath10k: Fix return value in ath10k_pci_init()

[ Upstream commit 2af7749047d8d6ad43feff69f555a13a6a6c2831 ]

This driver is attempting to register to support two different buses.
if either of these is successful then ath10k_pci_init() should return 0
so that hardware attached to the successful bus can be probed and
supported. only if both of these are unsuccessful should ath10k_pci_init()
return an errno.

Fixes: 0b523ced9a3c ("ath10k: add basic skeleton to support ahb")
Signed-off-by: Xiu Jianfeng <xiujianfeng@huawei.com>
Reviewed-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20221110061926.18163-1-xiujianfeng@huawei.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

selftests/bpf: fix memory leak of lsm_cgroup

[ Upstream commit c453e64cbc9532c0c2edfa999c35d29dad16b8bb ]

kmemleak reports this issue:

unreferenced object 0xffff88810b7835c0 (size 32):
  comm "test_progs", pid 270, jiffies 4294969007 (age 1621.315s)
  hex dump (first 32 bytes):
    00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
    03 00 00 00 03 00 00 00 0f 00 00 00 00 00 00 00  ................
  backtrace:
    [<00000000376cdeab>] kmalloc_trace+0x27/0x110
    [<000000003bcdb3b6>] selinux_sk_alloc_security+0x66/0x110
    [<000000003959008f>] security_sk_alloc+0x47/0x80
    [<00000000e7bc6668>] sk_prot_alloc+0xbd/0x1a0
    [<0000000002d6343a>] sk_alloc+0x3b/0x940
    [<000000009812a46d>] unix_create1+0x8f/0x3d0
    [<000000005ed0976b>] unix_create+0xa1/0x150
    [<0000000086a1d27f>] __sock_create+0x233/0x4a0
    [<00000000cffe3a73>] __sys_socket_create.part.0+0xaa/0x110
    [<0000000007c63f20>] __sys_socket+0x49/0xf0
    [<00000000b08753c8>] __x64_sys_socket+0x42/0x50
    [<00000000b56e26b3>] do_syscall_64+0x3b/0x90
    [<000000009b4871b8>] entry_SYSCALL_64_after_hwframe+0x63/0xcd

The issue occurs in the following scenarios:

unix_create1()
  sk_alloc()
    sk_prot_alloc()
      security_sk_alloc()
        call_int_hook()
          hlist_for_each_entry()
            entry1->hook.sk_alloc_security
            <-- selinux_sk_alloc_security() succeeded,
            <-- sk->security alloced here.
            entry2->hook.sk_alloc_security
            <-- bpf_lsm_sk_alloc_security() failed
      goto out_free;
        ...    <-- the sk->security not freed, memleak

The core problem is that the LSM is not yet fully stacked (work is
actively going on in this space) which means that some LSM hooks do
not support multiple LSMs at the same time. To fix, skip the
"EPERM" test when it runs in the environments that already have
non-bpf lsms installed

Fixes: dca85aac8895 ("selftests/bpf: lsm_cgroup functional test")
Signed-off-by: Wang Yufen <wangyufen@huawei.com>
Cc: Stanislav Fomichev <sdf@google.com>
Acked-by: Stanislav Fomichev <sdf@google.com>
Link: https://lore.kernel.org/r/1668482980-16163-1-git-send-email-wangyufen@huawei.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

dm: track per-add_disk holder relations in DM

[ Upstream commit 1a581b72169968f4154b5793828f3bc28b258b35 ]

dm is a bit special in that it opens the underlying devices.  Commit
89f871af1b26 ("dm: delay registering the gendisk") tried to accommodate
that by allowing to add the holder to the list before add_gendisk and
then just add them to sysfs once add_disk is called.  But that leads to
really odd lifetime problems and error handling problems as we can't
know the state of the kobjects and don't unwind properly.  To fix this
switch to just registering all existing table_devices with the holder
code right after add_disk, and remove them before calling del_gendisk.

Fixes: 89f871af1b26 ("dm: delay registering the gendisk")
Reported-by: Yu Kuai <yukuai3@huawei.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Yu Kuai <yukuai3@huawei.com>
Reviewed-by: Mike Snitzer <snitzer@kernel.org>
Link: https://lore.kernel.org/r/20221115141054.1051801-7-yukuai1@huaweicloud.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>

dm: make sure create and remove dm device won't race with open and close table

[ Upstream commit d563792c8933a810d28ce0f2831f0726c2b15a31 ]

open_table_device() and close_table_device() is protected by
table_devices_lock, hence use it to protect add_disk() and
del_gendisk().

Prepare to track per-add_disk holder relations in dm.

Signed-off-by: Yu Kuai <yukuai3@huawei.com>
Reviewed-by: Mike Snitzer <snitzer@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20221115141054.1051801-6-yukuai1@huaweicloud.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Stable-dep-of: 1a581b721699 ("dm: track per-add_disk holder relations in DM")
Signed-off-by: Sasha Levin <sashal@kernel.org>

dm: cleanup close_table_device

[ Upstream commit 7b5865831c1003122f737df5e16adaa583f1a595 ]

Take the list unlink and free into close_table_device so that no half
torn down table_devices exist.  Also remove the check for a NULL bdev
as that can't happen - open_table_device never adds a table_device to
the list that does not have a valid block_device.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Yu Kuai <yukuai3@huawei.com>
Reviewed-by: Mike Snitzer <snitzer@kernel.org>
Link: https://lore.kernel.org/r/20221115141054.1051801-5-yukuai1@huaweicloud.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Stable-dep-of: 1a581b721699 ("dm: track per-add_disk holder relations in DM")
Signed-off-by: Sasha Levin <sashal@kernel.org>

dm: cleanup open_table_device

[ Upstream commit b9a785d2dc6567b2fd9fc60057a6a945a276927a ]

Move all the logic for allocation the table_device and linking it into
the list into the open_table_device.  This keeps the code tidy and
ensures that the table_devices only exist in fully initialized state.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Yu Kuai <yukuai3@huawei.com>
Reviewed-by: Mike Snitzer <snitzer@kernel.org>
Link: https://lore.kernel.org/r/20221115141054.1051801-4-yukuai1@huaweicloud.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Stable-dep-of: 1a581b721699 ("dm: track per-add_disk holder relations in DM")
Signed-off-by: Sasha Levin <sashal@kernel.org>

block: clear ->slave_dir when dropping the main slave_dir reference

[ Upstream commit d90db3b1c8676bc88b4309c5a571333de2263b8e ]

Zero out the pointer to ->slave_dir so that the holder code doesn't
incorrectly treat the object as alive when add_disk failed or after
del_gendisk was called.

Fixes: 89f871af1b26 ("dm: delay registering the gendisk")
Reported-by: Yu Kuai <yukuai3@huawei.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Yu Kuai <yukuai3@huawei.com>
Reviewed-by: Mike Snitzer <snitzer@kernel.org>
Link: https://lore.kernel.org/r/20221115141054.1051801-2-yukuai1@huaweicloud.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>

ima: Fix misuse of dereference of pointer in template_desc_init_fields()

[ Upstream commit 25369175ce84813dd99d6604e710dc2491f68523 ]

The input parameter @fields is type of struct ima_template_field ***, so
when allocates array memory for @fields, the size of element should be
sizeof(**field) instead of sizeof(*field).

Actually the original code would not cause any runtime error, but it's
better to make it logically right.

Fixes: adf53a778a0a ("ima: new templates management mechanism")
Signed-off-by: Xiu Jianfeng <xiujianfeng@huawei.com>
Reviewed-by: Roberto Sassu <roberto.sassu@huawei.com>
Signed-off-by: Mimi Zohar <zohar@linux.ibm.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

integrity: Fix memory leakage in keyring allocation error path

[ Upstream commit 39419ef7af0916cc3620ecf1ed42d29659109bf3 ]

Key restriction is allocated in integrity_init_keyring(). However, if
keyring allocation failed, it is not freed, causing memory leaks.

Fixes: 2b6aa412ff23 ("KEYS: Use structure to capture key restriction function and data")
Signed-off-by: GUO Zihua <guozihua@huawei.com>
Signed-off-by: Mimi Zohar <zohar@linux.ibm.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

ALSA: memalloc: Allocate more contiguous pages for fallback case

[ Upstream commit cc26516374065a34e10c9a8bf3e940e42cd96e2a ]

Currently the fallback SG allocation tries to allocate each single
page, and this tends to result in the reverse order of memory
addresses when large space is available at boot, as the kernel takes a
free page from the top to the bottom in the zone.  The end result
looks as if non-contiguous (although it actually is).  What's worse is
that it leads to an overflow of BDL entries for HD-audio.

For avoiding such a problem, this patch modifies the allocation code
slightly; now it tries to allocate the larger contiguous chunks as
much as possible, then reduces to the smaller chunks only if the
allocation failed -- a similar strategy as the existing
snd_dma_alloc_pages_fallback() function.

Along with the trick, drop the unused address array from
snd_dma_sg_fallback object.  It was needed in the past when
dma_alloc_coherent() was used, but with the standard page allocator,
it became superfluous and never referred.

Fixes: a8d302a0b770 ("ALSA: memalloc: Revive x86-specific WC page allocations again")
Reviewed-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Link: https://lore.kernel.org/r/20221114141658.29620-1-tiwai@suse.de
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Sasha Levin <sashal@kernel.org>

drm/fourcc: Fix vsub/hsub for Q410 and Q401

[ Upstream commit b230555f3257f197dd98641ef6ebaf778b52dd51 ]

These formats are not subsampled, but that means hsub and vsub should be
1, not 0.

Fixes: 94b292b27734 ("drm: drm_fourcc: add NV15, Q410, Q401 YUV formats")
Reported-by: George Kennedy <george.kennedy@oracle.com>
Reported-by: butt3rflyh4ck <butterflyhuangxx@gmail.com>
Signed-off-by: Brian Starkey <brian.starkey@arm.com>
Reviewed-by: Liviu Dudau <liviu.dudau@arm.com>
Signed-off-by: Liviu Dudau <liviu.dudau@arm.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220913144306.17279-1-brian.starkey@arm.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

regulator: qcom-rpmh: Fix PMR735a S3 regulator spec

[ Upstream commit dd801b2265c81bf0c8b0b4b8f7c1e7bfed078403 ]

PMR735a has a wider range than previously defined. Fix it.

Fixes: c4e5aa3dbee5 ("regulator: qcom-rpmh: Add PM7325/PMR735A regulator support")
Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Link: https://lore.kernel.org/r/20221110210706.80301-1-konrad.dybcio@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

wifi: rtw89: Fix some error handling path in rtw89_core_sta_assoc()

[ Upstream commit 81c0b8928437ddfe87b0024c5bd817bc331d8524 ]

'ret' is not updated after a function call in rtw89_core_sta_assoc().
This prevent error handling from working.

Add the missing assignment.

Fixes: e3ec7017f6a2 ("rtw89: add Realtek 802.11ax driver")
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Acked-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/7b1d82594635e4406d3438f33d8da29eaa056c5a.1668354547.git.christophe.jaillet@wanadoo.fr
Signed-off-by: Sasha Levin <sashal@kernel.org>

nvme: return err on nvme_init_non_mdts_limits fail

[ Upstream commit bcaf434b8f04e1ee82a8b1e1bce0de99fbff67fa ]

In nvme_init_non_mdts_limits function we were returning 0 when kzalloc
failed; it now returns -ENOMEM.

Fixes: 5befc7c26e5a ("nvme: implement non-mdts command limits")
Signed-off-by: Joel Granados <j.granados@samsung.com>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Sasha Levin <sashal@kernel.org>

amdgpu/pm: prevent array underflow in vega20_odn_edit_dpm_table()

[ Upstream commit d27252b5706e51188aed7647126e44dcf9e940c1 ]

In the PP_OD_EDIT_VDDC_CURVE case the "input_index" variable is capped at
2 but not checked for negative values so it results in an out of bounds
read.  This value comes from the user via sysfs.

Fixes: d5bf26539494 ("drm/amd/powerplay: added vega20 overdrive support V3")
Signed-off-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

regulator: core: fix unbalanced of node refcount in regulator_dev_lookup()

[ Upstream commit f2b41b748c19962b82709d9f23c6b2b0ce9d2f91 ]

I got the the following report:

  OF: ERROR: memory leak, expected refcount 1 instead of 2,
  of_node_get()/of_node_put() unbalanced - destroy cset entry:
  attach overlay node /i2c/pmic@62/regulators/exten

In of_get_regulator(), the node is returned from of_parse_phandle()
with refcount incremented, after using it, of_node_put() need be called.

Fixes: 69511a452e6d ("regulator: map consumer regulator based on device tree")
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Link: https://lore.kernel.org/r/20221115091508.900752-1-yangyingliang@huawei.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

nvmet: only allocate a single slab for bvecs

[ Upstream commit fa8f9ac42350edd3ce82d0d148a60f0fa088f995 ]

There is no need to have a separate slab cache for each namespace,
and having separate ones creates duplicate debugs file names as well.

Fixes: d5eff33ee6f8 ("nvmet: add simple file backed ns support")
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Keith Busch <kbusch@kernel.org>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

ASoC: pxa: fix null-pointer dereference in filter()

[ Upstream commit ec7bf231aaa1bdbcb69d23bc50c753c80fb22429 ]

kasprintf() would return NULL pointer when kmalloc() fail to allocate.
Need to check the return pointer before calling strcmp().

Fixes: 7a824e214e25 ("ASoC: mmp: add audio dma support")
Signed-off-by: Zeng Heng <zengheng4@huawei.com>
Link: https://lore.kernel.org/r/20221114085629.1910435-1-zengheng4@huawei.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

drm/mediatek: Modify dpi power on/off sequence.

[ Upstream commit ff446c0f6290185cefafe3b376bb86063a3a9f6a ]

Modify dpi power on/off sequence so that the first gpio operation will
take effect.

Fixes: 6bd4763fd532 ("drm/mediatek: set dpi pin mode to gpio low to avoid leakage current")
Signed-off-by: Xinlei Lee <xinlei.lee@mediatek.com>
Signed-off-by: Chun-Kuang Hu <chunkuang.hu@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

selftests/bpf: Fix incorrect ASSERT in the tcp_hdr_options test

[ Upstream commit 52929912d7bda040b43538e8d88e8d231b76eb4e ]

This patch fixes the incorrect ASSERT test in tcp_hdr_options during
the CHECK to ASSERT macro cleanup.

Fixes: 3082f8cd4ba3 ("selftests/bpf: Convert tcp_hdr_options test to ASSERT_* macros")
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Wang Yufen <wangyufen@huawei.com>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20221107230420.4192307-3-martin.lau@linux.dev
Signed-off-by: Sasha Levin <sashal@kernel.org>

selftests/bpf: Fix xdp_synproxy compilation failure in 32-bit arch

[ Upstream commit e4c9cf0ce8c413c2030e8fb215551d7e0582ee7b ]

xdp_synproxy fails to be compiled in the 32-bit arch, log is as follows:

  xdp_synproxy.c: In function 'parse_options':
  xdp_synproxy.c:175:36: error: left shift count >= width of type [-Werror=shift-count-overflow]
    175 |                 *tcpipopts = (mss6 << 32) | (ttl << 24) | (wscale << 16) | mss4;
        |                                    ^~
  xdp_synproxy.c: In function 'syncookie_open_bpf_maps':
  xdp_synproxy.c:289:28: error: cast from pointer to integer of different size [-Werror=pointer-to-int-cast]
    289 |                 .map_ids = (__u64)map_ids,
        |                            ^

Fix it.

Fixes: fb5cd0ce70d4 ("selftests/bpf: Add selftests for raw syncookie helpers")
Signed-off-by: Yang Jihong <yangjihong1@huawei.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20221111030836.37632-1-yangjihong1@huawei.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

ASoC: codecs: wsa883x: use correct header file

[ Upstream commit 5f52ceddc40cd61b1dd2ecf735624deaf05f779f ]

Fix build errors when GPIOLIB is not set/enabled:

../sound/soc/codecs/wsa883x.c: In function 'wsa883x_probe':
../sound/soc/codecs/wsa883x.c:1394:25: error: implicit declaration of function 'devm_gpiod_get_optional'; did you mean 'devm_regulator_get_optional'? [-Werror=implicit-function-declaration]
         wsa883x->sd_n = devm_gpiod_get_optional(&pdev->dev, "powerdown",
../sound/soc/codecs/wsa883x.c:1395:49: error: 'GPIOD_FLAGS_BIT_NONEXCLUSIVE' undeclared (first use in this function)
         GPIOD_FLAGS_BIT_NONEXCLUSIVE);
../sound/soc/codecs/wsa883x.c:1414:9: error: implicit declaration of function 'gpiod_direction_output'; did you mean 'gpio_direction_output'? [-Werror=implicit-function-declaration]
         gpiod_direction_output(wsa883x->sd_n, 1);

Fixes: 43b8c7dc85a1 ("ASoC: codecs: add wsa883x amplifier support")
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Reported-by: kernel test robot <lkp@intel.com>
Cc: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Cc: Banajit Goswami <bgoswami@quicinc.com>
Cc: Mark Brown <broonie@kernel.org>
Cc: Liam Girdwood <lgirdwood@gmail.com>
Cc: alsa-devel@alsa-project.org
Cc: Jaroslav Kysela <perex@perex.cz>
Cc: Takashi Iwai <tiwai@suse.com>
Reviewed-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Link: https://lore.kernel.org/r/20221108001829.5100-1-rdunlap@infradead.org
Signed-off-by: Mark Brown <broonie@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

ASoC: codecs: wsa883x: Use proper shutdown GPIO polarity

[ Upstream commit ec5dba73f7ba10797904cf18092d2e6975a22147 ]

The shutdown GPIO is active low (SD_N), but this depends on actual board
layout.  Linux drivers should only care about logical state, where high
(1) means shutdown and low (0) means do not shutdown.

Invert the GPIO to match logical value.

Fixes: 43b8c7dc85a1 ("ASoC: codecs: add wsa883x amplifier support")
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Link: https://lore.kernel.org/r/20221110133512.478831-2-krzysztof.kozlowski@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

module: Fix NULL vs IS_ERR checking for module_get_next_page

[ Upstream commit 45af1d7aae7d5520d2858f8517a1342646f015db ]

The module_get_next_page() function return error pointers on error
instead of NULL.
Use IS_ERR() to check the return value to fix this.

Fixes: b1ae6dc41eaa ("module: add in-kernel support for decompressing")
Signed-off-by: Miaoqian Lin <linmq006@gmail.com>
Reviewed-by: Dmitry Torokhov <dmitry.torokhov@gmail.com
Signed-off-by: Luis Chamberlain <mcgrof@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

wifi: iwlwifi: mei: fix potential NULL-ptr deref after clone

[ Upstream commit d3df49dda431f7ae4132a9a0ac25a5134c04e812 ]

If cloning the SKB fails, don't try to use it, but rather return
as if we should pass it.

Coverity CID: 1503456

Fixes: 2da4366f9e2c ("iwlwifi: mei: add the driver to allow cooperation with CSME")
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20221030191011.0ce03ba99601.I87960b7cb0a3d16b9fd8d9144027e7e2587f5a58@changeid
Signed-off-by: Sasha Levin <sashal@kernel.org>

wifi: iwlwifi: mei: avoid blocking sap messages handling due to rtnl lock

[ Upstream commit d288067ede4b375e72daf7f9a98d937ede11a311 ]

The AMT_STATE sap message handler tries to take the rtnl lock.
This means that in case the rtnl lock is already taken, sap messages
will not be processed.
When an interface is brought up, the host requests ownership from
csme. However, since the rtnl lock is already held, if there is a
pending amt state message, the host will not be able to read the
ownership confirm message because the amt state message handler
is pending. As a result, the host fails to get ownership although
csme granted it.
Fix it by moving the part that needs the rtnl lock into a dedicated
worker, so handling sap messages can continue.

Fixes: 2da4366f9e2c ("iwlwifi: mei: add the driver to allow cooperation with CSME")
Signed-off-by: Avraham Stern <avraham.stern@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20221030191011.8599f2b4e9dd.I518f79e9099bf815c5f8d90235b4ce3250f59970@changeid
Signed-off-by: Sasha Levin <sashal@kernel.org>

wifi: iwlwifi: mei: fix tx DHCP packet for devices with new Tx API

[ Upstream commit bcd68b3dbe78b7b0f7b6b55162cf1eff1e7fff9e ]

Devices with new Tx API have the IV introduced by the HW and it is not
present in the skb at all. Hence we don't need to tell
iwl_mvm_mei_tx_copy_to_csme to jump over 8 bytes to get to the ethernet
header.

Fixes: 2da4366f9e2c ("iwlwifi: mei: add the driver to allow cooperation with CSME")
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20221030191011.12dc42133502.Idd744ffeeb84b880eb497963ee02563cbb959a42@changeid
Signed-off-by: Sasha Levin <sashal@kernel.org>

wifi: iwlwifi: mei: don't send SAP commands if AMT is disabled

[ Upstream commit 95170a46b7dddbc3ac31b20ef2e8fa9d556d783d ]

We should not send any SAP command to CSME if AMT is disabled.

Reported-by: Toke Høiland-Jørgensen <toke@toke.dk>
Fixes: 2da4366f9e2c ("iwlwifi: mei: add the driver to allow cooperation with CSME")
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20221030191011.ea222d41c781.Ifc90ddc3e35187683ff7f59371d792b61c8854c8@changeid
Signed-off-by: Sasha Levin <sashal@kernel.org>

wifi: iwlwifi: mei: make sure ownership confirmed message is sent

[ Upstream commit 5aa7ce31bd84c2f4f059200f06c537c920cbb458 ]

It is possible that CSME will try to take ownership while the driver
is stopping. In this case, if the CSME takes ownership message arrives
after the driver started unregistering, the iwl_mei_cache->ops is
already invalid, so the host will not answer with the ownership
confirmed message.
Similarly, if the take ownership message arrived after the mac was
stopped or when iwl_mvm_up() failed, setting rfkill will not trigger
sending the confirm message. As a result, CSME will not take
ownership, which will result in a disconnection.

Fix it by sending the ownership confirmed message immediately in such
cases.

Fixes: 2da4366f9e2c ("iwlwifi: mei: add the driver to allow cooperation with CSME")
Signed-off-by: Avraham Stern <avraham.stern@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20221030191011.b2a4c009e3e6.I7f931b7ee8b168e8ac88b11f23bff98b7ed3cb19@changeid
Signed-off-by: Sasha Levin <sashal@kernel.org>

pinctrl: mediatek: fix the pinconf register offset of some pins

[ Upstream commit 3476b354c65db442580ef355885c69e60c546ef0 ]

Correct the bias-pull-up, bias-pull-down and bias-disable register
offset of mt7986 pin-42 to pin-49, in the original driver, the
relative offset value was erroneously decremented by 1.

Fixes: 360de6728064 ("pinctrl: mediatek: add support for MT7986 SoC")
Signed-off-by: Sam Shih <sam.shih@mediatek.com>
Signed-off-by: Frank Wunderlich <frank-w@public-files.de>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Link: https://lore.kernel.org/r/20221106080114.7426-5-linux@fw-web.de
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

dt-bindings: pinctrl: update uart/mmc bindings for MT7986 SoC

[ Upstream commit c115e7f51e685536ecb885854bdd4b3f225ff3e4 ]

Fix mmc and uart pins after uart splitting.

Some pinmux pins of the mt7986 pinctrl driver is composed of multiple
pinctrl groups, the original binding only allows one pinctrl group
per dts node, this patch sets "maxItems" for these groups and add new
examples to the binding documentation.

Fixes: 65916a1ca90a ("dt-bindings: pinctrl: update bindings for MT7986 SoC")
Signed-off-by: Sam Shih <sam.shih@mediatek.com>
Signed-off-by: Frank Wunderlich <frank-w@public-files.de>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/r/20221106080114.7426-3-linux@fw-web.de
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

drm/radeon: Add the missed acpi_put_table() to fix memory leak

[ Upstream commit 10276a20be1115e1f76c189330da2992df980eee ]

When the radeon driver reads the bios information from ACPI
table in radeon_acpi_vfct_bios(), it misses to call acpi_put_table()
to release the ACPI memory after the init, so add acpi_put_table()
properly to fix the memory leak.

v2: fix text formatting (Alex)

Fixes: 268ba0a99f89 ("drm/radeon: implement ACPI VFCT vbios fetch (v3)")
Signed-off-by: Hanjun Guo <guohanjun@huawei.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

bfq: fix waker_bfqq inconsistency crash

[ Upstream commit a1795c2ccb1e4c49220d2a0d381540024d71647c ]

This fixes crashes in bfq_add_bfqq_busy due to waker_bfqq being NULL,
but woken_list_node still being hashed. This would happen when
bfq_init_rq() expects a brand new allocated queue to be returned from
bfq_get_bfqq_handle_split() and unconditionally updates waker_bfqq
without resetting woken_list_node. Since we can always return oom_bfqq
when attempting to allocate, we cannot assume waker_bfqq starts as NULL.

Avoid setting woken_bfqq for oom_bfqq entirely, as it's not useful.

Crashes would have a stacktrace like:
[160595.656560]  bfq_add_bfqq_busy+0x110/0x1ec
[160595.661142]  bfq_add_request+0x6bc/0x980
[160595.666602]  bfq_insert_request+0x8ec/0x1240
[160595.671762]  bfq_insert_requests+0x58/0x9c
[160595.676420]  blk_mq_sched_insert_request+0x11c/0x198
[160595.682107]  blk_mq_submit_bio+0x270/0x62c
[160595.686759]  __submit_bio_noacct_mq+0xec/0x178
[160595.691926]  submit_bio+0x120/0x184
[160595.695990]  ext4_mpage_readpages+0x77c/0x7c8
[160595.701026]  ext4_readpage+0x60/0xb0
[160595.705158]  filemap_read_page+0x54/0x114
[160595.711961]  filemap_fault+0x228/0x5f4
[160595.716272]  do_read_fault+0xe0/0x1f0
[160595.720487]  do_fault+0x40/0x1c8

Tested by injecting random failures into bfq_get_queue, crashes go away
completely.

Fixes: 8ef3fc3a043c ("block, bfq: make shared queues inherit wakers")
Signed-off-by: Khazhismel Kumykov <khazhy@google.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20221108181030.1611703-1-khazhy@google.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>

drbd: use blk_queue_max_discard_sectors helper

[ Upstream commit 258bea6388ac93f34561fd91064232d14e174bff ]

We currently only set q->limits.max_discard_sectors, but that is not
enough. Another field, max_hw_discard_sectors, was introduced in
commit 0034af036554 ("block: make /sys/block/<dev>/queue/discard_max_bytes
writeable").

The difference is that max_discard_sectors can be changed from user
space via sysfs, while max_hw_discard_sectors is the "hardware" upper
limit.

So use this helper, which sets both.

This is also a fixup for commit 998e9cbcd615 ("drbd: cleanup
decide_on_discard_support"): if discards are not supported, that does
not necessarily mean we also want to disable write_zeroes.

Fixes: 998e9cbcd615 ("drbd: cleanup decide_on_discard_support")
Reviewed-by: Joel Colledge <joel.colledge@linbit.com>
Signed-off-by: Christoph Böhmwalder <christoph.boehmwalder@linbit.com>
Link: https://lore.kernel.org/r/20221109133453.51652-2-christoph.boehmwalder@linbit.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>

regmap-irq: Use the new num_config_regs property in regmap_add_irq_chip_fwnode

[ Upstream commit 84498d1fb35de6ab71bdfdb6270a464fb4a0951b ]

Commit faa87ce9196d ("regmap-irq: Introduce config registers for irq
types") added the num_config_regs, then commit 9edd4f5aee84 ("regmap-irq:
Deprecate type registers and virtual registers") suggested to replace
num_type_reg with it. However, regmap_add_irq_chip_fwnode wasn't modified
to use the new property. Later on, commit 255a03bb1bb3 ("ASoC: wcd9335:
Convert irq chip to config regs") removed the old num_type_reg property
from the WCD9335 driver's struct regmap_irq_chip, causing a null pointer
dereference in regmap_irq_set_type when it tried to index d->type_buf as
it was never allocated in regmap_add_irq_chip_fwnode:

[   39.199374] Unable to handle kernel NULL pointer dereference at virtual address 0000000000000000

[   39.200006] Call trace:
[   39.200014]  regmap_irq_set_type+0x84/0x1c0
[   39.200026]  __irq_set_trigger+0x60/0x1c0
[   39.200040]  __setup_irq+0x2f4/0x78c
[   39.200051]  request_threaded_irq+0xe8/0x1a0

Use num_config_regs in regmap_add_irq_chip_fwnode instead of num_type_reg,
and fall back to it if num_config_regs isn't defined to maintain backward
compatibility.

Fixes: faa87ce9196d ("regmap-irq: Introduce config registers for irq types")
Signed-off-by: Yassine Oudjana <y.oudjana@protonmail.com>
Link: https://lore.kernel.org/r/20221107202114.823975-1-y.oudjana@protonmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

drm: rcar-du: Drop leftovers dependencies from Kconfig

[ Upstream commit 1760eb547276299ab0c6a6cd3d29469e54ade615 ]

Commit 841281fe52a7 ("drm: rcar-du: Drop LVDS device tree backward
compatibility") has removed device tree overlay sources used for
backward compatibility with old bindings, but forgot to remove related
dependencies from Kconfig. Fix it.

Fixes: 841281fe52a7 ("drm: rcar-du: Drop LVDS device tree backward compatibility")
Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>
Reviewed-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

wifi: rtw89: use u32_encode_bits() to fill MAC quota value

[ Upstream commit 525c06c81d75690a9b795cc62a758838c1a6b6fe ]

Coverity reported shift 16 bits could cause sign extension and might get
an unexpected value. Since the input values are predefined and no this
kind of case, original code is safe so far. But, still changing them to
use u32_encode_bits() will be more clear and prevent mistakes in the
future.

The original message of Coverity is:
  Suspicious implicit sign extension: "max_cfg->cma0_dma" with type "u16"
  (16 bits, unsigned) is promoted in "max_cfg->cma0_dma << 16" to type
  "int" (32 bits, signed), then sign-extended to type "unsigned long"
  (64 bits, unsigned).  If "max_cfg->cma0_dma << 16" is greater than
  0x7FFFFFFF, the upper bits of the result will all be 1."

Reported-by: coverity-bot <keescook+coverity-bot@chromium.org>
Addresses-Coverity-ID: 1527095 ("Integer handling issues")
Fixes: e3ec7017f6a2 ("rtw89: add Realtek 802.11ax driver")
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20221108013858.10806-1-pkshih@realtek.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

drm: lcdif: Set and enable FIFO Panic threshold

[ Upstream commit e3cac8f7749f78dacdf19c00ed5862a1db52239f ]

In case the LCDIFv3 is used to drive a 4k panel via i.MX8MP HDMI bridge,
the LCDIFv3 becomes susceptible to FIFO underflows, these lead to nasty
flicker of the image on the panel, or image being shifted by half frame
horizontally every second frame. The flicker can be easily triggered by
running 3D application on top of weston compositor, like neverball or
chromium. Surprisingly glmark2-es2-wayland or glmark2-es2-drm does not
trigger this effect so easily.

Configure the FIFO Panic threshold register and enable the FIFO Panic
mode, which internally boosts the NoC interconnect priority for LCDIFv3
transactions in case of possible underflow. This mitigates the flicker
effect on 4k panels as well.

Fixes: 9db35bb349a0 ("drm: lcdif: Add support for i.MX8MP LCDIF variant")
Signed-off-by: Marek Vasut <marex@denx.de>
Tested-by: Liu Ying <victor.liu@nxp.com> # i.MX8mp EVK
Reviewed-by: Liu Ying <victor.liu@nxp.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20221101152629.21768-1-marex@denx.de
Signed-off-by: Sasha Levin <sashal@kernel.org>

rxrpc: Fix ack.bufferSize to be 0 when generating an ack

[ Upstream commit 8889a711f9b4dcf4dd1330fa493081beebd118c9 ]

ack.bufferSize should be set to 0 when generating an ack.

Fixes: 8d94aa381dab ("rxrpc: Calls shouldn't hold socket refs")
Reported-by: Jeffrey Altman <jaltman@auristor.com>
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org
Signed-off-by: Sasha Levin <sashal@kernel.org>

net, proc: Provide PROC_FS=n fallback for proc_create_net_single_write()

[ Upstream commit c3d96f690a790074b508fe183a41e36a00cd7ddd ]

Provide a CONFIG_PROC_FS=n fallback for proc_create_net_single_write().

Also provide a fallback for proc_create_net_data_write().

Fixes: 564def71765c ("proc: Add a way to make network proc files writable")
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org
cc: netdev@vger.kernel.org
Signed-off-by: Sasha Levin <sashal@kernel.org>

virt/sev-guest: Add a MODULE_ALIAS

[ Upstream commit 2874529b3513bdc90299c90f40713602da685e35 ]

Autoload the driver when, for example, SNP init code creates the
corresponding platform device.

  [ bp: Rewrite commit message. ]

Fixes: fce96cf04430 ("virt: Add SEV-SNP guest driver")
Signed-off-by: Cole Robinson <crobinso@redhat.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Acked-by: Tom Lendacky <thomas.lendacky@amd.com>
Link: https://lore.kernel.org/r/ff480c5e688eb0a72a4db0a29c7b1bb54c45bfd4.1667594253.git.crobinso@redhat.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

clk: renesas: r8a779f0: Fix SCIF parent clocks

[ Upstream commit 2e0d7d3eabce3babae1fd66d7650e00c848a3b45 ]

As serial communication requires a clean clock signal, the Serial
Communication Interfaces with FIFO (SCIF) are clocked by a clock that is
not affected by Spread Spectrum or Fractional Multiplication.

Hence change the parent clocks for the SCIF modules from the S0D12_PER
clock to the SASYNCPERD4 clock (which has the same clock rate), cfr.
R-Car S4-8 Hardware User's Manual rev. 0.81.

Fixes: 24aaff6a6ce4 ("clk: renesas: cpg-mssr: Add support for R-Car S4-8")
Reported-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Link: https://lore.kernel.org/r/20221103143440.46449-3-wsa+renesas@sang-engineering.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Sasha Levin <sashal@kernel.org>

clk: renesas: r8a779f0: Fix HSCIF parent clocks

[ Upstream commit c258e3ab639112d8f5ae9df9a873750ae2623ce2 ]

As serial communication requires a clean clock signal, the High Speed
Serial Communication Interfaces with FIFO (HSCIF) are clocked by a clock
that is not affected by Spread Spectrum or Fractional Multiplication.

Hence change the parent clocks for the HSCIF modules from the S0D3_PER
clock to the SASYNCPERD1 clock (which has the same clock rate), cfr.
R-Car S4-8 Hardware User's Manual rev. 0.81.

Fixes: 080bcd8d5997 ("clk: renesas: r8a779f0: Add HSCIF clocks")
Reported-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Link: https://lore.kernel.org/r/20221103143440.46449-2-wsa+renesas@sang-engineering.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Sasha Levin <sashal@kernel.org>

media: camss: Do not attach an already attached power domain on MSM8916 platform

[ Upstream commit 3d658980e6dac2af6a024fdb6ded3d7bc44dc9ff ]

The change to dynamically allocated power domains neglected a case of
CAMSS on MSM8916 platform, where a single VFE power domain is neither
attached, linked or managed in runtime in any way explicitly.

This is a special case and it shall be kept as is, because the power
domain management is done outside of the driver, and it's very different
in comparison to all other platforms supported by CAMSS.

Fixes: 6b1814e26989 ("media: camss: Allocate power domain resources dynamically")
Signed-off-by: Vladimir Zapolskiy <vladimir.zapolskiy@linaro.org>
Reviewed-by: Robert Foss <robert.foss@linaro.org>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

media: camss: Clean up received buffers on failed start of streaming

[ Upstream commit c8f3582345e6a69da65ab588f7c4c2d1685b0e80 ]

It is required to return the received buffers, if streaming can not be
started. For instance media_pipeline_start() may fail with EPIPE, if
a link validation between entities is not passed, and in such a case
a user gets a kernel warning:

  WARNING: CPU: 1 PID: 520 at drivers/media/common/videobuf2/videobuf2-core.c:1592 vb2_start_streaming+0xec/0x160
  <snip>
  Call trace:
   vb2_start_streaming+0xec/0x160
   vb2_core_streamon+0x9c/0x1a0
   vb2_ioctl_streamon+0x68/0xbc
   v4l_streamon+0x30/0x3c
   __video_do_ioctl+0x184/0x3e0
   video_usercopy+0x37c/0x7b0
   video_ioctl2+0x24/0x40
   v4l2_ioctl+0x4c/0x70

The fix is to correct the error path in video_start_streaming() of camss.

Fixes: 0ac2586c410f ("media: camss: Add files which handle the video device nodes")
Signed-off-by: Vladimir Zapolskiy <vladimir.zapolskiy@linaro.org>
Reviewed-by: Robert Foss <robert.foss@linaro.org>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

wifi: rsi: Fix handling of 802.3 EAPOL frames sent via control port

[ Upstream commit b8f6efccbb9dc0ff5dee7e20d69a4747298ee603 ]

When using wpa_supplicant v2.10, this driver is no longer able to
associate with any AP and fails in the EAPOL 4-way handshake while
sending the 2/4 message to the AP. The problem is not present in
wpa_supplicant v2.9 or older. The problem stems from HostAP commit
144314eaa ("wpa_supplicant: Send EAPOL frames over nl80211 where available")
which changes the way EAPOL frames are sent, from them being send
at L2 frames to them being sent via nl80211 control port.

An EAPOL frame sent as L2 frame is passed to the WiFi driver with
skb->protocol ETH_P_PAE, while EAPOL frame sent via nl80211 control
port has skb->protocol set to ETH_P_802_3 . The later happens in
ieee80211_tx_control_port(), where the EAPOL frame is encapsulated
into 802.3 frame.

The rsi_91x driver handles ETH_P_PAE EAPOL frames as high-priority
frames and sends them via highest-priority transmit queue, while
the ETH_P_802_3 frames are sent as regular frames. The EAPOL 4-way
handshake frames must be sent as highest-priority, otherwise the
4-way handshake times out.

Therefore, to fix this problem, inspect the skb control flags and
if flag IEEE80211_TX_CTRL_PORT_CTRL_PROTO is set, assume this is
an EAPOL frame and transmit the frame via high-priority queue just
like other ETH_P_PAE frames.

Fixes: 0eb42586cf87 ("rsi: data packet descriptor enhancements")
Signed-off-by: Marek Vasut <marex@denx.de>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20221104163339.227432-1-marex@denx.de
Signed-off-by: Sasha Levin <sashal@kernel.org>

Input: joystick - fix Kconfig warning for JOYSTICK_ADC

[ Upstream commit 6100a19c4fcfe154dd32f8a8ef4e8c0b1f607c75 ]

Fix a Kconfig warning for JOYSTICK_ADC by also selecting
IIO_BUFFER.

WARNING: unmet direct dependencies detected for IIO_BUFFER_CB
  Depends on [n]: IIO [=y] && IIO_BUFFER [=n]
  Selected by [y]:
  - JOYSTICK_ADC [=y] && INPUT [=y] && INPUT_JOYSTICK [=y] && IIO [=y]

Fixes: 2c2b364fddd5 ("Input: joystick - add ADC attached joystick driver.")
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Link: https://lore.kernel.org/r/20221104201238.31628-1-rdunlap@infradead.org
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

mtd: core: fix possible resource leak in init_mtd()

[ Upstream commit 1aadf01e5076b9ab6bf294b9622335c651314895 ]

I got the error report while inject fault in init_mtd():

sysfs: cannot create duplicate filename '/devices/virtual/bdi/mtd-0'
Call Trace:
 <TASK>
 dump_stack_lvl+0x67/0x83
 sysfs_warn_dup+0x60/0x70
 sysfs_create_dir_ns+0x109/0x120
 kobject_add_internal+0xce/0x2f0
 kobject_add+0x98/0x110
 device_add+0x179/0xc00
 device_create_groups_vargs+0xf4/0x100
 device_create+0x7b/0xb0
 bdi_register_va.part.13+0x58/0x2d0
 bdi_register+0x9b/0xb0
 init_mtd+0x62/0x171 [mtd]
 do_one_initcall+0x6c/0x3c0
 do_init_module+0x58/0x222
 load_module+0x268e/0x27d0
 __do_sys_finit_module+0xd5/0x140
 do_syscall_64+0x37/0x90
 entry_SYSCALL_64_after_hwframe+0x63/0xcd
 </TASK>
kobject_add_internal failed for mtd-0 with -EEXIST, don't try to register
things with the same name in the same directory.
Error registering mtd class or bdi: -17

If init_mtdchar() fails in init_mtd(), mtd_bdi will not be unregistered,
as a result, we can't load the mtd module again, to fix this by calling
bdi_unregister(mtd_bdi) after out_procfs label.

Fixes: 445caaa20c4d ("mtd: Allocate bdi objects dynamically")
Signed-off-by: Gaosheng Cui <cuigaosheng1@huawei.com>
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/linux-mtd/20221024065109.2050705-1-cuigaosheng1@huawei.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

mtd: Fix device name leak when register device failed in add_mtd_device()

[ Upstream commit 895d68a39481a75c680aa421546931fb11942fa6 ]

There is a kmemleak when register device failed:
  unreferenced object 0xffff888101aab550 (size 8):
    comm "insmod", pid 3922, jiffies 4295277753 (age 925.408s)
    hex dump (first 8 bytes):
      6d 74 64 30 00 88 ff ff                          mtd0....
    backtrace:
      [<00000000bde26724>] __kmalloc_node_track_caller+0x4e/0x150
      [<000000003c32b416>] kvasprintf+0xb0/0x130
      [<000000001f7a8f15>] kobject_set_name_vargs+0x2f/0xb0
      [<000000006e781163>] dev_set_name+0xab/0xe0
      [<00000000e30d0c78>] add_mtd_device+0x4bb/0x700
      [<00000000f3d34de7>] mtd_device_parse_register+0x2ac/0x3f0
      [<00000000c0d88488>] 0xffffffffa0238457
      [<00000000b40d0922>] 0xffffffffa02a008f
      [<0000000023d17b9d>] do_one_initcall+0x87/0x2a0
      [<00000000770f6ca6>] do_init_module+0xdf/0x320
      [<000000007b6768fe>] load_module+0x2f98/0x3330
      [<00000000346bed5a>] __do_sys_finit_module+0x113/0x1b0
      [<00000000674c2290>] do_syscall_64+0x35/0x80
      [<000000004c6a8d97>] entry_SYSCALL_64_after_hwframe+0x46/0xb0

If register device failed, should call put_device() to give up the
reference.

Fixes: 1f24b5a8ecbb ("[MTD] driver model updates")
Signed-off-by: Zhang Xiaoxu <zhangxiaoxu5@huawei.com>
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/linux-mtd/20221022121352.2534682-1-zhangxiaoxu5@huawei.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

clk: qcom: gcc-sm8250: Use retention mode for USB GDSCs

[ Upstream commit ac1c5a03d3772b1db25e8092f771aa33f6ae2f7e ]

USB controllers on SM8250 doesn't work after coming back from suspend.
This can be fixed by keeping the USB GDSCs in retention mode so that
hardware can keep them ON and put into rentention mode once the parent
domain goes to a low power state.

Fixes: 3e5770921a88 ("clk: qcom: gcc: Add global clock controller driver for SM8250")
Signed-off-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
Link: https://lore.kernel.org/r/20221102091320.66007-1-manivannan.sadhasivam@linaro.org
Signed-off-by: Sasha Levin <sashal@kernel.org>

clk: qcom: dispcc-sm6350: Add CLK_OPS_PARENT_ENABLE to pixel&byte src

[ Upstream commit 92039e8c080c63748f8e133e7cfad33a75daefb6 ]

Add the CLK_OPS_PARENT_ENABLE flag to pixel and byte clk srcs to
ensure set_rate can succeed.

Signed-off-by: Konrad Dybcio <konrad.dybcio@somainline.org>
Fixes: 837519775f1d ("clk: qcom: Add display clock controller driver for  SM6350")
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
Link: https://lore.kernel.org/r/20221010155546.73884-1-konrad.dybcio@somainline.org
Signed-off-by: Sasha Levin <sashal@kernel.org>

clk: qcom: gcc-ipq806x: use parent_data for the last remaining entry

[ Upstream commit 55307e522cc7a4dddc3d231ca5cb7e68e9668f66 ]

Use parent_data for the last remaining entry (pll4). This clock is
provided by the lcc device.

Fixes: cb02866f9a74 ("clk: qcom: gcc-ipq806x: convert parent_names to parent_data")
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Konrad Dybcio <konrad.dybcio@somainline.org>
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
Link: https://lore.kernel.org/r/20220927113826.246241-3-dmitry.baryshkov@linaro.org
Signed-off-by: Sasha Levin <sashal@kernel.org>

bpf: propagate precision across all frames, not just the last one

[ Upstream commit 529409ea92d590659be487ba0839710329bd8074 ]

When equivalent completed state is found and it has additional precision
restrictions, BPF verifier propagates precision to
currently-being-verified state chain (i.e., including parent states) so
that if some of the states in the chain are not yet completed, necessary
precision restrictions are enforced.

Unfortunately, right now this happens only for the last frame (deepest
active subprogram's frame), not all the frames. This can lead to
incorrect matching of states due to missing precision marker. Currently
this doesn't seem possible as BPF verifier forces everything to precise
when validated BPF program has any subprograms. But with the next patch
lifting this restriction, this becomes problematic.

In fact, without this fix, we'll start getting failure in one of the
existing test_verifier test cases:

  #906/p precise: cross frame pruning FAIL
  Unexpected success to load!
  verification time 48 usec
  stack depth 0+0
  processed 26 insns (limit 1000000) max_states_per_insn 3 total_states 17 peak_states 17 mark_read 8

This patch adds precision propagation across all frames.

Fixes: a3ce685dd01a ("bpf: fix precision tracking")
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20221104163649.121784-3-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

bpf: propagate precision in ALU/ALU64 operations

[ Upstream commit a3b666bfa9c9edc05bca62a87abafe0936bd7f97 ]

When processing ALU/ALU64 operations (apart from BPF_MOV, which is
handled correctly already; and BPF_NEG and BPF_END are special and don't
have source register), if destination register is already marked
precise, this causes problem with potentially missing precision tracking
for the source register. E.g., when we have r1 >>= r5 and r1 is marked
precise, but r5 isn't, this will lead to r5 staying as imprecise. This
is due to the precision backtracking logic stopping early when it sees
r1 is already marked precise. If r1 wasn't precise, we'd keep
backtracking and would add r5 to the set of registers that need to be
marked precise. So there is a discrepancy here which can lead to invalid
and incompatible states matched due to lack of precision marking on r5.
If r1 wasn't precise, precision backtracking would correctly mark both
r1 and r5 as precise.

This is simple to fix, though. During the forward instruction simulation
pass, for arithmetic operations of `scalar <op>= scalar` form (where
<op> is ALU or ALU64 operations), if destination register is already
precise, mark source register as precise. This applies only when both
involved registers are SCALARs. `ptr += scalar` and `scalar += ptr`
cases are already handled correctly.

This does have (negative) effect on some selftest programs and few
Cilium programs.  ~/baseline-tmp-results.csv are veristat results with
this patch, while ~/baseline-results.csv is without it. See post
scriptum for instructions on how to make Cilium programs testable with
veristat. Correctness has a price.

$ ./veristat -C -e file,prog,insns,states ~/baseline-results.csv ~/baseline-tmp-results.csv | grep -v '+0'
File                     Program               Total insns (A)  Total insns (B)  Total insns (DIFF)  Total states (A)  Total states (B)  Total states (DIFF)
-----------------------  --------------------  ---------------  ---------------  ------------------  ----------------  ----------------  -------------------
bpf_cubic.bpf.linked1.o  bpf_cubic_cong_avoid              997             1700      +703 (+70.51%)                62                90        +28 (+45.16%)
test_l4lb.bpf.linked1.o  balancer_ingress                 4559             5469      +910 (+19.96%)               118               126          +8 (+6.78%)
-----------------------  --------------------  ---------------  ---------------  ------------------  ----------------  ----------------  -------------------

$ ./veristat -C -e file,prog,verdict,insns,states ~/baseline-results-cilium.csv ~/baseline-tmp-results-cilium.csv | grep -v '+0'
File           Program                         Total insns (A)  Total insns (B)  Total insns (DIFF)  Total states (A)  Total states (B)  Total states (DIFF)
-------------  ------------------------------  ---------------  ---------------  ------------------  ----------------  ----------------  -------------------
bpf_host.o     tail_nodeport_nat_ingress_ipv6             4448             5261      +813 (+18.28%)               234               247         +13 (+5.56%)
bpf_host.o     tail_nodeport_nat_ipv6_egress              3396             3446        +50 (+1.47%)               201               203          +2 (+1.00%)
bpf_lxc.o      tail_nodeport_nat_ingress_ipv6             4448             5261      +813 (+18.28%)               234               247         +13 (+5.56%)
bpf_overlay.o  tail_nodeport_nat_ingress_ipv6             4448             5261      +813 (+18.28%)               234               247         +13 (+5.56%)
bpf_xdp.o      tail_lb_ipv4                              71736            73442      +1706 (+2.38%)              4295              4370         +75 (+1.75%)
-------------  ------------------------------  ---------------  ---------------  ------------------  ----------------  ----------------  -------------------

P.S. To make Cilium ([0]) programs libbpf-compatible and thus
veristat-loadable, apply changes from topmost commit in [1], which does
minimal changes to Cilium source code, mostly around SEC() annotations
and BPF map definitions.

  [0] https://github.com/cilium/cilium/
  [1] https://github.com/anakryiko/cilium/commits/libbpf-friendliness

Fixes: b5dc0163d8fd ("bpf: precise scalar_value tracking")
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20221104163649.121784-2-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

media: platform: exynos4-is: fix return value check in fimc_md_probe()

[ Upstream commit e38e42c078da4af962d322b97e726dcb2f184e3f ]

devm_pinctrl_get() may return ERR_PTR(-EPROBE_DEFER), add a minus sign
to fix it.

Fixes: 4163851f7b99 ("[media] s5p-fimc: Use pinctrl API for camera ports configuration")
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Sasha Levin <sashal@kernel.org>