hwmon: (adt7475) Display smoothing attributes in correct order

[ Upstream commit 5f8d1e3b6f9b5971f9c06d5846ce00c49e3a8d94 ]

Throughout the ADT7475 driver, attributes relating to the temperature
sensors are displayed in the order Remote 1, Local, Remote 2.  Make
temp_st_show() conform to this expectation so that values set by
temp_st_store() can be displayed using the correct attribute.

Fixes: 8f05bcc33e74 ("hwmon: (adt7475) temperature smoothing")
Signed-off-by: Tony O'Brien <tony.obrien@alliedtelesis.co.nz>
Link: https://lore.kernel.org/r/20230222005228.158661-2-tony.obrien@alliedtelesis.co.nz
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>

bonding: restore bond's IFF_SLAVE flag if a non-eth dev enslave fails

[ Upstream commit e667d469098671261d558be0cd93dca4d285ce1e ]

syzbot reported a warning[1] where the bond device itself is a slave and
we try to enslave a non-ethernet device as the first slave which fails
but then in the error path when ether_setup() restores the bond device
it also clears all flags. In my previous fix[2] I restored the
IFF_MASTER flag, but I didn't consider the case that the bond device
itself might also be a slave with IFF_SLAVE set, so we need to restore
that flag as well. Use the bond_ether_setup helper which does the right
thing and restores the bond's flags properly.

Steps to reproduce using a nlmon dev:
 $ ip l add nlmon0 type nlmon
 $ ip l add bond1 type bond
 $ ip l add bond2 type bond
 $ ip l set bond1 master bond2
 $ ip l set dev nlmon0 master bond1
 $ ip -d l sh dev bond1
 22: bond1: <BROADCAST,MULTICAST,MASTER> mtu 1500 qdisc noqueue master bond2 state DOWN mode DEFAULT group default qlen 1000
 (now bond1's IFF_SLAVE flag is gone and we'll hit a warning[3] if we
  try to delete it)

[1] https://syzkaller.appspot.com/bug?id=391c7b1f6522182899efba27d891f1743e8eb3ef
[2] commit 7d5cd2ce5292 ("bonding: correctly handle bonding type change on enslave failure")
[3] example warning:
 [   27.008664] bond1: (slave nlmon0): The slave device specified does not support setting the MAC address
 [   27.008692] bond1: (slave nlmon0): Error -95 calling set_mac_address
 [   32.464639] bond1 (unregistering): Released all slaves
 [   32.464685] ------------[ cut here ]------------
 [   32.464686] WARNING: CPU: 1 PID: 2004 at net/core/dev.c:10829 unregister_netdevice_many+0x72a/0x780
 [   32.464694] Modules linked in: br_netfilter bridge bonding virtio_net
 [   32.464699] CPU: 1 PID: 2004 Comm: ip Kdump: loaded Not tainted 5.18.0-rc3+ #47
 [   32.464703] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.1-2.fc37 04/01/2014
 [   32.464704] RIP: 0010:unregister_netdevice_many+0x72a/0x780
 [   32.464707] Code: 99 fd ff ff ba 90 1a 00 00 48 c7 c6 f4 02 66 96 48 c7 c7 20 4d 35 96 c6 05 fa c7 2b 02 01 e8 be 6f 4a 00 0f 0b e9 73 fd ff ff <0f> 0b e9 5f fd ff ff 80 3d e3 c7 2b 02 00 0f 85 3b fd ff ff ba 59
 [   32.464710] RSP: 0018:ffffa006422d7820 EFLAGS: 00010206
 [   32.464712] RAX: ffff8f6e077140a0 RBX: ffffa006422d7888 RCX: 0000000000000000
 [   32.464714] RDX: ffff8f6e12edbe58 RSI: 0000000000000296 RDI: ffffffff96d4a520
 [   32.464716] RBP: ffff8f6e07714000 R08: ffffffff96d63600 R09: ffffa006422d7728
 [   32.464717] R10: 0000000000000ec0 R11: ffffffff9698c988 R12: ffff8f6e12edb140
 [   32.464719] R13: dead000000000122 R14: dead000000000100 R15: ffff8f6e12edb140
 [   32.464723] FS:  00007f297c2f1740(0000) GS:ffff8f6e5d900000(0000) knlGS:0000000000000000
 [   32.464725] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
 [   32.464726] CR2: 00007f297bf1c800 CR3: 00000000115e8000 CR4: 0000000000350ee0
 [   32.464730] Call Trace:
 [   32.464763]  <TASK>
 [   32.464767]  rtnl_dellink+0x13e/0x380
 [   32.464776]  ? cred_has_capability.isra.0+0x68/0x100
 [   32.464780]  ? __rtnl_unlock+0x33/0x60
 [   32.464783]  ? bpf_lsm_capset+0x10/0x10
 [   32.464786]  ? security_capable+0x36/0x50
 [   32.464790]  rtnetlink_rcv_msg+0x14e/0x3b0
 [   32.464792]  ? _copy_to_iter+0xb1/0x790
 [   32.464796]  ? post_alloc_hook+0xa0/0x160
 [   32.464799]  ? rtnl_calcit.isra.0+0x110/0x110
 [   32.464802]  netlink_rcv_skb+0x50/0xf0
 [   32.464806]  netlink_unicast+0x216/0x340
 [   32.464809]  netlink_sendmsg+0x23f/0x480
 [   32.464812]  sock_sendmsg+0x5e/0x60
 [   32.464815]  ____sys_sendmsg+0x22c/0x270
 [   32.464818]  ? import_iovec+0x17/0x20
 [   32.464821]  ? sendmsg_copy_msghdr+0x59/0x90
 [   32.464823]  ? do_set_pte+0xa0/0xe0
 [   32.464828]  ___sys_sendmsg+0x81/0xc0
 [   32.464832]  ? mod_objcg_state+0xc6/0x300
 [   32.464835]  ? refill_obj_stock+0xa9/0x160
 [   32.464838]  ? memcg_slab_free_hook+0x1a5/0x1f0
 [   32.464842]  __sys_sendmsg+0x49/0x80
 [   32.464847]  do_syscall_64+0x3b/0x90
 [   32.464851]  entry_SYSCALL_64_after_hwframe+0x44/0xae
 [   32.464865] RIP: 0033:0x7f297bf2e5e7
 [   32.464868] Code: 64 89 02 48 c7 c0 ff ff ff ff eb bb 0f 1f 80 00 00 00 00 f3 0f 1e fa 64 8b 04 25 18 00 00 00 85 c0 75 10 b8 2e 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 51 c3 48 83 ec 28 89 54 24 1c 48 89 74 24 10
 [   32.464869] RSP: 002b:00007ffd96c824c8 EFLAGS: 00000246 ORIG_RAX: 000000000000002e
 [   32.464872] RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f297bf2e5e7
 [   32.464874] RDX: 0000000000000000 RSI: 00007ffd96c82540 RDI: 0000000000000003
 [   32.464875] RBP: 00000000640f19de R08: 0000000000000001 R09: 000000000000007c
 [   32.464876] R10: 00007f297bffabe0 R11: 0000000000000246 R12: 0000000000000001
 [   32.464877] R13: 00007ffd96c82d20 R14: 00007ffd96c82610 R15: 000055bfe38a7020
 [   32.464881]  </TASK>
 [   32.464882] ---[ end trace 0000000000000000 ]---

Fixes: 7d5cd2ce5292 ("bonding: correctly handle bonding type change on enslave failure")
Reported-by: syzbot+9dfc3f3348729cc82277@syzkaller.appspotmail.com
Link: https://syzkaller.appspot.com/bug?id=391c7b1f6522182899efba27d891f1743e8eb3ef
Signed-off-by: Nikolay Aleksandrov <razor@blackwall.org>
Reviewed-by: Michal Kubiak <michal.kubiak@intel.com>
Acked-by: Jonathan Toppins <jtoppins@redhat.com>
Acked-by: Jay Vosburgh <jay.vosburgh@canonical.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>

bonding: restore IFF_MASTER/SLAVE flags on bond enslave ether type change

[ Upstream commit 9ec7eb60dcbcb6c41076defbc5df7bbd95ceaba5 ]

Add bond_ether_setup helper which is used to fix ether_setup() calls in the
bonding driver. It takes care of both IFF_MASTER and IFF_SLAVE flags, the
former is always restored and the latter only if it was set.
If the bond enslaves non-ARPHRD_ETHER device (changes its type), then
releases it and enslaves ARPHRD_ETHER device (changes back) then we
use ether_setup() to restore the bond device type but it also resets its
flags and removes IFF_MASTER and IFF_SLAVE[1]. Use the bond_ether_setup
helper to restore both after such transition.

[1] reproduce (nlmon is non-ARPHRD_ETHER):
 $ ip l add nlmon0 type nlmon
 $ ip l add bond2 type bond mode active-backup
 $ ip l set nlmon0 master bond2
 $ ip l set nlmon0 nomaster
 $ ip l add bond1 type bond
 (we use bond1 as ARPHRD_ETHER device to restore bond2's mode)
 $ ip l set bond1 master bond2
 $ ip l sh dev bond2
 37: bond2: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000
    link/ether be:d7:c5:40:5b:cc brd ff:ff:ff:ff:ff:ff promiscuity 0 minmtu 68 maxmtu 1500
 (notice bond2's IFF_MASTER is missing)

Fixes: e36b9d16c6a6 ("bonding: clean muticast addresses when device changes type")
Signed-off-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>

ethernet: sun: add check for the mdesc_grab()

[ Upstream commit 90de546d9a0b3c771667af18bb3f80567eabb89b ]

In vnet_port_probe() and vsw_port_probe(), we should
check the return value of mdesc_grab() as it may
return NULL which can caused NPD bugs.

Fixes: 5d01fa0c6bd8 ("ldmvsw: Add ldmvsw.c driver code")
Fixes: 43fdf27470b2 ("[SPARC64]: Abstract out mdesc accesses for better MD update handling.")
Signed-off-by: Liang He <windhl@126.com>
Reviewed-by: Piotr Raczynski <piotr.raczynski@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>

qed/qed_mng_tlv: correctly zero out ->min instead of ->hour

[ Upstream commit 470efd68a4653d9819d391489886432cd31bcd0b ]

This fixes an issue where ->hour would erroneously get zeroed out
instead of ->min because of a bad copy paste.

Found by Linux Verification Center (linuxtesting.org) with the SVACE
static analysis tool.

Fixes: f240b6882211 ("qed: Add support for processing fcoe tlv request.")
Signed-off-by: Daniil Tatianin <d-tatianin@yandex-team.ru>
Link: https://lore.kernel.org/r/20230315194618.579286-1-d-tatianin@yandex-team.ru
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

selftests: net: devlink_port_split.py: skip test if no suitable device available

[ Upstream commit 24994513ad13ff2c47ba91d2b5df82c3d496c370 ]

The `devlink -j port show` command output may not contain the "flavour"
key, an example from Ubuntu 22.10 s390x LPAR(5.19.0-37-generic), with
mlx4 driver and iproute2-5.15.0:
  {"port":{"pci/0001:00:00.0/1":{"type":"eth","netdev":"ens301"},
           "pci/0001:00:00.0/2":{"type":"eth","netdev":"ens301d1"},
           "pci/0002:00:00.0/1":{"type":"eth","netdev":"ens317"},
           "pci/0002:00:00.0/2":{"type":"eth","netdev":"ens317d1"}}}

This will cause a KeyError exception.

Create a validate_devlink_output() to check for this "flavour" from
devlink command output to avoid this KeyError exception. Also let
it handle the check for `devlink -j dev show` output in main().

Apart from this, if the test was not started because the max lanes of
the designated device is 0. The script will still return 0 and thus
causing a false-negative test result.

Use a found_max_lanes flag to determine if these tests were skipped
due to this reason and return KSFT_SKIP to make it more clear.

Link: https://bugs.launchpad.net/bugs/1937133
Fixes: f3348a82e727 ("selftests: net: Add port split test")
Signed-off-by: Po-Hsu Lin <po-hsu.lin@canonical.com>
Link: https://lore.kernel.org/r/20230315165353.229590-1-po-hsu.lin@canonical.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

net/iucv: Fix size of interrupt data

[ Upstream commit 3d87debb8ed2649608ff432699e7c961c0c6f03b ]

iucv_irq_data needs to be 4 bytes larger.
These bytes are not used by the iucv module, but written by
the z/VM hypervisor in case a CPU is deconfigured.

Reported as:
BUG dma-kmalloc-64 (Not tainted): kmalloc Redzone overwritten
-----------------------------------------------------------------------------
0x0000000000400564-0x0000000000400567 @offset=1380. First byte 0x80 instead of 0xcc
Allocated in iucv_cpu_prepare+0x44/0xd0 age=167839 cpu=2 pid=1
__kmem_cache_alloc_node+0x166/0x450
kmalloc_node_trace+0x3a/0x70
iucv_cpu_prepare+0x44/0xd0
cpuhp_invoke_callback+0x156/0x2f0
cpuhp_issue_call+0xf0/0x298
__cpuhp_setup_state_cpuslocked+0x136/0x338
__cpuhp_setup_state+0xf4/0x288
iucv_init+0xf4/0x280
do_one_initcall+0x78/0x390
do_initcalls+0x11a/0x140
kernel_init_freeable+0x25e/0x2a0
kernel_init+0x2e/0x170
__ret_from_fork+0x3c/0x58
ret_from_fork+0xa/0x40
Freed in iucv_init+0x92/0x280 age=167839 cpu=2 pid=1
__kmem_cache_free+0x308/0x358
iucv_init+0x92/0x280
do_one_initcall+0x78/0x390
do_initcalls+0x11a/0x140
kernel_init_freeable+0x25e/0x2a0
kernel_init+0x2e/0x170
__ret_from_fork+0x3c/0x58
ret_from_fork+0xa/0x40
Slab 0x0000037200010000 objects=32 used=30 fp=0x0000000000400640 flags=0x1ffff00000010200(slab|head|node=0|zone=0|
Object 0x0000000000400540 @offset=1344 fp=0x0000000000000000
Redzone  0000000000400500: cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc  ................
Redzone  0000000000400510: cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc  ................
Redzone  0000000000400520: cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc  ................
Redzone  0000000000400530: cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc  ................
Object   0000000000400540: 00 01 00 03 00 00 00 00 00 00 00 00 00 00 00 00  ................
Object   0000000000400550: f3 86 81 f2 f4 82 f8 82 f0 f0 f0 f0 f0 f0 f0 f2  ................
Object   0000000000400560: 00 00 00 00 80 00 00 00 cc cc cc cc cc cc cc cc  ................
Object   0000000000400570: cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc  ................
Redzone  0000000000400580: cc cc cc cc cc cc cc cc                          ........
Padding  00000000004005d4: 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a  ZZZZZZZZZZZZZZZZ
Padding  00000000004005e4: 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a  ZZZZZZZZZZZZZZZZ
Padding  00000000004005f4: 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a              ZZZZZZZZZZZZ
CPU: 6 PID: 121030 Comm: 116-pai-crypto. Not tainted 6.3.0-20230221.rc0.git4.99b8246b2d71.300.fc37.s390x+debug #1
Hardware name: IBM 3931 A01 704 (z/VM 7.3.0)
Call Trace:
[<000000032aa034ec>] dump_stack_lvl+0xac/0x100
[<0000000329f5a6cc>] check_bytes_and_report+0x104/0x140
[<0000000329f5aa78>] check_object+0x370/0x3c0
[<0000000329f5ede6>] free_debug_processing+0x15e/0x348
[<0000000329f5f06a>] free_to_partial_list+0x9a/0x2f0
[<0000000329f5f4a4>] __slab_free+0x1e4/0x3a8
[<0000000329f61768>] __kmem_cache_free+0x308/0x358
[<000000032a91465c>] iucv_cpu_dead+0x6c/0x88
[<0000000329c2fc66>] cpuhp_invoke_callback+0x156/0x2f0
[<000000032aa062da>] _cpu_down.constprop.0+0x22a/0x5e0
[<0000000329c3243e>] cpu_device_down+0x4e/0x78
[<000000032a61dee0>] device_offline+0xc8/0x118
[<000000032a61e048>] online_store+0x60/0xe0
[<000000032a08b6b0>] kernfs_fop_write_iter+0x150/0x1e8
[<0000000329fab65c>] vfs_write+0x174/0x360
[<0000000329fab9fc>] ksys_write+0x74/0x100
[<000000032aa03a5a>] __do_syscall+0x1da/0x208
[<000000032aa177b2>] system_call+0x82/0xb0
INFO: lockdep is turned off.
FIX dma-kmalloc-64: Restoring kmalloc Redzone 0x0000000000400564-0x0000000000400567=0xcc
FIX dma-kmalloc-64: Object at 0x0000000000400540 not freed

Fixes: 2356f4cb1911 ("[S390]: Rewrite of the IUCV base code, part 2")
Signed-off-by: Alexandra Winter <wintera@linux.ibm.com>
Link: https://lore.kernel.org/r/20230315131435.4113889-1-wintera@linux.ibm.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

net: usb: smsc75xx: Move packet length check to prevent kernel panic in skb_pull

[ Upstream commit 43ffe6caccc7a1bb9d7442fbab521efbf6c1378c ]

Packet length check needs to be located after size and align_count
calculation to prevent kernel panic in skb_pull() in case
rx_cmd_a & RX_CMD_A_RED evaluates to true.

Fixes: d8b228318935 ("net: usb: smsc75xx: Limit packet length to skb->len")
Signed-off-by: Szymon Heidrich <szymon.heidrich@gmail.com>
Link: https://lore.kernel.org/r/20230316110540.77531-1-szymon.heidrich@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

ipv4: Fix incorrect table ID in IOCTL path

[ Upstream commit 8a2618e14f81604a9b6ad305d57e0c8da939cd65 ]

Commit f96a3d74554d ("ipv4: Fix incorrect route flushing when source
address is deleted") started to take the table ID field in the FIB info
structure into account when determining if two structures are identical
or not. This field is initialized using the 'fc_table' field in the
route configuration structure, which is not set when adding a route via
IOCTL.

The above can result in user space being able to install two identical
routes that only differ in the table ID field of their associated FIB
info.

Fix by initializing the table ID field in the route configuration
structure in the IOCTL path.

Before the fix:

 # ip route add default via 192.0.2.2
 # route add default gw 192.0.2.2
 # ip -4 r show default
 # default via 192.0.2.2 dev dummy10
 # default via 192.0.2.2 dev dummy10

After the fix:

 # ip route add default via 192.0.2.2
 # route add default gw 192.0.2.2
 SIOCADDRT: File exists
 # ip -4 r show default
 default via 192.0.2.2 dev dummy10

Audited the code paths to ensure there are no other paths that do not
properly initialize the route configuration structure when installing a
route.

Fixes: 5a56a0b3a45d ("net: Don't delete routes in different VRFs")
Fixes: f96a3d74554d ("ipv4: Fix incorrect route flushing when source address is deleted")
Reported-by: gaoxingwang <gaoxingwang1@huawei.com>
Link: https://lore.kernel.org/netdev/20230314144159.2354729-1-gaoxingwang1@huawei.com/
Tested-by: gaoxingwang <gaoxingwang1@huawei.com>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Link: https://lore.kernel.org/r/20230315124009.4015212-1-idosch@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

sh_eth: avoid PHY being resumed when interface is not up

[ Upstream commit c6be7136afb224a01d4cde2983ddebac8da98693 ]

SH_ETH doesn't need mdiobus suspend/resume, that's why it sets
'mac_managed_pm'. However, setting it needs to be moved from init to
probe, so mdiobus PM functions will really never be called (e.g. when
the interface is not up yet during suspend/resume).

Fixes: 6a1dbfefdae4 ("net: sh_eth: Fix PHY state warning splat during system resume")
Suggested-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Reviewed-by: Michal Kubiak <michal.kubiak@intel.com>
Reviewed-by: Sergey Shtylyov <s.shtylyov@omp.ru>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

ravb: avoid PHY being resumed when interface is not up

[ Upstream commit 7f5ebf5dae42e710162f1c481ebcf28ab7b741c7 ]

RAVB doesn't need mdiobus suspend/resume, that's why it sets
'mac_managed_pm'. However, setting it needs to be moved from init to
probe, so mdiobus PM functions will really never be called (e.g. when
the interface is not up yet during suspend/resume).

Fixes: 4924c0cdce75 ("net: ravb: Fix PHY state warning splat during system resume")
Suggested-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Reviewed-by: Michal Kubiak <michal.kubiak@intel.com>
Reviewed-by: Sergey Shtylyov <s.shtylyov@omp.ru>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

net: dsa: mv88e6xxx: fix max_mtu of 1492 on 6165, 6191, 6220, 6250, 6290

[ Upstream commit 7e9517375a14f44ee830ca1c3278076dd65fcc8f ]

There are 3 classes of switch families that the driver is aware of, as
far as mv88e6xxx_change_mtu() is concerned:

- MTU configuration is available per port. Here, the
  chip->info->ops->port_set_jumbo_size() method will be present.

- MTU configuration is global to the switch. Here, the
  chip->info->ops->set_max_frame_size() method will be present.

- We don't know how to change the MTU. Here, none of the above methods
  will be present.

Switch families MV88E6165, MV88E6191, MV88E6220, MV88E6250 and MV88E6290
fall in category 3.

The blamed commit has adjusted the MTU for all 3 categories by EDSA_HLEN
(8 bytes), resulting in a new maximum MTU of 1492 being reported by the
driver for these switches.

I don't have the hardware to test, but I do have a MV88E6390 switch on
which I can simulate this by commenting out its .port_set_jumbo_size
definition from mv88e6390_ops. The result is this set of messages at
probe time:

mv88e6085 d0032004.mdio-mii:10: nonfatal error -34 setting MTU to 1500 on port 1
mv88e6085 d0032004.mdio-mii:10: nonfatal error -34 setting MTU to 1500 on port 2
mv88e6085 d0032004.mdio-mii:10: nonfatal error -34 setting MTU to 1500 on port 3
mv88e6085 d0032004.mdio-mii:10: nonfatal error -34 setting MTU to 1500 on port 4
mv88e6085 d0032004.mdio-mii:10: nonfatal error -34 setting MTU to 1500 on port 5
mv88e6085 d0032004.mdio-mii:10: nonfatal error -34 setting MTU to 1500 on port 6
mv88e6085 d0032004.mdio-mii:10: nonfatal error -34 setting MTU to 1500 on port 7
mv88e6085 d0032004.mdio-mii:10: nonfatal error -34 setting MTU to 1500 on port 8

It is highly implausible that there exist Ethernet switches which don't
support the standard MTU of 1500 octets, and this is what the DSA
framework says as well - the error comes from dsa_slave_create() ->
dsa_slave_change_mtu(slave_dev, ETH_DATA_LEN).

But the error messages are alarming, and it would be good to suppress
them.

As a consequence of this unlikeliness, we reimplement mv88e6xxx_get_max_mtu()
and mv88e6xxx_change_mtu() on switches from the 3rd category as follows:
the maximum supported MTU is 1500, and any request to set the MTU to a
value larger than that fails in dev_validate_mtu().

Fixes: b9c587fed61c ("dsa: mv88e6xxx: Include tagger overhead when setting MTU for DSA and CPU ports")
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>

ice: xsk: disable txq irq before flushing hw

[ Upstream commit b830c9642386867863ac64295185f896ff2928ac ]

ice_qp_dis() intends to stop a given queue pair that is a target of xsk
pool attach/detach. One of the steps is to disable interrupts on these
queues. It currently is broken in a way that txq irq is turned off
*after* HW flush which in turn takes no effect.

ice_qp_dis():
-> ice_qvec_dis_irq()
--> disable rxq irq
--> flush hw
-> ice_vsi_stop_tx_ring()
-->disable txq irq

Below splat can be triggered by following steps:
- start xdpsock WITHOUT loading xdp prog
- run xdp_rxq_info with XDP_TX action on this interface
- start traffic
- terminate xdpsock

[  256.312485] BUG: kernel NULL pointer dereference, address: 0000000000000018
[  256.319560] #PF: supervisor read access in kernel mode
[  256.324775] #PF: error_code(0x0000) - not-present page
[  256.329994] PGD 0 P4D 0
[  256.332574] Oops: 0000 [#1] PREEMPT SMP NOPTI
[  256.337006] CPU: 3 PID: 32 Comm: ksoftirqd/3 Tainted: G           OE      6.2.0-rc5+ #51
[  256.345218] Hardware name: Intel Corporation S2600WFT/S2600WFT, BIOS SE5C620.86B.02.01.0008.031920191559 03/19/2019
[  256.355807] RIP: 0010:ice_clean_rx_irq_zc+0x9c/0x7d0 [ice]
[  256.361423] Code: b7 8f 8a 00 00 00 66 39 ca 0f 84 f1 04 00 00 49 8b 47 40 4c 8b 24 d0 41 0f b7 45 04 66 25 ff 3f 66 89 04 24 0f 84 85 02 00 00 <49> 8b 44 24 18 0f b7 14 24 48 05 00 01 00 00 49 89 04 24 49 89 44
[  256.380463] RSP: 0018:ffffc900088bfd20 EFLAGS: 00010206
[  256.385765] RAX: 000000000000003c RBX: 0000000000000035 RCX: 000000000000067f
[  256.393012] RDX: 0000000000000775 RSI: 0000000000000000 RDI: ffff8881deb3ac80
[  256.400256] RBP: 000000000000003c R08: ffff889847982710 R09: 0000000000010000
[  256.407500] R10: ffffffff82c060c0 R11: 0000000000000004 R12: 0000000000000000
[  256.414746] R13: ffff88811165eea0 R14: ffffc9000d255000 R15: ffff888119b37600
[  256.421990] FS:  0000000000000000(0000) GS:ffff8897e0cc0000(0000) knlGS:0000000000000000
[  256.430207] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  256.436036] CR2: 0000000000000018 CR3: 0000000005c0a006 CR4: 00000000007706e0
[  256.443283] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[  256.450527] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[  256.457770] PKRU: 55555554
[  256.460529] Call Trace:
[  256.463015]  <TASK>
[  256.465157]  ? ice_xmit_zc+0x6e/0x150 [ice]
[  256.469437]  ice_napi_poll+0x46d/0x680 [ice]
[  256.473815]  ? _raw_spin_unlock_irqrestore+0x1b/0x40
[  256.478863]  __napi_poll+0x29/0x160
[  256.482409]  net_rx_action+0x136/0x260
[  256.486222]  __do_softirq+0xe8/0x2e5
[  256.489853]  ? smpboot_thread_fn+0x2c/0x270
[  256.494108]  run_ksoftirqd+0x2a/0x50
[  256.497747]  smpboot_thread_fn+0x1c1/0x270
[  256.501907]  ? __pfx_smpboot_thread_fn+0x10/0x10
[  256.506594]  kthread+0xea/0x120
[  256.509785]  ? __pfx_kthread+0x10/0x10
[  256.513597]  ret_from_fork+0x29/0x50
[  256.517238]  </TASK>

In fact, irqs were not disabled and napi managed to be scheduled and run
while xsk_pool pointer was still valid, but SW ring of xdp_buff pointers
was already freed.

To fix this, call ice_qvec_dis_irq() after ice_vsi_stop_tx_ring(). Also
while at it, remove redundant ice_clean_rx_ring() call - this is handled
in ice_qp_clean_rings().

Fixes: 2d4238f55697 ("ice: Add support for AF_XDP")
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Reviewed-by: Larysa Zaremba <larysa.zaremba@intel.com>
Tested-by: Chandan Kumar Rout <chandanx.rout@intel.com> (A Contingent Worker at Intel)
Acked-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>

block: sunvdc: add check for mdesc_grab() returning NULL

[ Upstream commit 6030363199e3a6341afb467ddddbed56640cbf6a ]

In vdc_port_probe(), we should check the return value of mdesc_grab() as
it may return NULL, which can cause potential NPD bug.

Fixes: 43fdf27470b2 ("[SPARC64]: Abstract out mdesc accesses for better MD update handling.")
Signed-off-by: Liang He <windhl@126.com>
Link: https://lore.kernel.org/r/20230315062032.1741692-1-windhl@126.com
[axboe: style cleanup]
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>

nvmet: avoid potential UAF in nvmet_req_complete()

[ Upstream commit 6173a77b7e9d3e202bdb9897b23f2a8afe7bf286 ]

An nvme target ->queue_response() operation implementation may free the
request passed as argument. Such implementation potentially could result
in a use after free of the request pointer when percpu_ref_put() is
called in nvmet_req_complete().

Avoid such problem by using a local variable to save the sq pointer
before calling __nvmet_req_complete(), thus avoiding dereferencing the
req pointer after that function call.

Fixes: a07b4970f464 ("nvmet: add a generic NVMe target")
Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Sasha Levin <sashal@kernel.org>

nvme: fix handling single range discard request

[ Upstream commit 37f0dc2ec78af0c3f35dd05578763de059f6fe77 ]

When investigating one customer report on warning in nvme_setup_discard,
we observed the controller(nvme/tcp) actually exposes
queue_max_discard_segments(req->q) == 1.

Obviously the current code can't handle this situation, since contiguity
merge like normal RW request is taken.

Fix the issue by building range from request sector/nr_sectors directly.

Fixes: b35ba01ea697 ("nvme: support ranged discard requests")
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Sasha Levin <sashal@kernel.org>

block: null_blk: Fix handling of fake timeout request

[ Upstream commit 63f886597085f346276e3b3c8974de0100d65f32 ]

When injecting a fake timeout into the null_blk driver using
fail_io_timeout, the request timeout handler does not execute
blk_mq_complete_request(), so the complete callback is never executed
for a timedout request.

The null_blk driver also has a driver-specific fake timeout mechanism
which does not have this problem. Fix the problem with fail_io_timeout
by using the same meachanism as null_blk internal timeout feature, using
the fake_timeout field of null_blk commands.

Reported-by: Akinobu Mita <akinobu.mita@gmail.com>
Fixes: de3510e52b0a ("null_blk: fix command timeout completion handling")
Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Link: https://lore.kernel.org/r/20230314041106.19173-2-damien.lemoal@opensource.wdc.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>

drm/bridge: Fix returned array size name for atomic_get_input_bus_fmts kdoc

[ Upstream commit 0d3c9333d976af41d7dbc6bf4d9d2e95fbdf9c89 ]

The returned array size for input formats is set through
atomic_get_input_bus_fmts()'s 'num_input_fmts' argument, so use
'num_input_fmts' to represent the array size in the function's kdoc,
not 'num_output_fmts'.

Fixes: 91ea83306bfa ("drm/bridge: Fix the bridge kernel doc")
Fixes: f32df58acc68 ("drm/bridge: Add the necessary bits to support bus format negotiation")
Signed-off-by: Liu Ying <victor.liu@nxp.com>
Reviewed-by: Robert Foss <rfoss@kernel.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230314055035.3731179-1-victor.liu@nxp.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

net: usb: smsc75xx: Limit packet length to skb->len

[ Upstream commit d8b228318935044dafe3a5bc07ee71a1f1424b8d ]

Packet length retrieved from skb data may be larger than
the actual socket buffer length (up to 9026 bytes). In such
case the cloned skb passed up the network stack will leak
kernel memory contents.

Fixes: d0cad871703b ("smsc75xx: SMSC LAN75xx USB gigabit ethernet adapter driver")
Signed-off-by: Szymon Heidrich <szymon.heidrich@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>

net/smc: fix deadlock triggered by cancel_delayed_work_syn()

[ Upstream commit 13085e1b5cab8ad802904d72e6a6dae85ae0cd20 ]

The following LOCKDEP was detected:
Workqueue: events smc_lgr_free_work [smc]
WARNING: possible circular locking dependency detected
6.1.0-20221027.rc2.git8.56bc5b569087.300.fc36.s390x+debug #1 Not tainted
------------------------------------------------------
kworker/3:0/176251 is trying to acquire lock:
00000000f1467148 ((wq_completion)smc_tx_wq-00000000#2){+.+.}-{0:0},
at: __flush_workqueue+0x7a/0x4f0
but task is already holding lock:
0000037fffe97dc8 ((work_completion)(&(&lgr->free_work)->work)){+.+.}-{0:0},
at: process_one_work+0x232/0x730
which lock already depends on the new lock.
the existing dependency chain (in reverse order) is:
-> #4 ((work_completion)(&(&lgr->free_work)->work)){+.+.}-{0:0}:
       __lock_acquire+0x58e/0xbd8
       lock_acquire.part.0+0xe2/0x248
       lock_acquire+0xac/0x1c8
       __flush_work+0x76/0xf0
       __cancel_work_timer+0x170/0x220
       __smc_lgr_terminate.part.0+0x34/0x1c0 [smc]
       smc_connect_rdma+0x15e/0x418 [smc]
       __smc_connect+0x234/0x480 [smc]
       smc_connect+0x1d6/0x230 [smc]
       __sys_connect+0x90/0xc0
       __do_sys_socketcall+0x186/0x370
       __do_syscall+0x1da/0x208
       system_call+0x82/0xb0
-> #3 (smc_client_lgr_pending){+.+.}-{3:3}:
       __lock_acquire+0x58e/0xbd8
       lock_acquire.part.0+0xe2/0x248
       lock_acquire+0xac/0x1c8
       __mutex_lock+0x96/0x8e8
       mutex_lock_nested+0x32/0x40
       smc_connect_rdma+0xa4/0x418 [smc]
       __smc_connect+0x234/0x480 [smc]
       smc_connect+0x1d6/0x230 [smc]
       __sys_connect+0x90/0xc0
       __do_sys_socketcall+0x186/0x370
       __do_syscall+0x1da/0x208
       system_call+0x82/0xb0
-> #2 (sk_lock-AF_SMC){+.+.}-{0:0}:
       __lock_acquire+0x58e/0xbd8
       lock_acquire.part.0+0xe2/0x248
       lock_acquire+0xac/0x1c8
       lock_sock_nested+0x46/0xa8
       smc_tx_work+0x34/0x50 [smc]
       process_one_work+0x30c/0x730
       worker_thread+0x62/0x420
       kthread+0x138/0x150
       __ret_from_fork+0x3c/0x58
       ret_from_fork+0xa/0x40
-> #1 ((work_completion)(&(&smc->conn.tx_work)->work)){+.+.}-{0:0}:
       __lock_acquire+0x58e/0xbd8
       lock_acquire.part.0+0xe2/0x248
       lock_acquire+0xac/0x1c8
       process_one_work+0x2bc/0x730
       worker_thread+0x62/0x420
       kthread+0x138/0x150
       __ret_from_fork+0x3c/0x58
       ret_from_fork+0xa/0x40
-> #0 ((wq_completion)smc_tx_wq-00000000#2){+.+.}-{0:0}:
       check_prev_add+0xd8/0xe88
       validate_chain+0x70c/0xb20
       __lock_acquire+0x58e/0xbd8
       lock_acquire.part.0+0xe2/0x248
       lock_acquire+0xac/0x1c8
       __flush_workqueue+0xaa/0x4f0
       drain_workqueue+0xaa/0x158
       destroy_workqueue+0x44/0x2d8
       smc_lgr_free+0x9e/0xf8 [smc]
       process_one_work+0x30c/0x730
       worker_thread+0x62/0x420
       kthread+0x138/0x150
       __ret_from_fork+0x3c/0x58
       ret_from_fork+0xa/0x40
other info that might help us debug this:
Chain exists of:
  (wq_completion)smc_tx_wq-00000000#2
     --> smc_client_lgr_pending
  --> (work_completion)(&(&lgr->free_work)->work)
 Possible unsafe locking scenario:
       CPU0                    CPU1
       ----                    ----
  lock((work_completion)(&(&lgr->free_work)->work));
                   lock(smc_client_lgr_pending);
                   lock((work_completion)
(&(&lgr->free_work)->work));
  lock((wq_completion)smc_tx_wq-00000000#2);
 *** DEADLOCK ***
2 locks held by kworker/3:0/176251:
 #0: 0000000080183548
((wq_completion)events){+.+.}-{0:0},
at: process_one_work+0x232/0x730
 #1: 0000037fffe97dc8
((work_completion)
 (&(&lgr->free_work)->work)){+.+.}-{0:0},
at: process_one_work+0x232/0x730
stack backtrace:
CPU: 3 PID: 176251 Comm: kworker/3:0 Not tainted
Hardware name: IBM 8561 T01 701 (z/VM 7.2.0)
Call Trace:
 [<000000002983c3e4>] dump_stack_lvl+0xac/0x100
 [<0000000028b477ae>] check_noncircular+0x13e/0x160
 [<0000000028b48808>] check_prev_add+0xd8/0xe88
 [<0000000028b49cc4>] validate_chain+0x70c/0xb20
 [<0000000028b4bd26>] __lock_acquire+0x58e/0xbd8
 [<0000000028b4cf6a>] lock_acquire.part.0+0xe2/0x248
 [<0000000028b4d17c>] lock_acquire+0xac/0x1c8
 [<0000000028addaaa>] __flush_workqueue+0xaa/0x4f0
 [<0000000028addf9a>] drain_workqueue+0xaa/0x158
 [<0000000028ae303c>] destroy_workqueue+0x44/0x2d8
 [<000003ff8029af26>] smc_lgr_free+0x9e/0xf8 [smc]
 [<0000000028adf3d4>] process_one_work+0x30c/0x730
 [<0000000028adf85a>] worker_thread+0x62/0x420
 [<0000000028aeac50>] kthread+0x138/0x150
 [<0000000028a63914>] __ret_from_fork+0x3c/0x58
 [<00000000298503da>] ret_from_fork+0xa/0x40
INFO: lockdep is turned off.
===================================================================

This deadlock occurs because cancel_delayed_work_sync() waits for
the work(&lgr->free_work) to finish, while the &lgr->free_work
waits for the work(lgr->tx_wq), which needs the sk_lock-AF_SMC, that
is already used under the mutex_lock.

The solution is to use cancel_delayed_work() instead, which kills
off a pending work.

Fixes: a52bcc919b14 ("net/smc: improve termination processing")
Signed-off-by: Wenjia Zhang <wenjia@linux.ibm.com>
Reviewed-by: Jan Karcher <jaka@linux.ibm.com>
Reviewed-by: Karsten Graul <kgraul@linux.ibm.com>
Reviewed-by: Tony Lu <tonylu@linux.alibaba.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>

nfc: st-nci: Fix use after free bug in ndlc_remove due to race condition

[ Upstream commit 5000fe6c27827a61d8250a7e4a1d26c3298ef4f6 ]

This bug influences both st_nci_i2c_remove and st_nci_spi_remove.
Take st_nci_i2c_remove as an example.

In st_nci_i2c_probe, it called ndlc_probe and bound &ndlc->sm_work
with llt_ndlc_sm_work.

When it calls ndlc_recv or timeout handler, it will finally call
schedule_work to start the work.

When we call st_nci_i2c_remove to remove the driver, there
may be a sequence as follows:

Fix it by finishing the work before cleanup in ndlc_remove

CPU0                  CPU1

                    |llt_ndlc_sm_work
st_nci_i2c_remove   |
  ndlc_remove       |
     st_nci_remove  |
     nci_free_device|
     kfree(ndev)    |
//free ndlc->ndev   |
                    |llt_ndlc_rcv_queue
                    |nci_recv_frame
                    |//use ndlc->ndev

Fixes: 35630df68d60 ("NFC: st21nfcb: Add driver for STMicroelectronics ST21NFCB NFC chip")
Signed-off-by: Zheng Wang <zyytlz.wz@163.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/r/20230312160837.2040857-1-zyytlz.wz@163.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

net: phy: smsc: bail out in lan87xx_read_status if genphy_read_status fails

[ Upstream commit c22c3bbf351e4ce905f082649cffa1ff893ea8c1 ]

If genphy_read_status fails then further access to the PHY may result
in unpredictable behavior. To prevent this bail out immediately if
genphy_read_status fails.

Fixes: 4223dbffed9f ("net: phy: smsc: Re-enable EDPD mode for LAN87xx")
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Link: https://lore.kernel.org/r/026aa4f2-36f5-1c10-ab9f-cdb17dda6ac4@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

net: tunnels: annotate lockless accesses to dev->needed_headroom

[ Upstream commit 4b397c06cb987935b1b097336532aa6b4210e091 ]

IP tunnels can apparently update dev->needed_headroom
in their xmit path.

This patch takes care of three tunnels xmit, and also the
core LL_RESERVED_SPACE() and LL_RESERVED_SPACE_EXTRA()
helpers.

More changes might be needed for completeness.

BUG: KCSAN: data-race in ip_tunnel_xmit / ip_tunnel_xmit

read to 0xffff88815b9da0ec of 2 bytes by task 888 on cpu 1:
ip_tunnel_xmit+0x1270/0x1730 net/ipv4/ip_tunnel.c:803
__gre_xmit net/ipv4/ip_gre.c:469 [inline]
ipgre_xmit+0x516/0x570 net/ipv4/ip_gre.c:661
__netdev_start_xmit include/linux/netdevice.h:4881 [inline]
netdev_start_xmit include/linux/netdevice.h:4895 [inline]
xmit_one net/core/dev.c:3580 [inline]
dev_hard_start_xmit+0x127/0x400 net/core/dev.c:3596
__dev_queue_xmit+0x1007/0x1eb0 net/core/dev.c:4246
dev_queue_xmit include/linux/netdevice.h:3051 [inline]
neigh_direct_output+0x17/0x20 net/core/neighbour.c:1623
neigh_output include/net/neighbour.h:546 [inline]
ip_finish_output2+0x740/0x840 net/ipv4/ip_output.c:228
ip_finish_output+0xf4/0x240 net/ipv4/ip_output.c:316
NF_HOOK_COND include/linux/netfilter.h:291 [inline]
ip_output+0xe5/0x1b0 net/ipv4/ip_output.c:430
dst_output include/net/dst.h:444 [inline]
ip_local_out+0x64/0x80 net/ipv4/ip_output.c:126
iptunnel_xmit+0x34a/0x4b0 net/ipv4/ip_tunnel_core.c:82
ip_tunnel_xmit+0x1451/0x1730 net/ipv4/ip_tunnel.c:813
__gre_xmit net/ipv4/ip_gre.c:469 [inline]
ipgre_xmit+0x516/0x570 net/ipv4/ip_gre.c:661
__netdev_start_xmit include/linux/netdevice.h:4881 [inline]
netdev_start_xmit include/linux/netdevice.h:4895 [inline]
xmit_one net/core/dev.c:3580 [inline]
dev_hard_start_xmit+0x127/0x400 net/core/dev.c:3596
__dev_queue_xmit+0x1007/0x1eb0 net/core/dev.c:4246
dev_queue_xmit include/linux/netdevice.h:3051 [inline]
neigh_direct_output+0x17/0x20 net/core/neighbour.c:1623
neigh_output include/net/neighbour.h:546 [inline]
ip_finish_output2+0x740/0x840 net/ipv4/ip_output.c:228
ip_finish_output+0xf4/0x240 net/ipv4/ip_output.c:316
NF_HOOK_COND include/linux/netfilter.h:291 [inline]
ip_output+0xe5/0x1b0 net/ipv4/ip_output.c:430
dst_output include/net/dst.h:444 [inline]
ip_local_out+0x64/0x80 net/ipv4/ip_output.c:126
iptunnel_xmit+0x34a/0x4b0 net/ipv4/ip_tunnel_core.c:82
ip_tunnel_xmit+0x1451/0x1730 net/ipv4/ip_tunnel.c:813
__gre_xmit net/ipv4/ip_gre.c:469 [inline]
ipgre_xmit+0x516/0x570 net/ipv4/ip_gre.c:661
__netdev_start_xmit include/linux/netdevice.h:4881 [inline]
netdev_start_xmit include/linux/netdevice.h:4895 [inline]
xmit_one net/core/dev.c:3580 [inline]
dev_hard_start_xmit+0x127/0x400 net/core/dev.c:3596
__dev_queue_xmit+0x1007/0x1eb0 net/core/dev.c:4246
dev_queue_xmit include/linux/netdevice.h:3051 [inline]
neigh_direct_output+0x17/0x20 net/core/neighbour.c:1623
neigh_output include/net/neighbour.h:546 [inline]
ip_finish_output2+0x740/0x840 net/ipv4/ip_output.c:228
ip_finish_output+0xf4/0x240 net/ipv4/ip_output.c:316
NF_HOOK_COND include/linux/netfilter.h:291 [inline]
ip_output+0xe5/0x1b0 net/ipv4/ip_output.c:430
dst_output include/net/dst.h:444 [inline]
ip_local_out+0x64/0x80 net/ipv4/ip_output.c:126
iptunnel_xmit+0x34a/0x4b0 net/ipv4/ip_tunnel_core.c:82
ip_tunnel_xmit+0x1451/0x1730 net/ipv4/ip_tunnel.c:813
__gre_xmit net/ipv4/ip_gre.c:469 [inline]
ipgre_xmit+0x516/0x570 net/ipv4/ip_gre.c:661
__netdev_start_xmit include/linux/netdevice.h:4881 [inline]
netdev_start_xmit include/linux/netdevice.h:4895 [inline]
xmit_one net/core/dev.c:3580 [inline]
dev_hard_start_xmit+0x127/0x400 net/core/dev.c:3596
__dev_queue_xmit+0x1007/0x1eb0 net/core/dev.c:4246
dev_queue_xmit include/linux/netdevice.h:3051 [inline]
neigh_direct_output+0x17/0x20 net/core/neighbour.c:1623
neigh_output include/net/neighbour.h:546 [inline]
ip_finish_output2+0x740/0x840 net/ipv4/ip_output.c:228
ip_finish_output+0xf4/0x240 net/ipv4/ip_output.c:316
NF_HOOK_COND include/linux/netfilter.h:291 [inline]
ip_output+0xe5/0x1b0 net/ipv4/ip_output.c:430
dst_output include/net/dst.h:444 [inline]
ip_local_out+0x64/0x80 net/ipv4/ip_output.c:126
iptunnel_xmit+0x34a/0x4b0 net/ipv4/ip_tunnel_core.c:82
ip_tunnel_xmit+0x1451/0x1730 net/ipv4/ip_tunnel.c:813
__gre_xmit net/ipv4/ip_gre.c:469 [inline]
ipgre_xmit+0x516/0x570 net/ipv4/ip_gre.c:661
__netdev_start_xmit include/linux/netdevice.h:4881 [inline]
netdev_start_xmit include/linux/netdevice.h:4895 [inline]
xmit_one net/core/dev.c:3580 [inline]
dev_hard_start_xmit+0x127/0x400 net/core/dev.c:3596
__dev_queue_xmit+0x1007/0x1eb0 net/core/dev.c:4246
dev_queue_xmit include/linux/netdevice.h:3051 [inline]
neigh_direct_output+0x17/0x20 net/core/neighbour.c:1623
neigh_output include/net/neighbour.h:546 [inline]
ip_finish_output2+0x740/0x840 net/ipv4/ip_output.c:228
ip_finish_output+0xf4/0x240 net/ipv4/ip_output.c:316
NF_HOOK_COND include/linux/netfilter.h:291 [inline]
ip_output+0xe5/0x1b0 net/ipv4/ip_output.c:430
dst_output include/net/dst.h:444 [inline]
ip_local_out+0x64/0x80 net/ipv4/ip_output.c:126
iptunnel_xmit+0x34a/0x4b0 net/ipv4/ip_tunnel_core.c:82
ip_tunnel_xmit+0x1451/0x1730 net/ipv4/ip_tunnel.c:813
__gre_xmit net/ipv4/ip_gre.c:469 [inline]
ipgre_xmit+0x516/0x570 net/ipv4/ip_gre.c:661
__netdev_start_xmit include/linux/netdevice.h:4881 [inline]
netdev_start_xmit include/linux/netdevice.h:4895 [inline]
xmit_one net/core/dev.c:3580 [inline]
dev_hard_start_xmit+0x127/0x400 net/core/dev.c:3596
__dev_queue_xmit+0x1007/0x1eb0 net/core/dev.c:4246
dev_queue_xmit include/linux/netdevice.h:3051 [inline]
neigh_direct_output+0x17/0x20 net/core/neighbour.c:1623
neigh_output include/net/neighbour.h:546 [inline]
ip_finish_output2+0x740/0x840 net/ipv4/ip_output.c:228
ip_finish_output+0xf4/0x240 net/ipv4/ip_output.c:316
NF_HOOK_COND include/linux/netfilter.h:291 [inline]
ip_output+0xe5/0x1b0 net/ipv4/ip_output.c:430
dst_output include/net/dst.h:444 [inline]
ip_local_out+0x64/0x80 net/ipv4/ip_output.c:126
iptunnel_xmit+0x34a/0x4b0 net/ipv4/ip_tunnel_core.c:82
ip_tunnel_xmit+0x1451/0x1730 net/ipv4/ip_tunnel.c:813
__gre_xmit net/ipv4/ip_gre.c:469 [inline]
ipgre_xmit+0x516/0x570 net/ipv4/ip_gre.c:661
__netdev_start_xmit include/linux/netdevice.h:4881 [inline]
netdev_start_xmit include/linux/netdevice.h:4895 [inline]
xmit_one net/core/dev.c:3580 [inline]
dev_hard_start_xmit+0x127/0x400 net/core/dev.c:3596
__dev_queue_xmit+0x1007/0x1eb0 net/core/dev.c:4246

write to 0xffff88815b9da0ec of 2 bytes by task 2379 on cpu 0:
ip_tunnel_xmit+0x1294/0x1730 net/ipv4/ip_tunnel.c:804
__gre_xmit net/ipv4/ip_gre.c:469 [inline]
ipgre_xmit+0x516/0x570 net/ipv4/ip_gre.c:661
__netdev_start_xmit include/linux/netdevice.h:4881 [inline]
netdev_start_xmit include/linux/netdevice.h:4895 [inline]
xmit_one net/core/dev.c:3580 [inline]
dev_hard_start_xmit+0x127/0x400 net/core/dev.c:3596
__dev_queue_xmit+0x1007/0x1eb0 net/core/dev.c:4246
dev_queue_xmit include/linux/netdevice.h:3051 [inline]
neigh_direct_output+0x17/0x20 net/core/neighbour.c:1623
neigh_output include/net/neighbour.h:546 [inline]
ip6_finish_output2+0x9bc/0xc50 net/ipv6/ip6_output.c:134
__ip6_finish_output net/ipv6/ip6_output.c:195 [inline]
ip6_finish_output+0x39a/0x4e0 net/ipv6/ip6_output.c:206
NF_HOOK_COND include/linux/netfilter.h:291 [inline]
ip6_output+0xeb/0x220 net/ipv6/ip6_output.c:227
dst_output include/net/dst.h:444 [inline]
NF_HOOK include/linux/netfilter.h:302 [inline]
mld_sendpack+0x438/0x6a0 net/ipv6/mcast.c:1820
mld_send_cr net/ipv6/mcast.c:2121 [inline]
mld_ifc_work+0x519/0x7b0 net/ipv6/mcast.c:2653
process_one_work+0x3e6/0x750 kernel/workqueue.c:2390
worker_thread+0x5f2/0xa10 kernel/workqueue.c:2537
kthread+0x1ac/0x1e0 kernel/kthread.c:376
ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:308

value changed: 0x0dd4 -> 0x0e14

Reported by Kernel Concurrency Sanitizer on:
CPU: 0 PID: 2379 Comm: kworker/0:0 Not tainted 6.3.0-rc1-syzkaller-00002-g8ca09d5fa354-dirty #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 03/02/2023
Workqueue: mld mld_ifc_work

Fixes: 8eb30be0352d ("ipv6: Create ip6_tnl_xmit")
Reported-by: syzbot <syzkaller@googlegroups.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20230310191109.2384387-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

loop: Fix use-after-free issues

[ Upstream commit 9b0cb770f5d7b1ff40bea7ca385438ee94570eec ]

do_req_filebacked() calls blk_mq_complete_request() synchronously or
asynchronously when using asynchronous I/O unless memory allocation fails.
Hence, modify loop_handle_cmd() such that it does not dereference 'cmd' nor
'rq' after do_req_filebacked() finished unless we are sure that the request
has not yet been completed. This patch fixes the following kernel crash:

Unable to handle kernel NULL pointer dereference at virtual address 0000000000000054
Call trace:
 css_put.42938+0x1c/0x1ac
 loop_process_work+0xc8c/0xfd4
 loop_rootcg_workfn+0x24/0x34
 process_one_work+0x244/0x558
 worker_thread+0x400/0x8fc
 kthread+0x16c/0x1e0
 ret_from_fork+0x10/0x20

Cc: Christoph Hellwig <hch@lst.de>
Cc: Ming Lei <ming.lei@redhat.com>
Cc: Jan Kara <jack@suse.cz>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Dan Schatzberg <schatzberg.dan@gmail.com>
Fixes: c74d40e8b5e2 ("loop: charge i/o to mem and blk cg")
Fixes: bc07c10a3603 ("block: loop: support DIO & AIO")
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Link: https://lore.kernel.org/r/20230314182155.80625-1-bvanassche@acm.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>

net: dsa: mt7530: set PLL frequency and trgmii only when trgmii is used

[ Upstream commit 0b086d76e7b011772b0ac214c6e5fd5816eff2df ]

As my testing on the MCM MT7530 switch on MT7621 SoC shows, setting the PLL
frequency does not affect MII modes other than trgmii on port 5 and port 6.
So the assumption is that the operation here called "setting the PLL
frequency" actually sets the frequency of the TRGMII TX clock.

Make it so that it and the rest of the trgmii setup run only when the
trgmii mode is used.

Tested rgmii and trgmii modes of port 6 on MCM MT7530 on MT7621AT Unielec
U7621-06 and standalone MT7530 on MT7623NI Bananapi BPI-R2.

Fixes: b8f126a8d543 ("net-next: dsa: add dsa support for Mediatek MT7530 switch")
Tested-by: Arınç ÜNAL <arinc.unal@arinc9.com>
Signed-off-by: Arınç ÜNAL <arinc.unal@arinc9.com>
Link: https://lore.kernel.org/r/20230310073338.5836-2-arinc.unal@arinc9.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

net: dsa: mt7530: remove now incorrect comment regarding port 5

[ Upstream commit feb03fd11c5616f3a47e4714d2f9917d0f1a2edd ]

Remove now incorrect comment regarding port 5 as GMAC5. This is supposed to
be supported since commit 38f790a80560 ("net: dsa: mt7530: Add support for
port 5") under mt7530_setup_port5().

Fixes: 38f790a80560 ("net: dsa: mt7530: Add support for port 5")
Signed-off-by: Arınç ÜNAL <arinc.unal@arinc9.com>
Link: https://lore.kernel.org/r/20230310073338.5836-1-arinc.unal@arinc9.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

qed/qed_dev: guard against a possible division by zero

[ Upstream commit 1a9dc5610ef89d807acdcfbff93a558f341a44da ]

Previously we would divide total_left_rate by zero if num_vports
happened to be 1 because non_requested_count is calculated as
num_vports - req_count. Guard against this by validating num_vports at
the beginning and returning an error otherwise.

Found by Linux Verification Center (linuxtesting.org) with the SVACE
static analysis tool.

Fixes: bcd197c81f63 ("qed: Add vport WFQ configuration APIs")
Signed-off-by: Daniil Tatianin <d-tatianin@yandex-team.ru>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Link: https://lore.kernel.org/r/20230309201556.191392-1-d-tatianin@yandex-team.ru
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

net/smc: fix NULL sndbuf_desc in smc_cdc_tx_handler()

[ Upstream commit 22a825c541d775c1dbe7b2402786025acad6727b ]

When performing a stress test on SMC-R by rmmod mlx5_ib driver
during the wrk/nginx test, we found that there is a probability
of triggering a panic while terminating all link groups.

This issue dues to the race between smc_smcr_terminate_all()
and smc_buf_create().

smc_smcr_terminate_all

smc_buf_create
/* init */
conn->sndbuf_desc = NULL;
...

__smc_lgr_terminate
smc_conn_kill
smc_close_abort
smc_cdc_get_slot_and_msg_send

__softirqentry_text_start
smc_wr_tx_process_cqe
smc_cdc_tx_handler
READ(conn->sndbuf_desc->len);
/* panic dues to NULL sndbuf_desc */

conn->sndbuf_desc = xxx;

This patch tries to fix the issue by always to check the sndbuf_desc
before send any cdc msg, to make sure that no null pointer is
seen during cqe processing.

Fixes: 0b29ec643613 ("net/smc: immediate termination for SMCR link groups")
Signed-off-by: D. Wythe <alibuda@linux.alibaba.com>
Reviewed-by: Tony Lu <tonylu@linux.alibaba.com>
Reviewed-by: Wenjia Zhang <wenjia@linux.ibm.com>
Link: https://lore.kernel.org/r/1678263432-17329-1-git-send-email-alibuda@linux.alibaba.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

drm/i915/psr: Use calculated io and fast wake lines

[ Upstream commit 71c602103c74b277bef3d20a308874a33ec8326d ]

Currently we are using hardcoded 7 for io and fast wake lines.

According to Bspec io and fast wake times are both 42us for
DISPLAY_VER >= 12 and 50us and 32us for older platforms.

Calculate line counts for these and configure them into PSR2_CTL
accordingly

Use 45 us for the fast wake calculation as 42 seems to be too
tight based on testing.

Bspec: 49274, 4289

Cc: Mika Kahola <mika.kahola@intel.com>
Cc: José Roberto de Souza <jose.souza@intel.com>
Fixes: 64cf40a125ff ("drm/i915/psr: Program default IO buffer Wake and Fast Wake")
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/7725
Signed-off-by: Jouni Högander <jouni.hogander@intel.com>
Reviewed-by: Stanislav Lisovskiy <stanislav.lisovskiy@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230221085304.3382297-1-jouni.hogander@intel.com
(cherry picked from commit cb42e8ede5b475c096e473b86c356b1158b4bc3b)
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

drm/i915/display: clean up comments

[ Upstream commit 3461b040a90d723c93c9d1c7c11e3464f5cadc0e ]

spelling changes
resoluition -> resolution
dont        -> don't
commmit     -> commit
Invalidade  -> Invalidate

Signed-off-by: Tom Rix <trix@redhat.com>
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220701203236.1871668-1-trix@redhat.com
Stable-dep-of: 71c602103c74 ("drm/i915/psr: Use calculated io and fast wake lines")
Signed-off-by: Sasha Levin <sashal@kernel.org>

drm/i915/display/psr: Handle plane and pipe restrictions at every page flip

[ Upstream commit ac220f5f754b1d2f4a69428f515c3f1b10d1fad0 ]

PSR2 selective is not supported over rotated and scaled planes.
We had the rotation check in intel_psr2_sel_fetch_config_valid()
but that code path is only execute when a modeset is needed and
those plane parameters can change without a modeset.

Pipe selective fetch restrictions are also needed, it could be added
in intel_psr_compute_config() but pippe scaling is computed after
it is executed, so leaving as is for now.
There is no much loss in this approach as it would cause selective
fetch to not enabled as for alderlake-P and newer will cause it to
switch to PSR1 that will have the same power-savings as do full pipe
fetch.

Also need to check those restricions in the second
for_each_oldnew_intel_plane_in_state() loop because the state could
only have a plane that is not affected by those restricitons but
the damaged area intersect with planes that has those restrictions,
so a full pipe fetch is required.

v2:
- also handling pipe restrictions

BSpec: 55229
Reviewed-by: Gwan-gyeong Mun <gwan-gyeong.mun@intel.com> # v1
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Gwan-gyeong Mun <gwan-gyeong.mun@intel.com>
Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20210930001409.254817-1-jose.souza@intel.com
Stable-dep-of: 71c602103c74 ("drm/i915/psr: Use calculated io and fast wake lines")
Signed-off-by: Sasha Levin <sashal@kernel.org>

drm/i915/display/psr: Use drm damage helpers to calculate plane damaged area

[ Upstream commit af7ea1e22afc7ce7773b2e4562df4370c8c711ea ]

drm_atomic_helper_damage_iter_init() + drm_atomic_for_each_plane_damage()
returns the full plane area in case no damaged area was set by
userspace or it was discarted by driver.

This is important to fix the rendering of userspace applications that
does frontbuffer rendering and notify driver about dirty areas but do
not set any dirty clips.

With this we don't need to worry about to check and mark the whole
area as damaged in page flips.

Another important change here is the move of
drm_atomic_add_affected_planes() call, it needs to called late
otherwise the area of all the planes would be added to pipe_clip and
not saving power.

Cc: Daniel Vetter <daniel@ffwll.ch>
Cc: Gwan-gyeong Mun <gwan-gyeong.mun@intel.com>
Reviewed-by: Gwan-gyeong Mun <gwan-gyeong.mun@intel.com>
Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20210914212507.177511-4-jose.souza@intel.com
Stable-dep-of: 71c602103c74 ("drm/i915/psr: Use calculated io and fast wake lines")
Signed-off-by: Sasha Levin <sashal@kernel.org>

drm/i915/display: Workaround cursor left overs with PSR2 selective fetch enabled

[ Upstream commit 1f3a11c341ab211d6ba55ef3d58026b7b5319945 ]

Not sure why but when moving the cursor fast it causes some artifacts
of the cursor to be left in the cursor path, adding some pixels above
the cursor to the damaged area fixes the issue, so leaving this as a
workaround until proper fix is found.

This is reproducile on TGL and ADL-P.

Cc: Gwan-gyeong Mun <gwan-gyeong.mun@intel.com>
Reviewed-by: Gwan-gyeong Mun <gwan-gyeong.mun@intel.com>
Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20210914212507.177511-3-jose.souza@intel.com
Stable-dep-of: 71c602103c74 ("drm/i915/psr: Use calculated io and fast wake lines")
Signed-off-by: Sasha Levin <sashal@kernel.org>

PCI: s390: Fix use-after-free of PCI resources with per-function hotplug

[ Upstream commit ab909509850b27fd39b8ba99e44cda39dbc3858c ]

On s390 PCI functions may be hotplugged individually even when they
belong to a multi-function device. In particular on an SR-IOV device VFs
may be removed and later re-added.

In commit a50297cf8235 ("s390/pci: separate zbus creation from
scanning") it was missed however that struct pci_bus and struct
zpci_bus's resource list retained a reference to the PCI functions MMIO
resources even though those resources are released and freed on
hot-unplug. These stale resources may subsequently be claimed when the
PCI function re-appears resulting in use-after-free.

One idea of fixing this use-after-free in s390 specific code that was
investigated was to simply keep resources around from the moment a PCI
function first appeared until the whole virtual PCI bus created for
a multi-function device disappears. The problem with this however is
that due to the requirement of artificial MMIO addreesses (address
cookies) extra logic is then needed to keep the address cookies
compatible on re-plug. At the same time the MMIO resources semantically
belong to the PCI function so tying their lifecycle to the function
seems more logical.

Instead a simpler approach is to remove the resources of an individually
hot-unplugged PCI function from the PCI bus's resource list while
keeping the resources of other PCI functions on the PCI bus untouched.

This is done by introducing pci_bus_remove_resource() to remove an
individual resource. Similarly the resource also needs to be removed
from the struct zpci_bus's resource list. It turns out however, that
there is really no need to add the MMIO resources to the struct
zpci_bus's resource list at all and instead we can simply use the
zpci_bar_struct's resource pointer directly.

Fixes: a50297cf8235 ("s390/pci: separate zbus creation from scanning")
Signed-off-by: Niklas Schnelle <schnelle@linux.ibm.com>
Reviewed-by: Matthew Rosato <mjrosato@linux.ibm.com>
Acked-by: Bjorn Helgaas <bhelgaas@google.com>
Link: https://lore.kernel.org/r/20230306151014.60913-2-schnelle@linux.ibm.com
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

vdpa_sim: set last_used_idx as last_avail_idx in vdpasim_queue_ready

[ Upstream commit b4cca6d48eb3fa6f0d9caba4329b1a2b0ff67a77 ]

Starting from an used_idx different than 0 is needed in use cases like
virtual machine migration.  Not doing so and letting the caller set an
avail idx different than 0 causes destination device to try to use old
buffers that source driver already recover and are not available
anymore.

Since vdpa_sim does not support receive inflight descriptors as a
destination of a migration, let's set both avail_idx and used_idx the
same at vq start.  This is how vhost-user works in a
VHOST_SET_VRING_BASE call.

Although the simple fix is to set last_used_idx at vdpasim_set_vq_state,
it would be reset at vdpasim_queue_ready.  The last_avail_idx case is
fixed with commit 0e84f918fac8 ("vdpa_sim: not reset state in
vdpasim_queue_ready").  Since the only option is to make it equal to
last_avail_idx, adding the only change needed here.

This was discovered and tested live migrating the vdpa_sim_net device.

Fixes: 2c53d0f64c06 ("vdpasim: vDPA device simulator")
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Message-Id: <20230302181857.925374-1-eperezma@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

vdpa_sim: not reset state in vdpasim_queue_ready

[ Upstream commit 0e84f918fac8ae61dcb790534fad5e3555ca2930 ]

vdpasim_queue_ready calls vringh_init_iotlb, which resets split indexes.
But it can be called after setting a ring base with
vdpasim_set_vq_state.

Fix it by stashing them. They're still resetted in vdpasim_vq_reset.

This was discovered and tested live migrating the vdpa_sim_net device.

Fixes: 2c53d0f64c06 ("vdpasim: vDPA device simulator")
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Message-Id: <20230118164359.1523760-2-eperezma@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Tested-by: Lei Yang <leiyang@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

i40e: Fix kernel crash during reboot when adapter is in recovery mode

[ Upstream commit 7e4f8a0c495413a50413e8c9f1032ce1bc633bae ]

If the driver detects during probe that firmware is in recovery
mode then i40e_init_recovery_mode() is called and the rest of
probe function is skipped including pci_set_drvdata(). Subsequent
i40e_shutdown() called during shutdown/reboot dereferences NULL
pointer as pci_get_drvdata() returns NULL.

To fix call pci_set_drvdata() also during entering to recovery mode.

Reproducer:
1) Lets have i40e NIC with firmware in recovery mode
2) Run reboot

Result:
[  139.084698] i40e: Intel(R) Ethernet Connection XL710 Network Driver
[  139.090959] i40e: Copyright (c) 2013 - 2019 Intel Corporation.
[  139.108438] i40e 0000:02:00.0: Firmware recovery mode detected. Limiting functionality.
[  139.116439] i40e 0000:02:00.0: Refer to the Intel(R) Ethernet Adapters and Devices User Guide for details on firmware recovery mode.
[  139.129499] i40e 0000:02:00.0: fw 8.3.64775 api 1.13 nvm 8.30 0x8000b78d 1.3106.0 [8086:1583] [15d9:084a]
[  139.215932] i40e 0000:02:00.0 enp2s0f0: renamed from eth0
[  139.223292] i40e 0000:02:00.1: Firmware recovery mode detected. Limiting functionality.
[  139.231292] i40e 0000:02:00.1: Refer to the Intel(R) Ethernet Adapters and Devices User Guide for details on firmware recovery mode.
[  139.244406] i40e 0000:02:00.1: fw 8.3.64775 api 1.13 nvm 8.30 0x8000b78d 1.3106.0 [8086:1583] [15d9:084a]
[  139.329209] i40e 0000:02:00.1 enp2s0f1: renamed from eth0
...
[  156.311376] BUG: kernel NULL pointer dereference, address: 00000000000006c2
[  156.318330] #PF: supervisor write access in kernel mode
[  156.323546] #PF: error_code(0x0002) - not-present page
[  156.328679] PGD 0 P4D 0
[  156.331210] Oops: 0002 [#1] PREEMPT SMP NOPTI
[  156.335567] CPU: 26 PID: 15119 Comm: reboot Tainted: G            E      6.2.0+ #1
[  156.343126] Hardware name: Abacus electric, s.r.o. - servis@abacus.cz Super Server/H12SSW-iN, BIOS 2.4 04/13/2022
[  156.353369] RIP: 0010:i40e_shutdown+0x15/0x130 [i40e]
[  156.358430] Code: c1 fc ff ff 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 f3 0f 1e fa 0f 1f 44 00 00 55 48 89 fd 53 48 8b 9f 48 01 00 00 <f0> 80 8b c2 06 00 00 04 f0 80 8b c0 06 00 00 08 48 8d bb 08 08 00
[  156.377168] RSP: 0018:ffffb223c8447d90 EFLAGS: 00010282
[  156.382384] RAX: ffffffffc073ee70 RBX: 0000000000000000 RCX: 0000000000000001
[  156.389510] RDX: 0000000080000001 RSI: 0000000000000246 RDI: ffff95db49988000
[  156.396634] RBP: ffff95db49988000 R08: ffffffffffffffff R09: ffffffff8bd17d40
[  156.403759] R10: 0000000000000001 R11: ffffffff8a5e3d28 R12: ffff95db49988000
[  156.410882] R13: ffffffff89a6fe17 R14: ffff95db49988150 R15: 0000000000000000
[  156.418007] FS:  00007fe7c0cc3980(0000) GS:ffff95ea8ee80000(0000) knlGS:0000000000000000
[  156.426083] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  156.431819] CR2: 00000000000006c2 CR3: 00000003092fc005 CR4: 0000000000770ee0
[  156.438944] PKRU: 55555554
[  156.441647] Call Trace:
[  156.444096]  <TASK>
[  156.446199]  pci_device_shutdown+0x38/0x60
[  156.450297]  device_shutdown+0x163/0x210
[  156.454215]  kernel_restart+0x12/0x70
[  156.457872]  __do_sys_reboot+0x1ab/0x230
[  156.461789]  ? vfs_writev+0xa6/0x1a0
[  156.465362]  ? __pfx_file_free_rcu+0x10/0x10
[  156.469635]  ? __call_rcu_common.constprop.85+0x109/0x5a0
[  156.475034]  do_syscall_64+0x3e/0x90
[  156.478611]  entry_SYSCALL_64_after_hwframe+0x72/0xdc
[  156.483658] RIP: 0033:0x7fe7bff37ab7

Fixes: 4ff0ee1af016 ("i40e: Introduce recovery mode support")
Signed-off-by: Ivan Vecera <ivecera@redhat.com>
Tested-by: Arpana Arland <arpanax.arland@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Link: https://lore.kernel.org/r/20230309184509.984639-1-anthony.l.nguyen@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

ipvlan: Make skb->skb_iif track skb->dev for l3s mode

[ Upstream commit 59a0b022aa249e3f5735d93de0849341722c4754 ]

For l3s mode, skb->dev is set to ipvlan interface in ipvlan_nf_input():
  skb->dev = addr->master->dev
but, skb->skb_iif remain unchanged, this will cause socket lookup failed
if a target socket is bound to a interface, like the following example:

  ip link add ipvlan0 link eth0 type ipvlan mode l3s
  ip addr add dev ipvlan0 192.168.124.111/24
  ip link set ipvlan0 up

  ping -c 1 -I ipvlan0 8.8.8.8
  100% packet loss

This is because there is no match sk in __raw_v4_lookup() as sk->sk_bound_dev_if != dif(skb->skb_iif).
Fix this by make skb->skb_iif track skb->dev in ipvlan_nf_input().

Fixes: c675e06a98a4 ("ipvlan: decouple l3s mode dependencies from other modes")
Signed-off-by: Jianguo Wu <wujianguo@chinatelecom.cn>
Reviewed-by: Jiri Pirko <jiri@nvidia.com>
Link: https://lore.kernel.org/r/29865b1f-6db7-c07a-de89-949d3721ea30@163.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

nfc: pn533: initialize struct pn533_out_arg properly

[ Upstream commit 484b7059796e3bc1cb527caa61dfc60da649b4f6 ]

struct pn533_out_arg used as a temporary context for out_urb is not
initialized properly. Its uninitialized 'phy' field can be dereferenced in
error cases inside pn533_out_complete() callback function. It causes the
following failure:

general protection fault, probably for non-canonical address 0xdffffc0000000000: 0000 [#1] PREEMPT SMP KASAN
KASAN: null-ptr-deref in range [0x0000000000000000-0x0000000000000007]
CPU: 1 PID: 0 Comm: swapper/1 Not tainted 6.2.0-rc3-next-20230110-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 10/26/2022
RIP: 0010:pn533_out_complete.cold+0x15/0x44 drivers/nfc/pn533/usb.c:441
Call Trace:
 <IRQ>
 __usb_hcd_giveback_urb+0x2b6/0x5c0 drivers/usb/core/hcd.c:1671
 usb_hcd_giveback_urb+0x384/0x430 drivers/usb/core/hcd.c:1754
 dummy_timer+0x1203/0x32d0 drivers/usb/gadget/udc/dummy_hcd.c:1988
 call_timer_fn+0x1da/0x800 kernel/time/timer.c:1700
 expire_timers+0x234/0x330 kernel/time/timer.c:1751
 __run_timers kernel/time/timer.c:2022 [inline]
 __run_timers kernel/time/timer.c:1995 [inline]
 run_timer_softirq+0x326/0x910 kernel/time/timer.c:2035
 __do_softirq+0x1fb/0xaf6 kernel/softirq.c:571
 invoke_softirq kernel/softirq.c:445 [inline]
 __irq_exit_rcu+0x123/0x180 kernel/softirq.c:650
 irq_exit_rcu+0x9/0x20 kernel/softirq.c:662
 sysvec_apic_timer_interrupt+0x97/0xc0 arch/x86/kernel/apic/apic.c:1107

Initialize the field with the pn533_usb_phy currently used.

Found by Linux Verification Center (linuxtesting.org) with Syzkaller.

Fixes: 9dab880d675b ("nfc: pn533: Wait for out_urb's completion in pn533_usb_send_frame()")
Reported-by: syzbot+1e608ba4217c96d1952f@syzkaller.appspotmail.com
Signed-off-by: Fedor Pchelkin <pchelkin@ispras.ru>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Link: https://lore.kernel.org/r/20230309165050.207390-1-pchelkin@ispras.ru
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

tcp: tcp_make_synack() can be called from process context

[ Upstream commit bced3f7db95ff2e6ca29dc4d1c9751ab5e736a09 ]

tcp_rtx_synack() now could be called in process context as explained in
0a375c822497 ("tcp: tcp_rtx_synack() can be called from process
context").

tcp_rtx_synack() might call tcp_make_synack(), which will touch per-CPU
variables with preemption enabled. This causes the following BUG:

    BUG: using __this_cpu_add() in preemptible [00000000] code: ThriftIO1/5464
    caller is tcp_make_synack+0x841/0xac0
    Call Trace:
     <TASK>
     dump_stack_lvl+0x10d/0x1a0
     check_preemption_disabled+0x104/0x110
     tcp_make_synack+0x841/0xac0
     tcp_v6_send_synack+0x5c/0x450
     tcp_rtx_synack+0xeb/0x1f0
     inet_rtx_syn_ack+0x34/0x60
     tcp_check_req+0x3af/0x9e0
     tcp_rcv_state_process+0x59b/0x2030
     tcp_v6_do_rcv+0x5f5/0x700
     release_sock+0x3a/0xf0
     tcp_sendmsg+0x33/0x40
     ____sys_sendmsg+0x2f2/0x490
     __sys_sendmsg+0x184/0x230
     do_syscall_64+0x3d/0x90

Avoid calling __TCP_INC_STATS() with will touch per-cpu variables. Use
TCP_INC_STATS() which is safe to be called from context switch.

Fixes: 8336886f786f ("tcp: TCP Fast Open Server - support TFO listeners")
Signed-off-by: Breno Leitao <leitao@debian.org>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20230308190745.780221-1-leitao@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

scsi: core: Fix a procfs host directory removal regression

[ Upstream commit be03df3d4bfe7e8866d4aa43d62e648ffe884f5f ]

scsi_proc_hostdir_rm() decreases a reference counter and hence must only be
called once per host that is removed. This change does not require a
scsi_add_host_with_dma() change since scsi_add_host_with_dma() will return
0 (success) if scsi_proc_host_add() is called.

Fixes: fc663711b944 ("scsi: core: Remove the /proc/scsi/${proc_name} directory earlier")
Cc: John Garry <john.g.garry@oracle.com>
Reported-by: John Garry <john.g.garry@oracle.com>
Link: https://lore.kernel.org/all/ed6b8027-a9d9-1b45-be8e-df4e8c6c4605@oracle.com/
Reported-by: syzbot+645a4616b87a2f10e398@syzkaller.appspotmail.com
Link: https://lore.kernel.org/linux-scsi/000000000000890fab05f65342b6@google.com/
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
Link: https://lore.kernel.org/r/20230307214428.3703498-1-bvanassche@acm.org
Tested-by: John Garry <john.g.garry@oracle.com>
Tested-by: Shin'ichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

netfilter: nft_redir: correct value of inet type `.maxattrs`

[ Upstream commit 493924519b1fe3faab13ee621a43b0d0939abab1 ]

`nft_redir_inet_type.maxattrs` was being set, presumably because of a
cut-and-paste error, to `NFTA_MASQ_MAX`, instead of `NFTA_REDIR_MAX`.

Fixes: 63ce3940f3ab ("netfilter: nft_redir: add inet support")
Signed-off-by: Jeremy Sowden <jeremy@azazel.net>
Reviewed-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

netfilter: nft_redir: correct length for loading protocol registers

[ Upstream commit 1f617b6b4c7a3d5ea7a56abb83a4c27733b60c2f ]

The values in the protocol registers are two bytes wide.  However, when
parsing the register loads, the code currently uses the larger 16-byte
size of a `union nf_inet_addr`.  Change it to use the (correct) size of
a `union nf_conntrack_man_proto` instead.

Fixes: d07db9884a5f ("netfilter: nf_tables: introduce nft_validate_register_load()")
Signed-off-by: Jeremy Sowden <jeremy@azazel.net>
Reviewed-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

netfilter: nft_masq: correct length for loading protocol registers

[ Upstream commit ec2c5917eb858428b2083d1c74f445aabbe8316b ]

The values in the protocol registers are two bytes wide.  However, when
parsing the register loads, the code currently uses the larger 16-byte
size of a `union nf_inet_addr`.  Change it to use the (correct) size of
a `union nf_conntrack_man_proto` instead.

Fixes: 8a6bf5da1aef ("netfilter: nft_masq: support port range")
Signed-off-by: Jeremy Sowden <jeremy@azazel.net>
Reviewed-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

netfilter: nft_nat: correct length for loading protocol registers

[ Upstream commit 068d82e75d537b444303b8c449a11e51ea659565 ]

The values in the protocol registers are two bytes wide.  However, when
parsing the register loads, the code currently uses the larger 16-byte
size of a `union nf_inet_addr`.  Change it to use the (correct) size of
a `union nf_conntrack_man_proto` instead.

Fixes: d07db9884a5f ("netfilter: nf_tables: introduce nft_validate_register_load()")
Signed-off-by: Jeremy Sowden <jeremy@azazel.net>
Reviewed-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

ALSA: hda: Match only Intel devices with CONTROLLER_IN_GPU()

[ Upstream commit ff447886e675979d66b2bc01810035d3baea1b3a ]

CONTROLLER_IN_GPU() is clearly intended to match only Intel devices, but
previously it checked only the PCI Device ID, not the Vendor ID, so it
could match devices from other vendors that happened to use the same Device
ID.

Update CONTROLLER_IN_GPU() so it matches only Intel devices.

Fixes: 535115b5ff51 ("ALSA: hda - Abort the probe without i915 binding for HSW/B")
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Link: https://lore.kernel.org/r/20230307214054.886721-1-helgaas@kernel.org
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Sasha Levin <sashal@kernel.org>

scsi: mpt3sas: Fix NULL pointer access in mpt3sas_transport_port_add()

[ Upstream commit d3c57724f1569311e4b81e98fad0931028b9bdcd ]

Port is allocated by sas_port_alloc_num() and rphy is allocated by either
sas_end_device_alloc() or sas_expander_alloc(), all of which may return
NULL. So we need to check the rphy to avoid possible NULL pointer access.

If sas_rphy_add() returned with failure, rphy is set to NULL. We would
access the rphy in the following lines which would also result NULL pointer
access.

Fixes: 78316e9dfc24 ("scsi: mpt3sas: Fix possible resource leaks in mpt3sas_transport_port_add()")
Signed-off-by: Wenchao Hao <haowenchao2@huawei.com>
Link: https://lore.kernel.org/r/20230225100135.2109330-1-haowenchao2@huawei.com
Acked-by: Sathya Prakash Veerichetty <sathya.prakash@broadcom.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

docs: Correct missing "d_" prefix for dentry_operations member d_weak_revalidate

[ Upstream commit 74596085796fae0cfce3e42ee46bf4f8acbdac55 ]

The details for struct dentry_operations member d_weak_revalidate is
missing a "d_" prefix.

Fixes: af96c1e304f7 ("docs: filesystems: vfs: Convert vfs.txt to RST")
Signed-off-by: Glenn Washburn <development@efficientek.com>
Reviewed-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Link: https://lore.kernel.org/r/20230227184042.2375235-1-development@efficientek.com
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>

clk: HI655X: select REGMAP instead of depending on it

[ Upstream commit 0ffad67784a097beccf34d297ddd1b0773b3b8a3 ]

REGMAP is a hidden (not user visible) symbol. Users cannot set it
directly thru "make *config", so drivers should select it instead of
depending on it if they need it.

Consistently using "select" or "depends on" can also help reduce
Kconfig circular dependency issues.

Therefore, change the use of "depends on REGMAP" to "select REGMAP".

Fixes: 3a49afb84ca0 ("clk: enable hi655x common clk automatically")
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Cc: Riku Voipio <riku.voipio@linaro.org>
Cc: Stephen Boyd <sboyd@kernel.org>
Cc: Michael Turquette <mturquette@baylibre.com>
Cc: linux-clk@vger.kernel.org
Link: https://lore.kernel.org/r/20230226053953.4681-3-rdunlap@infradead.org
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

drm/meson: fix 1px pink line on GXM when scaling video overlay

[ Upstream commit 5c8cf1664f288098a971a1d1e65716a2b6a279e1 ]

Playing media with a resolution smaller than the crtc size requires the
video overlay to be scaled for output and GXM boards display a 1px pink
line on the bottom of the scaled overlay. Comparing with the downstream
vendor driver revealed VPP_DUMMY_DATA not being set [0].

Setting VPP_DUMMY_DATA prevents the 1px pink line from being seen.

[0] https://github.com/endlessm/linux-s905x/blob/master/drivers/amlogic/amports/video.c#L7869

Fixes: bbbe775ec5b5 ("drm: Add support for Amlogic Meson Graphic Controller")
Suggested-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Signed-off-by: Christian Hewitt <christianshewitt@gmail.com>
Acked-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230303123312.155164-1-christianshewitt@gmail.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

cifs: Move the in_send statistic to __smb_send_rqst()

[ Upstream commit d0dc41119905f740e8d5594adce277f7c0de8c92 ]

When send SMB_COM_NT_CANCEL and RFC1002_SESSION_REQUEST, the
in_send statistic was lost.

Let's move the in_send statistic to the send function to avoid
this scenario.

Fixes: 7ee1af765dfa ("[CIFS]")
Signed-off-by: Zhang Xiaoxu <zhangxiaoxu5@huawei.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

drm/panfrost: Don't sync rpm suspension after mmu flushing

[ Upstream commit ba3be66f11c3c49afaa9f49b99e21d88756229ef ]

Lockdep warns about potential circular locking dependency of devfreq
with the fs_reclaim caused by immediate device suspension when mapping is
released by shrinker. Fix it by doing the suspension asynchronously.

Reviewed-by: Steven Price <steven.price@arm.com>
Fixes: ec7eba47da86 ("drm/panfrost: Rework page table flushing and runtime PM interaction")
Signed-off-by: Dmitry Osipenko <dmitry.osipenko@collabora.com>
Link: https://lore.kernel.org/all/20230108210445.3948344-3-dmitry.osipenko@collabora.com/
Signed-off-by: Sasha Levin <sashal@kernel.org>

xfrm: Allow transport-mode states with AF_UNSPEC selector

[ Upstream commit c276a706ea1f51cf9723ed8484feceaf961b8f89 ]

xfrm state selectors are matched against the inner-most flow
which can be of any address family.  Therefore middle states
in nested configurations need to carry a wildcard selector in
order to work at all.

However, this is currently forbidden for transport-mode states.

Fix this by removing the unnecessary check.

Fixes: 13996378e658 ("[IPSEC]: Rename mode to outer_mode and add inner_mode")
Reported-by: David George <David.George@sophos.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

Linux 5.15.103

Link: https://lore.kernel.org/r/20230315115738.951067403@linuxfoundation.org
Tested-by: Chris Paterson (CIP) <chris.paterson2@renesas.com>
Tested-by: Florian Fainelli <f.fainelli@gmail.com>
Tested-by: Shuah Khan <skhan@linuxfoundation.org>
Tested-by: Bagas Sanjaya <bagasdotme@gmail.com>
Link: https://lore.kernel.org/r/20230316083443.411936182@linuxfoundation.org
Tested-by: Chris Paterson (CIP) <chris.paterson2@renesas.com>
Tested-by: Florian Fainelli <f.fainelli@gmail.com>
Tested-by: Harshit Mogalapalli <harshit.m.mogalapalli@oracle.com>
Tested-by: Ron Economos <re@w6rz.net>
Tested-by: Tom Saeger <tom.saeger@oracle.com>
Tested-by: Guenter Roeck <linux@roeck-us.net>
Tested-by: Linux Kernel Functional Testing <lkft@linaro.org>
Tested-by: Bagas Sanjaya <bagasdotme@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Makefile: use -gdwarf-{4|5} for assembler for DEBUG_INFO_DWARF{4|5}

This is _not_ an upstream commit and just for 5.15.y only. It is based
on upstream
commit 32ef9e5054ec ("Makefile.debug: re-enable debug info for .S files").

When the user has chosen not to use their compiler's implicit default
DWARF version (which changes over time) via selecting
- CONFIG_DEBUG_INFO_DWARF4 or
- CONFIG_DEBUG_INFO_DWARF5
we need to tell the compiler this for Asm sources as well as C sources.
(We use the compiler to drive assembler jobs in kbuild, since most asm
needs to be preprocessed first).  Otherwise, we will get object files
built from Asm sources with the compiler's implicit default DWARF
version.

For example, selecting CONFIG_DEBUG_INFO_DWARF4 would produce a DWARFv5
vmlinux, since it was a mix of DWARFv4 object files from C sources and
DWARFv5 object files from Asm sources when using Clang as the assembler
(ex. `make LLVM=1`).

Fixes: 0ee2f0567a56 ("Makefile.debug: re-enable debug info for .S files")
Reported-by: Lecopzer Chen <lecopzer.chen@mediatek.com>
Signed-off-by: Nick Desaulniers <ndesaulniers@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

KVM: VMX: Fix crash due to uninitialized current_vmcs

commit 93827a0a36396f2fd6368a54a020f420c8916e9b upstream.

KVM enables 'Enlightened VMCS' and 'Enlightened MSR Bitmap' when running as
a nested hypervisor on top of Hyper-V. When MSR bitmap is updated,
evmcs_touch_msr_bitmap function uses current_vmcs per-cpu variable to mark
that the msr bitmap was changed.

vmx_vcpu_create() modifies the msr bitmap via vmx_disable_intercept_for_msr
-> vmx_msr_bitmap_l01_changed which in the end calls this function. The
function checks for current_vmcs if it is null but the check is
insufficient because current_vmcs is not initialized. Because of this, the
code might incorrectly write to the structure pointed by current_vmcs value
left by another task. Preemption is not disabled, the current task can be
preempted and moved to another CPU while current_vmcs is accessed multiple
times from evmcs_touch_msr_bitmap() which leads to crash.

The manipulation of MSR bitmaps by callers happens only for vmcs01 so the
solution is to use vmx->vmcs01.vmcs instead of current_vmcs.

  BUG: kernel NULL pointer dereference, address: 0000000000000338
  PGD 4e1775067 P4D 0
  Oops: 0002 [#1] PREEMPT SMP NOPTI
  ...
  RIP: 0010:vmx_msr_bitmap_l01_changed+0x39/0x50 [kvm_intel]
  ...
  Call Trace:
   vmx_disable_intercept_for_msr+0x36/0x260 [kvm_intel]
   vmx_vcpu_create+0xe6/0x540 [kvm_intel]
   kvm_arch_vcpu_create+0x1d1/0x2e0 [kvm]
   kvm_vm_ioctl_create_vcpu+0x178/0x430 [kvm]
   kvm_vm_ioctl+0x53f/0x790 [kvm]
   __x64_sys_ioctl+0x8a/0xc0
   do_syscall_64+0x5c/0x90
   entry_SYSCALL_64_after_hwframe+0x63/0xcd

Fixes: ceef7d10dfb6 ("KVM: x86: VMX: hyper-v: Enlightened MSR-Bitmap support")
Cc: stable@vger.kernel.org
Suggested-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Alexandru Matei <alexandru.matei@uipath.com>
Link: https://lore.kernel.org/r/20230123221208.4964-1-alexandru.matei@uipath.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
[manual backport: evmcs.h got renamed to hyperv.h in a later
version, modified in evmcs.h instead]
Signed-off-by: Alexandru Matei <alexandru.matei@uipath.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

KVM: VMX: Introduce vmx_msr_bitmap_l01_changed() helper

commit b84155c38076b36d625043a06a2f1c90bde62903 upstream.

In preparation to enabling 'Enlightened MSR Bitmap' feature for Hyper-V
guests move MSR bitmap update tracking to a dedicated helper.

Note: vmx_msr_bitmap_l01_changed() is called when MSR bitmap might be
updated. KVM doesn't check if the bit we're trying to set is already set
(or the bit it's trying to clear is already cleared). Such situations
should not be common and a few false positives should not be a problem.

No functional change intended.

Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
Reviewed-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20211129094704.326635-3-vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Alexandru Matei <alexandru.matei@uipath.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

KVM: nVMX: Don't use Enlightened MSR Bitmap for L3

commit 250552b925ce400c17d166422fde9bb215958481 upstream.

When KVM runs as a nested hypervisor on top of Hyper-V it uses Enlightened
VMCS and enables Enlightened MSR Bitmap feature for its L1s and L2s (which
are actually L2s and L3s from Hyper-V's perspective). When MSR bitmap is
updated, KVM has to reset HV_VMX_ENLIGHTENED_CLEAN_FIELD_MSR_BITMAP from
clean fields to make Hyper-V aware of the change. For KVM's L1s, this is
done in vmx_disable_intercept_for_msr()/vmx_enable_intercept_for_msr().
MSR bitmap for L2 is build in nested_vmx_prepare_msr_bitmap() by blending
MSR bitmap for L1 and L1's idea of MSR bitmap for L2. KVM, however, doesn't
check if the resulting bitmap is different and never cleans
HV_VMX_ENLIGHTENED_CLEAN_FIELD_MSR_BITMAP in eVMCS02. This is incorrect and
may result in Hyper-V missing the update.

The issue could've been solved by calling evmcs_touch_msr_bitmap() for
eVMCS02 from nested_vmx_prepare_msr_bitmap() unconditionally but doing so
would not give any performance benefits (compared to not using Enlightened
MSR Bitmap at all). 3-level nesting is also not a very common setup
nowadays.

Don't enable 'Enlightened MSR Bitmap' feature for KVM's L2s (real L3s) for
now.

Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20211129094704.326635-2-vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Alexandru Matei <alexandru.matei@uipath.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

fs: hold writers when changing mount's idmapping

commit e1bbcd277a53e08d619ffeec56c5c9287f2bf42f upstream.

Hold writers when changing a mount's idmapping to make it more robust.

The vfs layer takes care to retrieve the idmapping of a mount once
ensuring that the idmapping used for vfs permission checking is
identical to the idmapping passed down to the filesystem.

For ioctl codepaths the filesystem itself is responsible for taking the
idmapping into account if they need to. While all filesystems with
FS_ALLOW_IDMAP raised take the same precautions as the vfs we should
enforce it explicitly by making sure there are no active writers on the
relevant mount while changing the idmapping.

This is similar to turning a mount ro with the difference that in
contrast to turning a mount ro changing the idmapping can only ever be
done once while a mount can transition between ro and rw as much as it
wants.

This is a minor user-visible change. But it is extremely unlikely to
matter. The caller must've created a detached mount via OPEN_TREE_CLONE
and then handed that O_PATH fd to another process or thread which then
must've gotten a writable fd for that mount and started creating files
in there while the caller is still changing mount properties. While not
impossible it will be an extremely rare corner-case and should in
general be considered a bug in the application. Consider making a mount
MOUNT_ATTR_NOEXEC or MOUNT_ATTR_NODEV while allowing someone else to
perform lookups or exec'ing in parallel by handing them a copy of the
OPEN_TREE_CLONE fd or another fd beneath that mount.

Link: https://lore.kernel.org/r/20220510095840.152264-1-brauner@kernel.org
Cc: Seth Forshee <seth.forshee@digitalocean.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: linux-fsdevel@vger.kernel.org
Signed-off-by: Christian Brauner (Microsoft) <brauner@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

UML: define RUNTIME_DISCARD_EXIT

commit b99ddbe8336ee680257c8ab479f75051eaa49dcf upstream.

With CONFIG_VIRTIO_UML=y, GNU ld < 2.36 fails to link UML vmlinux
(w/wo CONFIG_LD_SCRIPT_STATIC).

  `.exit.text' referenced in section `.uml.exitcall.exit' of arch/um/drivers/virtio_uml.o: defined in discarded section `.exit.text' of arch/um/drivers/virtio_uml.o
  collect2: error: ld returned 1 exit status

This fix is similar to the following commits:

- 4b9880dbf3bd ("powerpc/vmlinux.lds: Define RUNTIME_DISCARD_EXIT")
- a494398bde27 ("s390: define RUNTIME_DISCARD_EXIT to fix link error
  with GNU ld < 2.36")
- c1c551bebf92 ("sh: define RUNTIME_DISCARD_EXIT")

Fixes: 99cb0d917ffa ("arch: fix broken BuildID for arm64 and riscv")
Reported-by: SeongJae Park <sj@kernel.org>
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
Tested-by: SeongJae Park <sj@kernel.org>
Signed-off-by: Richard Weinberger <richard@nod.at>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

xfs: remove xfs_setattr_time() declaration

commit b0463b9dd7030a766133ad2f1571f97f204d7bdf upstream.

xfs_setattr_time() has been removed since
commit e014f37db1a2 ("xfs: use setattr_copy to set vfs inode
attributes"), so remove it.

Signed-off-by: Gaosheng Cui <cuigaosheng1@huawei.com>
Reviewed-by: Carlos Maiolino <cmaiolino@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

KVM: fix memoryleak in kvm_init()

commit 5a2a961be2ad6a16eb388a80442443b353c11d16 upstream.

When alloc_cpumask_var_node() fails for a certain cpu, there might be some
allocated cpumasks for percpu cpu_kick_mask. We should free these cpumasks
or memoryleak will occur.

Fixes: baff59ccdc65 ("KVM: Pre-allocate cpumasks for kvm_make_all_cpus_request_except()")
Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
Link: https://lore.kernel.org/r/20220823063414.59778-1-linmiaohe@huawei.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

tools bpftool: Fix compilation error with new binutils

commit 600b7b26c07a070d0153daa76b3806c1e52c9e00 upstream.

binutils changed the signature of init_disassemble_info(), which now causes
compilation to fail for tools/bpf/bpftool/jit_disasm.c, e.g. on debian
unstable.

Relevant binutils commit:

  https://sourceware.org/git/?p=binutils-gdb.git;a=commit;h=60a3da00bd5407f07

Wire up the feature test and switch to init_disassemble_info_compat(),
which were introduced in prior commits, fixing the compilation failure.

I verified that bpftool can still disassemble bpf programs, both with an
old and new dis-asm.h API. There are no output changes for plain and json
formats. When comparing the output from old binutils (2.35)
to new bintuils with the patch (upstream snapshot) there are a few output
differences, but they are unrelated to this patch. An example hunk is:

     2f: pop    %r14
     31: pop    %r13
     33: pop    %rbx
  -  34: leaveq
  -  35: retq
  +  34: leave
  +  35: ret

Signed-off-by: Andres Freund <andres@anarazel.de>
Acked-by: Quentin Monnet <quentin@isovalent.com>
Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Ben Hutchings <benh@debian.org>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Quentin Monnet <quentin@isovalent.com>
Cc: Sedat Dilek <sedat.dilek@gmail.com>
Cc: bpf@vger.kernel.org
Link: http://lore.kernel.org/lkml/20220622181918.ykrs5rsnmx3og4sv@alap3.anarazel.de
Link: https://lore.kernel.org/r/20220801013834.156015-8-andres@anarazel.de
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

tools bpf_jit_disasm: Fix compilation error with new binutils

commit 96ed066054abf11c7d3e106e3011a51f3f1227a3 upstream.

binutils changed the signature of init_disassemble_info(), which now causes
compilation to fail for tools/bpf/bpf_jit_disasm.c, e.g. on debian
unstable.

Relevant binutils commit:

  https://sourceware.org/git/?p=binutils-gdb.git;a=commit;h=60a3da00bd5407f07

Wire up the feature test and switch to init_disassemble_info_compat(),
which were introduced in prior commits, fixing the compilation failure.

I verified that bpf_jit_disasm can still disassemble bpf programs, both
with the old and new dis-asm.h API. With old binutils there's no change in
output before/after this patch. When comparing the output from old
binutils (2.35) to new bintuils with the patch (upstream snapshot) there
are a few output differences, but they are unrelated to this patch. An
example hunk is:

     f4: mov    %r14,%rsi
     f7: mov    %r15,%rdx
     fa: mov    $0x2a,%ecx
  -  ff: callq  0xffffffffea8c4988
  +  ff: call   0xffffffffea8c4988
    104: test   %rax,%rax
    107: jge    0x0000000000000110
    109: xor    %eax,%eax
  - 10b: jmpq   0x0000000000000073
  + 10b: jmp    0x0000000000000073
    110: cmp    $0x16,%rax

However, I had to use an older kernel to generate the bpf_jit_enabled =
2 output, as that has been broken since 5.18 / 1022a5498f6f745c ("bpf,
x86_64: Use bpf_jit_binary_pack_alloc").

  https://lore.kernel.org/20220703030210.pmjft7qc2eajzi6c@alap3.anarazel.de

Signed-off-by: Andres Freund <andres@anarazel.de>
Acked-by: Quentin Monnet <quentin@isovalent.com>
Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Ben Hutchings <benh@debian.org>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Quentin Monnet <quentin@isovalent.com>
Cc: Sedat Dilek <sedat.dilek@gmail.com>
Cc: bpf@vger.kernel.org
Link: http://lore.kernel.org/lkml/20220622181918.ykrs5rsnmx3og4sv@alap3.anarazel.de
Link: https://lore.kernel.org/r/20220801013834.156015-6-andres@anarazel.de
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

tools perf: Fix compilation error with new binutils

commit 83aa0120487e8bc3f231e72c460add783f71f17c upstream.

binutils changed the signature of init_disassemble_info(), which now causes
compilation failures for tools/perf/util/annotate.c, e.g. on debian
unstable.

Relevant binutils commit:

  https://sourceware.org/git/?p=binutils-gdb.git;a=commit;h=60a3da00bd5407f07

Wire up the feature test and switch to init_disassemble_info_compat(),
which were introduced in prior commits, fixing the compilation failure.

I verified that perf can still disassemble bpf programs by using bpftrace
under load, recording a perf trace, and then annotating the bpf "function"
with and without the changes. With old binutils there's no change in output
before/after this patch. When comparing the output from old binutils (2.35)
to new bintuils with the patch (upstream snapshot) there are a few output
differences, but they are unrelated to this patch. An example hunk is:

       1.15 :   55:mov    %rbp,%rdx
       0.00 :   58:add    $0xfffffffffffffff8,%rdx
       0.00 :   5c:xor    %ecx,%ecx
  -    1.03 :   5e:callq  0xffffffffe12aca3c
  +    1.03 :   5e:call   0xffffffffe12aca3c
       0.00 :   63:xor    %eax,%eax
  -    2.18 :   65:leaveq
  -    2.82 :   66:retq
  +    2.18 :   65:leave
  +    2.82 :   66:ret

Signed-off-by: Andres Freund <andres@anarazel.de>
Acked-by: Quentin Monnet <quentin@isovalent.com>
Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Ben Hutchings <benh@debian.org>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Sedat Dilek <sedat.dilek@gmail.com>
Cc: bpf@vger.kernel.org
Link: http://lore.kernel.org/lkml/20220622181918.ykrs5rsnmx3og4sv@alap3.anarazel.de
Link: https://lore.kernel.org/r/20220801013834.156015-5-andres@anarazel.de
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

tools include: add dis-asm-compat.h to handle version differences

commit a45b3d6926231c3d024ea0de4f7bd967f83709ee upstream.

binutils changed the signature of init_disassemble_info(), which now causes
compilation failures for tools/{perf,bpf}, e.g. on debian unstable.

Relevant binutils commit:

  https://sourceware.org/git/?p=binutils-gdb.git;a=commit;h=60a3da00bd5407f07

This commit introduces a wrapper for init_disassemble_info(), to avoid
spreading #ifdef DISASM_INIT_STYLED to a bunch of places. Subsequent
commits will use it to fix the build failures.

It likely is worth adding a wrapper for disassember(), to avoid the already
existing DISASM_FOUR_ARGS_SIGNATURE ifdefery.

Signed-off-by: Andres Freund <andres@anarazel.de>
Signed-off-by: Ben Hutchings <benh@debian.org>
Acked-by: Quentin Monnet <quentin@isovalent.com>
Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Ben Hutchings <benh@debian.org>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Quentin Monnet <quentin@isovalent.com>
Cc: Sedat Dilek <sedat.dilek@gmail.com>
Cc: bpf@vger.kernel.org
Link: http://lore.kernel.org/lkml/20220622181918.ykrs5rsnmx3og4sv@alap3.anarazel.de
Link: https://lore.kernel.org/r/20220801013834.156015-4-andres@anarazel.de
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

tools build: Add feature test for init_disassemble_info API changes

commit cfd59ca91467056bb2c36907b2fa67b8e1af9952 upstream.

binutils changed the signature of init_disassemble_info(), which now causes
compilation failures for tools/{perf,bpf}, e.g. on debian unstable.

Relevant binutils commit:

  https://sourceware.org/git/?p=binutils-gdb.git;a=commit;h=60a3da00bd5407f07

This commit adds a feature test to detect the new signature.  Subsequent
commits will use it to fix the build failures.

Signed-off-by: Andres Freund <andres@anarazel.de>
Acked-by: Quentin Monnet <quentin@isovalent.com>
Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Ben Hutchings <benh@debian.org>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Quentin Monnet <quentin@isovalent.com>
Cc: Sedat Dilek <sedat.dilek@gmail.com>
Cc: bpf@vger.kernel.org
Link: http://lore.kernel.org/lkml/20220622181918.ykrs5rsnmx3og4sv@alap3.anarazel.de
Link: https://lore.kernel.org/r/20220801013834.156015-2-andres@anarazel.de
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

sh: define RUNTIME_DISCARD_EXIT

commit c1c551bebf928889e7a8fef7415b44f9a64975f4 upstream.

sh vmlinux fails to link with GNU ld < 2.40 (likely < 2.36) since
commit 99cb0d917ffa ("arch: fix broken BuildID for arm64 and riscv").

This is similar to fixes for powerpc and s390:
commit 4b9880dbf3bd ("powerpc/vmlinux.lds: Define RUNTIME_DISCARD_EXIT").
commit a494398bde27 ("s390: define RUNTIME_DISCARD_EXIT to fix link error
with GNU ld < 2.36").

  $ sh4-linux-gnu-ld --version | head -n1
  GNU ld (GNU Binutils for Debian) 2.35.2

  $ make ARCH=sh CROSS_COMPILE=sh4-linux-gnu- microdev_defconfig
  $ make ARCH=sh CROSS_COMPILE=sh4-linux-gnu-

  `.exit.text' referenced in section `__bug_table' of crypto/algboss.o:
  defined in discarded section `.exit.text' of crypto/algboss.o
  `.exit.text' referenced in section `__bug_table' of
  drivers/char/hw_random/core.o: defined in discarded section
  `.exit.text' of drivers/char/hw_random/core.o
  make[2]: *** [scripts/Makefile.vmlinux:34: vmlinux] Error 1
  make[1]: *** [Makefile:1252: vmlinux] Error 2

arch/sh/kernel/vmlinux.lds.S keeps EXIT_TEXT:

/*
 * .exit.text is discarded at runtime, not link time, to deal with
 * references from __bug_table
 */
.exit.text : AT(ADDR(.exit.text)) { EXIT_TEXT }

However, EXIT_TEXT is thrown away by
DISCARD(include/asm-generic/vmlinux.lds.h) because
sh does not define RUNTIME_DISCARD_EXIT.

GNU ld 2.40 does not have this issue and builds fine.
This corresponds with Masahiro's comments in a494398bde27:
"Nathan [Chancellor] also found that binutils
commit 21401fc7bf67 ("Duplicate output sections in scripts") cured this
issue, so we cannot reproduce it with binutils 2.36+, but it is better
to not rely on it."

Link: https://lkml.kernel.org/r/9166a8abdc0f979e50377e61780a4bba1dfa2f52.1674518464.git.tom.saeger@oracle.com
Fixes: 99cb0d917ffa ("arch: fix broken BuildID for arm64 and riscv")
Link: https://lore.kernel.org/all/Y7Jal56f6UBh1abE@dev-arch.thelio-3990X/
Link: https://lore.kernel.org/all/20230123194218.47ssfzhrpnv3xfez@oracle.com/
Signed-off-by: Tom Saeger <tom.saeger@oracle.com>
Tested-by: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Cc: Ard Biesheuvel <ardb@kernel.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Dennis Gilmore <dennis@ausil.us>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Masahiro Yamada <masahiroy@kernel.org>
Cc: Naresh Kamboju <naresh.kamboju@linaro.org>
Cc: Nathan Chancellor <nathan@kernel.org>
Cc: Palmer Dabbelt <palmer@rivosinc.com>
Cc: Rich Felker <dalias@libc.org>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Tom Saeger <tom.saeger@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

s390: define RUNTIME_DISCARD_EXIT to fix link error with GNU ld < 2.36

commit a494398bde273143c2352dd373cad8211f7d94b2 upstream.

Nathan Chancellor reports that the s390 vmlinux fails to link with
GNU ld < 2.36 since commit 99cb0d917ffa ("arch: fix broken BuildID
for arm64 and riscv").

It happens for defconfig, or more specifically for CONFIG_EXPOLINE=y.

  $ s390x-linux-gnu-ld --version | head -n1
  GNU ld (GNU Binutils for Debian) 2.35.2
  $ make -s ARCH=s390 CROSS_COMPILE=s390x-linux-gnu- allnoconfig
  $ ./scripts/config -e CONFIG_EXPOLINE
  $ make -s ARCH=s390 CROSS_COMPILE=s390x-linux-gnu- olddefconfig
  $ make -s ARCH=s390 CROSS_COMPILE=s390x-linux-gnu-
  `.exit.text' referenced in section `.s390_return_reg' of drivers/base/dd.o: defined in discarded section `.exit.text' of drivers/base/dd.o
  make[1]: *** [scripts/Makefile.vmlinux:34: vmlinux] Error 1
  make: *** [Makefile:1252: vmlinux] Error 2

arch/s390/kernel/vmlinux.lds.S wants to keep EXIT_TEXT:

        .exit.text : {
                EXIT_TEXT
        }

But, at the same time, EXIT_TEXT is thrown away by DISCARD because
s390 does not define RUNTIME_DISCARD_EXIT.

I still do not understand why the latter wins after 99cb0d917ffa,
but defining RUNTIME_DISCARD_EXIT seems correct because the comment
line in arch/s390/kernel/vmlinux.lds.S says:

        /*
         * .exit.text is discarded at runtime, not link time,
         * to deal with references from __bug_table
         */

Nathan also found that binutils commit 21401fc7bf67 ("Duplicate output
sections in scripts") cured this issue, so we cannot reproduce it with
binutils 2.36+, but it is better to not rely on it.

Fixes: 99cb0d917ffa ("arch: fix broken BuildID for arm64 and riscv")
Link: https://lore.kernel.org/all/Y7Jal56f6UBh1abE@dev-arch.thelio-3990X/
Reported-by: Nathan Chancellor <nathan@kernel.org>
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
Link: https://lore.kernel.org/r/20230105031306.1455409-1-masahiroy@kernel.org
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Tom Saeger <tom.saeger@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

powerpc/vmlinux.lds: Don't discard .rela* for relocatable builds

commit 07b050f9290ee012a407a0f64151db902a1520f5 upstream.

Relocatable kernels must not discard relocations, they need to be
processed at runtime. As such they are included for CONFIG_RELOCATABLE
builds in the powerpc linker script (line 340).

However they are also unconditionally discarded later in the
script (line 414). Previously that worked because the earlier inclusion
superseded the discard.

However commit 99cb0d917ffa ("arch: fix broken BuildID for arm64 and
riscv") introduced an earlier use of DISCARD as part of the RO_DATA
macro (line 137). With binutils < 2.36 that causes the DISCARD
directives later in the script to be applied earlier, causing .rela* to
actually be discarded at link time, leading to build warnings and a
kernel that doesn't boot:

  ld: warning: discarding dynamic section .rela.init.rodata

Fix it by conditionally discarding .rela* only when CONFIG_RELOCATABLE
is disabled.

Fixes: 99cb0d917ffa ("arch: fix broken BuildID for arm64 and riscv")
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20230105132349.384666-2-mpe@ellerman.id.au
Signed-off-by: Tom Saeger <tom.saeger@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

powerpc/vmlinux.lds: Define RUNTIME_DISCARD_EXIT

commit 4b9880dbf3bdba3a7c56445137c3d0e30aaa0a40 upstream.

The powerpc linker script explicitly includes .exit.text, because
otherwise the link fails due to references from __bug_table and
__ex_table. The code is freed (discarded) at runtime along with
.init.text and data.

That has worked in the past despite powerpc not defining
RUNTIME_DISCARD_EXIT because DISCARDS appears late in the powerpc linker
script (line 410), and the explicit inclusion of .exit.text
earlier (line 280) supersedes the discard.

However commit 99cb0d917ffa ("arch: fix broken BuildID for arm64 and
riscv") introduced an earlier use of DISCARD as part of the RO_DATA
macro (line 136). With binutils < 2.36 that causes the DISCARD
directives later in the script to be applied earlier [1], causing
.exit.text to actually be discarded at link time, leading to build
errors:

  '.exit.text' referenced in section '__bug_table' of crypto/algboss.o: defined in
  discarded section '.exit.text' of crypto/algboss.o
  '.exit.text' referenced in section '__ex_table' of drivers/nvdimm/core.o: defined in
  discarded section '.exit.text' of drivers/nvdimm/core.o

Fix it by defining RUNTIME_DISCARD_EXIT, which causes the generic
DISCARDS macro to not include .exit.text at all.

1: https://lore.kernel.org/lkml/87fscp2v7k.fsf@igel.home/

Fixes: 99cb0d917ffa ("arch: fix broken BuildID for arm64 and riscv")
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20230105132349.384666-1-mpe@ellerman.id.au
Signed-off-by: Tom Saeger <tom.saeger@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

arch: fix broken BuildID for arm64 and riscv

commit 99cb0d917ffa1ab628bb67364ca9b162c07699b1 upstream.

Dennis Gilmore reports that the BuildID is missing in the arm64 vmlinux
since commit 994b7ac1697b ("arm64: remove special treatment for the
link order of head.o").

The issue is that the type of .notes section, which contains the BuildID,
changed from NOTES to PROGBITS.

Ard Biesheuvel figured out that whichever object gets linked first gets
to decide the type of a section. The PROGBITS type is the result of the
compiler emitting .note.GNU-stack as PROGBITS rather than NOTE.

While Ard provided a fix for arm64, I want to fix this globally because
the same issue is happening on riscv since commit 2348e6bf4421 ("riscv:
remove special treatment for the link order of head.o"). This problem
will happen in general for other architectures if they start to drop
unneeded entries from scripts/head-object-list.txt.

Discard .note.GNU-stack in include/asm-generic/vmlinux.lds.h.

Link: https://lore.kernel.org/lkml/CAABkxwuQoz1CTbyb57n0ZX65eSYiTonFCU8-LCQc=74D=xE=rA@mail.gmail.com/
Fixes: 994b7ac1697b ("arm64: remove special treatment for the link order of head.o")
Fixes: 2348e6bf4421 ("riscv: remove special treatment for the link order of head.o")
Reported-by: Dennis Gilmore <dennis@ausil.us>
Suggested-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
Acked-by: Palmer Dabbelt <palmer@rivosinc.com>
[Tom: stable backport 5.15.y, 5.10.y, 5.4.y]

Though the above "Fixes:" commits are not in this kernel, the conditions
which lead to a missing Build ID in arm64 vmlinux are similar.

Evidence points to these conditions:
1. ld version > 2.36 (exact binutils commit documented in a494398bde27)
2. first object which gets linked (head.o) has a PROGBITS .note.GNU-stack segment

These conditions can be observed when:
- 5.15.60+ OR 5.10.136+ OR 5.4.210+
- AND ld version > 2.36
- AND arch=arm64
- AND CONFIG_MODVERSIONS=y

There are notable differences in the vmlinux elf files produced
before(bad) and after(good) applying this series.

Good: p_type:PT_NOTE segment exists.
 Bad: p_type:PT_NOTE segment is missing.

Good: sh_name_str:.notes section has sh_type:SHT_NOTE
 Bad: sh_name_str:.notes section has sh_type:SHT_PROGBITS

`readelf -n` (as of v2.40) searches for Build Id
by processing only the very first note in sh_type:SHT_NOTE sections.

This was previously bisected to the stable backport of 0d362be5b142.
Follow-up experiments were discussed here: https://lore.kernel.org/all/20221221235413.xaisboqmr7dkqwn6@oracle.com/
which strongly hints at condition 2.
Signed-off-by: Tom Saeger <tom.saeger@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

ext4: block range must be validated before use in ext4_mb_clear_bb()

commit 1e1c2b86ef86a8477fd9b9a4f48a6bfe235606f6 upstream.

Block range to free is validated in ext4_free_blocks() using
ext4_inode_block_valid() and then it's passed to ext4_mb_clear_bb().
However in some situations on bigalloc file system the range might be
adjusted after the validation in ext4_free_blocks() which can lead to
troubles on corrupted file systems such as one found by syzkaller that
resulted in the following BUG

kernel BUG at fs/ext4/ext4.h:3319!
PREEMPT SMP NOPTI
CPU: 28 PID: 4243 Comm: repro Kdump: loaded Not tainted 5.19.0-rc6+ #1
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.15.0-1.fc35 04/01/2014
RIP: 0010:ext4_free_blocks+0x95e/0xa90
Call Trace:
 <TASK>
 ? lock_timer_base+0x61/0x80
 ? __es_remove_extent+0x5a/0x760
 ? __mod_timer+0x256/0x380
 ? ext4_ind_truncate_ensure_credits+0x90/0x220
 ext4_clear_blocks+0x107/0x1b0
 ext4_free_data+0x15b/0x170
 ext4_ind_truncate+0x214/0x2c0
 ? _raw_spin_unlock+0x15/0x30
 ? ext4_discard_preallocations+0x15a/0x410
 ? ext4_journal_check_start+0xe/0x90
 ? __ext4_journal_start_sb+0x2f/0x110
 ext4_truncate+0x1b5/0x460
 ? __ext4_journal_start_sb+0x2f/0x110
 ext4_evict_inode+0x2b4/0x6f0
 evict+0xd0/0x1d0
 ext4_enable_quotas+0x11f/0x1f0
 ext4_orphan_cleanup+0x3de/0x430
 ? proc_create_seq_private+0x43/0x50
 ext4_fill_super+0x295f/0x3ae0
 ? snprintf+0x39/0x40
 ? sget_fc+0x19c/0x330
 ? ext4_reconfigure+0x850/0x850
 get_tree_bdev+0x16d/0x260
 vfs_get_tree+0x25/0xb0
 path_mount+0x431/0xa70
 __x64_sys_mount+0xe2/0x120
 do_syscall_64+0x5b/0x80
 ? do_user_addr_fault+0x1e2/0x670
 ? exc_page_fault+0x70/0x170
 entry_SYSCALL_64_after_hwframe+0x46/0xb0
RIP: 0033:0x7fdf4e512ace

Fix it by making sure that the block range is properly validated before
used every time it changes in ext4_free_blocks() or ext4_mb_clear_bb().

Link: https://syzkaller.appspot.com/bug?id=5266d464285a03cee9dbfda7d2452a72c3c2ae7c
Reported-by: syzbot+15cd994e273307bf5cfa@syzkaller.appspotmail.com
Signed-off-by: Lukas Czerner <lczerner@redhat.com>
Cc: Tadeusz Struk <tadeusz.struk@linaro.org>
Tested-by: Tadeusz Struk <tadeusz.struk@linaro.org>
Link: https://lore.kernel.org/r/20220714165903.58260-1-lczerner@redhat.com
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Tudor Ambarus <tudor.ambarus@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

ext4: add strict range checks while freeing blocks

commit a00b482b82fb098956a5bed22bd7873e56f152f1 upstream.

Currently ext4_mb_clear_bb() & ext4_group_add_blocks() only checks
whether the given block ranges (which is to be freed) belongs to any FS
metadata blocks or not, of the block's respective block group.
But to detect any FS error early, it is better to add more strict
checkings in those functions which checks whether the given blocks
belongs to any critical FS metadata or not within system-zone.

Suggested-by: Jan Kara <jack@suse.cz>
Signed-off-by: Ritesh Harjani <riteshh@linux.ibm.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/ddd9143d064774e32d6364a99667817c6e8bfdc0.1644992610.git.riteshh@linux.ibm.com
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Tudor Ambarus <tudor.ambarus@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

ext4: add ext4_sb_block_valid() refactored out of ext4_inode_block_valid()

commit 6bc6c2bdf1baca6522b8d9ba976257d722423085 upstream.

This API will be needed at places where we don't have an inode
for e.g. while freeing blocks in ext4_group_add_blocks()

Suggested-by: Jan Kara <jack@suse.cz>
Signed-off-by: Ritesh Harjani <riteshh@linux.ibm.com>
Link: https://lore.kernel.org/r/dd34a236543ad5ae7123eeebe0cb69e6bdd44f34.1644992610.git.riteshh@linux.ibm.com
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Tudor Ambarus <tudor.ambarus@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

ext4: refactor ext4_free_blocks() to pull out ext4_mb_clear_bb()

commit 8ac3939db99f99667b8eb670cf4baf292896e72d upstream.

ext4_free_blocks() function became too long and confusing, this patch
just pulls out the ext4_mb_clear_bb() function logic from it
which clears the block bitmap and frees it.

No functionality change in this patch

Signed-off-by: Ritesh Harjani <riteshh@linux.ibm.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/22c30fbb26ba409cf8aa5f0c7912970272c459e8.1644992610.git.riteshh@linux.ibm.com
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Tudor Ambarus <tudor.ambarus@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

filelocks: use mount idmapping for setlease permission check

commit 42d0c4bdf753063b6eec55415003184d3ca24f6e upstream.

A user should be allowed to take out a lease via an idmapped mount if
the fsuid matches the mapped uid of the inode. generic_setlease() is
checking the unmapped inode uid, causing these operations to be denied.

Fix this by comparing against the mapped inode uid instead of the
unmapped uid.

Fixes: 9caccd41541a ("fs: introduce MOUNT_ATTR_IDMAP")
Cc: stable@vger.kernel.org
Signed-off-by: Seth Forshee (DigitalOcean) <sforshee@kernel.org>
Signed-off-by: Christian Brauner (Microsoft) <brauner@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

media: rc: gpio-ir-recv: add remove function

[ Upstream commit 30040818b338b8ebc956ce0ebd198f8d593586a6 ]

In case runtime PM is enabled, do runtime PM clean up to remove
cpu latency qos request, otherwise driver removal may have below
kernel dump:

[   19.463299] Unable to handle kernel NULL pointer dereference at
virtual address 0000000000000048
[   19.472161] Mem abort info:
[   19.474985]   ESR = 0x0000000096000004
[   19.478754]   EC = 0x25: DABT (current EL), IL = 32 bits
[   19.484081]   SET = 0, FnV = 0
[   19.487149]   EA = 0, S1PTW = 0
[   19.490361]   FSC = 0x04: level 0 translation fault
[   19.495256] Data abort info:
[   19.498149]   ISV = 0, ISS = 0x00000004
[   19.501997]   CM = 0, WnR = 0
[   19.504977] user pgtable: 4k pages, 48-bit VAs, pgdp=0000000049f81000
[   19.511432] [0000000000000048] pgd=0000000000000000,
p4d=0000000000000000
[   19.518245] Internal error: Oops: 0000000096000004 [#1] PREEMPT SMP
[   19.524520] Modules linked in: gpio_ir_recv(+) rc_core [last
unloaded: rc_core]
[   19.531845] CPU: 0 PID: 445 Comm: insmod Not tainted
6.2.0-rc1-00028-g2c397a46d47c #72
[   19.531854] Hardware name: FSL i.MX8MM EVK board (DT)
[   19.531859] pstate: 80000005 (Nzcv daif -PAN -UAO -TCO -DIT -SSBS
BTYPE=--)
[   19.551777] pc : cpu_latency_qos_remove_request+0x20/0x110
[   19.557277] lr : gpio_ir_recv_runtime_suspend+0x18/0x30
[gpio_ir_recv]
[   19.557294] sp : ffff800008ce3740
[   19.557297] x29: ffff800008ce3740 x28: 0000000000000000 x27:
ffff800008ce3d50
[   19.574270] x26: ffffc7e3e9cea100 x25: 00000000000f4240 x24:
ffffc7e3f9ef0e30
[   19.574284] x23: 0000000000000000 x22: ffff0061803820f4 x21:
0000000000000008
[   19.574296] x20: ffffc7e3fa75df30 x19: 0000000000000020 x18:
ffffffffffffffff
[   19.588570] x17: 0000000000000000 x16: ffffc7e3f9efab70 x15:
ffffffffffffffff
[   19.595712] x14: ffff800008ce37b8 x13: ffff800008ce37aa x12:
0000000000000001
[   19.602853] x11: 0000000000000001 x10: ffffcbe3ec0dff87 x9 :
0000000000000008
[   19.609991] x8 : 0101010101010101 x7 : 0000000000000000 x6 :
000000000f0bfe9f
[   19.624261] x5 : 00ffffffffffffff x4 : 0025ab8e00000000 x3 :
ffff006180382010
[   19.631405] x2 : ffffc7e3e9ce8030 x1 : ffffc7e3fc3eb810 x0 :
0000000000000020
[   19.638548] Call trace:
[   19.640995]  cpu_latency_qos_remove_request+0x20/0x110
[   19.646142]  gpio_ir_recv_runtime_suspend+0x18/0x30 [gpio_ir_recv]
[   19.652339]  pm_generic_runtime_suspend+0x2c/0x44
[   19.657055]  __rpm_callback+0x48/0x1dc
[   19.660807]  rpm_callback+0x6c/0x80
[   19.664301]  rpm_suspend+0x10c/0x640
[   19.667880]  rpm_idle+0x250/0x2d0
[   19.671198]  update_autosuspend+0x38/0xe0
[   19.675213]  pm_runtime_set_autosuspend_delay+0x40/0x60
[   19.680442]  gpio_ir_recv_probe+0x1b4/0x21c [gpio_ir_recv]
[   19.685941]  platform_probe+0x68/0xc0
[   19.689610]  really_probe+0xc0/0x3dc
[   19.693189]  __driver_probe_device+0x7c/0x190
[   19.697550]  driver_probe_device+0x3c/0x110
[   19.701739]  __driver_attach+0xf4/0x200
[   19.705578]  bus_for_each_dev+0x70/0xd0
[   19.709417]  driver_attach+0x24/0x30
[   19.712998]  bus_add_driver+0x17c/0x240
[   19.716834]  driver_register+0x78/0x130
[   19.720676]  __platform_driver_register+0x28/0x34
[   19.725386]  gpio_ir_recv_driver_init+0x20/0x1000 [gpio_ir_recv]
[   19.731404]  do_one_initcall+0x44/0x2ac
[   19.735243]  do_init_module+0x48/0x1d0
[   19.739003]  load_module+0x19fc/0x2034
[   19.742759]  __do_sys_finit_module+0xac/0x12c
[   19.747124]  __arm64_sys_finit_module+0x20/0x30
[   19.751664]  invoke_syscall+0x48/0x114
[   19.755420]  el0_svc_common.constprop.0+0xcc/0xec
[   19.760132]  do_el0_svc+0x38/0xb0
[   19.763456]  el0_svc+0x2c/0x84
[   19.766516]  el0t_64_sync_handler+0xf4/0x120
[   19.770789]  el0t_64_sync+0x190/0x194
[   19.774460] Code: 910003fd a90153f3 aa0003f3 91204021 (f9401400)
[   19.780556] ---[ end trace 0000000000000000 ]---

Signed-off-by: Li Jun <jun.li@nxp.com>
Signed-off-by: Sean Young <sean@mess.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

media: ov5640: Fix analogue gain control

[ Upstream commit afa4805799c1d332980ad23339fdb07b5e0cf7e0 ]

Gain control is badly documented in publicly available (including
leaked) documentation.

There is an AGC pre-gain in register 0x3a13, expressed as a 6-bit value
(plus an enable bit in bit 6). The driver hardcodes it to 0x43, which
one application note states is equal to x1.047. The documentation also
states that 0x40 is equel to x1.000. The pre-gain thus seems to be
expressed as in 1/64 increments, and thus ranges from x1.00 to x1.984.
What the pre-gain does is however unspecified.

There is then an AGC gain limit, in registers 0x3a18 and 0x3a19,
expressed as a 10-bit "real gain format" value. One application note
sets it to 0x00f8 and states it is equal to x15.5, so it appears to be
expressed in 1/16 increments, up to x63.9375.

The manual gain is stored in registers 0x350a and 0x350b, also as a
10-bit "real gain format" value. It is documented in the application
note as a Q6.4 values, up to x63.9375.

One version of the datasheet indicates that the sensor supports a
digital gain:

  The OV5640 supports 1/2/4 digital gain. Normally, the gain is
  controlled automatically by the automatic gain control (AGC) block.

It isn't clear how that would be controlled manually.

There appears to be no indication regarding whether the gain controlled
through registers 0x350a and 0x350b is an analogue gain only or also
includes digital gain. The words "real gain" don't necessarily mean
"combined analogue and digital gains". Some OmniVision sensors (such as
the OV8858) are documented as supoprting different formats for the gain
values, selectable through a register bit, and they are called "real
gain format" and "sensor gain format". For that sensor, we have (one of)
the gain registers documented as

  0x3503[2]=0, gain[7:0] is real gain format, where low 4 bits are
  fraction bits, for example, 0x10 is 1x gain, 0x28 is 2.5x gain

  If 0x3503[2]=1, gain[7:0] is sensor gain format, gain[7:4] is coarse
  gain, 00000: 1x, 00001: 2x, 00011: 4x, 00111: 8x, gain[7] is 1,
  gain[3:0] is fine gain. For example, 0x10 is 1x gain, 0x30 is 2x gain,
  0x70 is 4x gain

(The second part of the text makes little sense)

"Real gain" may thus refer to the combination of the coarse and fine
analogue gains as a single value.

The OV5640 0x350a and 0x350b registers thus appear to control analogue
gain. The driver incorrectly uses V4L2_CID_GAIN as V4L2 has a specific
control for analogue gain, V4L2_CID_ANALOGUE_GAIN. Use it.

If registers 0x350a and 0x350b are later found to control digital gain
as well, the driver could then restrict the range of the analogue gain
control value to lower than x64 and add a separate digital gain control.

Signed-off-by: Paul Elder <paul.elder@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Jacopo Mondi <jacopo.mondi@ideasonboard.com>
Reviewed-by: Jai Luthra <j-luthra@ti.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

scripts: handle BrokenPipeError for python scripts

[ Upstream commit 87c7ee67deb7fce9951a5f9d80641138694aad17 ]

In the follow-up of commit fb3041d61f68 ("kbuild: fix SIGPIPE error
message for AR=gcc-ar and AR=llvm-ar"), Kees Cook pointed out that
tools should _not_ catch their own SIGPIPEs [1] [2].

Based on his feedback, LLVM was fixed [3].

However, Python's default behavior is to show noisy bracktrace when
SIGPIPE is sent. So, scripts written in Python are basically in the
same situation as the buggy llvm tools.

Example:

  $ make -s allnoconfig
  $ make -s allmodconfig
  $ scripts/diffconfig .config.old .config | head -n1
  -ALIX n
  Traceback (most recent call last):
    File "/home/masahiro/linux/scripts/diffconfig", line 132, in <module>
      main()
    File "/home/masahiro/linux/scripts/diffconfig", line 130, in main
      print_config("+", config, None, b[config])
    File "/home/masahiro/linux/scripts/diffconfig", line 64, in print_config
      print("+%s %s" % (config, new_value))
  BrokenPipeError: [Errno 32] Broken pipe

Python documentation [4] notes how to make scripts die immediately and
silently:

  """
  Piping output of your program to tools like head(1) will cause a
  SIGPIPE signal to be sent to your process when the receiver of its
  standard output closes early. This results in an exception like
  BrokenPipeError: [Errno 32] Broken pipe. To handle this case,
  wrap your entry point to catch this exception as follows:

    import os
    import sys

    def main():
        try:
            # simulate large output (your code replaces this loop)
            for x in range(10000):
                print("y")
            # flush output here to force SIGPIPE to be triggered
            # while inside this try block.
            sys.stdout.flush()
        except BrokenPipeError:
            # Python flushes standard streams on exit; redirect remaining output
            # to devnull to avoid another BrokenPipeError at shutdown
            devnull = os.open(os.devnull, os.O_WRONLY)
            os.dup2(devnull, sys.stdout.fileno())
            sys.exit(1)  # Python exits with error code 1 on EPIPE

    if __name__ == '__main__':
        main()

  Do not set SIGPIPE’s disposition to SIG_DFL in order to avoid
  BrokenPipeError. Doing that would cause your program to exit
  unexpectedly whenever any socket connection is interrupted while
  your program is still writing to it.
  """

Currently, tools/perf/scripts/python/intel-pt-events.py seems to be the
only script that fixes the issue that way.

tools/perf/scripts/python/compaction-times.py uses another approach
signal.signal(signal.SIGPIPE, signal.SIG_DFL) but the Python
documentation clearly says "Don't do it".

I cannot fix all Python scripts since there are so many.
I fixed some in the scripts/ directory.

[1]: https://lore.kernel.org/all/202211161056.1B9611A@keescook/
[2]: https://github.com/llvm/llvm-project/issues/59037
[3]: https://github.com/llvm/llvm-project/commit/4787efa38066adb51e2c049499d25b3610c0877b
[4]: https://docs.python.org/3/library/signal.html#note-on-sigpipe

Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
Reviewed-by: Nick Desaulniers <ndesaulniers@google.com>
Reviewed-by: Nicolas Schier <nicolas@fjasle.eu>
Signed-off-by: Sasha Levin <sashal@kernel.org>

PCI: Add SolidRun vendor ID

[ Upstream commit db6c4dee4c104f50ed163af71c53bfdb878a8318 ]

Add SolidRun vendor ID to pci_ids.h

The vendor ID is used in 2 different source files, the SNET vDPA driver
and PCI quirks.

Signed-off-by: Alvaro Karsz <alvaro.karsz@solid-run.com>
Acked-by: Bjorn Helgaas <bhelgaas@google.com>
Message-Id: <20230110165638.123745-2-alvaro.karsz@solid-run.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

macintosh: windfarm: Use unsigned type for 1-bit bitfields

[ Upstream commit 748ea32d2dbd813d3bd958117bde5191182f909a ]

Clang warns:

  drivers/macintosh/windfarm_lm75_sensor.c:63:14: error: implicit truncation from 'int' to a one-bit wide bit-field changes value from 1 to -1 [-Werror,-Wsingle-bit-bitfield-constant-conversion]
                  lm->inited = 1;
                             ^ ~

  drivers/macintosh/windfarm_smu_sensors.c:356:19: error: implicit truncation from 'int' to a one-bit wide bit-field changes value from 1 to -1 [-Werror,-Wsingle-bit-bitfield-constant-conversion]
                  pow->fake_volts = 1;
                                  ^ ~
  drivers/macintosh/windfarm_smu_sensors.c:368:18: error: implicit truncation from 'int' to a one-bit wide bit-field changes value from 1 to -1 [-Werror,-Wsingle-bit-bitfield-constant-conversion]
                  pow->quadratic = 1;
                                 ^ ~

There is no bug here since no code checks the actual value of these
fields, just whether or not they are zero (boolean context), but this
can be easily fixed by switching to an unsigned type.

Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20230215-windfarm-wsingle-bit-bitfield-constant-conversion-v1-1-26415072e855@kernel.org
Signed-off-by: Sasha Levin <sashal@kernel.org>

alpha: fix R_ALPHA_LITERAL reloc for large modules

[ Upstream commit b6b17a8b3ecd878d98d5472a9023ede9e669ca72 ]

Previously, R_ALPHA_LITERAL relocations would overflow for large kernel
modules.

This was because the Alpha's apply_relocate_add was relying on the kernel's
module loader to have sorted the GOT towards the very end of the module as it
was mapped into memory in order to correctly assign the global pointer. While
this behavior would mostly work fine for small kernel modules, this approach
would overflow on kernel modules with large GOT's since the global pointer
would be very far away from the GOT, and thus, certain entries would be out of
range.

This patch fixes this by instead using the Tru64 behavior of assigning the
global pointer to be 32KB away from the start of the GOT. The change made
in this patch won't work for multi-GOT kernel modules as it makes the
assumption the module only has one GOT located at the beginning of .got,
although for the vast majority kernel modules, this should be fine. Of the
kernel modules that would previously result in a relocation error, none of
them, even modules like nouveau, have even come close to filling up a single
GOT, and they've all worked fine under this patch.

Signed-off-by: Edward Humes <aurxenon@lunos.org>
Signed-off-by: Matt Turner <mattst88@gmail.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

powerpc/kcsan: Exclude udelay to prevent recursive instrumentation

[ Upstream commit 2a7ce82dc46c591c9244057d89a6591c9639b9b9 ]

In order for KCSAN to increase its likelihood of observing a data race,
it sets a watchpoint on memory accesses and stalls, allowing for
detection of conflicting accesses by other kernel threads or interrupts.

Stalls are implemented by injecting a call to udelay in instrumented code.
To prevent recursive instrumentation, exclude udelay from being instrumented.

Signed-off-by: Rohan McLure <rmclure@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20230206021801.105268-3-rmclure@linux.ibm.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

powerpc/iommu: fix memory leak with using debugfs_lookup()

[ Upstream commit b505063910c134778202dfad9332dfcecb76bab3 ]

When calling debugfs_lookup() the result must have dput() called on it,
otherwise the memory will leak over time.  To make things simpler, just
call debugfs_lookup_and_remove() instead which handles all of the logic
at once.

Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20230202141919.2298821-1-gregkh@linuxfoundation.org
Signed-off-by: Sasha Levin <sashal@kernel.org>

MIPS: Fix a compilation issue

[ Upstream commit 109d587a4b4d7ccca2200ab1f808f43ae23e2585 ]

arch/mips/include/asm/mach-rc32434/pci.h:377:
cc1: error: result of ‘-117440512 << 16’ requires 44 bits to represent, but ‘int’ only has 32 bits [-Werror=shift-overflow=]

All bits in KORINA_STAT are already at the correct position, so there is
no addtional shift needed.

Signed-off-by: xurui <xurui@kylinos.cn>
Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Signed-off-by: Sasha Levin <sashal@kernel.org>

fs: use consistent setgid checks in is_sxid()

commit 8d84e39d76bd83474b26cb44f4b338635676e7e8 upstream.

Now that we made the VFS setgid checking consistent an inode can't be
marked security irrelevant even if the setgid bit is still set. Make
this function consistent with all other helpers.

Note that enforcing consistent setgid stripping checks for file
modification and mode- and ownership changes will cause the setgid bit
to be lost in more cases than useed to be the case. If an unprivileged
user wrote to a non-executable setgid file that they don't have
privilege over the setgid bit will be dropped. This will lead to
temporary failures in some xfstests until they have been updated.

Reported-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Christian Brauner (Microsoft) <brauner@kernel.org>
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Tested-by: Leah Rumancik <leah.rumancik@gmail.com>
Acked-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

attr: use consistent sgid stripping checks

commit ed5a7047d2011cb6b2bf84ceb6680124cc6a7d95 upstream.

[backport to 5.15.y, prior to vfsgid_t]

Currently setgid stripping in file_remove_privs()'s should_remove_suid()
helper is inconsistent with other parts of the vfs. Specifically, it only
raises ATTR_KILL_SGID if the inode is S_ISGID and S_IXGRP but not if the
inode isn't in the caller's groups and the caller isn't privileged over the
inode although we require this already in setattr_prepare() and
setattr_copy() and so all filesystem implement this requirement implicitly
because they have to use setattr_{prepare,copy}() anyway.

But the inconsistency shows up in setgid stripping bugs for overlayfs in
xfstests (e.g., generic/673, generic/683, generic/685, generic/686,
generic/687). For example, we test whether suid and setgid stripping works
correctly when performing various write-like operations as an unprivileged
user (fallocate, reflink, write, etc.):

echo "Test 1 - qa_user, non-exec file $verb"
setup_testfile
chmod a+rws $junk_file
commit_and_check "$qa_user" "$verb" 64k 64k

The test basically creates a file with 6666 permissions. While the file has
the S_ISUID and S_ISGID bits set it does not have the S_IXGRP set. On a
regular filesystem like xfs what will happen is:

sys_fallocate()
-> vfs_fallocate()
   -> xfs_file_fallocate()
      -> file_modified()
         -> __file_remove_privs()
            -> dentry_needs_remove_privs()
               -> should_remove_suid()
            -> __remove_privs()
               newattrs.ia_valid = ATTR_FORCE | kill;
               -> notify_change()
                  -> setattr_copy()

In should_remove_suid() we can see that ATTR_KILL_SUID is raised
unconditionally because the file in the test has S_ISUID set.

But we also see that ATTR_KILL_SGID won't be set because while the file
is S_ISGID it is not S_IXGRP (see above) which is a condition for
ATTR_KILL_SGID being raised.

So by the time we call notify_change() we have attr->ia_valid set to
ATTR_KILL_SUID | ATTR_FORCE. Now notify_change() sees that
ATTR_KILL_SUID is set and does:

ia_valid = attr->ia_valid |= ATTR_MODE
attr->ia_mode = (inode->i_mode & ~S_ISUID);

which means that when we call setattr_copy() later we will definitely
update inode->i_mode. Note that attr->ia_mode still contains S_ISGID.

Now we call into the filesystem's ->setattr() inode operation which will
end up calling setattr_copy(). Since ATTR_MODE is set we will hit:

if (ia_valid & ATTR_MODE) {
        umode_t mode = attr->ia_mode;
        vfsgid_t vfsgid = i_gid_into_vfsgid(mnt_userns, inode);
        if (!vfsgid_in_group_p(vfsgid) &&
            !capable_wrt_inode_uidgid(mnt_userns, inode, CAP_FSETID))
                mode &= ~S_ISGID;
        inode->i_mode = mode;
}

and since the caller in the test is neither capable nor in the group of the
inode the S_ISGID bit is stripped.

But assume the file isn't suid then ATTR_KILL_SUID won't be raised which
has the consequence that neither the setgid nor the suid bits are stripped
even though it should be stripped because the inode isn't in the caller's
groups and the caller isn't privileged over the inode.

If overlayfs is in the mix things become a bit more complicated and the bug
shows up more clearly. When e.g., ovl_setattr() is hit from
ovl_fallocate()'s call to file_remove_privs() then ATTR_KILL_SUID and
ATTR_KILL_SGID might be raised but because the check in notify_change() is
questioning the ATTR_KILL_SGID flag again by requiring S_IXGRP for it to be
stripped the S_ISGID bit isn't removed even though it should be stripped:

sys_fallocate()
-> vfs_fallocate()
   -> ovl_fallocate()
      -> file_remove_privs()
         -> dentry_needs_remove_privs()
            -> should_remove_suid()
         -> __remove_privs()
            newattrs.ia_valid = ATTR_FORCE | kill;
            -> notify_change()
               -> ovl_setattr()
                  // TAKE ON MOUNTER'S CREDS
                  -> ovl_do_notify_change()
                     -> notify_change()
                  // GIVE UP MOUNTER'S CREDS
     // TAKE ON MOUNTER'S CREDS
     -> vfs_fallocate()
        -> xfs_file_fallocate()
           -> file_modified()
              -> __file_remove_privs()
                 -> dentry_needs_remove_privs()
                    -> should_remove_suid()
                 -> __remove_privs()
                    newattrs.ia_valid = attr_force | kill;
                    -> notify_change()

The fix for all of this is to make file_remove_privs()'s
should_remove_suid() helper to perform the same checks as we already
require in setattr_prepare() and setattr_copy() and have notify_change()
not pointlessly requiring S_IXGRP again. It doesn't make any sense in the
first place because the caller must calculate the flags via
should_remove_suid() anyway which would raise ATTR_KILL_SGID.

While we're at it we move should_remove_suid() from inode.c to attr.c
where it belongs with the rest of the iattr helpers. Especially since it
returns ATTR_KILL_S{G,U}ID flags. We also rename it to
setattr_should_drop_suidgid() to better reflect that it indicates both
setuid and setgid bit removal and also that it returns attr flags.

Running xfstests with this doesn't report any regressions. We should really
try and use consistent checks.

Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Christian Brauner (Microsoft) <brauner@kernel.org>
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Tested-by: Leah Rumancik <leah.rumancik@gmail.com>
Acked-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

attr: add setattr_should_drop_sgid()

commit 72ae017c5451860443a16fb2a8c243bff3e396b8 upstream.

[backport to 5.15.y, prior to vfsgid_t]

The current setgid stripping logic during write and ownership change
operations is inconsistent and strewn over multiple places. In order to
consolidate it and make more consistent we'll add a new helper
setattr_should_drop_sgid(). The function retains the old behavior where
we remove the S_ISGID bit unconditionally when S_IXGRP is set but also
when it isn't set and the caller is neither in the group of the inode
nor privileged over the inode.

We will use this helper both in write operation permission removal such
as file_remove_privs() as well as in ownership change operations.

Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Christian Brauner (Microsoft) <brauner@kernel.org>
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Tested-by: Leah Rumancik <leah.rumancik@gmail.com>
Acked-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

fs: move should_remove_suid()

commit e243e3f94c804ecca9a8241b5babe28f35258ef4 upstream.

Move the helper from inode.c to attr.c. This keeps the the core of the
set{g,u}id stripping logic in one place when we add follow-up changes.
It is the better place anyway, since should_remove_suid() returns
ATTR_KILL_S{G,U}ID flags.

Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Christian Brauner (Microsoft) <brauner@kernel.org>
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Tested-by: Leah Rumancik <leah.rumancik@gmail.com>
Acked-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

attr: add in_group_or_capable()

commit 11c2a8700cdcabf9b639b7204a1e38e2a0b6798e upstream.

[backport to 5.15.y, prior to vfsgid_t]

In setattr_{copy,prepare}() we need to perform the same permission
checks to determine whether we need to drop the setgid bit or not.
Instead of open-coding it twice add a simple helper the encapsulates the
logic. We will reuse this helpers to make dropping the setgid bit during
write operations more consistent in a follow up patch.

Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Christian Brauner (Microsoft) <brauner@kernel.org>
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Tested-by: Leah Rumancik <leah.rumancik@gmail.com>
Acked-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

fs: move S_ISGID stripping into the vfs_*() helpers

commit 1639a49ccdce58ea248841ed9b23babcce6dbb0b upsream.

Move setgid handling out of individual filesystems and into the VFS
itself to stop the proliferation of setgid inheritance bugs.

Creating files that have both the S_IXGRP and S_ISGID bit raised in
directories that themselves have the S_ISGID bit set requires additional
privileges to avoid security issues.

When a filesystem creates a new inode it needs to take care that the
caller is either in the group of the newly created inode or they have
CAP_FSETID in their current user namespace and are privileged over the
parent directory of the new inode. If any of these two conditions is
true then the S_ISGID bit can be raised for an S_IXGRP file and if not
it needs to be stripped.

However, there are several key issues with the current implementation:

* S_ISGID stripping logic is entangled with umask stripping.

  If a filesystem doesn't support or enable POSIX ACLs then umask
  stripping is done directly in the vfs before calling into the
  filesystem.
  If the filesystem does support POSIX ACLs then unmask stripping may be
  done in the filesystem itself when calling posix_acl_create().

  Since umask stripping has an effect on S_ISGID inheritance, e.g., by
  stripping the S_IXGRP bit from the file to be created and all relevant
  filesystems have to call posix_acl_create() before inode_init_owner()
  where we currently take care of S_ISGID handling S_ISGID handling is
  order dependent. IOW, whether or not you get a setgid bit depends on
  POSIX ACLs and umask and in what order they are called.

  Note that technically filesystems are free to impose their own
  ordering between posix_acl_create() and inode_init_owner() meaning
  that there's additional ordering issues that influence S_SIGID
  inheritance.

* Filesystems that don't rely on inode_init_owner() don't get S_ISGID
  stripping logic.

  While that may be intentional (e.g. network filesystems might just
  defer setgid stripping to a server) it is often just a security issue.

This is not just ugly it's unsustainably messy especially since we do
still have bugs in this area years after the initial round of setgid
bugfixes.

So the current state is quite messy and while we won't be able to make
it completely clean as posix_acl_create() is still a filesystem specific
call we can improve the S_SIGD stripping situation quite a bit by
hoisting it out of inode_init_owner() and into the vfs creation
operations. This means we alleviate the burden for filesystems to handle
S_ISGID stripping correctly and can standardize the ordering between
S_ISGID and umask stripping in the vfs.

We add a new helper vfs_prepare_mode() so S_ISGID handling is now done
in the VFS before umask handling. This has S_ISGID handling is
unaffected unaffected by whether umask stripping is done by the VFS
itself (if no POSIX ACLs are supported or enabled) or in the filesystem
in posix_acl_create() (if POSIX ACLs are supported).

The vfs_prepare_mode() helper is called directly in vfs_*() helpers that
create new filesystem objects. We need to move them into there to make
sure that filesystems like overlayfs hat have callchains like:

sys_mknod()
-> do_mknodat(mode)
   -> .mknod = ovl_mknod(mode)
      -> ovl_create(mode)
         -> vfs_mknod(mode)

get S_ISGID stripping done when calling into lower filesystems via
vfs_*() creation helpers. Moving vfs_prepare_mode() into e.g.
vfs_mknod() takes care of that. This is in any case semantically cleaner
because S_ISGID stripping is VFS security requirement.

Security hooks so far have seen the mode with the umask applied but
without S_ISGID handling done. The relevant hooks are called outside of
vfs_*() creation helpers so by calling vfs_prepare_mode() from vfs_*()
helpers the security hooks would now see the mode without umask
stripping applied. For now we fix this by passing the mode with umask
settings applied to not risk any regressions for LSM hooks. IOW, nothing
changes for LSM hooks. It is worth pointing out that security hooks
never saw the mode that is seen by the filesystem when actually creating
the file. They have always been completely misplaced for that to work.

The following filesystems use inode_init_owner() and thus relied on
S_ISGID stripping: spufs, 9p, bfs, btrfs, ext2, ext4, f2fs, hfsplus,
hugetlbfs, jfs, minix, nilfs2, ntfs3, ocfs2, omfs, overlayfs, ramfs,
reiserfs, sysv, ubifs, udf, ufs, xfs, zonefs, bpf, tmpfs.

All of the above filesystems end up calling inode_init_owner() when new
filesystem objects are created through the ->mkdir(), ->mknod(),
->create(), ->tmpfile(), ->rename() inode operations.

Since directories always inherit the S_ISGID bit with the exception of
xfs when irix_sgid_inherit mode is turned on S_ISGID stripping doesn't
apply. The ->symlink() and ->link() inode operations trivially inherit
the mode from the target and the ->rename() inode operation inherits the
mode from the source inode. All other creation inode operations will get
S_ISGID handling via vfs_prepare_mode() when called from their relevant
vfs_*() helpers.

In addition to this there are filesystems which allow the creation of
filesystem objects through ioctl()s or - in the case of spufs -
circumventing the vfs in other ways. If filesystem objects are created
through ioctl()s the vfs doesn't know about it and can't apply regular
permission checking including S_ISGID logic. Therfore, a filesystem
relying on S_ISGID stripping in inode_init_owner() in their ioctl()
callpath will be affected by moving this logic into the vfs. We audited
those filesystems:

* btrfs allows the creation of filesystem objects through various
  ioctls(). Snapshot creation literally takes a snapshot and so the mode
  is fully preserved and S_ISGID stripping doesn't apply.

  Creating a new subvolum relies on inode_init_owner() in
  btrfs_new_subvol_inode() but only creates directories and doesn't
  raise S_ISGID.

* ocfs2 has a peculiar implementation of reflinks. In contrast to e.g.
  xfs and btrfs FICLONE/FICLONERANGE ioctl() that is only concerned with
  the actual extents ocfs2 uses a separate ioctl() that also creates the
  target file.

  Iow, ocfs2 circumvents the vfs entirely here and did indeed rely on
  inode_init_owner() to strip the S_ISGID bit. This is the only place
  where a filesystem needs to call mode_strip_sgid() directly but this
  is self-inflicted pain.

* spufs doesn't go through the vfs at all and doesn't use ioctl()s
  either. Instead it has a dedicated system call spufs_create() which
  allows the creation of filesystem objects. But spufs only creates
  directories and doesn't allo S_SIGID bits, i.e. it specifically only
  allows 0777 bits.

* bpf uses vfs_mkobj() but also doesn't allow S_ISGID bits to be created.

The patch will have an effect on ext2 when the EXT2_MOUNT_GRPID mount
option is used, on ext4 when the EXT4_MOUNT_GRPID mount option is used,
and on xfs when the XFS_FEAT_GRPID mount option is used. When any of
these filesystems are mounted with their respective GRPID option then
newly created files inherit the parent directories group
unconditionally. In these cases non of the filesystems call
inode_init_owner() and thus did never strip the S_ISGID bit for newly
created files. Moving this logic into the VFS means that they now get
the S_ISGID bit stripped. This is a user visible change. If this leads
to regressions we will either need to figure out a better way or we need
to revert. However, given the various setgid bugs that we found just in
the last two years this is a regression risk we should take.

Associated with this change is a new set of fstests to enforce the
semantics for all new filesystems.

Link: https://lore.kernel.org/ceph-devel/20220427092201.wvsdjbnc7b4dttaw@wittgenstein
Link: e014f37db1a2 ("xfs: use setattr_copy to set vfs inode attributes") [2]
Link: 01ea173e103e ("xfs: fix up non-directory creation in SGID directories") [3]
Link: fd84bfdddd16 ("ceph: fix up non-directory creation in SGID directories") [4]
Link: https://lore.kernel.org/r/1657779088-2242-3-git-send-email-xuyang2018.jy@fujitsu.com
Suggested-by: Dave Chinner <david@fromorbit.com>
Suggested-by: Christian Brauner (Microsoft) <brauner@kernel.org>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-and-Tested-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: Yang Xu <xuyang2018.jy@fujitsu.com>
[<brauner@kernel.org>: rewrote commit message]
Signed-off-by: Christian Brauner (Microsoft) <brauner@kernel.org>
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Tested-by: Leah Rumancik <leah.rumancik@gmail.com>
Acked-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

fs: add mode_strip_sgid() helper

commit 2b3416ceff5e6bd4922f6d1c61fb68113dd82302 upsream.

Add a dedicated helper to handle the setgid bit when creating a new file
in a setgid directory. This is a preparatory patch for moving setgid
stripping into the vfs. The patch contains no functional changes.

Currently the setgid stripping logic is open-coded directly in
inode_init_owner() and the individual filesystems are responsible for
handling setgid inheritance. Since this has proven to be brittle as
evidenced by old issues we uncovered over the last months (see [1] to
[3] below) we will try to move this logic into the vfs.

Link: e014f37db1a2 ("xfs: use setattr_copy to set vfs inode attributes") [1]
Link: 01ea173e103e ("xfs: fix up non-directory creation in SGID directories") [2]
Link: fd84bfdddd16 ("ceph: fix up non-directory creation in SGID directories") [3]
Link: https://lore.kernel.org/r/1657779088-2242-1-git-send-email-xuyang2018.jy@fujitsu.com
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christian Brauner (Microsoft) <brauner@kernel.org>
Reviewed-and-Tested-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: Yang Xu <xuyang2018.jy@fujitsu.com>
Signed-off-by: Christian Brauner (Microsoft) <brauner@kernel.org>
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Tested-by: Leah Rumancik <leah.rumancik@gmail.com>
Acked-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

xfs: set prealloc flag in xfs_alloc_file_space()

commit 0b02c8c0d75a738c98c35f02efb36217c170d78c upsream.

Now that we only call xfs_update_prealloc_flags() from
xfs_file_fallocate() in the case where we need to set the
preallocation flag, do this in xfs_alloc_file_space() where we
already have the inode joined into a transaction and get
rid of the call to xfs_update_prealloc_flags() from the fallocate
code.

This also means that we now correctly avoid setting the
XFS_DIFLAG_PREALLOC flag when xfs_is_always_cow_inode() is true, as
these inodes will never have preallocated extents.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Tested-by: Leah Rumancik <leah.rumancik@gmail.com>
Acked-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

xfs: fallocate() should call file_modified()

commit fbe7e520036583a783b13ff9744e35c2a329d9a4 upsream.

In XFS, we always update the inode change and modification time when
any fallocate() operation succeeds.  Furthermore, as various
fallocate modes can change the file contents (extending EOF,
punching holes, zeroing things, shifting extents), we should drop
file privileges like suid just like we do for a regular write().
There's already a VFS helper that figures all this out for us, so
use that.

The net effect of this is that we no longer drop suid/sgid if the
caller is root, but we also now drop file capabilities.

We also move the xfs_update_prealloc_flags() function so that it now
is only called by the scope that needs to set the the prealloc flag.

Based on a patch from Darrick Wong.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Tested-by: Leah Rumancik <leah.rumancik@gmail.com>
Acked-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

xfs: remove XFS_PREALLOC_SYNC

commit 472c6e46f589c26057596dcba160712a5b3e02c5 upstream.

[partial backport for dependency -
 xfs_ioc_space() still uses XFS_PREALLOC_SYNC]

Callers can acheive the same thing by calling xfs_log_force_inode()
after making their modifications. There is no need for
xfs_update_prealloc_flags() to do this.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Tested-by: Leah Rumancik <leah.rumancik@gmail.com>
Acked-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

xfs: use setattr_copy to set vfs inode attributes

commit e014f37db1a2d109afa750042ac4d69cf3e3d88e upsream.

Filipe Manana pointed out that XFS' behavior w.r.t. setuid/setgid
revocation isn't consistent with btrfs[1] or ext4.  Those two
filesystems use the VFS function setattr_copy to convey certain
attributes from struct iattr into the VFS inode structure.

Andrey Zhadchenko reported[2] that XFS uses the wrong user namespace to
decide if it should clear setgid and setuid on a file attribute update.
This is a second symptom of the problem that Filipe noticed.

XFS, on the other hand, open-codes setattr_copy in xfs_setattr_mode,
xfs_setattr_nonsize, and xfs_setattr_time.  Regrettably, setattr_copy is
/not/ a simple copy function; it contains additional logic to clear the
setgid bit when setting the mode, and XFS' version no longer matches.

The VFS implements its own setuid/setgid stripping logic, which
establishes consistent behavior.  It's a tad unfortunate that it's
scattered across notify_change, should_remove_suid, and setattr_copy but
XFS should really follow the Linux VFS.  Adapt XFS to use the VFS
functions and get rid of the old functions.

[1] https://lore.kernel.org/fstests/CAL3q7H47iNQ=Wmk83WcGB-KBJVOEtR9+qGczzCeXJ9Y2KCV25Q@mail.gmail.com/
[2] https://lore.kernel.org/linux-xfs/20220221182218.748084-1-andrey.zhadchenko@virtuozzo.com/

Fixes: 7fa294c8991c ("userns: Allow chown and setgid preservation")
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Christian Brauner <brauner@kernel.org>
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Tested-by: Leah Rumancik <leah.rumancik@gmail.com>
Acked-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

tpm/eventlog: Don't abort tpm_read_log on faulty ACPI address

[ Upstream commit 80a6c216b16d7f5c584d2148c2e4345ea4eb06ce ]

tpm_read_log_acpi() should return -ENODEV when no eventlog from the ACPI
table is found. If the firmware vendor includes an invalid log address
we are unable to map from the ACPI memory and tpm_read_log() returns -EIO
which would abort discovery of the eventlog.

Change the return value from -EIO to -ENODEV when acpi_os_map_iomem()
fails to map the event log.

The following hardware was used to test this issue:
    Framework Laptop (Pre-production)
    BIOS: INSYDE Corp, Revision: 3.2
    TPM Device: NTC, Firmware Revision: 7.2

Dump of the faulty ACPI TPM2 table:
    [000h 0000   4]                    Signature : "TPM2"    [Trusted Platform Module hardware interface Table]
    [004h 0004   4]                 Table Length : 0000004C
    [008h 0008   1]                     Revision : 04
    [009h 0009   1]                     Checksum : 2B
    [00Ah 0010   6]                       Oem ID : "INSYDE"
    [010h 0016   8]                 Oem Table ID : "TGL-ULT"
    [018h 0024   4]                 Oem Revision : 00000002
    [01Ch 0028   4]              Asl Compiler ID : "ACPI"
    [020h 0032   4]        Asl Compiler Revision : 00040000

    [024h 0036   2]               Platform Class : 0000
    [026h 0038   2]                     Reserved : 0000
    [028h 0040   8]              Control Address : 0000000000000000
    [030h 0048   4]                 Start Method : 06 [Memory Mapped I/O]

    [034h 0052  12]            Method Parameters : 00 00 00 00 00 00 00 00 00 00 00 00
    [040h 0064   4]           Minimum Log Length : 00010000
    [044h 0068   8]                  Log Address : 000000004053D000

Fixes: 0cf577a03f21 ("tpm: Fix handling of missing event log")
Tested-by: Erkki Eilonen <erkki@bearmetal.eu>
Signed-off-by: Morten Linderud <morten@linderud.pw>
Reviewed-by: Jarkko Sakkinen <jarkko@kernel.org>
Signed-off-by: Jarkko Sakkinen <jarkko@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

watch_queue: fix IOC_WATCH_QUEUE_SET_SIZE alloc error paths

[ Upstream commit 03e1d60e177eedbd302b77af4ea5e21b5a7ade31 ]

The watch_queue_set_size() allocation error paths return the ret value
set via the prior pipe_resize_ring() call, which will always be zero.

As a result, IOC_WATCH_QUEUE_SET_SIZE callers such as "keyctl watch"
fail to detect kernel wqueue->notes allocation failures and proceed to
KEYCTL_WATCH_KEY, with any notifications subsequently lost.

Fixes: c73be61cede58 ("pipe: Add general notification queue support")
Signed-off-by: David Disseldorp <ddiss@suse.de>
Signed-off-by: Christian Brauner (Microsoft) <brauner@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

staging: rtl8723bs: Fix key-store index handling

[ Upstream commit 05cbcc415c9b8c8bc4f9a09f8e03610a89042f03 ]

There are 2 issues with the key-store index handling

1. The non WEP key stores can store keys with indexes 0 - BIP_MAX_KEYID,
   this means that they should be an array with BIP_MAX_KEYID + 1
   entries. But some of the arrays where just BIP_MAX_KEYID entries
   big. While one other array was hardcoded to a size of 6 entries,
   instead of using the BIP_MAX_KEYID define.

2. The rtw_cfg80211_set_encryption() and wpa_set_encryption() functions
   index check where checking that the passed in key-index would fit
   inside both the WEP key store (which only has 4 entries) as well as
   in the non WEP key stores. This breaks any attempts to set non WEP
   keys with index 4 or 5.

Issue 2. specifically breaks wifi connection with some access points
which advertise PMF support. Without this fix connecting to these
access points fails with the following wpa_supplicant messages:

 nl80211: kernel reports: key addition failed
 wlan0: WPA: Failed to configure IGTK to the driver
 wlan0: RSN: Failed to configure IGTK
 wlan0: CTRL-EVENT-DISCONNECTED bssid=... reason=1 locally_generated=1

Fix 1. by using the right size for the key-stores. After this 2. can
safely be fixed by checking the right max-index value depending on the
used algorithm, fixing wifi not working with some PMF capable APs.

Cc: stable@vger.kernel.org
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20230306153512.162104-1-hdegoede@redhat.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>