Commit graph

1,308,922 commits

Author SHA1 Message Date
Gil Fine
ccd8450211 thunderbolt: Calculate DisplayPort tunnel bandwidth after DPRX capabilities read
According to USB4 Connection Manager guide, after DisplayPort tunnel was
setup, the DPRX capabilities read is performed by the DPTX. According to
VESA spec, this shall be completed within 5 seconds after the DisplayPort
tunnel was setup. Hence, if the bit: DPRX Capabilities Read Done, was
not set to '1' by this time, we timeout and fail calculating DisplayPort
tunnel consumed bandwidth.

Signed-off-by: Gil Fine <gil.fine@linux.intel.com>
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
2024-02-16 12:29:23 +02:00
Mika Westerberg
52a4490e89 thunderbolt: Reserve released DisplayPort bandwidth for a group for 10 seconds
The USB4 spec says that the Connection Manager should reserve the
bandwidth that is released in the same group for 10 seconds before it
can be shared with other groups. Add support for this. We also delay the
symmetric transition by that same 10 seconds to avoid any unnecessary
transitions (i.e if the released bandwidth is used by another
DisplayPort tunnel in the same group the link can stay asymmetric the
whole time).

Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
2024-02-16 12:29:23 +02:00
Mika Westerberg
769da9708a thunderbolt: Introduce tb_tunnel_direction_downstream()
This helper takes tunnel as parameter. Convert existing code to call
this where possible.

No functional changes.

Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
2024-02-16 12:29:23 +02:00
Mika Westerberg
fca0f40ac7 thunderbolt: Re-order bandwidth group functions
This is needed by the following patches so that we do not have to add
forward declaratations for any of these. Separating the move and the
actual changes also makes it easier to review the code.

No functional changes.

Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
2024-02-16 12:29:23 +02:00
Mika Westerberg
37f49727cc thunderbolt: Fail the failed bandwidth request properly
The USB4 spec says that if the Connection Manager writes Allocated_BW
that is smaller than Requested_BW, the DisplayPort IN adapter signals
this failure back to the DPTX (graphics driver). Implement this by
rewriting the same allocated bandwidth values back.

Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
2024-02-16 12:29:22 +02:00
Mika Westerberg
2ec67a48cf thunderbolt: Log an error if DPTX request is not cleared
This helps debugging issues around DisplayPort bandwidth allocation
mode.

Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
2024-02-16 12:29:22 +02:00
Mika Westerberg
c7857ee68e thunderbolt: Handle bandwidth allocation mode disable request
Graphics can disable DisplayPort bandwidth allocation mode as well so if
this make sure to reset the tunnel state accordingly.

Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
2024-02-16 12:29:22 +02:00
Mika Westerberg
d44b9827e7 thunderbolt: Re-calculate estimated bandwidth when allocation mode is enabled
When we program the initial bandwidth estimation the DPTX (graphics
driver) has not yet read the capabilities of the monitor so the values
used are the highest possible of the involved DisplayPort IN and OUT
adapters, not the actual monitor capabilities. To allow the graphics
more accurate bandwidth estimation re-calculate it once we receive the
bandwidth allocation mode enabled notification.

Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
2024-02-16 12:29:22 +02:00
Mika Westerberg
7b5e0bfc03 thunderbolt: Use DP_LOCAL_CAP for maximum bandwidth calculation
The DisplayPort IN adapter DP_LOCAL_CAP holds the aggregated
capabilities and gets updated after graphics side does the DPRX
capabilities read so we should use this to figure out the maximum
possible bandwidth for the DisplayPort tunnel.

While there make the variable name to match better what it is used for
and add kernel-doc comment to the function.

Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
2024-02-16 12:29:22 +02:00
Jakub Kicinski
166c2c8a6a net/sched: act_mirred: don't override retval if we already lost the skb
If we're redirecting the skb, and haven't called tcf_mirred_forward(),
yet, we need to tell the core to drop the skb by setting the retcode
to SHOT. If we have called tcf_mirred_forward(), however, the skb
is out of our hands and returning SHOT will lead to UaF.

Move the retval override to the error path which actually need it.

Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Fixes: e5cf1baf92 ("act_mirred: use TC_ACT_REINSERT when possible")
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2024-02-16 10:13:31 +00:00
Jakub Kicinski
52f671db18 net/sched: act_mirred: use the backlog for mirred ingress
The test Davide added in commit ca22da2fbd ("act_mirred: use the backlog
for nested calls to mirred ingress") hangs our testing VMs every 10 or so
runs, with the familiar tcp_v4_rcv -> tcp_v4_rcv deadlock reported by
lockdep.

The problem as previously described by Davide (see Link) is that
if we reverse flow of traffic with the redirect (egress -> ingress)
we may reach the same socket which generated the packet. And we may
still be holding its socket lock. The common solution to such deadlocks
is to put the packet in the Rx backlog, rather than run the Rx path
inline. Do that for all egress -> ingress reversals, not just once
we started to nest mirred calls.

In the past there was a concern that the backlog indirection will
lead to loss of error reporting / less accurate stats. But the current
workaround does not seem to address the issue.

Fixes: 53592b3640 ("net/sched: act_mirred: Implement ingress actions")
Cc: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Suggested-by: Davide Caratti <dcaratti@redhat.com>
Link: https://lore.kernel.org/netdev/33dc43f587ec1388ba456b4915c75f02a8aae226.1663945716.git.dcaratti@redhat.com/
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2024-02-16 10:13:31 +00:00
Geert Uytterhoeven
21bd52ea38 tcp: Spelling s/curcuit/circuit/
Fix a misspelling of "circuit".

Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: David S. Miller <davem@davemloft.net>
2024-02-16 10:12:00 +00:00
Randy Dunlap
a9f80df4f5 net: ethernet: adi: requires PHYLIB support
This driver uses functions that are supplied by the Kconfig symbol
PHYLIB, so select it to ensure that they are built as needed.

When CONFIG_ADIN1110=y and CONFIG_PHYLIB=m, there are multiple build
(linker) errors that are resolved by this Kconfig change:

   ld: drivers/net/ethernet/adi/adin1110.o: in function `adin1110_net_open':
   drivers/net/ethernet/adi/adin1110.c:933: undefined reference to `phy_start'
   ld: drivers/net/ethernet/adi/adin1110.o: in function `adin1110_probe_netdevs':
   drivers/net/ethernet/adi/adin1110.c:1603: undefined reference to `get_phy_device'
   ld: drivers/net/ethernet/adi/adin1110.c:1609: undefined reference to `phy_connect'
   ld: drivers/net/ethernet/adi/adin1110.o: in function `adin1110_disconnect_phy':
   drivers/net/ethernet/adi/adin1110.c:1226: undefined reference to `phy_disconnect'
   ld: drivers/net/ethernet/adi/adin1110.o: in function `devm_mdiobus_alloc':
   include/linux/phy.h:455: undefined reference to `devm_mdiobus_alloc_size'
   ld: drivers/net/ethernet/adi/adin1110.o: in function `adin1110_register_mdiobus':
   drivers/net/ethernet/adi/adin1110.c:529: undefined reference to `__devm_mdiobus_register'
   ld: drivers/net/ethernet/adi/adin1110.o: in function `adin1110_net_stop':
   drivers/net/ethernet/adi/adin1110.c:958: undefined reference to `phy_stop'
   ld: drivers/net/ethernet/adi/adin1110.o: in function `adin1110_disconnect_phy':
   drivers/net/ethernet/adi/adin1110.c:1226: undefined reference to `phy_disconnect'
   ld: drivers/net/ethernet/adi/adin1110.o: in function `adin1110_adjust_link':
   drivers/net/ethernet/adi/adin1110.c:1077: undefined reference to `phy_print_status'
   ld: drivers/net/ethernet/adi/adin1110.o: in function `adin1110_ioctl':
   drivers/net/ethernet/adi/adin1110.c:790: undefined reference to `phy_do_ioctl'
   ld: drivers/net/ethernet/adi/adin1110.o:(.rodata+0xf60): undefined reference to `phy_ethtool_get_link_ksettings'
   ld: drivers/net/ethernet/adi/adin1110.o:(.rodata+0xf68): undefined reference to `phy_ethtool_set_link_ksettings'

Fixes: bc93e19d08 ("net: ethernet: adi: Add ADIN1110 support")
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202402070626.eZsfVHG5-lkp@intel.com/
Cc: Lennart Franzen <lennart@lfdomain.com>
Cc: Alexandru Tachici <alexandru.tachici@analog.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Eric Dumazet <edumazet@google.com>
Cc: Jakub Kicinski <kuba@kernel.org>
Cc: Paolo Abeni <pabeni@redhat.com>
Cc: netdev@vger.kernel.org
Reviewed-by: Nuno Sa <nuno.sa@analog.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2024-02-16 09:43:33 +00:00
Kuniyuki Iwashima
66b60b0c8c dccp/tcp: Unhash sk from ehash for tb2 alloc failure after check_estalblished().
syzkaller reported a warning [0] in inet_csk_destroy_sock() with no
repro.

  WARN_ON(inet_sk(sk)->inet_num && !inet_csk(sk)->icsk_bind_hash);

However, the syzkaller's log hinted that connect() failed just before
the warning due to FAULT_INJECTION.  [1]

When connect() is called for an unbound socket, we search for an
available ephemeral port.  If a bhash bucket exists for the port, we
call __inet_check_established() or __inet6_check_established() to check
if the bucket is reusable.

If reusable, we add the socket into ehash and set inet_sk(sk)->inet_num.

Later, we look up the corresponding bhash2 bucket and try to allocate
it if it does not exist.

Although it rarely occurs in real use, if the allocation fails, we must
revert the changes by check_established().  Otherwise, an unconnected
socket could illegally occupy an ehash entry.

Note that we do not put tw back into ehash because sk might have
already responded to a packet for tw and it would be better to free
tw earlier under such memory presure.

[0]:
WARNING: CPU: 0 PID: 350830 at net/ipv4/inet_connection_sock.c:1193 inet_csk_destroy_sock (net/ipv4/inet_connection_sock.c:1193)
Modules linked in:
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.16.0-0-gd239552ce722-prebuilt.qemu.org 04/01/2014
RIP: 0010:inet_csk_destroy_sock (net/ipv4/inet_connection_sock.c:1193)
Code: 41 5c 41 5d 41 5e e9 2d 4a 3d fd e8 28 4a 3d fd 48 89 ef e8 f0 cd 7d ff 5b 5d 41 5c 41 5d 41 5e e9 13 4a 3d fd e8 0e 4a 3d fd <0f> 0b e9 61 fe ff ff e8 02 4a 3d fd 4c 89 e7 be 03 00 00 00 e8 05
RSP: 0018:ffffc9000b21fd38 EFLAGS: 00010293
RAX: 0000000000000000 RBX: 0000000000009e78 RCX: ffffffff840bae40
RDX: ffff88806e46c600 RSI: ffffffff840bb012 RDI: ffff88811755cca8
RBP: ffff88811755c880 R08: 0000000000000003 R09: 0000000000000000
R10: 0000000000009e78 R11: 0000000000000000 R12: ffff88811755c8e0
R13: ffff88811755c892 R14: ffff88811755c918 R15: 0000000000000000
FS:  00007f03e5243800(0000) GS:ffff88811ae00000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000001b32f21000 CR3: 0000000112ffe001 CR4: 0000000000770ef0
PKRU: 55555554
Call Trace:
 <TASK>
 ? inet_csk_destroy_sock (net/ipv4/inet_connection_sock.c:1193)
 dccp_close (net/dccp/proto.c:1078)
 inet_release (net/ipv4/af_inet.c:434)
 __sock_release (net/socket.c:660)
 sock_close (net/socket.c:1423)
 __fput (fs/file_table.c:377)
 __fput_sync (fs/file_table.c:462)
 __x64_sys_close (fs/open.c:1557 fs/open.c:1539 fs/open.c:1539)
 do_syscall_64 (arch/x86/entry/common.c:52 arch/x86/entry/common.c:83)
 entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:129)
RIP: 0033:0x7f03e53852bb
Code: 03 00 00 00 0f 05 48 3d 00 f0 ff ff 77 41 c3 48 83 ec 18 89 7c 24 0c e8 43 c9 f5 ff 8b 7c 24 0c 41 89 c0 b8 03 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 35 44 89 c7 89 44 24 0c e8 a1 c9 f5 ff 8b 44
RSP: 002b:00000000005dfba0 EFLAGS: 00000293 ORIG_RAX: 0000000000000003
RAX: ffffffffffffffda RBX: 0000000000000004 RCX: 00007f03e53852bb
RDX: 0000000000000002 RSI: 0000000000000002 RDI: 0000000000000003
RBP: 0000000000000000 R08: 0000000000000000 R09: 000000000000167c
R10: 0000000008a79680 R11: 0000000000000293 R12: 00007f03e4e43000
R13: 00007f03e4e43170 R14: 00007f03e4e43178 R15: 00007f03e4e43170
 </TASK>

[1]:
FAULT_INJECTION: forcing a failure.
name failslab, interval 1, probability 0, space 0, times 0
CPU: 0 PID: 350833 Comm: syz-executor.1 Not tainted 6.7.0-12272-g2121c43f88f5 #9
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.16.0-0-gd239552ce722-prebuilt.qemu.org 04/01/2014
Call Trace:
 <TASK>
 dump_stack_lvl (lib/dump_stack.c:107 (discriminator 1))
 should_fail_ex (lib/fault-inject.c:52 lib/fault-inject.c:153)
 should_failslab (mm/slub.c:3748)
 kmem_cache_alloc (mm/slub.c:3763 mm/slub.c:3842 mm/slub.c:3867)
 inet_bind2_bucket_create (net/ipv4/inet_hashtables.c:135)
 __inet_hash_connect (net/ipv4/inet_hashtables.c:1100)
 dccp_v4_connect (net/dccp/ipv4.c:116)
 __inet_stream_connect (net/ipv4/af_inet.c:676)
 inet_stream_connect (net/ipv4/af_inet.c:747)
 __sys_connect_file (net/socket.c:2048 (discriminator 2))
 __sys_connect (net/socket.c:2065)
 __x64_sys_connect (net/socket.c:2072)
 do_syscall_64 (arch/x86/entry/common.c:52 arch/x86/entry/common.c:83)
 entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:129)
RIP: 0033:0x7f03e5284e5d
Code: ff c3 66 2e 0f 1f 84 00 00 00 00 00 90 f3 0f 1e fa 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 73 9f 1b 00 f7 d8 64 89 01 48
RSP: 002b:00007f03e4641cc8 EFLAGS: 00000246 ORIG_RAX: 000000000000002a
RAX: ffffffffffffffda RBX: 00000000004bbf80 RCX: 00007f03e5284e5d
RDX: 0000000000000010 RSI: 0000000020000000 RDI: 0000000000000003
RBP: 00000000004bbf80 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000001
R13: 000000000000000b R14: 00007f03e52e5530 R15: 0000000000000000
 </TASK>

Reported-by: syzkaller <syzkaller@googlegroups.com>
Fixes: 28044fc1d4 ("net: Add a bhash2 table hashed by port and address")
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2024-02-16 09:41:54 +00:00
David S. Miller
82a678e22d Merge branch 'bridge-mdb-events'
Tobias Waldekranz says:

====================
net: bridge: switchdev: Ensure MDB events are delivered exactly once

When a device is attached to a bridge, drivers will request a replay
of objects that were created before the device joined the bridge, that
are still of interest to the joining port. Typical examples include
FDB entries and MDB memberships on other ports ("foreign interfaces")
or on the bridge itself.

Conversely when a device is detached, the bridge will synthesize
deletion events for all those objects that are still live, but no
longer applicable to the device in question.

This series eliminates two races related to the synching and
unsynching phases of a bridge's MDB with a joining or leaving device,
that would cause notifications of such objects to be either delivered
twice (1/2), or not at all (2/2).

A similar race to the one solved by 1/2 still remains for the
FDB. This is much harder to solve, due to the lockless operation of
the FDB's rhashtable, and is therefore knowingly left out of this
series.

v1 -> v2:
- Squash the previously separate addition of
  switchdev_port_obj_act_is_deferred into first consumer.
- Use ether_addr_equal to compare MAC addresses.
- Document switchdev_port_obj_act_is_deferred (renamed from
  switchdev_port_obj_is_deferred in v1, to indicate that we also match
  on the action).
- Delay allocations of MDB objects until we know they're needed.
- Use non-RCU version of the hash list iterator, now that the MDB is
  not scanned while holding the RCU read lock.
- Add Fixes tag to commit message

v2 -> v3:
- Fix unlocking in error paths
- Access RCU protected port list via mlock_dereference, since MDB is
  guaranteed to remain constant for the duration of the scan.

v3 -> v4:
- Limit the search for exiting deferred events in 1/2 to only apply to
  additions, since the problem does not exist in the deletion case.
- Add 2/2, to plug a related race when unoffloading an indirectly
  associated device.

v4 -> v5:
- Fix grammatical errors in kerneldoc of
  switchdev_port_obj_act_is_deferred
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
2024-02-16 09:36:38 +00:00
Tobias Waldekranz
f7a70d650b net: bridge: switchdev: Ensure deferred event delivery on unoffload
When unoffloading a device, it is important to ensure that all
relevant deferred events are delivered to it before it disassociates
itself from the bridge.

Before this change, this was true for the normal case when a device
maps 1:1 to a net_bridge_port, i.e.

   br0
   /
swp0

When swp0 leaves br0, the call to switchdev_deferred_process() in
del_nbp() makes sure to process any outstanding events while the
device is still associated with the bridge.

In the case when the association is indirect though, i.e. when the
device is attached to the bridge via an intermediate device, like a
LAG...

    br0
    /
  lag0
  /
swp0

...then detaching swp0 from lag0 does not cause any net_bridge_port to
be deleted, so there was no guarantee that all events had been
processed before the device disassociated itself from the bridge.

Fix this by always synchronously processing all deferred events before
signaling completion of unoffloading back to the driver.

Fixes: 4e51bf44a0 ("net: bridge: move the switchdev object replay helpers to "push" mode")
Signed-off-by: Tobias Waldekranz <tobias@waldekranz.com>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2024-02-16 09:36:37 +00:00
Tobias Waldekranz
dc489f8625 net: bridge: switchdev: Skip MDB replays of deferred events on offload
Before this change, generation of the list of MDB events to replay
would race against the creation of new group memberships, either from
the IGMP/MLD snooping logic or from user configuration.

While new memberships are immediately visible to walkers of
br->mdb_list, the notification of their existence to switchdev event
subscribers is deferred until a later point in time. So if a replay
list was generated during a time that overlapped with such a window,
it would also contain a replay of the not-yet-delivered event.

The driver would thus receive two copies of what the bridge internally
considered to be one single event. On destruction of the bridge, only
a single membership deletion event was therefore sent. As a
consequence of this, drivers which reference count memberships (at
least DSA), would be left with orphan groups in their hardware
database when the bridge was destroyed.

This is only an issue when replaying additions. While deletion events
may still be pending on the deferred queue, they will already have
been removed from br->mdb_list, so no duplicates can be generated in
that scenario.

To a user this meant that old group memberships, from a bridge in
which a port was previously attached, could be reanimated (in
hardware) when the port joined a new bridge, without the new bridge's
knowledge.

For example, on an mv88e6xxx system, create a snooping bridge and
immediately add a port to it:

    root@infix-06-0b-00:~$ ip link add dev br0 up type bridge mcast_snooping 1 && \
    > ip link set dev x3 up master br0

And then destroy the bridge:

    root@infix-06-0b-00:~$ ip link del dev br0
    root@infix-06-0b-00:~$ mvls atu
    ADDRESS             FID  STATE      Q  F  0  1  2  3  4  5  6  7  8  9  a
    DEV:0 Marvell 88E6393X
    33:33:00:00:00:6a     1  static     -  -  0  .  .  .  .  .  .  .  .  .  .
    33:33:ff:87:e4:3f     1  static     -  -  0  .  .  .  .  .  .  .  .  .  .
    ff:ff:ff:ff:ff:ff     1  static     -  -  0  1  2  3  4  5  6  7  8  9  a
    root@infix-06-0b-00:~$

The two IPv6 groups remain in the hardware database because the
port (x3) is notified of the host's membership twice: once via the
original event and once via a replay. Since only a single delete
notification is sent, the count remains at 1 when the bridge is
destroyed.

Then add the same port (or another port belonging to the same hardware
domain) to a new bridge, this time with snooping disabled:

    root@infix-06-0b-00:~$ ip link add dev br1 up type bridge mcast_snooping 0 && \
    > ip link set dev x3 up master br1

All multicast, including the two IPv6 groups from br0, should now be
flooded, according to the policy of br1. But instead the old
memberships are still active in the hardware database, causing the
switch to only forward traffic to those groups towards the CPU (port
0).

Eliminate the race in two steps:

1. Grab the write-side lock of the MDB while generating the replay
   list.

This prevents new memberships from showing up while we are generating
the replay list. But it leaves the scenario in which a deferred event
was already generated, but not delivered, before we grabbed the
lock. Therefore:

2. Make sure that no deferred version of a replay event is already
   enqueued to the switchdev deferred queue, before adding it to the
   replay list, when replaying additions.

Fixes: 4f2673b3a2 ("net: bridge: add helper to replay port and host-joined mdb entries")
Signed-off-by: Tobias Waldekranz <tobias@waldekranz.com>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2024-02-16 09:36:37 +00:00
Alexander Gordeev
b4ea9b6a18 net/iucv: fix the allocation size of iucv_path_table array
iucv_path_table is a dynamically allocated array of pointers to
struct iucv_path items. Yet, its size is calculated as if it was
an array of struct iucv_path items.

Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
Reviewed-by: Alexandra Winter <wintera@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2024-02-16 09:25:09 +00:00
David S. Miller
d70a2a45b4 Merge branch 'ionic-xdp-support'
Shannon Nelson says:

====================
ionic: add XDP support

This patchset is new support in ionic for XDP processing,
including basic XDP on Rx packets, TX and REDIRECT, and frags
for jumbo frames.

Since ionic has not yet been converted to use the page_pool APIs,
this uses the simple MEM_TYPE_PAGE_ORDER0 buffering.  There are plans
to convert the driver in the near future.

v4:
 - removed "inline" from short utility functions
 - changed to use "goto err_out" in ionic_xdp_register_rxq_info()
 - added "continue" to reduce nesting in ionic_xdp_queues_config()
 - used xdp_prog in ionic_rx_clean() to flag whether or not to sync
   the rx buffer after calling ionix_xdp_run()
 - swapped order of XDP_TX and XDP_REDIRECT cases in ionic_xdp_run()
   to make patch 6 a little cleaner

v3:
https://lore.kernel.org/netdev/20240210004827.53814-1-shannon.nelson@amd.com/
 - removed budget==0 patch, sent it separately to net

v2:
https://lore.kernel.org/netdev/20240208005725.65134-1-shannon.nelson@amd.com/
 - added calls to txq_trans_cond_update()
 - added a new patch to catch NAPI budget==0

v1:
https://lore.kernel.org/netdev/20240130013042.11586-1-shannon.nelson@amd.com/

RFC:
https://lore.kernel.org/netdev/20240118192500.58665-1-shannon.nelson@amd.com/
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
2024-02-16 08:48:09 +00:00
Shannon Nelson
5377805dc1 ionic: implement xdp frags support
Add support for using scatter-gather / frags in XDP in both
Rx and Tx paths.

Co-developed-by: Brett Creeley <brett.creeley@amd.com>
Signed-off-by: Brett Creeley <brett.creeley@amd.com>
Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2024-02-16 08:48:08 +00:00
Shannon Nelson
26f5726a78 ionic: add ndo_xdp_xmit
When our ndo_xdp_xmit is called we mark the buffer with
XDP_REDIRECT so we know to return it to the XDP stack for
cleaning.

Co-developed-by: Brett Creeley <brett.creeley@amd.com>
Signed-off-by: Brett Creeley <brett.creeley@amd.com>
Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2024-02-16 08:48:08 +00:00
Shannon Nelson
587fc3f0dc ionic: Add XDP_REDIRECT support
The XDP_REDIRECT packets are given to the XDP stack and
we drop the use of the related page: it will get freed
by the driver that ends up doing the Tx.  Because we have
some hardware configurations with limited queue resources,
we use the existing datapath Tx queues rather than creating
and managing a separate set of xdp_tx queues.

Co-developed-by: Brett Creeley <brett.creeley@amd.com>
Signed-off-by: Brett Creeley <brett.creeley@amd.com>
Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2024-02-16 08:48:08 +00:00
Shannon Nelson
8eeed8373e ionic: Add XDP_TX support
The XDP_TX packets get fed back into the Rx queue's partnered
Tx queue as an xdp_frame.

Co-developed-by: Brett Creeley <brett.creeley@amd.com>
Signed-off-by: Brett Creeley <brett.creeley@amd.com>
Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2024-02-16 08:48:08 +00:00
Shannon Nelson
f81da39bf4 ionic: Add XDP packet headroom
If an xdp program is loaded, add headroom at the beginning
of the frame to allow for editing and insertions that an XDP
program might need room for, and tailroom used later for XDP
frame tracking.  These are only needed in the first Rx buffer
in a packet, not for any trailing frags.

Co-developed-by: Brett Creeley <brett.creeley@amd.com>
Signed-off-by: Brett Creeley <brett.creeley@amd.com>
Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2024-02-16 08:48:08 +00:00
Shannon Nelson
180e35cdf0 ionic: add initial framework for XDP support
Set up the basics for running Rx packets through XDP programs.
Add new queue setup and teardown steps for adding/removing an
XDP program, and add the call to run the XDP on a packet.

The XDP frame size needs to be the MTU plus standard ethernet
header, plus head room for XDP scribblings and tail room for a
struct skb_shared_info.  Also, at this point, we don't support
XDP frags, only a single contiguous Rx buffer.  This means
that our page splitting is not very useful, so when XDP is in
use we need to use the full Rx buffer size and not do sharing.

Co-developed-by: Brett Creeley <brett.creeley@amd.com>
Signed-off-by: Brett Creeley <brett.creeley@amd.com>
Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2024-02-16 08:48:08 +00:00
Shannon Nelson
d67ee210da ionic: use dma range APIs
Convert Rx datapath handling to use the DMA range APIs
in preparation for adding XDP handling.

Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Reviewed-by: Brett Creeley <brett.creeley@amd.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2024-02-16 08:48:08 +00:00
Shannon Nelson
97538c146c ionic: add helpers for accessing buffer info
These helpers clean up some of the code around DMA mapping
and other buffer references, and will be used in the next
few patches for the XDP support.

Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Reviewed-by: Brett Creeley <brett.creeley@amd.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2024-02-16 08:48:07 +00:00
Shannon Nelson
c699f35d65 ionic: set adminq irq affinity
We claim to have the AdminQ on our irq0 and thus cpu id 0,
but we need to be sure we set the affinity hint to try to
keep it there.

Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Reviewed-by: Brett Creeley <brett.creeley@amd.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2024-02-16 08:48:07 +00:00
Alexander Gordeev
2210c5485e net/iucv: fix virtual vs physical address confusion
Fix virtual vs physical address confusion. This does not fix a bug
since virtual and physical address spaces are currently the same.

Acked-by: Alexandra Winter <wintera@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2024-02-16 08:45:17 +00:00
Shrikanth Hegde
8cec3dd9e5 sched/core: Simplify code by removing duplicate #ifdefs
There's a few cases of nested #ifdefs in the scheduler code
that can be simplified:

  #ifdef DEFINE_A
  ...code block...
    #ifdef DEFINE_A       <-- This is a duplicate.
    ...code block...
    #endif
  #else
    #ifndef DEFINE_A     <-- This is also duplicate.
    ...code block...
    #endif
  #endif

More details about the script and methods used to find these code
patterns can be found at:

  https://lore.kernel.org/all/20240118080326.13137-1-sshegde@linux.ibm.com/

No change in functionality intended.

[ mingo: Clarified the changelog. ]

Signed-off-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20240216061433.535522-1-sshegde@linux.ibm.com
2024-02-16 09:37:15 +01:00
Mika Westerberg
d3d17e23d1 thunderbolt: Fix NULL pointer dereference in tb_port_update_credits()
Olliver reported that his system crashes when plugging in Thunderbolt 1
device:

 BUG: kernel NULL pointer dereference, address: 0000000000000020
 #PF: supervisor read access in kernel mode
 #PF: error_code(0x0000) - not-present page
 PGD 0 P4D 0
 Oops: 0000 [#1] PREEMPT SMP NOPTI
 RIP: 0010:tb_port_do_update_credits+0x1b/0x130 [thunderbolt]
 Call Trace:
  <TASK>
  ? __die+0x23/0x70
  ? page_fault_oops+0x171/0x4e0
  ? exc_page_fault+0x7f/0x180
  ? asm_exc_page_fault+0x26/0x30
  ? tb_port_do_update_credits+0x1b/0x130
  ? tb_switch_update_link_attributes+0x83/0xd0
  tb_switch_add+0x7a2/0xfe0
  tb_scan_port+0x236/0x6f0
  tb_handle_hotplug+0x6db/0x900
  process_one_work+0x171/0x340
  worker_thread+0x27b/0x3a0
  ? __pfx_worker_thread+0x10/0x10
  kthread+0xe5/0x120
  ? __pfx_kthread+0x10/0x10
  ret_from_fork+0x31/0x50
  ? __pfx_kthread+0x10/0x10
  ret_from_fork_asm+0x1b/0x30
  </TASK>

This is due the fact that some Thunderbolt 1 devices only have one lane
adapter. Fix this by checking for the lane 1 before we read its credits.

Reported-by: Olliver Schinagl <oliver@schinagl.nl>
Closes: https://lore.kernel.org/linux-usb/c24c7882-6254-4e68-8f22-f3e8f65dc84f@schinagl.nl/
Fixes: 81af2952e6 ("thunderbolt: Add support for asymmetric link")
Cc: stable@vger.kernel.org
Cc: Gil Fine <gil.fine@linux.intel.com>
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
2024-02-16 10:36:51 +02:00
David S. Miller
fc136368fd Merge branch 'ravb-rutime-PM-support'
Claudiu Beznea says:

====================
net: ravb: Add runtime PM support (part 2)

Series adds runtime PM support for the ravb driver. This is a continuation
of [1].

There are 5 more preparation patches (patches 1-5) and patch 6
adds runtime PM support.

Patches in this series were part of [2].

Changes in v4:
- remove unnecessary code from patch 4/6
- improve the code in patch 5/6

Changes in v3:
- fixed typos
- added patch "net: ravb: Move the update of ndev->features to
  ravb_set_features()"
- changes title of patch "net: ravb: Do not apply RX checksum
  settings to hardware if the interface is down" from v2 into
  "net: ravb: Do not apply features to hardware if the interface
  is down", changed patch description and updated the patch
- collected tags

Changes in v2:
- address review comments
- in patch 4/5 take into account the latest changes introduced
  in ravb_set_features_gbeth()

Changes since [2]:
- patch 1/5 is new
- use pm_runtime_get_noresume() and pm_runtime_active() in patches
  3/5, 4/5
- fixed higlighted typos in patch 4/5

[1] https://lore.kernel.org/all/20240202084136.3426492-1-claudiu.beznea.uj@bp.renesas.com/
[2] https://lore.kernel.org/all/20240105082339.1468817-1-claudiu.beznea.uj@bp.renesas.com/
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
2024-02-16 08:32:05 +00:00
Claudiu Beznea
48f894ab07 net: ravb: Add runtime PM support
Add runtime PM support for the ravb driver. As the driver is used by
different IP variants, with different behaviors, to be able to have the
runtime PM support available for all devices, the preparatory commits
moved all the resources parsing and allocations in the driver's probe
function and kept the settings for ravb_open(). This is due to the fact
that on some IP variants-platforms tuples disabling/enabling the clocks
will switch the IP to the reset operation mode where register contents is
lost and reconfiguration needs to be done. For this the rabv_open()
function enables the clocks, switches the IP to configuration mode, applies
all the register settings and switches the IP to the operational mode. At
the end of ravb_open() IP is ready to send/receive data.

In ravb_close() necessary reverts are done (compared with ravb_open()), the
IP is switched to reset mode and clocks are disabled.

The ethtool APIs or IOCTLs that might execute while the interface is down
are either cached (and applied in ravb_open()) or rejected (as at that time
the IP is in reset mode). Keeping the IP in the reset mode also increases
the power saved (according to the hardware manual).

Signed-off-by: Claudiu Beznea <claudiu.beznea.uj@bp.renesas.com>
Reviewed-by: Sergey Shtylyov <s.shtylyov@omp.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>
2024-02-16 08:32:04 +00:00
Claudiu Beznea
a71a50e391 net: ravb: Do not apply features to hardware if the interface is down
Do not apply features to hardware if the interface is down. In case runtime
PM is enabled, and while the interface is down, the IP will be in reset
mode (as for some platforms disabling the clocks will switch the IP to
reset mode, which will lead to losing register contents) and applying
settings in reset mode is not an option. Instead, cache the features and
apply them in ravb_open() through ravb_emac_init().

To avoid accessing the hardware while the interface is down
pm_runtime_active() check was introduced. Along with it the device runtime
PM usage counter has been incremented to avoid disabling the device clocks
while the check is in progress (if any).

Commit prepares for the addition of runtime PM.

Signed-off-by: Claudiu Beznea <claudiu.beznea.uj@bp.renesas.com>
Reviewed-by: Sergey Shtylyov <s.shtylyov@omp.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>
2024-02-16 08:32:04 +00:00
Claudiu Beznea
7bddccc991 net: ravb: Move the update of ndev->features to ravb_set_features()
Commit c2da940857 ("ravb: Add Rx checksum offload support for GbEth")
introduced support for setting GbEth features. With this the IP-specific
features update functions update the ndev->features individually.

Next commits add runtime PM support for the ravb driver. The runtime PM
implementation will enable/disable the IP clocks on
the ravb_open()/ravb_close() functions. Accessing the IP registers with
clocks disabled blocks the system.

The ravb_set_features() function could be executed when the Ethernet
interface is closed so we need to ensure we don't access IP registers while
the interface is down when runtime PM support will be in place.

For these, move the update of ndev->features to ravb_set_features(). In
this way we update the ndev->features only when the IP-specific features
set function returns success and we can avoid code duplication when
introducing runtime PM registers protection.

Signed-off-by: Claudiu Beznea <claudiu.beznea.uj@bp.renesas.com>
Reviewed-by: Sergey Shtylyov <s.shtylyov@omp.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>
2024-02-16 08:32:04 +00:00
Claudiu Beznea
bbf2345fa6 net: ravb: Return cached statistics if the interface is down
Return the cached statistics in case the interface is down. There should be
no drawback to this, as cached statistics are updated in ravb_close().

In order to avoid accessing the IP registers while the IP is runtime
suspended pm_runtime_active() check was introduced. The device runtime
PM usage counter has been incremented to avoid disabling the device clocks
while the check is in progress (if any).

The commit prepares the code for the addition of runtime PM support.

Suggested-by: Sergey Shtylyov <s.shtylyov@omp.ru>
Signed-off-by: Claudiu Beznea <claudiu.beznea.uj@bp.renesas.com>
Reviewed-by: Sergey Shtylyov <s.shtylyov@omp.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>
2024-02-16 08:32:04 +00:00
Claudiu Beznea
a5f149a97d net: ravb: Keep the reverse order of operations in ravb_close()
Keep the reverse order of operations in ravb_close() when compared with
ravb_open(). This is the recommended configuration sequence.

Signed-off-by: Claudiu Beznea <claudiu.beznea.uj@bp.renesas.com>
Reviewed-by: Sergey Shtylyov <s.shtylyov@omp.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>
2024-02-16 08:32:03 +00:00
Claudiu Beznea
a260f08066 net: ravb: Get rid of the temporary variable irq
The 4th argument of ravb_setup_irq() is used to save the IRQ number that
will be further used by the driver code. Not all ravb_setup_irqs() calls
need to save the IRQ number. The previous code used to pass a dummy
variable as the 4th argument in case the IRQ is not needed for further
usage. That is not necessary as the code from ravb_setup_irq() can detect
by itself if the IRQ needs to be saved. Thus, get rid of the code that is
not needed.

Reported-by: Sergey Shtylyov <s.shtylyov@omp.ru>
Signed-off-by: Claudiu Beznea <claudiu.beznea.uj@bp.renesas.com>
Reviewed-by: Sergey Shtylyov <s.shtylyov@omp.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>
2024-02-16 08:32:03 +00:00
Hangbin Liu
31f26e4fec selftests: bonding: make sure new active is not null
One of Jakub's tests[1] shows that there may be period all ports
are down and no active slave. This makes the new_active_slave null
and the test fails. Add a check to make sure the new active is not null.

 [  189.051966] br0: port 2(s1) entered disabled state
 [  189.317881] bond0: (slave eth1): link status definitely down, disabling slave
 [  189.318487] bond0: (slave eth2): making interface the new active one
 [  190.435430] br0: port 4(s2) entered disabled state
 [  190.773786] bond0: (slave eth0): link status definitely down, disabling slave
 [  190.774204] bond0: (slave eth2): link status definitely down, disabling slave
 [  190.774715] bond0: now running without any active interface!
 [  190.877760] bond0: (slave eth0): link status definitely up
 [  190.878098] bond0: (slave eth0): making interface the new active one
 [  190.878495] bond0: active interface up!
 [  191.802872] br0: port 4(s2) entered blocking state
 [  191.803157] br0: port 4(s2) entered forwarding state
 [  191.813756] bond0: (slave eth2): link status definitely up
 [  192.847095] br0: port 2(s1) entered blocking state
 [  192.847396] br0: port 2(s1) entered forwarding state
 [  192.853740] bond0: (slave eth1): link status definitely up
 # TEST: prio (active-backup ns_ip6_target primary_reselect 1)         [FAIL]
 # Current active slave is null but not eth0

[1] https://netdev-3.bots.linux.dev/vmksft-bonding/results/464481/1-bond-options-sh/stdout

Fixes: 45bf79bc56 ("selftests: bonding: reduce garp_test/arp_validate test time")
Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2024-02-16 08:02:39 +00:00
Cristian Ciocaltea
553be2830c phy: rockchip: Add Samsung HDMI/eDP Combo PHY driver
Add driver for the HDMI/eDP TX Combo PHY found on Rockchip RK3588 SoC.

The PHY is based on a Samsung IP block and supports HDMI 2.1 TMDS, FRL
and eDP links.  The maximum data rate is 12Gbps (FRL), while the minimum
is 250Mbps (TMDS).

Only the TMDS link is currently supported.

Co-developed-by: Algea Cao <algea.cao@rock-chips.com>
Signed-off-by: Algea Cao <algea.cao@rock-chips.com>
Tested-by: Heiko Stuebner <heiko@sntech.de>
Signed-off-by: Cristian Ciocaltea <cristian.ciocaltea@collabora.com>
Link: https://lore.kernel.org/r/20240214-phy-hdptx-v4-2-e7974f46c1a7@collabora.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
2024-02-16 12:26:56 +05:30
Cristian Ciocaltea
3312a0e8f6 dt-bindings: phy: Add Rockchip HDMI/eDP Combo PHY schema
Add dt-binding schema for the HDMI/eDP Transmitter Combo PHY found on
Rockchip RK3588 SoC.

Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Heiko Stuebner <heiko@sntech.de>
Signed-off-by: Cristian Ciocaltea <cristian.ciocaltea@collabora.com>
Link: https://lore.kernel.org/r/20240214-phy-hdptx-v4-1-e7974f46c1a7@collabora.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
2024-02-16 12:26:56 +05:30
Dave Airlie
ea69f782d0 Merge tag 'drm-msm-fixes-2024-02-15' of https://gitlab.freedesktop.org/drm/msm into drm-fixes
Fixes for v6.8-rc5

GPU:
- dmabuf vmap fix
- a610 UBWC corruption fix (incorrect hbb)
- revert a commit that was making GPU recovery unreliable
- tlb invalidation fix

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Rob Clark <robdclark@gmail.com>
Link: https://patchwork.freedesktop.org/patch/msgid/CAF6AEGszDSiw66+a=ttBr-hat+zrcBtfc_cZ4LQqXu89DJ0UeQ@mail.gmail.com
2024-02-16 15:47:15 +10:00
Dave Airlie
44395701ad amd-drm-fixes-6.8-2024-02-15-2:
amdgpu:
 - PSR fixes
 - Suspend/resume fixes
 - Link training fix
 - Aspect ratio fix
 - DCN 3.5 fixes
 - VCN 4.x fix
 - GFX 11 fix
 - Misc display fixes
 - Misc small fixes
 
 amdkfd:
 - Cache size reporting fix
 - SIMD distribution fix
 -----BEGIN PGP SIGNATURE-----
 
 iHUEABYKAB0WIQQgO5Idg2tXNTSZAr293/aFa7yZ2AUCZc5kRgAKCRC93/aFa7yZ
 2C9XAQDyLX03RVJYF60LuImEY+vcxkA8i5RhNzvmsLuy8CjJeQD/URQQCzE9Pi+b
 6vdJTHa4xPdB1NVhFu8c7N3AeLyXAQU=
 =pEYP
 -----END PGP SIGNATURE-----

Merge tag 'amd-drm-fixes-6.8-2024-02-15-2' of https://gitlab.freedesktop.org/agd5f/linux into drm-fixes

amd-drm-fixes-6.8-2024-02-15-2:

amdgpu:
- PSR fixes
- Suspend/resume fixes
- Link training fix
- Aspect ratio fix
- DCN 3.5 fixes
- VCN 4.x fix
- GFX 11 fix
- Misc display fixes
- Misc small fixes

amdkfd:
- Cache size reporting fix
- SIMD distribution fix

Signed-off-by: Dave Airlie <airlied@redhat.com>

From: Alex Deucher <alexander.deucher@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240215192452.11805-1-alexander.deucher@amd.com
2024-02-16 15:46:33 +10:00
Dave Airlie
427e337f7a Driver Changes:
- Fix an out-of-bounds shift.
 - Fix the display code thinking xe uses shmem
 - Fix a warning about index out-of-bound
 - Fix a clang-16 compilation warning
 -----BEGIN PGP SIGNATURE-----
 
 iHUEABYKAB0WIQRskUM7w1oG5rx2IZO4FpNVCsYGvwUCZc4GBQAKCRC4FpNVCsYG
 vxR5AQD8961ciqivB/Psn36f9lOctbJ41Hma46x5ESuQjfN+vAEA3qLplLgWuFrY
 HTaq3VG4kX1M5yF2b6p18NnTSAyyLAA=
 =tfli
 -----END PGP SIGNATURE-----

Merge tag 'drm-xe-fixes-2024-02-15' of https://gitlab.freedesktop.org/drm/xe/kernel into drm-fixes

Driver Changes:
- Fix an out-of-bounds shift.
- Fix the display code thinking xe uses shmem
- Fix a warning about index out-of-bound
- Fix a clang-16 compilation warning

Signed-off-by: Dave Airlie <airlied@redhat.com>

From: Thomas Hellstrom <thomas.hellstrom@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/Zc4GpcrbFVqdK9Ws@fedora
2024-02-16 15:45:12 +10:00
Dave Airlie
7e1c3be3f9 Merge tag 'drm-intel-fixes-2024-02-15' of git://anongit.freedesktop.org/drm/drm-intel into drm-fixes
Fix for #10172: Blank screen on JSL Chromebooks. Stable fix to limit DP SST link rate to <=8.1Gbps.

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/Zc37W27F5OvoeSkG@jlahtine-mobl.ger.corp.intel.com
2024-02-16 15:44:39 +10:00
Arınç ÜNAL
ae94dc25fd net: dsa: remove OF-based MDIO bus registration from DSA core
The code block under the "!ds->user_mii_bus && ds->ops->phy_read" check
under dsa_switch_setup() populates ds->user_mii_bus. The use of
ds->user_mii_bus is inappropriate when the MDIO bus of the switch is
described on the device tree [1].

For this reason, use this code block only for switches [with MDIO bus]
probed on platform_data, and OF which the switch MDIO bus isn't described
on the device tree. Therefore, remove OF-based MDIO bus registration as
it's useless for these cases.

These subdrivers which control switches [with MDIO bus] probed on OF, will
lose the ability to register the MDIO bus OF-based:

drivers/net/dsa/b53/b53_common.c
drivers/net/dsa/lan9303-core.c
drivers/net/dsa/vitesse-vsc73xx-core.c

These subdrivers let the DSA core driver register the bus:
- ds->ops->phy_read() and ds->ops->phy_write() are present.
- ds->user_mii_bus is not populated.

The commit fe7324b932 ("net: dsa: OF-ware slave_mii_bus") which brought
OF-based MDIO bus registration on the DSA core driver is reasonably recent
and, in this time frame, there have been no device trees in the Linux
repository that started describing the MDIO bus, or dt-bindings defining
the MDIO bus for the switches these subdrivers control. So I don't expect
any devices to be affected.

The logic we encourage is that all subdrivers should register the switch
MDIO bus on their own [2]. And, for subdrivers which control switches [with
MDIO bus] probed on OF, this logic must be followed to support all cases
properly:

No switch MDIO bus defined: Populate ds->user_mii_bus, register the MDIO
bus, set the interrupts for PHYs if "interrupt-controller" is defined at
the switch node. This case should only be covered for the switches which
their dt-bindings documentation didn't document the MDIO bus from the
start. This is to keep supporting the device trees that do not describe the
MDIO bus on the device tree but the MDIO bus is being used nonetheless.

Switch MDIO bus defined: Don't populate ds->user_mii_bus, register the MDIO
bus, set the interrupts for PHYs if ["interrupt-controller" is defined at
the switch node and "interrupts" is defined at the PHY nodes under the
switch MDIO bus node].

Switch MDIO bus defined but explicitly disabled: If the device tree says
status = "disabled" for the MDIO bus, we shouldn't need an MDIO bus at all.
Instead, just exit as early as possible and do not call any MDIO API.

After all subdrivers that control switches with MDIO buses are made to
register the MDIO buses on their own, we will be able to get rid of
dsa_switch_ops :: phy_read() and :: phy_write(), and the code block for
registering the MDIO bus on the DSA core driver.

Link: https://lore.kernel.org/netdev/20231213120656.x46fyad6ls7sqyzv@skbuf/ [1]
Link: https://lore.kernel.org/netdev/20240103184459.dcbh57wdnlox6w7d@skbuf/ [2]
Suggested-by: Luiz Angelo Daros de Luca <luizluca@gmail.com>
Acked-by: Luiz Angelo Daros de Luca <luizluca@gmail.com>
Signed-off-by: Arınç ÜNAL <arinc.unal@arinc9.com>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Link: https://lore.kernel.org/r/20240213-for-netnext-dsa-mdio-bus-v2-1-0ff6f4823a9e@arinc9.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-02-15 21:30:02 -08:00
Roberto Sassu
b6c0dec9f7 integrity: Remove LSM
Since now IMA and EVM use their own integrity metadata, it is safe to
remove the 'integrity' LSM, with its management of integrity metadata.

Keep the iint.c file only for loading IMA and EVM keys at boot, and for
creating the integrity directory in securityfs (we need to keep it for
retrocompatibility reasons).

Signed-off-by: Roberto Sassu <roberto.sassu@huawei.com>
Reviewed-by: Casey Schaufler <casey@schaufler-ca.com>
Reviewed-by: Stefan Berger <stefanb@linux.ibm.com>
Reviewed-by: Mimi Zohar <zohar@linux.ibm.com>
Acked-by: Mimi Zohar <zohar@linux.ibm.com>
Signed-off-by: Paul Moore <paul@paul-moore.com>
2024-02-15 23:43:48 -05:00
Roberto Sassu
4de2f084fb ima: Make it independent from 'integrity' LSM
Make the 'ima' LSM independent from the 'integrity' LSM by introducing IMA
own integrity metadata (ima_iint_cache structure, with IMA-specific fields
from the integrity_iint_cache structure), and by managing it directly from
the 'ima' LSM.

Create ima_iint.c and introduce the same integrity metadata management
functions found in iint.c (renamed with ima_). However, instead of putting
metadata in an rbtree, reserve space from IMA in the inode security blob
for a pointer, and introduce the ima_inode_set_iint()/ima_inode_get_iint()
primitives to store/retrieve that pointer. This improves search time from
logarithmic to constant.

Consequently, don't include the inode pointer as field in the
ima_iint_cache structure, since the association with the inode is clear.
Since the inode field is missing in ima_iint_cache, pass the extra inode
parameter to ima_get_verity_digest().

Prefer storing the pointer instead of the entire ima_iint_cache structure,
to avoid too much memory pressure. Use the same mechanism as before, a
cache named ima_iint_cache (renamed from iint_cache), to quickly allocate
a new ima_iint_cache structure when requested by the IMA policy.

Create the new ima_iint_cache in ima_iintcache_init(),
called by init_ima_lsm(), during the initialization of the 'ima' LSM. And,
register ima_inode_free_security() to free the ima_iint_cache structure, if
exists.

Replace integrity_iint_cache with ima_iint_cache in various places of the
IMA code. Also, replace integrity_inode_get() and integrity_iint_find(),
respectively with ima_inode_get() and ima_iint_find().

Finally, move the remaining IMA-specific flags
to security/integrity/ima/ima.h, since they are now unnecessary in the
common integrity layer.

Signed-off-by: Roberto Sassu <roberto.sassu@huawei.com>
Reviewed-by: Casey Schaufler <casey@schaufler-ca.com>
Reviewed-by: Stefan Berger <stefanb@linux.ibm.com>
Reviewed-by: Mimi Zohar <zohar@linux.ibm.com>
Acked-by: Mimi Zohar <zohar@linux.ibm.com>
Signed-off-by: Paul Moore <paul@paul-moore.com>
2024-02-15 23:43:47 -05:00
Roberto Sassu
75a323e604 evm: Make it independent from 'integrity' LSM
Define a new structure for EVM-specific metadata, called evm_iint_cache,
and embed it in the inode security blob. Introduce evm_iint_inode() to
retrieve metadata, and register evm_inode_alloc_security() for the
inode_alloc_security LSM hook, to initialize the structure (before
splitting metadata, this task was done by iint_init_always()).

Keep the non-NULL checks after calling evm_iint_inode() except in
evm_inode_alloc_security(), to take into account inodes for which
security_inode_alloc() was not called. When using shared metadata,
obtaining a NULL pointer from integrity_iint_find() meant that the file
wasn't in the IMA policy. Now, because IMA and EVM use disjoint metadata,
the EVM status has to be stored for every inode regardless of the IMA
policy.

Given that from now on EVM relies on its own metadata, remove the iint
parameter from evm_verifyxattr(). Also, directly retrieve the iint in
evm_verify_hmac(), called by both evm_verifyxattr() and
evm_verify_current_integrity(), since now there is no performance penalty
in retrieving EVM metadata (constant time).

Replicate the management of the IMA_NEW_FILE flag, by introducing
evm_post_path_mknod() and evm_file_release() to respectively set and clear
the newly introduced flag EVM_NEW_FILE, at the same time IMA does. Like for
IMA, select CONFIG_SECURITY_PATH when EVM is enabled, to ensure that files
are marked as new.

Unlike ima_post_path_mknod(), evm_post_path_mknod() cannot check if a file
must be appraised. Thus, it marks all affected files. Also, it does not
clear EVM_NEW_FILE depending on i_version, but that is not a problem
because IMA_NEW_FILE is always cleared when set in ima_check_last_writer().

Move the EVM-specific flag EVM_IMMUTABLE_DIGSIG to
security/integrity/evm/evm.h, since that definition is now unnecessary in
the common integrity layer.

Finally, switch to the LSM reservation mechanism for the EVM xattr, and
consequently decrement by one the number of xattrs to allocate in
security_inode_init_security().

Signed-off-by: Roberto Sassu <roberto.sassu@huawei.com>
Reviewed-by: Casey Schaufler <casey@schaufler-ca.com>
Reviewed-by: Stefan Berger <stefanb@linux.ibm.com>
Reviewed-by: Mimi Zohar <zohar@linux.ibm.com>
Acked-by: Mimi Zohar <zohar@linux.ibm.com>
Signed-off-by: Paul Moore <paul@paul-moore.com>
2024-02-15 23:43:47 -05:00
Roberto Sassu
9238311176 evm: Move to LSM infrastructure
As for IMA, move hardcoded EVM function calls from various places in the
kernel to the LSM infrastructure, by introducing a new LSM named 'evm'
(last and always enabled like 'ima'). The order in the Makefile ensures
that 'evm' hooks are executed after 'ima' ones.

Make EVM functions as static (except for evm_inode_init_security(), which
is exported), and register them as hook implementations in init_evm_lsm().
Also move the inline functions evm_inode_remove_acl(),
evm_inode_post_remove_acl(), and evm_inode_post_set_acl() from the public
evm.h header to evm_main.c.

Unlike before (see commit to move IMA to the LSM infrastructure),
evm_inode_post_setattr(), evm_inode_post_set_acl(),
evm_inode_post_remove_acl(), and evm_inode_post_removexattr() are not
executed for private inodes.

Finally, add the LSM_ID_EVM case in lsm_list_modules_test.c

Signed-off-by: Roberto Sassu <roberto.sassu@huawei.com>
Reviewed-by: Casey Schaufler <casey@schaufler-ca.com>
Acked-by: Christian Brauner <brauner@kernel.org>
Reviewed-by: Stefan Berger <stefanb@linux.ibm.com>
Reviewed-by: Mimi Zohar <zohar@linux.ibm.com>
Acked-by: Mimi Zohar <zohar@linux.ibm.com>
Signed-off-by: Paul Moore <paul@paul-moore.com>
2024-02-15 23:43:47 -05:00