Shinas Rasheed says:

====================
Cleanup and optimizations to transmit code

Pad small packets to ETH_ZLEN before transmit, cleanup dma sync calls,
add xmit_more functionality and then further remove atomic
variable usage in the prior.

Changes:
V3:
  - Stop returning NETDEV_TX_BUSY when ring is full in xmit_patch.
    Change to inspect early if next packet can fit in ring instead of
    current packet, and stop queue if not.
  - Add smp_mb between stopping tx queue and checking if tx queue has
    free entries again, in queue full check function to let reflect
    IQ process completions that might have happened on other cpus.
  - Update small packet padding patch changelog to give more info.
V2: https://lore.kernel.org/all/20231024145119.2366588-1-srasheed@marvell.com/
  - Added patch for padding small packets to ETH_ZLEN, part of
    optimization patches for transmit code missed out in V1
  - Updated changelog to provide more details for dma_sync remove patch
  - Updated changelog to use imperative tone in add xmit_more patch
V1: https://lore.kernel.org/all/20231023114449.2362147-1-srasheed@marvell.com/
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

Replace atomic variable "instr_pending" which represents number of
posted tx instructions pending completion, with host_write_idx and
flush_index variables in the xmit and completion processing respectively.
Signed-off-by: Shinas Rasheed <srasheed@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Add xmit_more handling in tx datapath for octeon_ep pf.
Signed-off-by: Shinas Rasheed <srasheed@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Cleanup dma sync calls for scatter gather
mappings, since they are coherent allocations
and do not need explicit sync to be called.
Signed-off-by: Shinas Rasheed <srasheed@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Pad small packets to ETH_ZLEN before transmit, as hardware
cannot pad and requires software padding to ensure
minimum ethernet frame length.
Signed-off-by: Shinas Rasheed <srasheed@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Cross-merge networking fixes after downstream PR.

No conflicts.
Signed-off-by: Paolo Abeni <pabeni@redhat.com>

Pull networking fixes from Paolo Abeni:
 "Including fixes from BPF and netfilter.

  Current release - regressions:

   - core: fix undefined behavior in netdev name allocation

   - bpf: do not allocate percpu memory at init stage

   - netfilter: nf_tables: split async and sync catchall in two
     functions

   - mptcp: fix possible NULL pointer dereference on close

  Current release - new code bugs:

   - eth: ice: dpll: fix initial lock status of dpll

  Previous releases - regressions:

   - bpf: fix precision backtracking instruction iteration

   - af_unix: fix use-after-free in unix_stream_read_actor()

   - tipc: fix kernel-infoleak due to uninitialized TLV value

   - eth: bonding: stop the device in bond_setup_by_slave()

   - eth: mlx5:
      - fix double free of encap_header
      - avoid referencing skb after free-ing in drop path

   - eth: hns3: fix VF reset

   - eth: mvneta: fix calls to page_pool_get_stats

  Previous releases - always broken:

   - core: set SOCK_RCU_FREE before inserting socket into hashtable

   - bpf: fix control-flow graph checking in privileged mode

   - eth: ppp: limit MRU to 64K

   - eth: stmmac: avoid rx queue overrun

   - eth: icssg-prueth: fix error cleanup on failing initialization

   - eth: hns3: fix out-of-bounds access may occur when coalesce info is
     read via debugfs

   - eth: cortina: handle large frames

  Misc:

   - selftests: gso: support CONFIG_MAX_SKB_FRAGS up to 45"

* tag 'net-6.7-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (78 commits)
  macvlan: Don't propagate promisc change to lower dev in passthru
  net: sched: do not offload flows with a helper in act_ct
  net/mlx5e: Check return value of snprintf writing to fw_version buffer for representors
  net/mlx5e: Check return value of snprintf writing to fw_version buffer
  net/mlx5e: Reduce the size of icosq_str
  net/mlx5: Increase size of irq name buffer
  net/mlx5e: Update doorbell for port timestamping CQ before the software counter
  net/mlx5e: Track xmit submission to PTP WQ after populating metadata map
  net/mlx5e: Avoid referencing skb after free-ing in drop path of mlx5e_sq_xmit_wqe
  net/mlx5e: Don't modify the peer sent-to-vport rules for IPSec offload
  net/mlx5e: Fix pedit endianness
  net/mlx5e: fix double free of encap_header in update funcs
  net/mlx5e: fix double free of encap_header
  net/mlx5: Decouple PHC .adjtime and .adjphase implementations
  net/mlx5: DR, Allow old devices to use multi destination FTE
  net/mlx5: Free used cpus mask when an IRQ is released
  Revert "net/mlx5: DR, Supporting inline WQE when possible"
  bpf: Do not allocate percpu memory at init stage
  net: Fix undefined behavior in netdev name allocation
  dt-bindings: net: ethernet-controller: Fix formatting error
  ...

Pull xen updates from Juergen Gross:

 - A fix in the Xen events driver avoiding the use of RCU after
   the call to rcu_report_dead() when taking a cpu down

 - A fix for running as Xen dom0 to line up ACPI's idea of power
   management capabilities with the one of Xen

 - A cleanup eliminating several kernel-doc warnings in Xen related
   code

 - A cleanup series of the Xen events driver

* tag 'for-linus-6.7a-rc2-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip:
  xen/events: remove some info_for_irq() calls in pirq handling
  xen/events: modify internal [un]bind interfaces
  xen/events: drop xen_allocate_irqs_dynamic()
  xen/events: remove some simple helpers from events_base.c
  xen/events: reduce externally visible helper functions
  xen/events: remove unused functions
  xen/events: fix delayed eoi list handling
  xen/shbuf: eliminate 17 kernel-doc warnings
  acpi/processor: sanitize _OSC/_PDC capabilities for Xen dom0
  xen/events: avoid using info_for_irq() in xen_send_IPI_one()

Pull virtio fixes from Michael Tsirkin:
 "Bugfixes all over the place"

* tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost:
  vhost-vdpa: fix use after free in vhost_vdpa_probe()
  virtio_pci: Switch away from deprecated irq_set_affinity_hint
  riscv, qemu_fw_cfg: Add support for RISC-V architecture
  vdpa_sim_blk: allocate the buffer zeroed
  virtio_pci: move structure to a header

Fix a small typo in the kerneldoc comment of the INDIRECT_CALL_$NR
macro.
Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://lore.kernel.org/r/20231114104202.4680-1-tklauser@distanz.chSigned-off-by: Paolo Abeni <pabeni@redhat.com>

Pablo Neira Ayuso says:

====================
Netfilter fixes for net

The following patchset contains Netfilter fixes for net:

1) Remove unused variable causing compilation warning in nft_set_rbtree,
   from Yang Li. This unused variable is a left over from previous
   merge window.

2) Possible return of uninitialized in nf_conntrack_bridge, from
   Linkui Xiao. This is there since nf_conntrack_bridge is available.

3) Fix incorrect pointer math in nft_byteorder, from Dan Carpenter.
   Problem has been there since 2016.

4) Fix bogus error in destroy set element command. Problem is there
   since this new destroy command was added.

5) Fix race condition in ipset between swap and destroy commands and
   add/del/test control plane. This problem is there since ipset was
   merged.

6) Split async and sync catchall GC in two function to fix unsafe
   iteration over RCU. This is a fix-for-fix that was included in
   the previous pull request.

* tag 'nf-23-11-15' of git://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf:
  netfilter: nf_tables: split async and sync catchall in two functions
  netfilter: ipset: fix race condition between swap/destroy and kernel side add/del/test
  netfilter: nf_tables: bogus ENOENT when destroying element which does not exist
  netfilter: nf_tables: fix pointer math issue in nft_byteorder_eval()
  netfilter: nf_conntrack_bridge: initialize err to 0
  netfilter: nft_set_rbtree: Remove unused variable nft_net
====================

Link: https://lore.kernel.org/r/20231115184514.8965-1-pablo@netfilter.orgSigned-off-by: Paolo Abeni <pabeni@redhat.com>

Macvlan device in passthru mode sets its lower device promiscuous mode
according to its MACVLAN_FLAG_NOPROMISC flag instead of synchronizing it to
its own promiscuity setting. However, macvlan_change_rx_flags() function
doesn't check the mode before propagating such changes to the lower device
which can cause net_device->promiscuity counter overflow as illustrated by
reproduction example [0] and resulting dmesg log [1]. Fix the issue by
first verifying the mode in macvlan_change_rx_flags() function before
propagating promiscuous mode change to the lower device.

[0]:
ip link add macvlan1 link enp8s0f0 type macvlan mode passthru
ip link set macvlan1 promisc on
ip l set dev macvlan1 up
ip link set macvlan1 promisc off
ip l set dev macvlan1 down
ip l set dev macvlan1 up

[1]:
[ 5156.281724] macvlan1: entered promiscuous mode
[ 5156.285467] mlx5_core 0000:08:00.0 enp8s0f0: entered promiscuous mode
[ 5156.287639] macvlan1: left promiscuous mode
[ 5156.288339] mlx5_core 0000:08:00.0 enp8s0f0: left promiscuous mode
[ 5156.290907] mlx5_core 0000:08:00.0 enp8s0f0: entered promiscuous mode
[ 5156.317197] mlx5_core 0000:08:00.0 enp8s0f0: promiscuity touches roof, set promiscuity failed. promiscuity feature of device might be broken.

Fixes: efdbd2b3 ("macvlan: Propagate promiscuity setting to lower devices.")
Reviewed-by: Gal Pressman <gal@nvidia.com>
Signed-off-by: Vlad Buslov <vladbu@nvidia.com>
Reviewed-by: Jiri Pirko <jiri@nvidia.com>
Link: https://lore.kernel.org/r/20231114175915.1649154-1-vladbu@nvidia.comSigned-off-by: Paolo Abeni <pabeni@redhat.com>

There is no hardware supporting ct helper offload. However, prior to this
patch, a flower filter with a helper in the ct action can be successfully
set into the HW, for example (eth1 is a bnxt NIC):

  # tc qdisc add dev eth1 ingress_block 22 ingress
  # tc filter add block 22 proto ip flower skip_sw ip_proto tcp \
    dst_port 21 ct_state -trk action ct helper ipv4-tcp-ftp
  # tc filter show dev eth1 ingress

    filter block 22 protocol ip pref 49152 flower chain 0 handle 0x1
      eth_type ipv4
      ip_proto tcp
      dst_port 21
      ct_state -trk
      skip_sw
      in_hw in_hw_count 1   <----
        action order 1: ct zone 0 helper ipv4-tcp-ftp pipe
         index 2 ref 1 bind 1
        used_hw_stats delayed

This might cause the flower filter not to work as expected in the HW.

This patch avoids this problem by simply returning -EOPNOTSUPP in
tcf_ct_offload_act_setup() to not allow to offload flows with a helper
in act_ct.

Fixes: a21b06e7 ("net: sched: add helper support in act_ct")
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Reviewed-by: Jamal Hadi Salim <jhs@mojatatu.com>
Link: https://lore.kernel.org/r/f8685ec7702c4a448a1371a8b34b43217b583b9d.1699898008.git.lucien.xin@gmail.comSigned-off-by: Paolo Abeni <pabeni@redhat.com>

Saeed Mahameed says:

====================
This series provides bug fixes to mlx5 driver.
====================

Link: https://lore.kernel.org/r/20231114215846.5902-1-saeed@kernel.org/Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Treat the operation as an error case when the return value is equivalent to
the size of the name buffer. Failed to write null terminator to the name
buffer, making the string malformed and should not be used. Provide a
string with only the firmware version when forming the string with the
board id fails. This logic for representors is identical to normal flow
with ethtool.

Without check, will trigger -Wformat-truncation with W=1.

    drivers/net/ethernet/mellanox/mlx5/core/en_rep.c: In function 'mlx5e_rep_get_drvinfo':
    drivers/net/ethernet/mellanox/mlx5/core/en_rep.c:78:31: warning: '%.16s' directive output may be truncated writing up to 16 bytes into a region of size between 13 and 22 [-Wformat-truncation=]
      78 |                  "%d.%d.%04d (%.16s)",
         |                               ^~~~~
    drivers/net/ethernet/mellanox/mlx5/core/en_rep.c:77:9: note: 'snprintf' output between 12 and 37 bytes into a destination of size 32
      77 |         snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version),
         |         ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
      78 |                  "%d.%d.%04d (%.16s)",
         |                  ~~~~~~~~~~~~~~~~~~~~~
      79 |                  fw_rev_maj(mdev), fw_rev_min(mdev),
         |                  ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
      80 |                  fw_rev_sub(mdev), mdev->board_id);
         |                  ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Fixes: cf83c8fd ("net/mlx5e: Add missing ethtool driver info for representors")
Link: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=6d4ab2e97dcfbcd748ae71761a9d8e5e41cc732cSigned-off-by: Rahul Rameshbabu <rrameshbabu@nvidia.com>
Reviewed-by: Dragos Tatulea <dtatulea@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Link: https://lore.kernel.org/r/20231114215846.5902-16-saeed@kernel.orgSigned-off-by: Jakub Kicinski <kuba@kernel.org>

Treat the operation as an error case when the return value is equivalent to
the size of the name buffer. Failed to write null terminator to the name
buffer, making the string malformed and should not be used. Provide a
string with only the firmware version when forming the string with the
board id fails.

Without check, will trigger -Wformat-truncation with W=1.

    drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c: In function 'mlx5e_ethtool_get_drvinfo':
    drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c:49:31: warning: '%.16s' directive output may be truncated writing up to 16 bytes into a region of size between 13 and 22 [-Wformat-truncation=]
      49 |                  "%d.%d.%04d (%.16s)",
         |                               ^~~~~
    drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c:48:9: note: 'snprintf' output between 12 and 37 bytes into a destination of size 32
      48 |         snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version),
         |         ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
      49 |                  "%d.%d.%04d (%.16s)",
         |                  ~~~~~~~~~~~~~~~~~~~~~
      50 |                  fw_rev_maj(mdev), fw_rev_min(mdev), fw_rev_sub(mdev),
         |                  ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
      51 |                  mdev->board_id);
         |                  ~~~~~~~~~~~~~~~

Fixes: 84e11edb ("net/mlx5e: Show board id in ethtool driver information")
Link: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=6d4ab2e97dcfbcd748ae71761a9d8e5e41cc732cSigned-off-by: Rahul Rameshbabu <rrameshbabu@nvidia.com>
Reviewed-by: Dragos Tatulea <dtatulea@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

icosq_str size is unnecessarily too long, and it causes a build warning
-Wformat-truncation with W=1. Looking closely, It doesn't need to be 255B,
hence this patch reduces the size to 32B which should be more than enough
to host the string: "ICOSQ: 0x%x, ".

While here, add a missing space in the formatted string.

This fixes the following build warning:

$ KCFLAGS='-Wall -Werror'
$ make O=/tmp/kbuild/linux W=1 -s -j12 drivers/net/ethernet/mellanox/mlx5/core/

drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c: In function 'mlx5e_reporter_rx_timeout':
drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c:718:56:
error: ', CQ: 0x' directive output may be truncated writing 8 bytes into a region of size between 0 and 255 [-Werror=format-truncation=]
  718 |                  "RX timeout on channel: %d, %sRQ: 0x%x, CQ: 0x%x",
      |                                                        ^~~~~~~~
drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c:717:9: note: 'snprintf' output between 43 and 322 bytes into a destination of size 288
  717 |         snprintf(err_str, sizeof(err_str),
      |         ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  718 |                  "RX timeout on channel: %d, %sRQ: 0x%x, CQ: 0x%x",
      |                  ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  719 |                  rq->ix, icosq_str, rq->rqn, rq->cq.mcq.cqn);
      |                  ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Fixes: 521f31af ("net/mlx5e: Allow RQ outside of channel context")
Link: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=6d4ab2e97dcfbcd748ae71761a9d8e5e41cc732cSigned-off-by: Saeed Mahameed <saeedm@nvidia.com>
Link: https://lore.kernel.org/r/20231114215846.5902-14-saeed@kernel.orgSigned-off-by: Jakub Kicinski <kuba@kernel.org>

Without increased buffer size, will trigger -Wformat-truncation with W=1
for the snprintf operation writing to the buffer.

    drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c: In function 'mlx5_irq_alloc':
    drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c:296:7: error: '@pci:' directive output may be truncated writing 5 bytes into a region of size between 1 and 32 [-Werror=format-truncation=]
      296 |    "%s@pci:%s", name, pci_name(dev->pdev));
          |       ^~~~~
    drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c:295:2: note: 'snprintf' output 6 or more bytes (assuming 37) into a destination of size 32
      295 |  snprintf(irq->name, MLX5_MAX_IRQ_NAME,
          |  ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
      296 |    "%s@pci:%s", name, pci_name(dev->pdev));
          |    ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Fixes: ada9f5d0 ("IB/mlx5: Fix eq names to display nicely in /proc/interrupts")
Link: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=6d4ab2e97dcfbcd748ae71761a9d8e5e41cc732cSigned-off-by: Rahul Rameshbabu <rrameshbabu@nvidia.com>
Reviewed-by: Dragos Tatulea <dtatulea@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Link: https://lore.kernel.org/r/20231114215846.5902-13-saeed@kernel.orgSigned-off-by: Jakub Kicinski <kuba@kernel.org>

Previously, mlx5e_ptp_poll_ts_cq would update the device doorbell with the
incremented consumer index after the relevant software counters in the
kernel were updated. In the mlx5e_sq_xmit_wqe context, this would lead to
either overrunning the device CQ or exceeding the expected software buffer
size in the device CQ if the device CQ size was greater than the software
buffer size. Update the relevant software counter only after updating the
device CQ consumer index in the port timestamping napi_poll context.

Log:
    mlx5_core 0000:08:00.0: cq_err_event_notifier:517:(pid 0): CQ error on CQN 0x487, syndrome 0x1
    mlx5_core 0000:08:00.0 eth2: mlx5e_cq_error_event: cqn=0x000487 event=0x04

Fixes: 1880bc4e ("net/mlx5e: Add TX port timestamp support")
Signed-off-by: Rahul Rameshbabu <rrameshbabu@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Link: https://lore.kernel.org/r/20231114215846.5902-12-saeed@kernel.orgSigned-off-by: Jakub Kicinski <kuba@kernel.org>

Ensure the skb is available in metadata mapping to skbs before tracking the
metadata index for detecting undelivered CQEs. If the metadata index is put
in the tracking list before putting the skb in the map, the metadata index
might be used for detecting undelivered CQEs before the relevant skb is
available in the map, which can lead to a null-ptr-deref.

Log:
    general protection fault, probably for non-canonical address 0xdffffc0000000005: 0000 [#1] SMP KASAN
    KASAN: null-ptr-deref in range [0x0000000000000028-0x000000000000002f]
    CPU: 0 PID: 1243 Comm: kworker/0:2 Not tainted 6.6.0-rc4+ #108
    Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014
    Workqueue: events mlx5e_rx_dim_work [mlx5_core]
    RIP: 0010:mlx5e_ptp_napi_poll+0x9a4/0x2290 [mlx5_core]
    Code: 8c 24 38 cc ff ff 4c 8d 3c c1 4c 89 f9 48 c1 e9 03 42 80 3c 31 00 0f 85 97 0f 00 00 4d 8b 3f 49 8d 7f 28 48 89 f9 48 c1 e9 03 <42> 80 3c 31 00 0f 85 8b 0f 00 00 49 8b 47 28 48 85 c0 0f 84 05 07
    RSP: 0018:ffff8884d3c09c88 EFLAGS: 00010206
    RAX: 0000000000000069 RBX: ffff8881160349d8 RCX: 0000000000000005
    RDX: ffffed10218f48cf RSI: 0000000000000004 RDI: 0000000000000028
    RBP: ffff888122707700 R08: 0000000000000001 R09: ffffed109a781383
    R10: 0000000000000003 R11: 0000000000000003 R12: ffff88810c7a7a40
    R13: ffff888122707700 R14: dffffc0000000000 R15: 0000000000000000
    FS:  0000000000000000(0000) GS:ffff8884d3c00000(0000) knlGS:0000000000000000
    CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
    CR2: 00007f4f878dd6e0 CR3: 000000014d108002 CR4: 0000000000370eb0
    DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
    DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
    Call Trace:
    <IRQ>
    ? die_addr+0x3c/0xa0
    ? exc_general_protection+0x144/0x210
    ? asm_exc_general_protection+0x22/0x30
    ? mlx5e_ptp_napi_poll+0x9a4/0x2290 [mlx5_core]
    ? mlx5e_ptp_napi_poll+0x8f6/0x2290 [mlx5_core]
    __napi_poll.constprop.0+0xa4/0x580
    net_rx_action+0x460/0xb80
    ? _raw_spin_unlock_irqrestore+0x32/0x60
    ? __napi_poll.constprop.0+0x580/0x580
    ? tasklet_action_common.isra.0+0x2ef/0x760
    __do_softirq+0x26c/0x827
    irq_exit_rcu+0xc2/0x100
    common_interrupt+0x7f/0xa0
    </IRQ>
    <TASK>
    asm_common_interrupt+0x22/0x40
    RIP: 0010:__kmem_cache_alloc_node+0xb/0x330
    Code: 41 5d 41 5e 41 5f c3 8b 44 24 14 8b 4c 24 10 09 c8 eb d5 e8 b7 43 ca 01 0f 1f 80 00 00 00 00 0f 1f 44 00 00 55 48 89 e5 41 57 <41> 56 41 89 d6 41 55 41 89 f5 41 54 49 89 fc 53 48 83 e4 f0 48 83
    RSP: 0018:ffff88812c4079c0 EFLAGS: 00000246
    RAX: 1ffffffff083c7fe RBX: ffff888100042dc0 RCX: 0000000000000218
    RDX: 00000000ffffffff RSI: 0000000000000dc0 RDI: ffff888100042dc0
    RBP: ffff88812c4079c8 R08: ffffffffa0289f96 R09: ffffed1025880ea9
    R10: ffff888138839f80 R11: 0000000000000002 R12: 0000000000000dc0
    R13: 0000000000000100 R14: 000000000000008c R15: ffff8881271fc450
    ? cmd_exec+0x796/0x2200 [mlx5_core]
    kmalloc_trace+0x26/0xc0
    cmd_exec+0x796/0x2200 [mlx5_core]
    mlx5_cmd_do+0x22/0xc0 [mlx5_core]
    mlx5_cmd_exec+0x17/0x30 [mlx5_core]
    mlx5_core_modify_cq_moderation+0x139/0x1b0 [mlx5_core]
    ? mlx5_add_cq_to_tasklet+0x280/0x280 [mlx5_core]
    ? lockdep_set_lock_cmp_fn+0x190/0x190
    ? process_one_work+0x659/0x1220
    mlx5e_rx_dim_work+0x9d/0x100 [mlx5_core]
    process_one_work+0x730/0x1220
    ? lockdep_hardirqs_on_prepare+0x400/0x400
    ? max_active_store+0xf0/0xf0
    ? assign_work+0x168/0x240
    worker_thread+0x70f/0x12d0
    ? __kthread_parkme+0xd1/0x1d0
    ? process_one_work+0x1220/0x1220
    kthread+0x2d9/0x3b0
    ? kthread_complete_and_exit+0x20/0x20
    ret_from_fork+0x2d/0x70
    ? kthread_complete_and_exit+0x20/0x20
    ret_from_fork_asm+0x11/0x20
    </TASK>
    Modules linked in: xt_conntrack xt_MASQUERADE nf_conntrack_netlink nfnetlink xt_addrtype iptable_nat nf_nat br_netfilter rpcsec_gss_krb5 auth_rpcgss oid_registry overlay mlx5_ib ib_uverbs ib_core zram zsmalloc mlx5_core fuse
    ---[ end trace 0000000000000000 ]---

Fixes: 3178308a ("net/mlx5e: Make tx_port_ts logic resilient to out-of-order CQEs")
Signed-off-by: Rahul Rameshbabu <rrameshbabu@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Link: https://lore.kernel.org/r/20231114215846.5902-11-saeed@kernel.orgSigned-off-by: Jakub Kicinski <kuba@kernel.org>

When SQ is a port timestamping SQ for PTP, do not access tx flags of skb
after free-ing the skb. Free the skb only after all references that depend
on it have been handled in the dropped WQE path.

Fixes: 3178308a ("net/mlx5e: Make tx_port_ts logic resilient to out-of-order CQEs")
Signed-off-by: Rahul Rameshbabu <rrameshbabu@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Link: https://lore.kernel.org/r/20231114215846.5902-10-saeed@kernel.orgSigned-off-by: Jakub Kicinski <kuba@kernel.org>

As IPSec packet offload in switchdev mode is not supported with LAG,
it's unnecessary to modify those sent-to-vport rules to the peer eswitch.

Fixes: c6c2bf5d ("net/mlx5e: Support IPsec packet offload for TX in switchdev mode")
Signed-off-by: Jianbo Liu <jianbol@nvidia.com>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Link: https://lore.kernel.org/r/20231114215846.5902-9-saeed@kernel.orgSigned-off-by: Jakub Kicinski <kuba@kernel.org>

Referenced commit addressed endianness issue in mlx5 pedit implementation
in ad hoc manner instead of systematically treating integer values
according to their types which left pedit fields of sizes not equal to 4
and where the bytes being modified are not least significant ones broken on
big endian machines since wrong bits will be consumed during parsing which
leads to following example error when applying pedit to source and
destination MAC addresses:

[Wed Oct 18 12:52:42 2023] mlx5_core 0001:00:00.1 p1v3_r: attempt to offload an unsupported field (cmd 0)
[Wed Oct 18 12:52:42 2023] mask: 00000000330c5b68: 00 00 00 00 ff ff 00 00 00 00 ff ff 00 00 00 00  ................
[Wed Oct 18 12:52:42 2023] mask: 0000000017d22fd9: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
[Wed Oct 18 12:52:42 2023] mask: 000000008186d717: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
[Wed Oct 18 12:52:42 2023] mask: 0000000029eb6149: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
[Wed Oct 18 12:52:42 2023] mask: 000000007ed103e4: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
[Wed Oct 18 12:52:42 2023] mask: 00000000db8101a6: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
[Wed Oct 18 12:52:42 2023] mask: 00000000ec3c08a9: 00 00 00 00 00 00 00 00 00 00 00 00              ............

Treat masks and values of pedit and filter match as network byte order,
refactor pointers to them to void pointers instead of confusing u32
pointers and only cast to pointer-to-integer when reading a value from
them. Treat pedit mlx5_fields->field_mask as host byte order according to
its type u32, change the constants in fields array accordingly.

Fixes: 82198d8b ("net/mlx5e: Fix endianness when calculating pedit mask first bit")
Signed-off-by: Vlad Buslov <vladbu@nvidia.com>
Reviewed-by: Gal Pressman <gal@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Link: https://lore.kernel.org/r/20231114215846.5902-8-saeed@kernel.orgSigned-off-by: Jakub Kicinski <kuba@kernel.org>

Follow up to the previous patch to fix the same issue for
mlx5e_tc_tun_update_header_ipv4{6} when mlx5_packet_reformat_alloc()
fails.

When mlx5_packet_reformat_alloc() fails, the encap_header allocated in
mlx5e_tc_tun_update_header_ipv4{6} will be released within it. However,
e->encap_header is already set to the previously freed encap_header
before mlx5_packet_reformat_alloc(). As a result, the later
mlx5e_encap_put() will free e->encap_header again, causing a double free
issue.

mlx5e_encap_put()
     --> mlx5e_encap_dealloc()
         --> kfree(e->encap_header)

This patch fix it by not setting e->encap_header until
mlx5_packet_reformat_alloc() success.

Fixes: a54e20b4 ("net/mlx5e: Add basic TC tunnel set action for SRIOV offloads")
Signed-off-by: Gavin Li <gavinl@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Link: https://lore.kernel.org/r/20231114215846.5902-7-saeed@kernel.orgSigned-off-by: Jakub Kicinski <kuba@kernel.org>

When mlx5_packet_reformat_alloc() fails, the encap_header allocated in
mlx5e_tc_tun_create_header_ipv4{6} will be released within it. However,
e->encap_header is already set to the previously freed encap_header
before mlx5_packet_reformat_alloc(). As a result, the later
mlx5e_encap_put() will free e->encap_header again, causing a double free
issue.

mlx5e_encap_put()
    --> mlx5e_encap_dealloc()
        --> kfree(e->encap_header)

This happens when cmd: MLX5_CMD_OP_ALLOC_PACKET_REFORMAT_CONTEXT fail.

This patch fix it by not setting e->encap_header until
mlx5_packet_reformat_alloc() success.

Fixes: d589e785 ("net/mlx5e: Allow concurrent creation of encap entries")
Reported-by: Cruz Zhao <cruzzhao@linux.alibaba.com>
Reported-by: Tianchen Ding <dtcccc@linux.alibaba.com>
Signed-off-by: Dust Li <dust.li@linux.alibaba.com>
Reviewed-by: Wojciech Drewek <wojciech.drewek@intel.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

When running a phase adjustment operation, the free running clock should
not be modified at all. The phase control keyword is intended to trigger an
internal servo on the device that will converge to the provided delta. A
free running counter cannot implement phase adjustment.

Fixes: 8e11a68e ("net/mlx5: Add adjphase function to support hardware-only offset control")
Signed-off-by: Rahul Rameshbabu <rrameshbabu@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Link: https://lore.kernel.org/r/20231114215846.5902-5-saeed@kernel.orgSigned-off-by: Jakub Kicinski <kuba@kernel.org>

The current check isn't aware of old devices that don't have the
relevant FW capability. This patch allows multi destination FTE
in old cards, as it was before this check.

Fixes: f6f46e71 ("net/mlx5: DR, Add check for multi destination FTE")
Signed-off-by: Erez Shitrit <erezsh@nvidia.com>
Reviewed-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Link: https://lore.kernel.org/r/20231114215846.5902-4-saeed@kernel.orgSigned-off-by: Jakub Kicinski <kuba@kernel.org>

Each EQ table maintains a cpumask of the already used CPUs that are mapped
to IRQs to ensure that each IRQ gets mapped to a unique CPU.

However, on IRQ release, the said cpumask is not updated by clearing the
CPU from the mask to allow future IRQ request, causing the following
error when a SF is reloaded after it has utilized all CPUs for its IRQs:

mlx5_irq_affinity_request:135:(pid 306010): Didn't find a matching IRQ.
err = -28

Thus, when releasing an IRQ, clear its mapped CPU from the used CPUs
mask, to prevent the case described above.

While at it, move the used cpumask update to the EQ layer as it is more
fitting and preserves symmetricity of the IRQ request/release API.

Fixes: a1772de7 ("net/mlx5: Refactor completion IRQ request/release API")
Signed-off-by: Maher Sanalla <msanalla@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Reviewed-by: Moshe Shemesh <moshe@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Link: https://lore.kernel.org/r/20231114215846.5902-3-saeed@kernel.orgSigned-off-by: Jakub Kicinski <kuba@kernel.org>

This reverts commit 95c337cc.
The revert is required due to the suspicion it cause some tests
fail and will be moved to further investigation.

Fixes: 95c337cc ("net/mlx5: DR, Supporting inline WQE when possible")
Signed-off-by: Itamar Gozlan <igozlan@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Link: https://lore.kernel.org/r/20231114215846.5902-2-saeed@kernel.orgSigned-off-by: Jakub Kicinski <kuba@kernel.org>

Alexei Starovoitov says:

====================
pull-request: bpf 2023-11-15

We've added 7 non-merge commits during the last 6 day(s) which contain
a total of 9 files changed, 200 insertions(+), 49 deletions(-).

The main changes are:

1) Do not allocate bpf specific percpu memory unconditionally, from Yonghong.

2) Fix precision backtracking instruction iteration, from Andrii.

3) Fix control flow graph checking, from Andrii.

4) Fix xskxceiver selftest build, from Anders.

* https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf:
  bpf: Do not allocate percpu memory at init stage
  selftests/bpf: add more test cases for check_cfg()
  bpf: fix control-flow graph checking in privileged mode
  selftests/bpf: add edge case backtracking logic test
  bpf: fix precision backtracking instruction iteration
  bpf: handle ldimm64 properly in check_cfg()
  selftests: bpf: xskxceiver: ksft_print_msg: fix format type error
====================

Link: https://lore.kernel.org/r/20231115214949.48854-1-alexei.starovoitov@gmail.comSigned-off-by: Jakub Kicinski <kuba@kernel.org>

Kirill Shutemov reported significant percpu memory consumption increase after
booting in 288-cpu VM ([1]) due to commit 41a5db8d ("bpf: Add support for
non-fix-size percpu mem allocation"). The percpu memory consumption is
increased from 111MB to 969MB. The number is from /proc/meminfo.

I tried to reproduce the issue with my local VM which at most supports upto
255 cpus. With 252 cpus, without the above commit, the percpu memory
consumption immediately after boot is 57MB while with the above commit the
percpu memory consumption is 231MB.

This is not good since so far percpu memory from bpf memory allocator is not
widely used yet. Let us change pre-allocation in init stage to on-demand
allocation when verifier detects there is a need of percpu memory for bpf
program. With this change, percpu memory consumption after boot can be reduced
signicantly.

  [1] https://lore.kernel.org/lkml/20231109154934.4saimljtqx625l3v@box.shutemov.name/

Fixes: 41a5db8d ("bpf: Add support for non-fix-size percpu mem allocation")
Reported-and-tested-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Signed-off-by: Yonghong Song <yonghong.song@linux.dev>
Acked-by: Hou Tao <houtao1@huawei.com>
Link: https://lore.kernel.org/r/20231111013928.948838-1-yonghong.song@linux.devSigned-off-by: Alexei Starovoitov <ast@kernel.org>

Cited commit removed the strscpy() call and kept the snprintf() only.

It is common to use 'dev->name' as the format string before a netdev is
registered, this results in 'res' and 'name' pointers being equal.
According to POSIX, if copying takes place between objects that overlap
as a result of a call to sprintf() or snprintf(), the results are
undefined.

Add back the strscpy() and use 'buf' as an intermediate buffer.

Fixes: 7ad17b04 ("net: trust the bitmap in __dev_alloc_name()")
Cc: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Vlad Buslov <vladbu@nvidia.com>
Signed-off-by: Gal Pressman <gal@nvidia.com>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Simon Horman <horms@kernel.org>
Reviewed-by: Jiri Pirko <jiri@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

The top syzbot report for networking (#14 for the entire kernel)
is the queue timeout splat. We kept it around for a long time,
because in real life it provides pretty strong signal that
something is wrong with the driver or the device.

Removing it is also likely to break monitoring for those who
track it as a kernel warning.

Nevertheless, WARN()ings are best suited for catching kernel
programming bugs. If a Tx queue gets starved due to a pause
storm, priority configuration, or other weirdness - that's
obviously a problem, but not a problem we can fix at
the kernel level.

Bite the bullet and convert the WARN() to a print.

Before:

  NETDEV WATCHDOG: eni1np1 (netdevsim): transmit queue 0 timed out 1975 ms
  WARNING: CPU: 0 PID: 0 at net/sched/sch_generic.c:525 dev_watchdog+0x39e/0x3b0
  [... completely pointless stack trace of a timer follows ...]

Now:

  netdevsim netdevsim1 eni1np1: NETDEV WATCHDOG: CPU: 0: transmit queue 0 timed out 1769 ms

Alternatively we could mark the drivers which syzbot has
learned to abuse as "print-instead-of-WARN" selectively.

Reported-by: syzbot+d55372214aff0faa1f1f@syzkaller.appspotmail.com
Reviewed-by: Jiri Pirko <jiri@nvidia.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

Michael Chan says:

====================
bnxt_en: TX path improvements

All patches in this patchset are related to improving the TX path.
There are 2 areas of improvements:

1. The TX interrupt logic currently counts the number of TX completions
to determine the number of TX SKBs to free.  We now change it so that
the TX completion will now contain the hardware consumer index
information.  The driver will keep track of the latest hardware
consumer index from the last TX completion and clean up all TX SKBs
up to that index.  This scheme aligns better with future chips and
allows xmit_more code path to be more optimized.

2. The current driver logic requires an additional MSIX for each
additional MQPRIO TX ring.  This scheme uses too many MSIX vectors if
the user enables a large number of MQPRIO TCs.  We now use a new scheme
that will use the same MSIX for all the MQPRIO TX rings for each
ethtool channel.  Each ethtool TX channel can have up to 8 MQPRIO
TX rings and now they all will share the same MSIX.

v2: Rebased

v1 posted on Oct 27 2023 right before the close of net-next:

https://lore.kernel.org/netdev/20231027232252.36111-1-michael.chan@broadcom.com/
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

Now that we use the cumulative consumer index scheme for TX completion,
we don't need to have one TX completion per TX packet in the xmit_more
code path.  Set the TX_BD_FLAGS_NO_CMPL flag if xmit_more is true.
Fallback to one interrupt per packet if the ring is filled beyond
bp->tx_wake_thresh.

Also, move the wmb() to bnxt_txr_db_kick().  When xmit_more is true,
we'll skip the bnxt_txr_db_kick() call and there is no need to call
wmb() to sync. the TX BD data.
Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
Reviewed-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

We can now fully support sharing the same MSIX for all mqprio TX rings
belonging to the same ethtool channel with the new infrastructure:

1. Allocate the proper entries for cp_ring_arr in struct bnxt_cp_ring_info
to support the additional TX rings.

2. Populate the tx_ring array in struct bnxt_napi for all TX rings
sharing the same NAPI.

3. bnxt_num_tx_to_cp() returns the proper NQ/completion rings to support
the TX rings in the input.

4. Adjust bnxt_get_num_ring_stats() for the reduced number of ring
counters with the new scheme.
Reviewed-by: Andy Gospodarek <andrew.gospodarek@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Add 3 macros that handle to conversions between TC numbers and TX
ring numbers.  These will help to clarify the existing logic and the
new logic in the next patch.
Reviewed-by: Andy Gospodarek <andrew.gospodarek@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Up until now, each TX ring always requires a completion ring/NQ/MSIX.
bnxt_trim_rings() and the assignment of bp->cp_nr_rings always make
this assumption.  This will no longer be true in the next patches, so
we refactor and add helper functions to determine the proper relationship
between TX rings and the required completion ring/NQ/MSIX.  This patch
does not change the 1:1 relationship yet.

Note that on P5 chips, each RX and TX ring still requires a completion
ring.  Only the number of NQs has been reduced.  We should no longer call
bnxt_trim_rings() to adjust the RX and TX rings on P5 chips.  Replace with
simple logic to check that RX + TX < CP and adjust accordingly.

bnxt_check_rings() should call _bnxt_get_max_rings() to get the raw
number of rings instead of bnxt_get_max_rings().  If we are about to
create TCs, bnxt_get_max_rings() would not be able to calculate the max
rings correctly.
Reviewed-by: Andy Gospodarek <andrew.gospodarek@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

For each mqprio TC, we allocate a set of TX rings to map to the new
hardware CoS queue.  Expand the tx_ring pointer in struct bnxt_napi
to an array of 8 to support up to 8 TX rings, one for each TC.
Only array entry 0 is used at this time.  The rest of the array
entries will be used in later patches.
Reviewed-by: Andy Gospodarek <andrew.gospodarek@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Add 2 helper functions to set coalescing for each RX and TX rings.  This
will make it easier to expand the number of TX rings per MSIX in the
next patches.
Reviewed-by: Andy Gospodarek <andrew.gospodarek@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>