Skip to content

L
linux
Switch branch/tag
  1. 22 Oct, 2021 14 commits
    • Xin Long's avatar
      sctp: add vtag check in sctp_sf_do_8_5_1_E_sa · ef16b173
      Xin Long authored 
      sctp_sf_do_8_5_1_E_sa() is called when processing SHUTDOWN_ACK chunk
in cookie_wait and cookie_echoed state.

The vtag in the chunk's sctphdr should be verified, otherwise, as
later in chunk length check, it may send abort with the existent
asoc's vtag, which can be exploited by one to cook a malicious
chunk to terminate a SCTP asoc.

Note that when fails to verify the vtag from SHUTDOWN-ACK chunk,
SHUTDOWN COMPLETE message will still be sent back to peer, but
with the vtag from SHUTDOWN-ACK chunk, as said in 5) of
rfc4960#section-8.4.

While at it, also remove the unnecessary chunk length check from
sctp_sf_shut_8_4_5(), as it's already done in both places where
it calls sctp_sf_shut_8_4_5().

Fixes: 1da177e4 ("Linux-2.6.12-rc2")
Signed-off-by: default avatarXin Long <lucien.xin@gmail.com>
Acked-by: default avatarMarcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: default avatarJakub Kicinski <kuba@kernel.org>
      ef16b173
    • Xin Long's avatar
      sctp: add vtag check in sctp_sf_violation · aa0f697e
      Xin Long authored 
      sctp_sf_violation() is called when processing HEARTBEAT_ACK chunk
in cookie_wait state, and some other places are also using it.

The vtag in the chunk's sctphdr should be verified, otherwise, as
later in chunk length check, it may send abort with the existent
asoc's vtag, which can be exploited by one to cook a malicious
chunk to terminate a SCTP asoc.

Fixes: 1da177e4 ("Linux-2.6.12-rc2")
Signed-off-by: default avatarXin Long <lucien.xin@gmail.com>
Acked-by: default avatarMarcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: default avatarJakub Kicinski <kuba@kernel.org>
      aa0f697e
    • Xin Long's avatar
      sctp: fix the processing for COOKIE_ECHO chunk · a64b341b
      Xin Long authored 
      1. In closed state: in sctp_sf_do_5_1D_ce():

  When asoc is NULL, making packet for abort will use chunk's vtag
  in sctp_ootb_pkt_new(). But when asoc exists, vtag from the chunk
  should be verified before using peer.i.init_tag to make packet
  for abort in sctp_ootb_pkt_new(), and just discard it if vtag is
  not correct.

2. In the other states: in sctp_sf_do_5_2_4_dupcook():

  asoc always exists, but duplicate cookie_echo's vtag will be
  handled by sctp_tietags_compare() and then take actions, so before
  that we only verify the vtag for the abort sent for invalid chunk
  length.

Fixes: 1da177e4 ("Linux-2.6.12-rc2")
Signed-off-by: default avatarXin Long <lucien.xin@gmail.com>
Acked-by: default avatarMarcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: default avatarJakub Kicinski <kuba@kernel.org>
      a64b341b
    • Xin Long's avatar
      sctp: fix the processing for INIT_ACK chunk · 438b95a7
      Xin Long authored 
      Currently INIT_ACK chunk in non-cookie_echoed state is processed in
sctp_sf_discard_chunk() to send an abort with the existent asoc's
vtag if the chunk length is not valid. But the vtag in the chunk's
sctphdr is not verified, which may be exploited by one to cook a
malicious chunk to terminal a SCTP asoc.

sctp_sf_discard_chunk() also is called in many other places to send
an abort, and most of those have this problem. This patch is to fix
it by sending abort with the existent asoc's vtag only if the vtag
from the chunk's sctphdr is verified in sctp_sf_discard_chunk().

Note on sctp_sf_do_9_1_abort() and sctp_sf_shutdown_pending_abort(),
the chunk length has been verified before sctp_sf_discard_chunk(),
so replace it with sctp_sf_discard(). On sctp_sf_do_asconf_ack() and
sctp_sf_do_asconf(), move the sctp_chunk_length_valid check ahead of
sctp_sf_discard_chunk(), then replace it with sctp_sf_discard().

Fixes: 1da177e4 ("Linux-2.6.12-rc2")
Signed-off-by: default avatarXin Long <lucien.xin@gmail.com>
Acked-by: default avatarMarcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: default avatarJakub Kicinski <kuba@kernel.org>
      438b95a7
    • Xin Long's avatar
      sctp: fix the processing for INIT chunk · eae57839
      Xin Long authored 
      This patch fixes the problems below:

1. In non-shutdown_ack_sent states: in sctp_sf_do_5_1B_init() and
   sctp_sf_do_5_2_2_dupinit():

  chunk length check should be done before any checks that may cause
  to send abort, as making packet for abort will access the init_tag
  from init_hdr in sctp_ootb_pkt_new().

2. In shutdown_ack_sent state: in sctp_sf_do_9_2_reshutack():

  The same checks as does in sctp_sf_do_5_2_2_dupinit() is needed
  for sctp_sf_do_9_2_reshutack().

Fixes: 1da177e4 ("Linux-2.6.12-rc2")
Signed-off-by: default avatarXin Long <lucien.xin@gmail.com>
Acked-by: default avatarMarcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: default avatarJakub Kicinski <kuba@kernel.org>
      eae57839
    • Xin Long's avatar
      sctp: use init_tag from inithdr for ABORT chunk · 4f7019c7
      Xin Long authored 
      Currently Linux SCTP uses the verification tag of the existing SCTP
asoc when failing to process and sending the packet with the ABORT
chunk. This will result in the peer accepting the ABORT chunk and
removing the SCTP asoc. One could exploit this to terminate a SCTP
asoc.

This patch is to fix it by always using the initiate tag of the
received INIT chunk for the ABORT chunk to be sent.

Fixes: 1da177e4 ("Linux-2.6.12-rc2")
Signed-off-by: default avatarXin Long <lucien.xin@gmail.com>
Acked-by: default avatarMarcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: default avatarJakub Kicinski <kuba@kernel.org>
      4f7019c7
    • Vasily Averin's avatar
      skb_expand_head() adjust skb->truesize incorrectly · 7f678def
      Vasily Averin authored 
      Christoph Paasch reports [1] about incorrect skb->truesize
after skb_expand_head() call in ip6_xmit.
This may happen because of two reasons:
- skb_set_owner_w() for newly cloned skb is called too early,
before pskb_expand_head() where truesize is adjusted for (!skb-sk) case.
- pskb_expand_head() does not adjust truesize in (skb->sk) case.
In this case sk->sk_wmem_alloc should be adjusted too.

[1] https://lkml.org/lkml/2021/8/20/1082

Fixes: f1260ff1 ("skbuff: introduce skb_expand_head()")
Fixes: 2d85a1b3 ("ipv6: ip6_finish_output2: set sk into newly allocated nskb")
Reported-by: default avatarChristoph Paasch <christoph.paasch@gmail.com>
Signed-off-by: default avatarVasily Averin <vvs@virtuozzo.com>
Reviewed-by: default avatarEric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/644330dd-477e-0462-83bf-9f514c41edd1@virtuozzo.comSigned-off-by: default avatarJakub Kicinski <kuba@kernel.org>
      7f678def
    • Jakub Kicinski's avatar
      Merge tag 'mac80211-for-net-2021-10-21' of... · 7fcb1c95
      Jakub Kicinski authored 
      Merge tag 'mac80211-for-net-2021-10-21' of git://git.kernel.org/pub/scm/linux/kernel/git/jberg/mac80211

Johannes Berg says:

====================
Two small fixes:
 * RCU misuse in scan processing in cfg80211
 * missing size check for HE data in mac80211 mesh

* tag 'mac80211-for-net-2021-10-21' of git://git.kernel.org/pub/scm/linux/kernel/git/jberg/mac80211:
  cfg80211: scan: fix RCU in cfg80211_add_nontrans_list()
  mac80211: mesh: fix HE operation element length check
====================

Link: https://lore.kernel.org/r/20211021154351.134297-1-johannes@sipsolutions.netSigned-off-by: default avatarJakub Kicinski <kuba@kernel.org>
      7fcb1c95
    • Linus Torvalds's avatar
      Merge tag 'drm-fixes-2021-10-22' of git://anongit.freedesktop.org/drm/drm · 64222515
      Linus Torvalds authored 
      Pull drm fixes from Dave Airlie:
 "Nothing too crazy at the end of the cycle, the kmb modesetting fixes
  are probably a bit large but it's not a major driver, and its fixing
  monitor doesn't turn on type problems.

  Otherwise it's just a few minor patches, one ast regression revert, an
  msm power stability fix.

  ast:
   - fix regression with connector detect

  msm:
   - fix power stability issue

  msxfb:
   - fix crash on unload

  panel:
   - sync fix

  kmb:
   - modesetting fixes"

* tag 'drm-fixes-2021-10-22' of git://anongit.freedesktop.org/drm/drm:
  Revert "drm/ast: Add detect function support"
  drm/kmb: Enable ADV bridge after modeset
  drm/kmb: Corrected typo in handle_lcd_irq
  drm/kmb: Disable change of plane parameters
  drm/kmb: Remove clearing DPHY regs
  drm/kmb: Limit supported mode to 1080p
  drm/kmb: Work around for higher system clock
  drm/panel: ilitek-ili9881c: Fix sync for Feixin K101-IM2BYL02 panel
  drm: mxsfb: Fix NULL pointer dereference crash on unload
  drm/msm/devfreq: Restrict idle clamping to a618 for now
      64222515
    • Mike Rapoport's avatar
      memblock: exclude MEMBLOCK_NOMAP regions from kmemleak · 658aafc8
      Mike Rapoport authored 
      Vladimir Zapolskiy reports:

Commit a7259df7 ("memblock: make memblock_find_in_range method
private") invokes a kernel panic while running kmemleak on OF platforms
with nomaped regions:

  Unable to handle kernel paging request at virtual address fff000021e00000
  [...]
    scan_block+0x64/0x170
    scan_gray_list+0xe8/0x17c
    kmemleak_scan+0x270/0x514
    kmemleak_write+0x34c/0x4ac

The memory allocated from memblock is registered with kmemleak, but if
it is marked MEMBLOCK_NOMAP it won't have linear map entries so an
attempt to scan such areas will fault.

Ideally, memblock_mark_nomap() would inform kmemleak to ignore
MEMBLOCK_NOMAP memory, but it can be called before kmemleak interfaces
operating on physical addresses can use __va() conversion.

Make sure that functions that mark allocated memory as MEMBLOCK_NOMAP
take care of informing kmemleak to ignore such memory.

Link: https://lore.kernel.org/all/8ade5174-b143-d621-8c8e-dc6a1898c6fb@linaro.org
Link: https://lore.kernel.org/all/c30ff0a2-d196-c50d-22f0-bd50696b1205@quicinc.com
Fixes: a7259df7 ("memblock: make memblock_find_in_range method private")
Reported-by: default avatarVladimir Zapolskiy <vladimir.zapolskiy@linaro.org>
Signed-off-by: default avatarMike Rapoport <rppt@linux.ibm.com>
Reviewed-by: default avatarCatalin Marinas <catalin.marinas@arm.com>
Tested-by: default avatarVladimir Zapolskiy <vladimir.zapolskiy@linaro.org>
Tested-by: default avatarQian Cai <quic_qiancai@quicinc.com>
Signed-off-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
      658aafc8
    • Mike Rapoport's avatar
      Revert "memblock: exclude NOMAP regions from kmemleak" · 6c9a5455
      Mike Rapoport authored 
      Commit 6e44bd6d ("memblock: exclude NOMAP regions from kmemleak")
breaks boot on EFI systems with kmemleak and VM_DEBUG enabled:

  efi: Processing EFI memory map:
  efi:   0x000090000000-0x000091ffffff [Conventional|   |  |  |  |  |  |  |  |  |   |WB|WT|WC|UC]
  efi:   0x000092000000-0x0000928fffff [Runtime Data|RUN|  |  |  |  |  |  |  |  |   |WB|WT|WC|UC]
  ------------[ cut here ]------------
  kernel BUG at mm/kmemleak.c:1140!
  Internal error: Oops - BUG: 0 [#1] SMP
  Modules linked in:
  CPU: 0 PID: 0 Comm: swapper Not tainted 5.15.0-rc6-next-20211019+ #104
  pstate: 600000c5 (nZCv daIF -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
  pc : kmemleak_free_part_phys+0x64/0x8c
  lr : kmemleak_free_part_phys+0x38/0x8c
  sp : ffff800011eafbc0
  x29: ffff800011eafbc0 x28: 1fffff7fffb41c0d x27: fffffbfffda0e068
  x26: 0000000092000000 x25: 1ffff000023d5f94 x24: ffff800011ed84d0
  x23: ffff800011ed84c0 x22: ffff800011ed83d8 x21: 0000000000900000
  x20: ffff800011782000 x19: 0000000092000000 x18: ffff800011ee0730
  x17: 0000000000000000 x16: 0000000000000000 x15: 1ffff0000233252c
  x14: ffff800019a905a0 x13: 0000000000000001 x12: ffff7000023d5ed7
  x11: 1ffff000023d5ed6 x10: ffff7000023d5ed6 x9 : dfff800000000000
  x8 : ffff800011eaf6b7 x7 : 0000000000000001 x6 : ffff800011eaf6b0
  x5 : 00008ffffdc2a12a x4 : ffff7000023d5ed7 x3 : 1ffff000023dbf99
  x2 : 1ffff000022f0463 x1 : 0000000000000000 x0 : ffffffffffffffff
  Call trace:
   kmemleak_free_part_phys+0x64/0x8c
   memblock_mark_nomap+0x5c/0x78
   reserve_regions+0x294/0x33c
   efi_init+0x2d0/0x490
   setup_arch+0x80/0x138
   start_kernel+0xa0/0x3ec
   __primary_switched+0xc0/0xc8
  Code: 34000041 97d526e7 f9418e80 36000040 (d4210000)
  random: get_random_bytes called from print_oops_end_marker+0x34/0x80 with crng_init=0
  ---[ end trace 0000000000000000 ]---

The crash happens because kmemleak_free_part_phys() tries to use __va()
before memstart_addr is initialized and this triggers a VM_BUG_ON() in
arch/arm64/include/asm/memory.h:

Revert 6e44bd6d ("memblock: exclude NOMAP regions from kmemleak"),
the issue it is fixing will be fixed differently.
Reported-by: default avatarQian Cai <quic_qiancai@quicinc.com>
Signed-off-by: default avatarMike Rapoport <rppt@linux.ibm.com>
Acked-by: default avatarCatalin Marinas <catalin.marinas@arm.com>
Signed-off-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
      6c9a5455
    • Linus Torvalds's avatar
      Merge branch 'ucount-fixes-for-v5.15' of... · 9d235ac0
      Linus Torvalds authored 
      Merge branch 'ucount-fixes-for-v5.15' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace

Pull ucounts fixes from Eric Biederman:
 "There has been one very hard to track down bug in the ucount code that
  we have been tracking since roughly v5.14 was released. Alex managed
  to find a reliable reproducer a few days ago and then I was able to
  instrument the code and figure out what the issue was.

  It turns out the sigqueue_alloc single atomic operation optimization
  did not play nicely with ucounts multiple level rlimits. It turned out
  that either sigqueue_alloc or sigqueue_free could be operating on
  multiple levels and trigger the conditions for the optimization on
  more than one level at the same time.

  To deal with that situation I have introduced inc_rlimit_get_ucounts
  and dec_rlimit_put_ucounts that just focuses on the optimization and
  the rlimit and ucount changes.

  While looking into the big bug I found I couple of other little issues
  so I am including those fixes here as well.

  When I have time I would very much like to dig into process ownership
  of the shared signal queue and see if we could pick a single owner for
  the entire queue so that all of the rlimits can count to that owner.
  That should entirely remove the need to call get_ucounts and
  put_ucounts in sigqueue_alloc and sigqueue_free. It is difficult
  because Linux unlike POSIX supports setuid that works on a single
  thread"

* 'ucount-fixes-for-v5.15' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace:
  ucounts: Move get_ucounts from cred_alloc_blank to key_change_session_keyring
  ucounts: Proper error handling in set_cred_ucounts
  ucounts: Pair inc_rlimit_ucounts with dec_rlimit_ucoutns in commit_creds
  ucounts: Fix signal ucount refcounting
      9d235ac0
    • Linus Torvalds's avatar
      Merge tag 'net-5.15-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net · 6c2c7127
      Linus Torvalds authored 
      Pull networking fixes from Jakub Kicinski:
 "Including fixes from netfilter, and can.

  We'll have one more fix for a socket accounting regression, it's still
  getting polished. Otherwise things look fine.

  Current release - regressions:

   - revert "vrf: reset skb conntrack connection on VRF rcv", there are
     valid uses for previous behavior

   - can: m_can: fix iomap_read_fifo() and iomap_write_fifo()

  Current release - new code bugs:

   - mlx5: e-switch, return correct error code on group creation failure

  Previous releases - regressions:

   - sctp: fix transport encap_port update in sctp_vtag_verify

   - stmmac: fix E2E delay mechanism (in PTP timestamping)

  Previous releases - always broken:

   - netfilter: ip6t_rt: fix out-of-bounds read of ipv6_rt_hdr

   - netfilter: xt_IDLETIMER: fix out-of-bound read caused by lack of
     init

   - netfilter: ipvs: make global sysctl read-only in non-init netns

   - tcp: md5: fix selection between vrf and non-vrf keys

   - ipv6: count rx stats on the orig netdev when forwarding

   - bridge: mcast: use multicast_membership_interval for IGMPv3

   - can:
      - j1939: fix UAF for rx_kref of j1939_priv abort sessions on
        receiving bad messages

      - isotp: fix TX buffer concurrent access in isotp_sendmsg() fix
        return error on FC timeout on TX path

   - ice: fix re-init of RDMA Tx queues and crash if RDMA was not inited

   - hns3: schedule the polling again when allocation fails, prevent
     stalls

   - drivers: add missing of_node_put() when aborting
     for_each_available_child_of_node()

   - ptp: fix possible memory leak and UAF in ptp_clock_register()

   - e1000e: fix packet loss in burst mode on Tiger Lake and later

   - mlx5e: ipsec: fix more checksum offload issues"

* tag 'net-5.15-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (75 commits)
  usbnet: sanity check for maxpacket
  net: enetc: make sure all traffic classes can send large frames
  net: enetc: fix ethtool counter name for PM0_TERR
  ptp: free 'vclock_index' in ptp_clock_release()
  sfc: Don't use netif_info before net_device setup
  sfc: Export fibre-specific supported link modes
  net/mlx5e: IPsec: Fix work queue entry ethernet segment checksum flags
  net/mlx5e: IPsec: Fix a misuse of the software parser's fields
  net/mlx5e: Fix vlan data lost during suspend flow
  net/mlx5: E-switch, Return correct error code on group creation failure
  net/mlx5: Lag, change multipath and bonding to be mutually exclusive
  ice: Add missing E810 device ids
  igc: Update I226_K device ID
  e1000e: Fix packet loss on Tiger Lake and later
  e1000e: Separate TGP board type from SPT
  ptp: Fix possible memory leak in ptp_clock_register()
  net: stmmac: Fix E2E delay mechanism
  nfc: st95hf: Make spi remove() callback return zero
  net: hns3: disable sriov before unload hclge layer
  net: hns3: fix vf reset workqueue cannot exit
  ...
      6c2c7127
    • Linus Torvalds's avatar
      Merge tag 'powerpc-5.15-5' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux · 0a3221b6
      Linus Torvalds authored 
      Pull powerpc fixes from Michael Ellerman:

 - Fix a bug exposed by a previous fix, where running guests with
   certain SMT topologies could crash the host on Power8.

 - Fix atomic sleep warnings when re-onlining CPUs, when PREEMPT is
   enabled.

Thanks to Nathan Lynch, Srikar Dronamraju, and Valentin Schneider.

* tag 'powerpc-5.15-5' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:
  powerpc/smp: do not decrement idle task preempt count in CPU offline
  powerpc/idle: Don't corrupt back chain when going idle
      0a3221b6
  3. 21 Oct, 2021 20 commits
  5. 20 Oct, 2021 6 commits
    • Linus Torvalds's avatar
      Merge tag 'ceph-for-5.15-rc7' of git://github.com/ceph/ceph-client · 2f111a6f
      Linus Torvalds authored 
      Pull ceph fixes from Ilya Dryomov:
 "Two important filesystem fixes, marked for stable.

  The blocklisted superblocks issue was particularly annoying because
  for unexperienced users it essentially exacted a reboot to establish a
  new functional mount in that scenario"

* tag 'ceph-for-5.15-rc7' of git://github.com/ceph/ceph-client:
  ceph: fix handling of "meta" errors
  ceph: skip existing superblocks that are blocklisted or shut down when mounting
      2f111a6f
    • Linus Torvalds's avatar
      Merge tag 'dma-mapping-5.15-2' of git://git.infradead.org/users/hch/dma-mapping · 515dcc2e
      Linus Torvalds authored 
      Pull dma-mapping fixes from Christoph Hellwig:

 - fix more dma-debug fallout (Gerald Schaefer, Hamza Mahfooz)

 - fix a kerneldoc warning (Logan Gunthorpe)

* tag 'dma-mapping-5.15-2' of git://git.infradead.org/users/hch/dma-mapping:
  dma-debug: teach add_dma_entry() about DMA_ATTR_SKIP_CPU_SYNC
  dma-debug: fix sg checks in debug_dma_map_sg()
  dma-mapping: fix the kerneldoc for dma_map_sgtable()
      515dcc2e
    • Emeel Hakim's avatar
      net/mlx5e: IPsec: Fix work queue entry ethernet segment checksum flags · 1d000323
      Emeel Hakim authored 
      Current Work Queue Entry (WQE) checksum (csum) flags in the ethernet
segment (eseg) in case of IPsec crypto offload datapath are not aligned
with PRM/HW expectations.

Currently the driver always sets the l3_inner_csum flag in case of IPsec
because of the wrong usage of skb->encapsulation as indicator for inner
IPsec header since skb->encapsulation is always ON for IPsec packets
since IPsec itself is an encapsulation protocol. The above forced a
failing attempts of calculating csum of non-existing segments (like in
the IP|ESP|TCP packet case which does not have an l3_inner) which led
to lots of packet drops hence the low throughput.

Fix by using xo->inner_ipproto as indicator for inner IPsec header
instead of skb->encapsulation in addition to setting the csum flags
as following:
* Tunnel Mode:
* Pkt: MAC  IP     ESP  IP    L4
* CSUM: l3_cs | l3_inner_cs | l4_inner_cs
*
* Transport Mode:
* Pkt: MAC  IP     ESP  L4
* CSUM: l3_cs [ | l4_cs (checksum partial case)]
*
* Tunnel(VXLAN TCP/UDP) over Transport Mode
* Pkt: MAC  IP     ESP  UDP  VXLAN  IP    L4
* CSUM: l3_cs | l3_inner_cs | l4_inner_cs

Fixes: f1267798 ("net/mlx5: Fix checksum issue of VXLAN and IPsec crypto offload")
Signed-off-by: default avatarEmeel Hakim <ehakim@nvidia.com>
Signed-off-by: default avatarSaeed Mahameed <saeedm@nvidia.com>
      1d000323
    • Emeel Hakim's avatar
      net/mlx5e: IPsec: Fix a misuse of the software parser's fields · d10457f8
      Emeel Hakim authored 
      IPsec crypto offload current Software Parser (SWP) fields settings in
the ethernet segment (eseg) are not aligned with PRM/HW expectations.
Among others in case of IP|ESP|TCP packet, current driver sets the
offsets for inner_l3 and inner_l4 although there is no inner l3/l4
headers relative to ESP header in such packets.

SWP provides the offsets for HW ,so it can be used to find csum fields
to offload the checksum, however these are not necessarily used by HW
and are used as fallback in case HW fails to parse the packet, e.g
when performing IPSec Transport Aware (IP | ESP | TCP) there is no
need to add SW parse on inner packet. So in some cases packets csum
was calculated correctly , whereas in other cases it failed. The later
faced csum errors (caused by wrong packet length calculations) which
led to lots of packet drops hence the low throughput.

Fix by setting the SWP fields as expected in a IP|ESP|TCP packet.

the following describe the expected SWP offsets:
* Tunnel Mode:
* SWP:      OutL3       InL3  InL4
* Pkt: MAC  IP     ESP  IP    L4
*
* Transport Mode:
* SWP:      OutL3       OutL4
* Pkt: MAC  IP     ESP  L4
*
* Tunnel(VXLAN TCP/UDP) over Transport Mode
* SWP:      OutL3                   InL3  InL4
* Pkt: MAC  IP     ESP  UDP  VXLAN  IP    L4

Fixes: f1267798 ("net/mlx5: Fix checksum issue of VXLAN and IPsec crypto offload")
Signed-off-by: default avatarEmeel Hakim <ehakim@nvidia.com>
Reviewed-by: default avatarRaed Salem <raeds@nvidia.com>
Signed-off-by: default avatarSaeed Mahameed <saeedm@nvidia.com>
      d10457f8
    • Moshe Shemesh's avatar
      net/mlx5e: Fix vlan data lost during suspend flow · 68e66e1a
      Moshe Shemesh authored 
      During suspend flow the driver calls mlx5e_destroy_vlan_table() which
does not only delete the vlans steering flow rules, but also frees the
data on currently active vlans, thus it is not restored during resume
flow.

This fix keeps the vlan data on suspend flow and frees it only on driver
remove flow.

Fixes: 6783f0a2 ("net/mlx5e: Dynamic alloc vlan table for netdev when needed")
Signed-off-by: default avatarMoshe Shemesh <moshe@nvidia.com>
Reviewed-by: default avatarTariq Toukan <tariqt@nvidia.com>
Signed-off-by: default avatarSaeed Mahameed <saeedm@nvidia.com>
      68e66e1a
    • Dmytro Linkin's avatar
      net/mlx5: E-switch, Return correct error code on group creation failure · a6f74333
      Dmytro Linkin authored 
      Dan Carpenter report:
The patch f47e04eb: "net/mlx5: E-switch, Allow setting share/max
tx rate limits of rate groups" from May 31, 2021, leads to the
following Smatch static checker warning:

	drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c:483 esw_qos_create_rate_group()
	warn: passing zero to 'ERR_PTR'

If min rate normalization failed then error code may be overwritten to 0
if scheduling element destruction succeed. Ignore this value and always
return initial one.

Fixes: f47e04eb ("net/mlx5: E-switch, Allow setting share/max tx rate limits of rate groups")
Reported-by: default avatarDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: default avatarDmytro Linkin <dlinkin@nvidia.com>
Signed-off-by: default avatarSaeed Mahameed <saeedm@nvidia.com>
      a6f74333
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7