linux-stable.git/net/xdp, branch linux-6.2.y

xsk: Fix unaligned descriptor validation

2023-05-11T14:11:12+00:00

[ Upstream commit d769ccaf957fe7391f357c0a923de71f594b8a2b ]

Make sure unaligned descriptors that straddle the end of the UMEM are
considered invalid. Currently, descriptor validation is broken for
zero-copy mode which only checks descriptors at page granularity.
For example, descriptors in zero-copy mode that overrun the end of the
UMEM but not a page boundary are (incorrectly) considered valid. The
UMEM boundary check needs to happen before the page boundary and
contiguity checks in xp_desc_crosses_non_contig_pg(). Do this check in
xp_unaligned_validate_desc() instead like xp_check_unaligned() already
does.

Fixes: 2b43470add8c ("xsk: Introduce AF_XDP buffer allocation API")
Signed-off-by: Kal Conley 
Acked-by: Magnus Karlsson 
Link: https://lore.kernel.org/r/20230405235920.7305-2-kal.conley@dectris.com
Signed-off-by: Martin KaFai Lau 
Signed-off-by: Sasha Levin

bpf: return long from bpf_map_ops funcs

2023-05-11T14:11:09+00:00

[ Upstream commit d7ba4cc900bf1eea2d8c807c6b1fc6bd61f41237 ]

This patch changes the return types of bpf_map_ops functions to long, where
previously int was returned. Using long allows for bpf programs to maintain
the sign bit in the absence of sign extension during situations where
inlined bpf helper funcs make calls to the bpf_map_ops funcs and a negative
error is returned.

The definitions of the helper funcs are generated from comments in the bpf
uapi header at `include/uapi/linux/bpf.h`. The return type of these
helpers was previously changed from int to long in commit bdb7b79b4ce8. For
any case where one of the map helpers call the bpf_map_ops funcs that are
still returning 32-bit int, a compiler might not include sign extension
instructions to properly convert the 32-bit negative value a 64-bit
negative value.

For example:
bpf assembly excerpt of an inlined helper calling a kernel function and
checking for a specific error:

; err = bpf_map_update_elem(&mymap, &key, &val, BPF_NOEXIST);
  ...
  46:	call   0xffffffffe103291c	; htab_map_update_elem
; if (err && err != -EEXIST) {
  4b:	cmp    $0xffffffffffffffef,%rax ; cmp -EEXIST,%rax

kernel function assembly excerpt of return value from
`htab_map_update_elem` returning 32-bit int:

movl $0xffffffef, %r9d
...
movl %r9d, %eax

...results in the comparison:
cmp $0xffffffffffffffef, $0x00000000ffffffef

Fixes: bdb7b79b4ce8 ("bpf: Switch most helper return values from 32-bit int to 64-bit long")
Tested-by: Eduard Zingerman 
Signed-off-by: JP Kobryn 
Link: https://lore.kernel.org/r/20230322194754.185781-3-inwardvessel@gmail.com
Signed-off-by: Alexei Starovoitov 
Signed-off-by: Sasha Levin

xsk: Add missing overflow check in xdp_umem_reg

2023-03-30T10:50:52+00:00

[ Upstream commit c7df4813b149362248d6ef7be41a311e27bf75fe ]

The number of chunks can overflow u32. Make sure to return -EINVAL on
overflow. Also remove a redundant u32 cast assigning umem->npgs.

Fixes: bbff2f321a86 ("xsk: new descriptor addressing scheme")
Signed-off-by: Kal Conley 
Signed-off-by: Daniel Borkmann 
Acked-by: Magnus Karlsson 
Link: https://lore.kernel.org/bpf/20230308174013.1114745-1-kal.conley@dectris.com
Signed-off-by: Sasha Levin

xsk: check IFF_UP earlier in Tx path

2023-03-10T08:28:23+00:00

[ Upstream commit 1596dae2f17ec5c6e8c8f0e3fec78c5ae55c1e0b ]

Xsk Tx can be triggered via either sendmsg() or poll() syscalls. These
two paths share a call to common function xsk_xmit() which has two
sanity checks within. A pseudo code example to show the two paths:

__xsk_sendmsg() :                       xsk_poll():
if (unlikely(!xsk_is_bound(xs)))        if (unlikely(!xsk_is_bound(xs)))
    return -ENXIO;                          return mask;
if (unlikely(need_wait))                (...)
    return -EOPNOTSUPP;                 xsk_xmit()
mark napi id
(...)
xsk_xmit()

xsk_xmit():
if (unlikely(!(xs->dev->flags & IFF_UP)))
	return -ENETDOWN;
if (unlikely(!xs->tx))
	return -ENOBUFS;

As it can be observed above, in sendmsg() napi id can be marked on
interface that was not brought up and this causes a NULL ptr
dereference:

[31757.505631] BUG: kernel NULL pointer dereference, address: 0000000000000018
[31757.512710] #PF: supervisor read access in kernel mode
[31757.517936] #PF: error_code(0x0000) - not-present page
[31757.523149] PGD 0 P4D 0
[31757.525726] Oops: 0000 [#1] PREEMPT SMP NOPTI
[31757.530154] CPU: 26 PID: 95641 Comm: xdpsock Not tainted 6.2.0-rc5+ #40
[31757.536871] Hardware name: Intel Corporation S2600WFT/S2600WFT, BIOS SE5C620.86B.02.01.0008.031920191559 03/19/2019
[31757.547457] RIP: 0010:xsk_sendmsg+0xde/0x180
[31757.551799] Code: 00 75 a2 48 8b 00 a8 04 75 9b 84 d2 74 69 8b 85 14 01 00 00 85 c0 75 1b 48 8b 85 28 03 00 00 48 8b 80 98 00 00 00 48 8b 40 20 <8b> 40 18 89 85 14 01 00 00 8b bd 14 01 00 00 81 ff 00 01 00 00 0f
[31757.570840] RSP: 0018:ffffc90034f27dc0 EFLAGS: 00010246
[31757.576143] RAX: 0000000000000000 RBX: ffffc90034f27e18 RCX: 0000000000000000
[31757.583389] RDX: 0000000000000001 RSI: ffffc90034f27e18 RDI: ffff88984cf3c100
[31757.590631] RBP: ffff88984714a800 R08: ffff88984714a800 R09: 0000000000000000
[31757.597877] R10: 0000000000000001 R11: 0000000000000000 R12: 00000000fffffffa
[31757.605123] R13: 0000000000000000 R14: 0000000000000003 R15: 0000000000000000
[31757.612364] FS:  00007fb4c5931180(0000) GS:ffff88afdfa00000(0000) knlGS:0000000000000000
[31757.620571] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[31757.626406] CR2: 0000000000000018 CR3: 000000184b41c003 CR4: 00000000007706e0
[31757.633648] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[31757.640894] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[31757.648139] PKRU: 55555554
[31757.650894] Call Trace:
[31757.653385]  
[31757.655524]  sock_sendmsg+0x8f/0xa0
[31757.659077]  ? sockfd_lookup_light+0x12/0x70
[31757.663416]  __sys_sendto+0xfc/0x170
[31757.667051]  ? do_sched_setscheduler+0xdb/0x1b0
[31757.671658]  __x64_sys_sendto+0x20/0x30
[31757.675557]  do_syscall_64+0x38/0x90
[31757.679197]  entry_SYSCALL_64_after_hwframe+0x72/0xdc
[31757.687969] Code: 8e f6 ff 44 8b 4c 24 2c 4c 8b 44 24 20 41 89 c4 44 8b 54 24 28 48 8b 54 24 18 b8 2c 00 00 00 48 8b 74 24 10 8b 7c 24 08 0f 05 <48> 3d 00 f0 ff ff 77 3a 44 89 e7 48 89 44 24 08 e8 b5 8e f6 ff 48
[31757.707007] RSP: 002b:00007ffd49c73c70 EFLAGS: 00000293 ORIG_RAX: 000000000000002c
[31757.714694] RAX: ffffffffffffffda RBX: 000055a996565380 RCX: 00007fb4c5727c16
[31757.721939] RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000003
[31757.729184] RBP: 0000000000000040 R08: 0000000000000000 R09: 0000000000000000
[31757.736429] R10: 0000000000000040 R11: 0000000000000293 R12: 0000000000000000
[31757.743673] R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000
[31757.754940]  

To fix this, let's make xsk_xmit a function that will be responsible for
generic Tx, where RCU is handled accordingly and pull out sanity checks
and xs->zc handling. Populate sanity checks to __xsk_sendmsg() and
xsk_poll().

Fixes: ca2e1a627035 ("xsk: Mark napi_id on sendmsg()")
Fixes: 18b1ab7aa76b ("xsk: Fix race at socket teardown")
Signed-off-by: Maciej Fijalkowski 
Reviewed-by: Alexander Lobakin 
Link: https://lore.kernel.org/r/20230215143309.13145-1-maciej.fijalkowski@intel.com
Signed-off-by: Martin KaFai Lau 
Signed-off-by: Daniel Borkmann 
Signed-off-by: Sasha Levin

bpf: Expand map key argument of bpf_redirect_map to u64

2022-11-15T17:00:27+00:00

For queueing packets in XDP we want to add a new redirect map type with
support for 64-bit indexes. To prepare fore this, expand the width of the
'key' argument to the bpf_redirect_map() helper. Since BPF registers are
always 64-bit, this should be safe to do after the fact.

Acked-by: Song Liu 
Reviewed-by: Stanislav Fomichev 
Signed-off-by: Toke Høiland-Jørgensen 
Link: https://lore.kernel.org/r/20221108140601.149971-3-toke@redhat.com
Signed-off-by: Alexei Starovoitov

Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net

2022-10-04T00:44:18+00:00

Merge in the left-over fixes before the net-next pull-request.

Conflicts:

drivers/net/ethernet/mediatek/mtk_ppe.c
  ae3ed15da588 ("net: ethernet: mtk_eth_soc: fix state in __mtk_foe_entry_clear")
  9d8cb4c096ab ("net: ethernet: mtk_eth_soc: add foe_entry_size to mtk_eth_soc")
https://lore.kernel.org/all/6cb6893b-4921-a068-4c30-1109795110bb@tessares.net/

kernel/bpf/helpers.c
  8addbfc7b308 ("bpf: Gate dynptr API behind CAP_BPF")
  5679ff2f138f ("bpf: Move bpf_loop and bpf_for_each_map_elem under CAP_BPF")
  8a67f2de9b1d ("bpf: expose bpf_strtol and bpf_strtoul to all program types")
https://lore.kernel.org/all/20221003201957.13149-1-daniel@iogearbox.net/

Signed-off-by: Jakub Kicinski

xsk: Expose min chunk size to drivers

2022-09-30T14:55:45+00:00

Drivers should be aware of the range of valid UMEM chunk sizes to be
able to allocate their internal structures of an appropriate size. It
will be used by mlx5e in the following patches.

Signed-off-by: Maxim Mikityanskiy 
Reviewed-by: Tariq Toukan 
CC: "Björn Töpel" 
CC: Magnus Karlsson 
CC: Maciej Fijalkowski 
Signed-off-by: Saeed Mahameed 
Signed-off-by: Jakub Kicinski

xsk: Inherit need_wakeup flag for shared sockets

2022-09-22T15:16:22+00:00

The flag for need_wakeup is not set for xsks with `XDP_SHARED_UMEM`
flag and of different queue ids and/or devices. They should inherit
the flag from the first socket buffer pool since no flags can be
specified once `XDP_SHARED_UMEM` is specified.

Fixes: b5aea28dca134 ("xsk: Add shared umem support between queue ids")
Signed-off-by: Jalal Mostafa 
Signed-off-by: Daniel Borkmann 
Acked-by: Magnus Karlsson 
Link: https://lore.kernel.org/bpf/20220921135701.10199-1-jalal.a.mostapha@gmail.com

xsk: Fix backpressure mechanism on Tx

2022-08-31T18:52:09+00:00

Commit d678cbd2f867 ("xsk: Fix handling of invalid descriptors in XSK TX
batching API") fixed batch API usage against set of descriptors with
invalid ones but introduced a problem when AF_XDP SW rings are smaller
than HW ones. Mismatch of reported Tx'ed frames between HW generator and
user space app was observed. It turned out that backpressure mechanism
became a bottleneck when the amount of produced descriptors to CQ is
lower than what we grabbed from XSK Tx ring.

Say that 512 entries had been taken from XSK Tx ring but we had only 490
free entries in CQ. Then callsite (ZC driver) will produce only 490
entries onto HW Tx ring but 512 entries will be released from Tx ring
and this is what will be seen by the user space.

In order to fix this case, mix XSK Tx/CQ ring interractions by moving
around internal functions and changing call order:

*  pull out xskq_prod_nb_free() from xskq_prod_reserve_addr_batch()
   up to xsk_tx_peek_release_desc_batch();
** move xskq_cons_release_n() into xskq_cons_read_desc_batch()

After doing so, algorithm can be described as follows:

1. lookup Tx entries
2. use value from 1. to reserve space in CQ (*)
3. Read from Tx ring as much descriptors as value from 2
 3a. release descriptors from XSK Tx ring (**)
4. Finally produce addresses to CQ

Fixes: d678cbd2f867 ("xsk: Fix handling of invalid descriptors in XSK TX batching API")
Signed-off-by: Magnus Karlsson 
Signed-off-by: Maciej Fijalkowski 
Signed-off-by: Daniel Borkmann 
Link: https://lore.kernel.org/bpf/20220830121705.8618-1-maciej.fijalkowski@intel.com

xsk: Fix corrupted packets for XDP_SHARED_UMEM

2022-08-15T15:26:07+00:00

Fix an issue in XDP_SHARED_UMEM mode together with aligned mode where
packets are corrupted for the second and any further sockets bound to
the same umem. In other words, this does not affect the first socket
bound to the umem. The culprit for this bug is that the initialization
of the DMA addresses for the pre-populated xsk buffer pool entries was
not performed for any socket but the first one bound to the umem. Only
the linear array of DMA addresses was populated. Fix this by populating
the DMA addresses in the xsk buffer pool for every socket bound to the
same umem.

Fixes: 94033cd8e73b8 ("xsk: Optimize for aligned case")
Reported-by: Alasdair McWilliam 
Reported-by: Intrusion Shield Team 
Signed-off-by: Magnus Karlsson 
Signed-off-by: Daniel Borkmann 
Tested-by: Alasdair McWilliam 
Acked-by: Maciej Fijalkowski 
Link: https://lore.kernel.org/xdp-newbies/6205E10C-292E-4995-9D10-409649354226@outlook.com/
Link: https://lore.kernel.org/bpf/20220812113259.531-1-magnus.karlsson@gmail.com