netdev - Re: [PATCH net-next] bnxt_en: fix kernel panic in queue api functions

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <d4b341f3-7c0c-47cd-b7cb-81f9fccd3d0f@davidwei.uk>
Date: Fri, 5 Jul 2024 11:42:37 -0700
From: David Wei <dw@...idwei.uk>
To: Taehee Yoo <ap420073@...il.com>, davem@...emloft.net, kuba@...nel.org,
 pabeni@...hat.com, edumazet@...gle.com, michael.chan@...adcom.com,
 netdev@...r.kernel.org
Cc: somnath.kotur@...adcom.com, horms@...nel.org
Subject: Re: [PATCH net-next] bnxt_en: fix kernel panic in queue api functions

On 2024-07-04 00:41, Taehee Yoo wrote:
> bnxt_queue_{mem_alloc,start,stop} access bp->rx_ring array and this is
> initialized while an interface is being up.
> The rings are initialized as a number of channels.
> 
> The queue API functions access rx_ring without checking both null and
> ring size.
> So, if the queue API functions are called when interface status is down,
> they access an uninitialized rx_ring array.
> Also if the queue index parameter value is larger than a ring, it
> would also access an uninitialized rx_ring.

Hi Taehee, I deliberately didn't do a check as it is too defensive. The
caller of the queue API e.g. netdev_rx_queue_restart() should be the one
doing the check instead.

> 
>  BUG: kernel NULL pointer dereference, address: 0000000000000000
>  #PF: supervisor read access in kernel mode
>  #PF: error_code(0x0000) - not-present page
>  PGD 0 P4D 0
>  Oops: Oops: 0000 [#1] PREEMPT SMP NOPTI
>  CPU: 1 PID: 1697 Comm: ncdevmem Not tainted 6.10.0-rc5+ #34
>  RIP: 0010:bnxt_queue_mem_alloc+0x38/0x410 [bnxt_en]
>  Code: 49 89 f5 41 54 4d 89 c4 4d 69 c0 c0 05 00 00 55 48 8d af 40 0a 00 00 53 48 89 fb 48 83 ec 05
>  RSP: 0018:ffffa1ad0449ba48 EFLAGS: 00010246
>  RAX: ffffffffc04c7710 RBX: ffff9b88aee48000 RCX: 0000000000000000
>  RDX: 0000000000000000 RSI: ffff9b8884ba0000 RDI: ffff9b8884ba0008
>  RBP: ffff9b88aee48a40 R08: 0000000000000000 R09: ffff9b8884ba6000
>  R10: ffffa1ad0449ba88 R11: ffff9b8884ba6000 R12: 0000000000000000
>  R13: ffff9b8884ba0000 R14: ffff9b8884ba0000 R15: ffff9b8884ba6000
>  FS:  00007f7b2a094740(0000) GS:ffff9b8f9f680000(0000) knlGS:0000000000000000
>  CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
>  CR2: 0000000000000000 CR3: 000000015f394000 CR4: 00000000007506f0
>  PKRU: 55555554
>  Call Trace:
>   <TASK>
>   ? __die+0x20/0x70
>   ? page_fault_oops+0x15a/0x460
>   ? __vmalloc_node_range_noprof+0x4f7/0x8e0
>   ? exc_page_fault+0x6e/0x180
>   ? asm_exc_page_fault+0x22/0x30
>   ? __pfx_bnxt_queue_mem_alloc+0x10/0x10 [bnxt_en 2b2843e995211f081639d5c0e74fe1cce7fed534]
>   ? bnxt_queue_mem_alloc+0x38/0x410 [bnxt_en 2b2843e995211f081639d5c0e74fe1cce7fed534]
>   netdev_rx_queue_restart+0xa9/0x1c0
>   net_devmem_bind_dmabuf_to_queue+0xcb/0x100
>   netdev_nl_bind_rx_doit+0x2f6/0x350
>   genl_family_rcv_msg_doit+0xd9/0x130
>   genl_rcv_msg+0x184/0x2b0
>   ? __pfx_netdev_nl_bind_rx_doit+0x10/0x10
>   ? __pfx_genl_rcv_msg+0x10/0x10
>   netlink_rcv_skb+0x54/0x100
>   genl_rcv+0x24/0x40
>   netlink_unicast+0x243/0x370
>   netlink_sendmsg+0x1bb/0x3e0
> 
> Fixes: 2d694c27d32e ("bnxt_en: implement netdev_queue_mgmt_ops")
> Signed-off-by: Taehee Yoo <ap420073@...il.com>
> ---
> 
> The branch is not net because the commit 2d694c27d32e is not
> yet merged into net branch.
> 
> devmem TCP causes this problem, but it is not yet merged.
> So, to test this patch, please patch the current devmem TCP.
> The /tools/testing/selftests/net/ncdevmem will immediately reproduce 
> this problem.
> 
>  drivers/net/ethernet/broadcom/bnxt/bnxt.c | 9 +++++++++
>  1 file changed, 9 insertions(+)
> 
> diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
> index 6fc34ccb86e3..e270fb6b2866 100644
> --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
> +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
> @@ -15022,6 +15022,9 @@ static int bnxt_queue_mem_alloc(struct net_device *dev, void *qmem, int idx)
>  	struct bnxt_ring_struct *ring;
>  	int rc;
>  
> +	if (!bp->rx_ring || idx >= bp->rx_nr_rings)
> +		return -EINVAL;
> +
>  	rxr = &bp->rx_ring[idx];
>  	clone = qmem;
>  	memcpy(clone, rxr, sizeof(*rxr));
> @@ -15156,6 +15159,9 @@ static int bnxt_queue_start(struct net_device *dev, void *qmem, int idx)
>  	struct bnxt_cp_ring_info *cpr;
>  	int rc;
>  
> +	if (!bp->rx_ring || idx >= bp->rx_nr_rings)
> +		return -EINVAL;
> +
>  	rxr = &bp->rx_ring[idx];
>  	clone = qmem;
>  
> @@ -15195,6 +15201,9 @@ static int bnxt_queue_stop(struct net_device *dev, void *qmem, int idx)
>  	struct bnxt *bp = netdev_priv(dev);
>  	struct bnxt_rx_ring_info *rxr;
>  
> +	if (!bp->rx_ring || idx >= bp->rx_nr_rings)
> +		return -EINVAL;
> +
>  	rxr = &bp->rx_ring[idx];
>  	napi_disable(&rxr->bnapi->napi);
>  	bnxt_hwrm_rx_ring_free(bp, rxr, false);