Re: [Intel-wired-lan] [PATCH net v4 1/1] i40e: xsk: advance next_to_clean on status descriptors

[Maciej Fijalkowski <maciej.fijalkowski@intel.com>]

On Tue, Nov 18, 2025 at 01:13:33PM +0100, Loktionov, Aleksandr wrote:
> 
> 
> > -----Original Message-----
> > From: Intel-wired-lan <intel-wired-lan-bounces@osuosl.org> On Behalf
> > Of Alessandro Decina
> > Sent: Tuesday, November 18, 2025 12:31 PM
> > To: netdev@vger.kernel.org
> > Cc: Fijalkowski, Maciej <maciej.fijalkowski@intel.com>; David S.
> > Miller <davem@davemloft.net>; Alexei Starovoitov <ast@kernel.org>;
> > Andrew Lunn <andrew+netdev@lunn.ch>; Daniel Borkmann
> > <daniel@iogearbox.net>; Eric Dumazet <edumazet@google.com>; Jakub
> > Kicinski <kuba@kernel.org>; Jesper Dangaard Brouer <hawk@kernel.org>;
> > John Fastabend <john.fastabend@gmail.com>; Paolo Abeni
> > <pabeni@redhat.com>; Kitszel, Przemyslaw
> > <przemyslaw.kitszel@intel.com>; Stanislav Fomichev <sdf@fomichev.me>;
> > Sarkar, Tirthendu <tirthendu.sarkar@intel.com>; Nguyen, Anthony L
> > <anthony.l.nguyen@intel.com>; bpf@vger.kernel.org; intel-wired-
> > lan@lists.osuosl.org; linux-kernel@vger.kernel.org; Alessandro Decina
> > <alessandro.d@gmail.com>
> > Subject: [Intel-wired-lan] [PATCH net v4 1/1] i40e: xsk: advance
> > next_to_clean on status descriptors
> > 
> > Whenever a status descriptor is received, i40e processes and skips
> > over it, correctly updating next_to_process but forgetting to update
> > next_to_clean. In the next iteration this accidentally causes the
> > creation of an invalid multi-buffer xdp_buff where the first fragment
> > is the status descriptor.
> > 
> > If then a skb is constructed from such an invalid buffer - because the
> > eBPF program returns XDP_PASS - a panic occurs:
> > 
> > [ 5866.367317] BUG: unable to handle page fault for address:
> > ffd31c37eab1c980 [ 5866.375050] #PF: supervisor read access in kernel
> > mode [ 5866.380825] #PF: error_code(0x0000) - not-present page [
> > 5866.386602] PGD 0 [ 5866.388867] Oops: Oops: 0000 [#1] SMP NOPTI [
> > 5866.393575] CPU: 34 UID: 0 PID: 0 Comm: swapper/34 Not tainted
> > 6.17.0-custom #1 PREEMPT(voluntary) [ 5866.403740] Hardware name:
> > Supermicro AS -2115GT-HNTR/H13SST-G, BIOS 3.2 03/20/2025 [
> > 5866.412339] RIP: 0010:memcpy+0x8/0x10 [ 5866.416454] Code: cc cc 90
> > cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc 90 90 90 90 90 90 90 90
> > 90 90 90 90 90 90 90 90 66 90 48 89 f8 48 89 d1 <f3> a4 e9 fc 26 c0 fe
> > 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 [ 5866.437538] RSP:
> > 0018:ff428d9ec0bb0ca8 EFLAGS: 00010286 [ 5866.443415] RAX:
> > ff2dd26dbd8f0000 RBX: ff2dd265ad161400 RCX: 00000000000004e1 [
> > 5866.451435] RDX: 00000000000004e1 RSI: ffd31c37eab1c980 RDI:
> > ff2dd26dbd8f0000 [ 5866.459454] RBP: ff428d9ec0bb0d40 R08:
> > 0000000000000000 R09: 0000000000000000 [ 5866.467470] R10:
> > 0000000000000000 R11: 0000000000000000 R12: ff428d9eec726ef8 [
> > 5866.475490] R13: ff2dd26dbd8f0000 R14: ff2dd265ca2f9fc0 R15:
> > ff2dd26548548b80 [ 5866.483509] FS:  0000000000000000(0000)
> > GS:ff2dd2c363592000(0000) knlGS:0000000000000000 [ 5866.492600] CS:
> > 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 5866.499060] CR2:
> > ffd31c37eab1c980 CR3: 0000000178d7b040 CR4: 0000000000f71ef0 [
> > 5866.507079] PKRU: 55555554 [ 5866.510125] Call Trace:
> > [ 5866.512867]  <IRQ>
> > [ 5866.515132]  ? i40e_clean_rx_irq_zc+0xc50/0xe60 [i40e] [
> > 5866.520921]  i40e_napi_poll+0x2d8/0x1890 [i40e] [ 5866.526022]  ?
> > srso_alias_return_thunk+0x5/0xfbef5
> > [ 5866.531408]  ? raise_softirq+0x24/0x70 [ 5866.535623]  ?
> > srso_alias_return_thunk+0x5/0xfbef5
> > [ 5866.541011]  ? srso_alias_return_thunk+0x5/0xfbef5
> > [ 5866.546397]  ? rcu_sched_clock_irq+0x225/0x1800 [ 5866.551493]
> > __napi_poll+0x30/0x230 [ 5866.555423]  net_rx_action+0x20b/0x3f0 [
> > 5866.559643]  handle_softirqs+0xe4/0x340 [ 5866.563962]
> > __irq_exit_rcu+0x10e/0x130 [ 5866.568283]  irq_exit_rcu+0xe/0x20 [
> > 5866.572110]  common_interrupt+0xb6/0xe0 [ 5866.576425]  </IRQ> [
> > 5866.578791]  <TASK>

Something is odd with your editor, it would be better to preserve the
formatting above.

> > 
> Commit message lacks test description.
> Please add how you validated the fix (kernel version, NIC model, reproducer steps).

Repro steps would be nice to have, rest would be rather redundant to me.

> 
> > Advance next_to_clean to ensure invalid xdp_buff(s) aren't created.
> > 
> > Move the common logic to i40e_clean_programming_status and update both
> > i40e_clean_rx_irq and i40e_clean_rx_irq_zc accordingly.
> > 
> > Fixes: 1c9ba9c14658 ("i40e: xsk: add RX multi-buffer support")
> > Signed-off-by: Alessandro Decina <alessandro.d@gmail.com>
> > ---
> >  drivers/net/ethernet/intel/i40e/i40e_txrx.c   | 59 +++++++++++-------
> > -
> >  .../ethernet/intel/i40e/i40e_txrx_common.h    |  5 +-
> >  drivers/net/ethernet/intel/i40e/i40e_xsk.c    | 14 ++---
> >  3 files changed, 42 insertions(+), 36 deletions(-)
> > 
> > diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
> > b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
> > index cc0b9efc2637..fe2190f4b9bc 100644
> > --- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
> > +++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
> > @@ -1386,22 +1386,36 @@ static void i40e_reuse_rx_page(struct
> > i40e_ring *rx_ring,
> >   * @rx_ring: the rx ring that has this descriptor
> >   * @qword0_raw: qword0
> >   * @qword1: qword1 representing status_error_len in CPU ordering
> > + * @next_to_process: pointer to next_to_process index
> > + * @next_to_clean: pointer to next_to_clean index
> >   *
> >   * Flow director should handle FD_FILTER_STATUS to check its filter
> > programming
> >   * status being successful or not and take actions accordingly. FCoE
> > should
> >   * handle its context/filter programming/invalidation status and take
> > actions.
> >   *
> > - * Returns an i40e_rx_buffer to reuse if the cleanup occurred,
> > otherwise NULL.
> > + * Returns false if what is passed is not a status descriptor.
> >   **/
> Please use kdoc Return: tag.
> Kernel-doc is incomplete, please also document what 'true' means (status descriptor processed).
> 
> > -void i40e_clean_programming_status(struct i40e_ring *rx_ring, u64
> > qword0_raw,
> > -				   u64 qword1)
> > +bool i40e_clean_programming_status(struct i40e_ring *rx_ring, u64
> > qword0_raw,
> > +				   u64 qword1, u16 *next_to_process,
> > +				   u16 *next_to_clean)
> >  {
> > +	u16 ntp = *next_to_process;
> >  	u8 id;
> > 
> > +	if (!i40e_rx_is_programming_status(qword1))
> > +		return false;
> > +
> >  	id = FIELD_GET(I40E_RX_PROG_STATUS_DESC_QW1_PROGID_MASK,
> > qword1);
> > 
> >  	if (id == I40E_RX_PROG_STATUS_DESC_FD_FILTER_STATUS)
> >  		i40e_fd_handle_status(rx_ring, qword0_raw, qword1, id);
> > +
> > +	if (++*next_to_process == rx_ring->count)
> > +		*next_to_process = 0;
> Consider using u16_inc_wrap() instead of manual wrap.

I don't see anything like that in kernel. Did AI hallucinated for you or
is it something from OOT realm? Or am I missing something.

This is a method we use throughout the whole driver. Even if such wrapper
exists, I'd rather see a refactor patch that addresses all the manual
wrappings, not a single one.

Let's keep it as-is.

> 
> > +	if (ntp == *next_to_clean)
> > +		*next_to_clean = *next_to_process;
> > +
> > +	return true;
> >  }
> > 
> >  /**
> > @@ -1971,19 +1985,18 @@ static void i40e_rx_buffer_flip(struct
> > i40e_rx_buffer *rx_buffer,  }
> > 
> >  /**
> > - * i40e_get_rx_buffer - Fetch Rx buffer and synchronize data for use
> > + * i40e_prepare_rx_buffer - Synchronize the buffer for use by the CPU
> >   * @rx_ring: rx descriptor ring to transact packets on
> > + * @rx_buffer: the rx buffer
> >   * @size: size of buffer to add to skb
> >   *
> > - * This function will pull an Rx buffer from the ring and synchronize
> > it
> > - * for use by the CPU.
> > + * This function will synchronize the given buffer for use by the
> > CPU.
> >   */
> > -static struct i40e_rx_buffer *i40e_get_rx_buffer(struct i40e_ring
> > *rx_ring,
> > -						 const unsigned int size)
> > +static struct i40e_rx_buffer *
> > +i40e_prepare_rx_buffer(struct i40e_ring *rx_ring,
> > +		       struct i40e_rx_buffer *rx_buffer,
> > +		       const unsigned int size)
> >  {
> > -	struct i40e_rx_buffer *rx_buffer;
> > -
> > -	rx_buffer = i40e_rx_bi(rx_ring, rx_ring->next_to_process);
> >  	rx_buffer->page_count =
> >  #if (PAGE_SIZE < 8192)
> >  		page_count(rx_buffer->page);
> > @@ -2450,6 +2463,7 @@ static int i40e_clean_rx_irq(struct i40e_ring
> > *rx_ring, int budget,
> > 
> >  	while (likely(total_rx_packets < (unsigned int)budget)) {
> >  		u16 ntp = rx_ring->next_to_process;
> > +		u16 ntc = rx_ring->next_to_clean;
> >  		struct i40e_rx_buffer *rx_buffer;
> >  		union i40e_rx_desc *rx_desc;
> >  		struct sk_buff *skb;
> > @@ -2480,21 +2494,15 @@ static int i40e_clean_rx_irq(struct i40e_ring
> > *rx_ring, int budget,
> >  		 */
> >  		dma_rmb();
> > 
> > -		if (i40e_rx_is_programming_status(qword)) {
> > -			i40e_clean_programming_status(rx_ring,
> > -						      rx_desc->raw.qword[0],
> > -						      qword);
> > -			rx_buffer = i40e_rx_bi(rx_ring, ntp);
> > -			i40e_inc_ntp(rx_ring);
> > +		rx_buffer = i40e_rx_bi(rx_ring, ntp);
> > +
> > +		if (i40e_clean_programming_status(rx_ring,
> > +						  rx_desc->raw.qword[0],
> > qword,
> > +						  &rx_ring->next_to_process,
> > +						  &rx_ring->next_to_clean)) {
> Wrap the call to i40e_clean_programming_status to stay within 80 columns.
> 
> >  			i40e_reuse_rx_page(rx_ring, rx_buffer);
> > -			/* Update ntc and bump cleaned count if not in
> > the
> > -			 * middle of mb packet.
> > -			 */
> > -			if (rx_ring->next_to_clean == ntp) {
> > -				rx_ring->next_to_clean =
> > -					rx_ring->next_to_process;
> > +			if (ntc != rx_ring->next_to_clean)
> >  				cleaned_count++;
> > -			}
> >  			continue;
> >  		}
> > 
> > @@ -2503,8 +2511,7 @@ static int i40e_clean_rx_irq(struct i40e_ring
> > *rx_ring, int budget,
> >  			break;
> > 
> >  		i40e_trace(clean_rx_irq, rx_ring, rx_desc, xdp);
> > -		/* retrieve a buffer from the ring */
> > -		rx_buffer = i40e_get_rx_buffer(rx_ring, size);
> > +		i40e_prepare_rx_buffer(rx_ring, rx_buffer, size);
> > 
> >  		neop = i40e_is_non_eop(rx_ring, rx_desc);
> >  		i40e_inc_ntp(rx_ring);
> > diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx_common.h
> > b/drivers/net/ethernet/intel/i40e/i40e_txrx_common.h
> > index e26807fd2123..21d9ed878bf0 100644
> > --- a/drivers/net/ethernet/intel/i40e/i40e_txrx_common.h
> > +++ b/drivers/net/ethernet/intel/i40e/i40e_txrx_common.h
> > @@ -7,8 +7,9 @@
> >  #include "i40e.h"
> > 
> >  int i40e_xmit_xdp_tx_ring(struct xdp_buff *xdp, struct i40e_ring
> > *xdp_ring); -void i40e_clean_programming_status(struct i40e_ring
> > *rx_ring, u64 qword0_raw,
> > -				   u64 qword1);
> > +bool i40e_clean_programming_status(struct i40e_ring *rx_ring, u64
> > qword0_raw,
> > +				   u64 qword1, u16 *next_to_clean,
> > +				   u16 *next_to_process);
> Argument order differs between header and implementation (next_to_clean vs next_to_process).
> This will compile but is misleading and dangerous for future maintenance.
> Please fix argument order to match.

Good catch

> 
> >  void i40e_process_skb_fields(struct i40e_ring *rx_ring,
> >  			     union i40e_rx_desc *rx_desc, struct sk_buff
> > *skb);  void i40e_xdp_ring_update_tail(struct i40e_ring *xdp_ring);
> > diff --git a/drivers/net/ethernet/intel/i40e/i40e_xsk.c
> > b/drivers/net/ethernet/intel/i40e/i40e_xsk.c
> > index 9f47388eaba5..f8accc266c2c 100644
> > --- a/drivers/net/ethernet/intel/i40e/i40e_xsk.c
> > +++ b/drivers/net/ethernet/intel/i40e/i40e_xsk.c
> > @@ -440,14 +440,13 @@ int i40e_clean_rx_irq_zc(struct i40e_ring
> > *rx_ring, int budget)
> >  		 */
> >  		dma_rmb();
> > 
> > -		if (i40e_rx_is_programming_status(qword)) {
> > -			i40e_clean_programming_status(rx_ring,
> > -						      rx_desc->raw.qword[0],
> > -						      qword);
> > -			bi = *i40e_rx_bi(rx_ring, next_to_process);
> > +		bi = *i40e_rx_bi(rx_ring, next_to_process);
> > +
> > +		if (i40e_clean_programming_status(rx_ring,
> > +						  rx_desc->raw.qword[0],
> > +						  qword, &next_to_process,
> > +						  &next_to_clean)) {
> >  			xsk_buff_free(bi);
> > -			if (++next_to_process == count)
> > -				next_to_process = 0;
> >  			continue;
> >  		}
> > 
> > @@ -455,7 +454,6 @@ int i40e_clean_rx_irq_zc(struct i40e_ring
> > *rx_ring, int budget)
> >  		if (!size)
> >  			break;
> > 
> > -		bi = *i40e_rx_bi(rx_ring, next_to_process);
> >  		xsk_buff_set_size(bi, size);
> >  		xsk_buff_dma_sync_for_cpu(bi);
> > 
> > --
> > 2.43.0
>