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: <e27cb384-8f4d-41cc-b87a-2b5c84ccbd66@amd.com>
Date: Thu, 25 Jan 2024 07:53:47 -0800
From: Brett Creeley <bcreeley@....com>
To: Alexei Starovoitov <alexei.starovoitov@...il.com>
Cc: Maciej Fijalkowski <maciej.fijalkowski@...el.com>,
 bpf <bpf@...r.kernel.org>, Alexei Starovoitov <ast@...nel.org>,
 Daniel Borkmann <daniel@...earbox.net>, Andrii Nakryiko <andrii@...nel.org>,
 Network Development <netdev@...r.kernel.org>,
 "Karlsson, Magnus" <magnus.karlsson@...el.com>,
 Björn Töpel <bjorn@...nel.org>,
 Eelco Chaudron <echaudro@...hat.com>, Lorenzo Bianconi <lorenzo@...nel.org>,
 Martin KaFai Lau <martin.lau@...ux.dev>,
 "Sarkar, Tirthendu" <tirthendu.sarkar@...el.com>,
 John Fastabend <john.fastabend@...il.com>, Simon Horman <horms@...nel.org>,
 Jakub Kicinski <kuba@...nel.org>
Subject: Re: [PATCH v6 bpf 04/11] ice: work on pre-XDP prog frag count



On 1/24/2024 6:34 PM, Alexei Starovoitov wrote:
> Caution: This message originated from an External Source. Use proper caution when opening attachments, clicking links, or responding.
> 
> 
> On Wed, Jan 24, 2024 at 6:26 PM Brett Creeley <bcreeley@....com> wrote:
>>
>>
>>
>> On 1/24/2024 11:15 AM, Maciej Fijalkowski wrote:
>>> Caution: This message originated from an External Source. Use proper caution when opening attachments, clicking links, or responding.
>>>
>>>
>>> Fix an OOM panic in XDP_DRV mode when a XDP program shrinks a
>>> multi-buffer packet by 4k bytes and then redirects it to an AF_XDP
>>> socket.
>>>
>>> Since support for handling multi-buffer frames was added to XDP, usage
>>> of bpf_xdp_adjust_tail() helper within XDP program can free the page
>>> that given fragment occupies and in turn decrease the fragment count
>>> within skb_shared_info that is embedded in xdp_buff struct. In current
>>> ice driver codebase, it can become problematic when page recycling logic
>>> decides not to reuse the page. In such case, __page_frag_cache_drain()
>>> is used with ice_rx_buf::pagecnt_bias that was not adjusted after
>>> refcount of page was changed by XDP prog which in turn does not drain
>>> the refcount to 0 and page is never freed.
>>>
>>> To address this, let us store the count of frags before the XDP program
>>> was executed on Rx ring struct. This will be used to compare with
>>> current frag count from skb_shared_info embedded in xdp_buff. A smaller
>>> value in the latter indicates that XDP prog freed frag(s). Then, for
>>> given delta decrement pagecnt_bias for XDP_DROP verdict.
>>>
>>> While at it, let us also handle the EOP frag within
>>> ice_set_rx_bufs_act() to make our life easier, so all of the adjustments
>>> needed to be applied against freed frags are performed in the single
>>> place.
>>>
>>> Fixes: 2fba7dc5157b ("ice: Add support for XDP multi-buffer on Rx side")
>>> Acked-by: Magnus Karlsson <magnus.karlsson@...el.com>
>>> Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@...el.com>
>>> ---
>>>    drivers/net/ethernet/intel/ice/ice_txrx.c     | 14 ++++++---
>>>    drivers/net/ethernet/intel/ice/ice_txrx.h     |  1 +
>>>    drivers/net/ethernet/intel/ice/ice_txrx_lib.h | 31 +++++++++++++------
>>>    3 files changed, 32 insertions(+), 14 deletions(-)
>>>
>>> diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.c b/drivers/net/ethernet/intel/ice/ice_txrx.c
>>> index 74d13cc5a3a7..0c9b4aa8a049 100644
>>> --- a/drivers/net/ethernet/intel/ice/ice_txrx.c
>>> +++ b/drivers/net/ethernet/intel/ice/ice_txrx.c
>>> @@ -603,9 +603,7 @@ ice_run_xdp(struct ice_rx_ring *rx_ring, struct xdp_buff *xdp,
>>>                   ret = ICE_XDP_CONSUMED;
>>>           }
>>>    exit:
>>> -       rx_buf->act = ret;
>>> -       if (unlikely(xdp_buff_has_frags(xdp)))
>>> -               ice_set_rx_bufs_act(xdp, rx_ring, ret);
>>> +       ice_set_rx_bufs_act(xdp, rx_ring, ret);
>>>    }
>>>
>>>    /**
>>> @@ -893,14 +891,17 @@ ice_add_xdp_frag(struct ice_rx_ring *rx_ring, struct xdp_buff *xdp,
>>>           }
>>>
>>>           if (unlikely(sinfo->nr_frags == MAX_SKB_FRAGS)) {
>>> -               if (unlikely(xdp_buff_has_frags(xdp)))
>>> -                       ice_set_rx_bufs_act(xdp, rx_ring, ICE_XDP_CONSUMED);
>>> +               ice_set_rx_bufs_act(xdp, rx_ring, ICE_XDP_CONSUMED);
>>>                   return -ENOMEM;
>>>           }
>>>
>>>           __skb_fill_page_desc_noacc(sinfo, sinfo->nr_frags++, rx_buf->page,
>>>                                      rx_buf->page_offset, size);
>>>           sinfo->xdp_frags_size += size;
>>> +       /* remember frag count before XDP prog execution; bpf_xdp_adjust_tail()
>>> +        * can pop off frags but driver has to handle it on its own
>>> +        */
>>> +       rx_ring->nr_frags = sinfo->nr_frags;
>>>
>>>           if (page_is_pfmemalloc(rx_buf->page))
>>>                   xdp_buff_set_frag_pfmemalloc(xdp);
>>> @@ -1251,6 +1252,7 @@ int ice_clean_rx_irq(struct ice_rx_ring *rx_ring, int budget)
>>>
>>>                   xdp->data = NULL;
>>>                   rx_ring->first_desc = ntc;
>>> +               rx_ring->nr_frags = 0;
>>>                   continue;
>>>    construct_skb:
>>>                   if (likely(ice_ring_uses_build_skb(rx_ring)))
>>> @@ -1266,10 +1268,12 @@ int ice_clean_rx_irq(struct ice_rx_ring *rx_ring, int budget)
>>>                                                       ICE_XDP_CONSUMED);
>>>                           xdp->data = NULL;
>>>                           rx_ring->first_desc = ntc;
>>> +                       rx_ring->nr_frags = 0;
>>>                           break;
>>>                   }
>>>                   xdp->data = NULL;
>>>                   rx_ring->first_desc = ntc;
>>> +               rx_ring->nr_frags = 0;
>>>
>>>                   stat_err_bits = BIT(ICE_RX_FLEX_DESC_STATUS0_RXE_S);
>>>                   if (unlikely(ice_test_staterr(rx_desc->wb.status_error0,
>>> diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.h b/drivers/net/ethernet/intel/ice/ice_txrx.h
>>> index b3379ff73674..af955b0e5dc5 100644
>>> --- a/drivers/net/ethernet/intel/ice/ice_txrx.h
>>> +++ b/drivers/net/ethernet/intel/ice/ice_txrx.h
>>> @@ -358,6 +358,7 @@ struct ice_rx_ring {
>>>           struct ice_tx_ring *xdp_ring;
>>>           struct ice_rx_ring *next;       /* pointer to next ring in q_vector */
>>>           struct xsk_buff_pool *xsk_pool;
>>> +       u32 nr_frags;
>>>           dma_addr_t dma;                 /* physical address of ring */
>>>           u16 rx_buf_len;
>>>           u8 dcb_tc;                      /* Traffic class of ring */
>>> diff --git a/drivers/net/ethernet/intel/ice/ice_txrx_lib.h b/drivers/net/ethernet/intel/ice/ice_txrx_lib.h
>>> index 762047508619..afcead4baef4 100644
>>> --- a/drivers/net/ethernet/intel/ice/ice_txrx_lib.h
>>> +++ b/drivers/net/ethernet/intel/ice/ice_txrx_lib.h
>>> @@ -12,26 +12,39 @@
>>>     * act: action to store onto Rx buffers related to XDP buffer parts
>>>     *
>>>     * Set action that should be taken before putting Rx buffer from first frag
>>> - * to one before last. Last one is handled by caller of this function as it
>>> - * is the EOP frag that is currently being processed. This function is
>>> - * supposed to be called only when XDP buffer contains frags.
>>> + * to the last.
>>>     */
>>>    static inline void
>>>    ice_set_rx_bufs_act(struct xdp_buff *xdp, const struct ice_rx_ring *rx_ring,
>>>                       const unsigned int act)
>>>    {
>>> -       const struct skb_shared_info *sinfo = xdp_get_shared_info_from_buff(xdp);
>>> -       u32 first = rx_ring->first_desc;
>>> -       u32 nr_frags = sinfo->nr_frags;
>>> +       u32 sinfo_frags = xdp_get_shared_info_from_buff(xdp)->nr_frags;
>>> +       u32 nr_frags = rx_ring->nr_frags + 1;
>>> +       u32 idx = rx_ring->first_desc;
>>>           u32 cnt = rx_ring->count;
>>>           struct ice_rx_buf *buf;
>>>
>>>           for (int i = 0; i < nr_frags; i++) {
>>> -               buf = &rx_ring->rx_buf[first];
>>> +               buf = &rx_ring->rx_buf[idx];
>>>                   buf->act = act;
>>>
>>> -               if (++first == cnt)
>>> -                       first = 0;
>>> +               if (++idx == cnt)
>>> +                       idx = 0;
>>> +       }
>>> +
>>> +       /* adjust pagecnt_bias on frags freed by XDP prog */
>>> +       if (sinfo_frags < rx_ring->nr_frags && act == ICE_XDP_CONSUMED) {
>>> +               u32 delta = rx_ring->nr_frags - sinfo_frags;
>>> +
>>> +               while (delta) {
>>> +                       if (idx == 0)
>>> +                               idx = cnt - 1;
>>> +                       else
>>> +                               idx--;
>>> +                       buf = &rx_ring->rx_buf[idx];
>>> +                       buf->pagecnt_bias--;
>>> +                       delta--;
>>> +               }
>>
>> Nit, but the function name ice_set_rx_bufs_act() doesn't completely
>> align with what it's doing anymore due to the additional pagecnt_bias
>> changes.
> 
> The patch set was applied. Please advise whether this can stay as-is
> or follow up is absolutely necessary.

I'm okay with it as is. If Maciej agrees he could always provide a 
follow up.

Thanks,

Brett

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ