search for: pskb_expand_head

On Thu, 23 Mar 2023 15:24:38 +0800, Yunsheng Lin <linyunsheng at huawei.com> wrote: > On 2023/3/23 13:40, Jason Wang wrote: > >>> > >>>> > >>>> Also, it seems better to split the xdp_linearize_page() to two functions > >>>> as pskb_expand_head() and __skb_linearize() do, one to expand the headroom, > >>>> the other one to do the linearizing. > >>> > >>> No skb here. > >> > >> I means following the semantics of pskb_expand_head() and __skb_linearize(), > >> not to combine t...

Looks like the GSO is involved? I got this while running Dom0 only (no guests), with a BOINC/Rosetta@home application running on all 4 cores. changeset: 10649:8e55c5c11475 Build: x86_32p (pae). ------------[ cut here ]------------ kernel BUG at net/core/dev.c:1133! invalid opcode: 0000 [#1] SMP CPU: 0 EIP: 0061:[<c04dceb0>] Not tainted VLI EFLAGS: 00210297 (2.6.16.13-xen

...734b>] dev_hard_start_xmit+0x1bb/0x1d0 [<c021896b>] dev_queue_xmit+0x11b/0x1b0 [<c025f1c6>] ip6_output2+0x276/0x2b0 [<c025ed30>] ip6_output_finish+0x0/0xf0 [<c025fc0a>] ip6_output+0x90a/0x940 [<c013e9e5>] cache_alloc_refill+0x2c5/0x3f0 [<c0212eed>] pskb_expand_head+0xdd/0x130 [<c02608d5>] ip6_forward+0x465/0x4b0 [<c02618c6>] ip6_rcv_finish+0x16/0x30 [<ce81a056>] nf_ct_frag6_output+0x86/0xb0 [nf_conntrack_ipv6] [<c02618b0>] ip6_rcv_finish+0x0/0x30 [<ce81911b>] ipv6_defrag+0x3b/0x50 [nf_conntrack_ipv6] [<c02618b0>...

...dle_frame_hook) = 0xbe03eb4d kernel(param_get_int) = 0x6980fe91 kernel(param_set_int) = 0xff964b25 kernel(per_cpu__current_task) = 0x4f1939c7 kernel(per_cpu__kernel_stack) = 0x6dcaeb88 kernel(per_cpu__this_cpu_off) = 0x57adf756 kernel(prepare_to_wait) = 0x33d92f9a kernel(printk) = 0xea147363 kernel(pskb_expand_head) = 0xf5f5d2d9 kernel(pv_irq_ops) = 0x78764f4e kernel(random32) = 0xb86e4ab9 kernel(rcu_barrier) = 0x60a13e90 kernel(register_netdevice) = 0xe85f2610 kernel(register_netdevice_notifier) = 0x63ecad53 kernel(register_pernet_gen_device) = 0x72428be7 kernel(rtnl_lock) = 0xc7a4fbed kernel(rtnl_unlock) =...

...uct ipv6hdr) - sizeof(struct iphdr); + + if (x->sel.family == AF_INET6) { + /* Here, the inner family is 6, therefore I have to + * substitute the IPhdr by enlarging it */ + if (skb_tailroom(skb) < delta){ + if (pskb_expand_head(skb, 0, delta, GFP_ATOMIC)) + goto out; //Just returning from here. + } + + skb->nh.raw -= delta; + } else if (x->sel.family == AF_INET) { + // We need to extract the PH + struct ip_beet_...

...g */ + struct vlan_ethhdr *vhdr; + if (skb_cloned(skb) || skb_shared(skb)) { + struct sk_buff *new_skb; + + new_skb = skb_copy(skb, GFP_ATOMIC); + kfree_skb(skb); + if (!new_skb) + return 1; + *pskb = skb = new_skb; + } + + if (skb_headroom(skb) < VLAN_HLEN) { + if (pskb_expand_head(skb, VLAN_HLEN, 0, + GFP_ATOMIC)) { + kfree_skb(skb); + return 1; + } + } + + skb_push(skb, VLAN_HLEN); + + skb->mac.raw -= VLAN_HLEN; + memmove(skb->mac.raw, skb->mac.raw + VLAN_HLEN, + ETH_ALEN * 2); + vhdr = (struct vlan_ethhdr *)skb->mac.raw; + vh...

...y estimate # of descriptors to use */ + count = VMXNET3_TXD_NEEDED(skb_headlen(skb)) + + skb_shinfo(skb)->nr_frags + 1; + + ctx.ipv4 = (skb->protocol == __constant_ntohs(ETH_P_IP)); + + ctx.mss = skb_shinfo(skb)->gso_size; + if (ctx.mss) { + if (skb_header_cloned(skb)) { + if (unlikely(pskb_expand_head(skb, 0, 0, + GFP_ATOMIC) != 0)) { + tq->stats.drop_tso++; + goto drop_pkt; + } + tq->stats.copy_skb_header++; + } + vmxnet3_prepare_tso(skb, &ctx); + } else { + if (unlikely(count > VMXNET3_MAX_TXD_PER_PKT)) { + + /* non-tso pkts must not use more than + *...

...y estimate # of descriptors to use */ + count = VMXNET3_TXD_NEEDED(skb_headlen(skb)) + + skb_shinfo(skb)->nr_frags + 1; + + ctx.ipv4 = (skb->protocol == __constant_ntohs(ETH_P_IP)); + + ctx.mss = skb_shinfo(skb)->gso_size; + if (ctx.mss) { + if (skb_header_cloned(skb)) { + if (unlikely(pskb_expand_head(skb, 0, 0, + GFP_ATOMIC) != 0)) { + tq->stats.drop_tso++; + goto drop_pkt; + } + tq->stats.copy_skb_header++; + } + vmxnet3_prepare_tso(skb, &ctx); + } else { + if (unlikely(count > VMXNET3_MAX_TXD_PER_PKT)) { + + /* non-tso pkts must not use more than + *...

...y estimate # of descriptors to use */ + count = VMXNET3_TXD_NEEDED(skb_headlen(skb)) + + skb_shinfo(skb)->nr_frags + 1; + + ctx.ipv4 = (skb->protocol == __constant_ntohs(ETH_P_IP)); + + ctx.mss = skb_shinfo(skb)->gso_size; + if (ctx.mss) { + if (skb_header_cloned(skb)) { + if (unlikely(pskb_expand_head(skb, 0, 0, + GFP_ATOMIC) != 0)) { + tq->stats.drop_tso++; + goto drop_pkt; + } + tq->stats.copy_skb_header++; + } + vmxnet3_prepare_tso(skb, &ctx); + } else { + if (unlikely(count > VMXNET3_MAX_TXD_PER_PKT)) { + + /* non-tso pkts must not use more than + *...

...y estimate # of descriptors to use */ + count = VMXNET3_TXD_NEEDED(skb_headlen(skb)) + + skb_shinfo(skb)->nr_frags + 1; + + ctx.ipv4 = (skb->protocol == __constant_ntohs(ETH_P_IP)); + + ctx.mss = skb_shinfo(skb)->gso_size; + if (ctx.mss) { + if (skb_header_cloned(skb)) { + if (unlikely(pskb_expand_head(skb, 0, 0, + GFP_ATOMIC) != 0)) { + tq->stats.drop_tso++; + goto drop_pkt; + } + tq->stats.copy_skb_header++; + } + vmxnet3_prepare_tso(skb, &ctx); + } else { + if (unlikely(count > VMXNET3_MAX_TXD_PER_PKT)) { + + /* non-tso pkts must not use more than + *...

...y estimate # of descriptors to use */ + count = VMXNET3_TXD_NEEDED(skb_headlen(skb)) + + skb_shinfo(skb)->nr_frags + 1; + + ctx.ipv4 = (skb->protocol == __constant_ntohs(ETH_P_IP)); + + ctx.mss = skb_shinfo(skb)->gso_size; + if (ctx.mss) { + if (skb_header_cloned(skb)) { + if (unlikely(pskb_expand_head(skb, 0, 0, + GFP_ATOMIC) != 0)) { + tq->stats.drop_tso++; + goto drop_pkt; + } + tq->stats.copy_skb_header++; + } + vmxnet3_prepare_tso(skb, &ctx); + } else { + if (unlikely(count > VMXNET3_MAX_TXD_PER_PKT)) { + + /* non-tso pkts must not use more than + *...

...y estimate # of descriptors to use */ + count = VMXNET3_TXD_NEEDED(skb_headlen(skb)) + + skb_shinfo(skb)->nr_frags + 1; + + ctx.ipv4 = (skb->protocol == __constant_ntohs(ETH_P_IP)); + + ctx.mss = skb_shinfo(skb)->gso_size; + if (ctx.mss) { + if (skb_header_cloned(skb)) { + if (unlikely(pskb_expand_head(skb, 0, 0, + GFP_ATOMIC) != 0)) { + tq->stats.drop_tso++; + goto drop_pkt; + } + tq->stats.copy_skb_header++; + } + vmxnet3_prepare_tso(skb, &ctx); + } else { + if (unlikely(count > VMXNET3_MAX_TXD_PER_PKT)) { + + /* non-tso pkts must not use more than + *...

...ED(skb_headlen(skb)) + + skb_shinfo(skb)->nr_frags + 1; + + ctx.ipv4 = (skb->protocol == __constant_ntohs(ETH_P_IP)); + + ctx.mss = skb_shinfo(skb)->gso_size; + if (ctx.mss) { + if (skb_header_cloned(skb)) { + if (unlikely(pskb_expand_head(skb, 0, 0, + GFP_ATOMIC) != 0)) { + tq->stats.drop_tso++; + goto drop_pkt; + } + tq->stats.copy_skb_header++; + } +...

...ED(skb_headlen(skb)) + + skb_shinfo(skb)->nr_frags + 1; + + ctx.ipv4 = (skb->protocol == __constant_ntohs(ETH_P_IP)); + + ctx.mss = skb_shinfo(skb)->gso_size; + if (ctx.mss) { + if (skb_header_cloned(skb)) { + if (unlikely(pskb_expand_head(skb, 0, 0, + GFP_ATOMIC) != 0)) { + tq->stats.drop_tso++; + goto drop_pkt; + } + tq->stats.copy_skb_header++; + } +...

Hi, Is the http://xenbits.xensource.com/linux-2.6-xen.hg tree still being updated? if not, what''s the preferred Linux tree to track that has all of the Xen bits? Thanks, Muli _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel