diff options
Diffstat (limited to 'net')
-rw-r--r-- | net/9p/protocol.c | 6 | ||||
-rw-r--r-- | net/9p/trans_fd.c | 3 | ||||
-rw-r--r-- | net/9p/trans_rdma.c | 52 | ||||
-rw-r--r-- | net/9p/trans_virtio.c | 5 | ||||
-rw-r--r-- | net/bluetooth/hidp/core.c | 3 | ||||
-rw-r--r-- | net/ceph/ceph_common.c | 37 | ||||
-rw-r--r-- | net/ceph/crush/crush.c | 14 | ||||
-rw-r--r-- | net/ceph/crush/crush_ln_table.h | 166 | ||||
-rw-r--r-- | net/ceph/crush/mapper.c | 118 | ||||
-rw-r--r-- | net/ceph/debugfs.c | 24 | ||||
-rw-r--r-- | net/ceph/messenger.c | 25 | ||||
-rw-r--r-- | net/ceph/osdmap.c | 25 | ||||
-rw-r--r-- | net/core/dev.c | 2 | ||||
-rw-r--r-- | net/core/filter.c | 41 | ||||
-rw-r--r-- | net/core/net_namespace.c | 1 | ||||
-rw-r--r-- | net/core/skbuff.c | 10 | ||||
-rw-r--r-- | net/dsa/dsa.c | 6 | ||||
-rw-r--r-- | net/ipv4/fou.c | 3 | ||||
-rw-r--r-- | net/ipv4/inet_diag.c | 28 | ||||
-rw-r--r-- | net/ipv4/ip_forward.c | 3 | ||||
-rw-r--r-- | net/ipv4/tcp.c | 14 | ||||
-rw-r--r-- | net/ipv4/tcp_dctcp.c | 5 | ||||
-rw-r--r-- | net/ipv4/tcp_illinois.c | 6 | ||||
-rw-r--r-- | net/ipv4/tcp_input.c | 2 | ||||
-rw-r--r-- | net/ipv4/tcp_vegas.c | 5 | ||||
-rw-r--r-- | net/ipv4/tcp_vegas.h | 2 | ||||
-rw-r--r-- | net/ipv4/tcp_westwood.c | 6 | ||||
-rw-r--r-- | net/sched/act_bpf.c | 3 | ||||
-rw-r--r-- | net/sched/act_mirred.c | 2 | ||||
-rw-r--r-- | net/sched/cls_bpf.c | 3 |
30 files changed, 545 insertions, 75 deletions
diff --git a/net/9p/protocol.c b/net/9p/protocol.c index e9d0f0c1a048..16d287565987 100644 --- a/net/9p/protocol.c +++ b/net/9p/protocol.c @@ -275,7 +275,7 @@ p9pdu_vreadf(struct p9_fcall *pdu, int proto_version, const char *fmt, } break; case 'R':{ - int16_t *nwqid = va_arg(ap, int16_t *); + uint16_t *nwqid = va_arg(ap, uint16_t *); struct p9_qid **wqids = va_arg(ap, struct p9_qid **); @@ -440,7 +440,7 @@ p9pdu_vwritef(struct p9_fcall *pdu, int proto_version, const char *fmt, stbuf->n_gid, stbuf->n_muid); } break; case 'V':{ - int32_t count = va_arg(ap, int32_t); + uint32_t count = va_arg(ap, uint32_t); struct iov_iter *from = va_arg(ap, struct iov_iter *); errcode = p9pdu_writef(pdu, proto_version, "d", @@ -471,7 +471,7 @@ p9pdu_vwritef(struct p9_fcall *pdu, int proto_version, const char *fmt, } break; case 'R':{ - int16_t nwqid = va_arg(ap, int); + uint16_t nwqid = va_arg(ap, int); struct p9_qid *wqids = va_arg(ap, struct p9_qid *); diff --git a/net/9p/trans_fd.c b/net/9p/trans_fd.c index 3e3d82d8ff70..bced8c074c12 100644 --- a/net/9p/trans_fd.c +++ b/net/9p/trans_fd.c @@ -734,6 +734,7 @@ static int parse_opts(char *params, struct p9_fd_opts *opts) opts->port = P9_PORT; opts->rfd = ~0; opts->wfd = ~0; + opts->privport = 0; if (!params) return 0; @@ -1013,7 +1014,6 @@ p9_fd_create(struct p9_client *client, const char *addr, char *args) { int err; struct p9_fd_opts opts; - struct p9_trans_fd *p; parse_opts(args, &opts); @@ -1026,7 +1026,6 @@ p9_fd_create(struct p9_client *client, const char *addr, char *args) if (err < 0) return err; - p = (struct p9_trans_fd *) client->trans; p9_conn_create(client); return 0; diff --git a/net/9p/trans_rdma.c b/net/9p/trans_rdma.c index 14ad43b5cf89..3533d2a53ab6 100644 --- a/net/9p/trans_rdma.c +++ b/net/9p/trans_rdma.c @@ -139,6 +139,7 @@ struct p9_rdma_opts { int sq_depth; int rq_depth; long timeout; + int privport; }; /* @@ -146,7 +147,10 @@ struct p9_rdma_opts { */ enum { /* Options that take integer arguments */ - Opt_port, Opt_rq_depth, Opt_sq_depth, Opt_timeout, Opt_err, + Opt_port, Opt_rq_depth, Opt_sq_depth, Opt_timeout, + /* Options that take no argument */ + Opt_privport, + Opt_err, }; static match_table_t tokens = { @@ -154,6 +158,7 @@ static match_table_t tokens = { {Opt_sq_depth, "sq=%u"}, {Opt_rq_depth, "rq=%u"}, {Opt_timeout, "timeout=%u"}, + {Opt_privport, "privport"}, {Opt_err, NULL}, }; @@ -175,6 +180,7 @@ static int parse_opts(char *params, struct p9_rdma_opts *opts) opts->sq_depth = P9_RDMA_SQ_DEPTH; opts->rq_depth = P9_RDMA_RQ_DEPTH; opts->timeout = P9_RDMA_TIMEOUT; + opts->privport = 0; if (!params) return 0; @@ -193,13 +199,13 @@ static int parse_opts(char *params, struct p9_rdma_opts *opts) if (!*p) continue; token = match_token(p, tokens, args); - if (token == Opt_err) - continue; - r = match_int(&args[0], &option); - if (r < 0) { - p9_debug(P9_DEBUG_ERROR, - "integer field, but no integer?\n"); - continue; + if ((token != Opt_err) && (token != Opt_privport)) { + r = match_int(&args[0], &option); + if (r < 0) { + p9_debug(P9_DEBUG_ERROR, + "integer field, but no integer?\n"); + continue; + } } switch (token) { case Opt_port: @@ -214,6 +220,9 @@ static int parse_opts(char *params, struct p9_rdma_opts *opts) case Opt_timeout: opts->timeout = option; break; + case Opt_privport: + opts->privport = 1; + break; default: continue; } @@ -607,6 +616,23 @@ static int rdma_cancelled(struct p9_client *client, struct p9_req_t *req) return 0; } +static int p9_rdma_bind_privport(struct p9_trans_rdma *rdma) +{ + struct sockaddr_in cl = { + .sin_family = AF_INET, + .sin_addr.s_addr = htonl(INADDR_ANY), + }; + int port, err = -EINVAL; + + for (port = P9_DEF_MAX_RESVPORT; port >= P9_DEF_MIN_RESVPORT; port--) { + cl.sin_port = htons((ushort)port); + err = rdma_bind_addr(rdma->cm_id, (struct sockaddr *)&cl); + if (err != -EADDRINUSE) + break; + } + return err; +} + /** * trans_create_rdma - Transport method for creating atransport instance * @client: client instance @@ -642,6 +668,16 @@ rdma_create_trans(struct p9_client *client, const char *addr, char *args) /* Associate the client with the transport */ client->trans = rdma; + /* Bind to a privileged port if we need to */ + if (opts.privport) { + err = p9_rdma_bind_privport(rdma); + if (err < 0) { + pr_err("%s (%d): problem binding to privport: %d\n", + __func__, task_pid_nr(current), -err); + goto error; + } + } + /* Resolve the server's address */ rdma->addr.sin_family = AF_INET; rdma->addr.sin_addr.s_addr = in_aton(addr); diff --git a/net/9p/trans_virtio.c b/net/9p/trans_virtio.c index e62bcbbabb5e..9dd49ca67dbc 100644 --- a/net/9p/trans_virtio.c +++ b/net/9p/trans_virtio.c @@ -525,7 +525,10 @@ static ssize_t p9_mount_tag_show(struct device *dev, vdev = dev_to_virtio(dev); chan = vdev->priv; - return snprintf(buf, chan->tag_len + 1, "%s", chan->tag); + memcpy(buf, chan->tag, chan->tag_len); + buf[chan->tag_len] = 0; + + return chan->tag_len + 1; } static DEVICE_ATTR(mount_tag, 0444, p9_mount_tag_show, NULL); diff --git a/net/bluetooth/hidp/core.c b/net/bluetooth/hidp/core.c index a05b9dbf14c9..9070dfd6b4ad 100644 --- a/net/bluetooth/hidp/core.c +++ b/net/bluetooth/hidp/core.c @@ -1313,7 +1313,8 @@ int hidp_connection_add(struct hidp_connadd_req *req, struct socket *ctrl_sock, struct socket *intr_sock) { - u32 valid_flags = 0; + u32 valid_flags = BIT(HIDP_VIRTUAL_CABLE_UNPLUG) | + BIT(HIDP_BOOT_PROTOCOL_MODE); struct hidp_session *session; struct l2cap_conn *conn; struct l2cap_chan *chan; diff --git a/net/ceph/ceph_common.c b/net/ceph/ceph_common.c index ec565508e904..79e8f71aef5b 100644 --- a/net/ceph/ceph_common.c +++ b/net/ceph/ceph_common.c @@ -490,6 +490,43 @@ out: } EXPORT_SYMBOL(ceph_parse_options); +int ceph_print_client_options(struct seq_file *m, struct ceph_client *client) +{ + struct ceph_options *opt = client->options; + size_t pos = m->count; + + if (opt->name) + seq_printf(m, "name=%s,", opt->name); + if (opt->key) + seq_puts(m, "secret=<hidden>,"); + + if (opt->flags & CEPH_OPT_FSID) + seq_printf(m, "fsid=%pU,", &opt->fsid); + if (opt->flags & CEPH_OPT_NOSHARE) + seq_puts(m, "noshare,"); + if (opt->flags & CEPH_OPT_NOCRC) + seq_puts(m, "nocrc,"); + if (opt->flags & CEPH_OPT_NOMSGAUTH) + seq_puts(m, "nocephx_require_signatures,"); + if ((opt->flags & CEPH_OPT_TCP_NODELAY) == 0) + seq_puts(m, "notcp_nodelay,"); + + if (opt->mount_timeout != CEPH_MOUNT_TIMEOUT_DEFAULT) + seq_printf(m, "mount_timeout=%d,", opt->mount_timeout); + if (opt->osd_idle_ttl != CEPH_OSD_IDLE_TTL_DEFAULT) + seq_printf(m, "osd_idle_ttl=%d,", opt->osd_idle_ttl); + if (opt->osd_keepalive_timeout != CEPH_OSD_KEEPALIVE_DEFAULT) + seq_printf(m, "osdkeepalivetimeout=%d,", + opt->osd_keepalive_timeout); + + /* drop redundant comma */ + if (m->count != pos) + m->count--; + + return 0; +} +EXPORT_SYMBOL(ceph_print_client_options); + u64 ceph_client_id(struct ceph_client *client) { return client->monc.auth->global_id; diff --git a/net/ceph/crush/crush.c b/net/ceph/crush/crush.c index 16bc199d9a62..9d84ce4ea0df 100644 --- a/net/ceph/crush/crush.c +++ b/net/ceph/crush/crush.c @@ -17,6 +17,7 @@ const char *crush_bucket_alg_name(int alg) case CRUSH_BUCKET_LIST: return "list"; case CRUSH_BUCKET_TREE: return "tree"; case CRUSH_BUCKET_STRAW: return "straw"; + case CRUSH_BUCKET_STRAW2: return "straw2"; default: return "unknown"; } } @@ -40,6 +41,8 @@ int crush_get_bucket_item_weight(const struct crush_bucket *b, int p) return ((struct crush_bucket_tree *)b)->node_weights[crush_calc_tree_node(p)]; case CRUSH_BUCKET_STRAW: return ((struct crush_bucket_straw *)b)->item_weights[p]; + case CRUSH_BUCKET_STRAW2: + return ((struct crush_bucket_straw2 *)b)->item_weights[p]; } return 0; } @@ -77,6 +80,14 @@ void crush_destroy_bucket_straw(struct crush_bucket_straw *b) kfree(b); } +void crush_destroy_bucket_straw2(struct crush_bucket_straw2 *b) +{ + kfree(b->item_weights); + kfree(b->h.perm); + kfree(b->h.items); + kfree(b); +} + void crush_destroy_bucket(struct crush_bucket *b) { switch (b->alg) { @@ -92,6 +103,9 @@ void crush_destroy_bucket(struct crush_bucket *b) case CRUSH_BUCKET_STRAW: crush_destroy_bucket_straw((struct crush_bucket_straw *)b); break; + case CRUSH_BUCKET_STRAW2: + crush_destroy_bucket_straw2((struct crush_bucket_straw2 *)b); + break; } } diff --git a/net/ceph/crush/crush_ln_table.h b/net/ceph/crush/crush_ln_table.h new file mode 100644 index 000000000000..6192c7fc958c --- /dev/null +++ b/net/ceph/crush/crush_ln_table.h @@ -0,0 +1,166 @@ +/* + * Ceph - scalable distributed file system + * + * Copyright (C) 2015 Intel Corporation All Rights Reserved + * + * This is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License version 2.1, as published by the Free Software + * Foundation. See file COPYING. + * + */ + +#if defined(__linux__) +#include <linux/types.h> +#elif defined(__FreeBSD__) +#include <sys/types.h> +#endif + +#ifndef CEPH_CRUSH_LN_H +#define CEPH_CRUSH_LN_H + + +// RH_LH_tbl[2*k] = 2^48/(1.0+k/128.0) +// RH_LH_tbl[2*k+1] = 2^48*log2(1.0+k/128.0) + +static int64_t __RH_LH_tbl[128*2+2] = { + 0x0001000000000000ll, 0x0000000000000000ll, 0x0000fe03f80fe040ll, 0x000002dfca16dde1ll, + 0x0000fc0fc0fc0fc1ll, 0x000005b9e5a170b4ll, 0x0000fa232cf25214ll, 0x0000088e68ea899all, + 0x0000f83e0f83e0f9ll, 0x00000b5d69bac77ell, 0x0000f6603d980f67ll, 0x00000e26fd5c8555ll, + 0x0000f4898d5f85bcll, 0x000010eb389fa29fll, 0x0000f2b9d6480f2cll, 0x000013aa2fdd27f1ll, + 0x0000f0f0f0f0f0f1ll, 0x00001663f6fac913ll, 0x0000ef2eb71fc435ll, 0x00001918a16e4633ll, + 0x0000ed7303b5cc0fll, 0x00001bc84240adabll, 0x0000ebbdb2a5c162ll, 0x00001e72ec117fa5ll, + 0x0000ea0ea0ea0ea1ll, 0x00002118b119b4f3ll, 0x0000e865ac7b7604ll, 0x000023b9a32eaa56ll, + 0x0000e6c2b4481cd9ll, 0x00002655d3c4f15cll, 0x0000e525982af70dll, 0x000028ed53f307eell, + 0x0000e38e38e38e39ll, 0x00002b803473f7adll, 0x0000e1fc780e1fc8ll, 0x00002e0e85a9de04ll, + 0x0000e070381c0e08ll, 0x0000309857a05e07ll, 0x0000dee95c4ca038ll, 0x0000331dba0efce1ll, + 0x0000dd67c8a60dd7ll, 0x0000359ebc5b69d9ll, 0x0000dbeb61eed19dll, 0x0000381b6d9bb29bll, + 0x0000da740da740dbll, 0x00003a93dc9864b2ll, 0x0000d901b2036407ll, 0x00003d0817ce9cd4ll, + 0x0000d79435e50d7all, 0x00003f782d7204d0ll, 0x0000d62b80d62b81ll, 0x000041e42b6ec0c0ll, + 0x0000d4c77b03531ell, 0x0000444c1f6b4c2dll, 0x0000d3680d3680d4ll, 0x000046b016ca47c1ll, + 0x0000d20d20d20d21ll, 0x000049101eac381cll, 0x0000d0b69fcbd259ll, 0x00004b6c43f1366all, + 0x0000cf6474a8819fll, 0x00004dc4933a9337ll, 0x0000ce168a772509ll, 0x0000501918ec6c11ll, + 0x0000cccccccccccdll, 0x00005269e12f346ell, 0x0000cb8727c065c4ll, 0x000054b6f7f1325all, + 0x0000ca4587e6b750ll, 0x0000570068e7ef5all, 0x0000c907da4e8712ll, 0x000059463f919deell, + 0x0000c7ce0c7ce0c8ll, 0x00005b8887367433ll, 0x0000c6980c6980c7ll, 0x00005dc74ae9fbecll, + 0x0000c565c87b5f9ell, 0x00006002958c5871ll, 0x0000c4372f855d83ll, 0x0000623a71cb82c8ll, + 0x0000c30c30c30c31ll, 0x0000646eea247c5cll, 0x0000c1e4bbd595f7ll, 0x000066a008e4788cll, + 0x0000c0c0c0c0c0c1ll, 0x000068cdd829fd81ll, 0x0000bfa02fe80bfbll, 0x00006af861e5fc7dll, + 0x0000be82fa0be830ll, 0x00006d1fafdce20all, 0x0000bd6910470767ll, 0x00006f43cba79e40ll, + 0x0000bc52640bc527ll, 0x00007164beb4a56dll, 0x0000bb3ee721a54ell, 0x000073829248e961ll, + 0x0000ba2e8ba2e8bbll, 0x0000759d4f80cba8ll, 0x0000b92143fa36f6ll, 0x000077b4ff5108d9ll, + 0x0000b81702e05c0cll, 0x000079c9aa879d53ll, 0x0000b70fbb5a19bfll, 0x00007bdb59cca388ll, + 0x0000b60b60b60b61ll, 0x00007dea15a32c1bll, 0x0000b509e68a9b95ll, 0x00007ff5e66a0ffell, + 0x0000b40b40b40b41ll, 0x000081fed45cbccbll, 0x0000b30f63528918ll, 0x00008404e793fb81ll, + 0x0000b21642c8590cll, 0x000086082806b1d5ll, 0x0000b11fd3b80b12ll, 0x000088089d8a9e47ll, + 0x0000b02c0b02c0b1ll, 0x00008a064fd50f2all, 0x0000af3addc680b0ll, 0x00008c01467b94bbll, + 0x0000ae4c415c9883ll, 0x00008df988f4ae80ll, 0x0000ad602b580ad7ll, 0x00008fef1e987409ll, + 0x0000ac7691840ac8ll, 0x000091e20ea1393ell, 0x0000ab8f69e2835all, 0x000093d2602c2e5fll, + 0x0000aaaaaaaaaaabll, 0x000095c01a39fbd6ll, 0x0000a9c84a47a080ll, 0x000097ab43af59f9ll, + 0x0000a8e83f5717c1ll, 0x00009993e355a4e5ll, 0x0000a80a80a80a81ll, 0x00009b79ffdb6c8bll, + 0x0000a72f0539782all, 0x00009d5d9fd5010bll, 0x0000a655c4392d7cll, 0x00009f3ec9bcfb80ll, + 0x0000a57eb50295fbll, 0x0000a11d83f4c355ll, 0x0000a4a9cf1d9684ll, 0x0000a2f9d4c51039ll, + 0x0000a3d70a3d70a4ll, 0x0000a4d3c25e68dcll, 0x0000a3065e3fae7dll, 0x0000a6ab52d99e76ll, + 0x0000a237c32b16d0ll, 0x0000a8808c384547ll, 0x0000a16b312ea8fdll, 0x0000aa5374652a1cll, + 0x0000a0a0a0a0a0a1ll, 0x0000ac241134c4e9ll, 0x00009fd809fd80a0ll, 0x0000adf26865a8a1ll, + 0x00009f1165e72549ll, 0x0000afbe7fa0f04dll, 0x00009e4cad23dd60ll, 0x0000b1885c7aa982ll, + 0x00009d89d89d89d9ll, 0x0000b35004723c46ll, 0x00009cc8e160c3fcll, 0x0000b5157cf2d078ll, + 0x00009c09c09c09c1ll, 0x0000b6d8cb53b0call, 0x00009b4c6f9ef03bll, 0x0000b899f4d8ab63ll, + 0x00009a90e7d95bc7ll, 0x0000ba58feb2703all, 0x000099d722dabde6ll, 0x0000bc15edfeed32ll, + 0x0000991f1a515886ll, 0x0000bdd0c7c9a817ll, 0x00009868c809868dll, 0x0000bf89910c1678ll, + 0x000097b425ed097cll, 0x0000c1404eadf383ll, 0x000097012e025c05ll, 0x0000c2f5058593d9ll, + 0x0000964fda6c0965ll, 0x0000c4a7ba58377cll, 0x000095a02568095bll, 0x0000c65871da59ddll, + 0x000094f2094f2095ll, 0x0000c80730b00016ll, 0x0000944580944581ll, 0x0000c9b3fb6d0559ll, + 0x0000939a85c4093all, 0x0000cb5ed69565afll, 0x000092f113840498ll, 0x0000cd07c69d8702ll, + 0x0000924924924925ll, 0x0000ceaecfea8085ll, 0x000091a2b3c4d5e7ll, 0x0000d053f6d26089ll, + 0x000090fdbc090fdcll, 0x0000d1f73f9c70c0ll, 0x0000905a38633e07ll, 0x0000d398ae817906ll, + 0x00008fb823ee08fcll, 0x0000d53847ac00a6ll, 0x00008f1779d9fdc4ll, 0x0000d6d60f388e41ll, + 0x00008e78356d1409ll, 0x0000d8720935e643ll, 0x00008dda5202376all, 0x0000da0c39a54804ll, + 0x00008d3dcb08d3ddll, 0x0000dba4a47aa996ll, 0x00008ca29c046515ll, 0x0000dd3b4d9cf24bll, + 0x00008c08c08c08c1ll, 0x0000ded038e633f3ll, 0x00008b70344a139cll, 0x0000e0636a23e2eell, + 0x00008ad8f2fba939ll, 0x0000e1f4e5170d02ll, 0x00008a42f870566all, 0x0000e384ad748f0ell, + 0x000089ae4089ae41ll, 0x0000e512c6e54998ll, 0x0000891ac73ae982ll, 0x0000e69f35065448ll, + 0x0000888888888889ll, 0x0000e829fb693044ll, 0x000087f78087f781ll, 0x0000e9b31d93f98ell, + 0x00008767ab5f34e5ll, 0x0000eb3a9f019750ll, 0x000086d905447a35ll, 0x0000ecc08321eb30ll, + 0x0000864b8a7de6d2ll, 0x0000ee44cd59ffabll, 0x000085bf37612cefll, 0x0000efc781043579ll, + 0x0000853408534086ll, 0x0000f148a170700all, 0x000084a9f9c8084bll, 0x0000f2c831e44116ll, + 0x0000842108421085ll, 0x0000f446359b1353ll, 0x0000839930523fbfll, 0x0000f5c2afc65447ll, + 0x000083126e978d50ll, 0x0000f73da38d9d4all, 0x0000828cbfbeb9a1ll, 0x0000f8b7140edbb1ll, + 0x0000820820820821ll, 0x0000fa2f045e7832ll, 0x000081848da8faf1ll, 0x0000fba577877d7dll, + 0x0000810204081021ll, 0x0000fd1a708bbe11ll, 0x0000808080808081ll, 0x0000fe8df263f957ll, + 0x0000800000000000ll, 0x0000ffff00000000ll, + }; + + + // LL_tbl[k] = 2^48*log2(1.0+k/2^15); +static int64_t __LL_tbl[256] = { + 0x0000000000000000ull, 0x00000002e2a60a00ull, 0x000000070cb64ec5ull, 0x00000009ef50ce67ull, + 0x0000000cd1e588fdull, 0x0000000fb4747e9cull, 0x0000001296fdaf5eull, 0x0000001579811b58ull, + 0x000000185bfec2a1ull, 0x0000001b3e76a552ull, 0x0000001e20e8c380ull, 0x0000002103551d43ull, + 0x00000023e5bbb2b2ull, 0x00000026c81c83e4ull, 0x00000029aa7790f0ull, 0x0000002c8cccd9edull, + 0x0000002f6f1c5ef2ull, 0x0000003251662017ull, 0x0000003533aa1d71ull, 0x0000003815e8571aull, + 0x0000003af820cd26ull, 0x0000003dda537faeull, 0x00000040bc806ec8ull, 0x000000439ea79a8cull, + 0x0000004680c90310ull, 0x0000004962e4a86cull, 0x0000004c44fa8ab6ull, 0x0000004f270aaa06ull, + 0x0000005209150672ull, 0x00000054eb19a013ull, 0x00000057cd1876fdull, 0x0000005aaf118b4aull, + 0x0000005d9104dd0full, 0x0000006072f26c64ull, 0x0000006354da3960ull, 0x0000006636bc441aull, + 0x0000006918988ca8ull, 0x0000006bfa6f1322ull, 0x0000006edc3fd79full, 0x00000071be0ada35ull, + 0x000000749fd01afdull, 0x00000077818f9a0cull, 0x0000007a6349577aull, 0x0000007d44fd535eull, + 0x0000008026ab8dceull, 0x00000083085406e3ull, 0x00000085e9f6beb2ull, 0x00000088cb93b552ull, + 0x0000008bad2aeadcull, 0x0000008e8ebc5f65ull, 0x0000009170481305ull, 0x0000009451ce05d3ull, + 0x00000097334e37e5ull, 0x0000009a14c8a953ull, 0x0000009cf63d5a33ull, 0x0000009fd7ac4a9dull, + 0x000000a2b07f3458ull, 0x000000a59a78ea6aull, 0x000000a87bd699fbull, 0x000000ab5d2e8970ull, + 0x000000ae3e80b8e3ull, 0x000000b11fcd2869ull, 0x000000b40113d818ull, 0x000000b6e254c80aull, + 0x000000b9c38ff853ull, 0x000000bca4c5690cull, 0x000000bf85f51a4aull, 0x000000c2671f0c26ull, + 0x000000c548433eb6ull, 0x000000c82961b211ull, 0x000000cb0a7a664dull, 0x000000cdeb8d5b82ull, + 0x000000d0cc9a91c8ull, 0x000000d3ada20933ull, 0x000000d68ea3c1ddull, 0x000000d96f9fbbdbull, + 0x000000dc5095f744ull, 0x000000df31867430ull, 0x000000e2127132b5ull, 0x000000e4f35632eaull, + 0x000000e7d43574e6ull, 0x000000eab50ef8c1ull, 0x000000ed95e2be90ull, 0x000000f076b0c66cull, + 0x000000f35779106aull, 0x000000f6383b9ca2ull, 0x000000f918f86b2aull, 0x000000fbf9af7c1aull, + 0x000000feda60cf88ull, 0x00000101bb0c658cull, 0x000001049bb23e3cull, 0x000001077c5259afull, + 0x0000010a5cecb7fcull, 0x0000010d3d81593aull, 0x000001101e103d7full, 0x00000112fe9964e4ull, + 0x00000115df1ccf7eull, 0x00000118bf9a7d64ull, 0x0000011ba0126eadull, 0x0000011e8084a371ull, + 0x0000012160f11bc6ull, 0x000001244157d7c3ull, 0x0000012721b8d77full, 0x0000012a02141b10ull, + 0x0000012ce269a28eull, 0x0000012fc2b96e0full, 0x00000132a3037daaull, 0x000001358347d177ull, + 0x000001386386698cull, 0x0000013b43bf45ffull, 0x0000013e23f266e9ull, 0x00000141041fcc5eull, + 0x00000143e4477678ull, 0x00000146c469654bull, 0x00000149a48598f0ull, 0x0000014c849c117cull, + 0x0000014f64accf08ull, 0x0000015244b7d1a9ull, 0x0000015524bd1976ull, 0x0000015804bca687ull, + 0x0000015ae4b678f2ull, 0x0000015dc4aa90ceull, 0x00000160a498ee31ull, 0x0000016384819134ull, + 0x00000166646479ecull, 0x000001694441a870ull, 0x0000016c24191cd7ull, 0x0000016df6ca19bdull, + 0x00000171e3b6d7aaull, 0x00000174c37d1e44ull, 0x00000177a33dab1cull, 0x0000017a82f87e49ull, + 0x0000017d62ad97e2ull, 0x00000180425cf7feull, 0x00000182b07f3458ull, 0x0000018601aa8c19ull, + 0x00000188e148c046ull, 0x0000018bc0e13b52ull, 0x0000018ea073fd52ull, 0x000001918001065dull, + 0x000001945f88568bull, 0x000001973f09edf2ull, 0x0000019a1e85ccaaull, 0x0000019cfdfbf2c8ull, + 0x0000019fdd6c6063ull, 0x000001a2bcd71593ull, 0x000001a59c3c126eull, 0x000001a87b9b570bull, + 0x000001ab5af4e380ull, 0x000001ae3a48b7e5ull, 0x000001b11996d450ull, 0x000001b3f8df38d9ull, + 0x000001b6d821e595ull, 0x000001b9b75eda9bull, 0x000001bc96961803ull, 0x000001bf75c79de3ull, + 0x000001c254f36c51ull, 0x000001c534198365ull, 0x000001c81339e336ull, 0x000001caf2548bd9ull, + 0x000001cdd1697d67ull, 0x000001d0b078b7f5ull, 0x000001d38f823b9aull, 0x000001d66e86086dull, + 0x000001d94d841e86ull, 0x000001dc2c7c7df9ull, 0x000001df0b6f26dfull, 0x000001e1ea5c194eull, + 0x000001e4c943555dull, 0x000001e7a824db23ull, 0x000001ea8700aab5ull, 0x000001ed65d6c42bull, + 0x000001f044a7279dull, 0x000001f32371d51full, 0x000001f60236cccaull, 0x000001f8e0f60eb3ull, + 0x000001fbbfaf9af3ull, 0x000001fe9e63719eull, 0x000002017d1192ccull, 0x000002045bb9fe94ull, + 0x000002073a5cb50dull, 0x00000209c06e6212ull, 0x0000020cf791026aull, 0x0000020fd622997cull, + 0x00000212b07f3458ull, 0x000002159334a8d8ull, 0x0000021871b52150ull, 0x0000021b502fe517ull, + 0x0000021d6a73a78full, 0x000002210d144eeeull, 0x00000223eb7df52cull, 0x00000226c9e1e713ull, + 0x00000229a84024bbull, 0x0000022c23679b4eull, 0x0000022f64eb83a8ull, 0x000002324338a51bull, + 0x00000235218012a9ull, 0x00000237ffc1cc69ull, 0x0000023a2c3b0ea4ull, 0x0000023d13ee805bull, + 0x0000024035e9221full, 0x00000243788faf25ull, 0x0000024656b4e735ull, 0x00000247ed646bfeull, + 0x0000024c12ee3d98ull, 0x0000024ef1025c1aull, 0x00000251cf10c799ull, 0x0000025492644d65ull, + 0x000002578b1c85eeull, 0x0000025a6919d8f0ull, 0x0000025d13ee805bull, 0x0000026025036716ull, + 0x0000026296453882ull, 0x00000265e0d62b53ull, 0x00000268beb701f3ull, 0x0000026b9c92265eull, + 0x0000026d32f798a9ull, 0x00000271583758ebull, 0x000002743601673bull, 0x0000027713c5c3b0ull, + 0x00000279f1846e5full, 0x0000027ccf3d6761ull, 0x0000027e6580aecbull, 0x000002828a9e44b3ull, + 0x0000028568462932ull, 0x00000287bdbf5255ull, 0x0000028b2384de4aull, 0x0000028d13ee805bull, + 0x0000029035e9221full, 0x0000029296453882ull, 0x0000029699bdfb61ull, 0x0000029902a37aabull, + 0x0000029c54b864c9ull, 0x0000029deabd1083ull, 0x000002a20f9c0bb5ull, 0x000002a4c7605d61ull, + 0x000002a7bdbf5255ull, 0x000002a96056dafcull, 0x000002ac3daf14efull, 0x000002af1b019ecaull, + 0x000002b296453882ull, 0x000002b5d022d80full, 0x000002b8fa471cb3ull, 0x000002ba9012e713ull, + 0x000002bd6d4901ccull, 0x000002c04a796cf6ull, 0x000002c327a428a6ull, 0x000002c61a5e8f4cull, + 0x000002c8e1e891f6ull, 0x000002cbbf023fc2ull, 0x000002ce9c163e6eull, 0x000002d179248e13ull, + 0x000002d4562d2ec6ull, 0x000002d73330209dull, 0x000002da102d63b0ull, 0x000002dced24f814ull, +}; + + + + +#endif diff --git a/net/ceph/crush/mapper.c b/net/ceph/crush/mapper.c index a1ef53c04415..5b47736d27d9 100644 --- a/net/ceph/crush/mapper.c +++ b/net/ceph/crush/mapper.c @@ -20,7 +20,7 @@ #include <linux/crush/crush.h> #include <linux/crush/hash.h> -#include <linux/crush/mapper.h> +#include "crush_ln_table.h" /* * Implement the core CRUSH mapping algorithm. @@ -238,6 +238,102 @@ static int bucket_straw_choose(struct crush_bucket_straw *bucket, return bucket->h.items[high]; } +// compute 2^44*log2(input+1) +uint64_t crush_ln(unsigned xin) +{ + unsigned x=xin, x1; + int iexpon, index1, index2; + uint64_t RH, LH, LL, xl64, result; + + x++; + + // normalize input + iexpon = 15; + while(!(x&0x18000)) { x<<=1; iexpon--; } + + index1 = (x>>8)<<1; + // RH ~ 2^56/index1 + RH = __RH_LH_tbl[index1 - 256]; + // LH ~ 2^48 * log2(index1/256) + LH = __RH_LH_tbl[index1 + 1 - 256]; + + // RH*x ~ 2^48 * (2^15 + xf), xf<2^8 + xl64 = (int64_t)x * RH; + xl64 >>= 48; + x1 = xl64; + + result = iexpon; + result <<= (12 + 32); + + index2 = x1 & 0xff; + // LL ~ 2^48*log2(1.0+index2/2^15) + LL = __LL_tbl[index2]; + + LH = LH + LL; + + LH >>= (48-12 - 32); + result += LH; + + return result; +} + + +/* + * straw2 + * + * for reference, see: + * + * http://en.wikipedia.org/wiki/Exponential_distribution#Distribution_of_the_minimum_of_exponential_random_variables + * + */ + +static int bucket_straw2_choose(struct crush_bucket_straw2 *bucket, + int x, int r) +{ + unsigned i, high = 0; + unsigned u; + unsigned w; + __s64 ln, draw, high_draw = 0; + + for (i = 0; i < bucket->h.size; i++) { + w = bucket->item_weights[i]; + if (w) { + u = crush_hash32_3(bucket->h.hash, x, + bucket->h.items[i], r); + u &= 0xffff; + + /* + * for some reason slightly less than 0x10000 produces + * a slightly more accurate distribution... probably a + * rounding effect. + * + * the natural log lookup table maps [0,0xffff] + * (corresponding to real numbers [1/0x10000, 1] to + * [0, 0xffffffffffff] (corresponding to real numbers + * [-11.090355,0]). + */ + ln = crush_ln(u) - 0x1000000000000ll; + + /* + * divide by 16.16 fixed-point weight. note + * that the ln value is negative, so a larger + * weight means a larger (less negative) value + * for draw. + */ + draw = div64_s64(ln, w); + } else { + draw = S64_MIN; + } + + if (i == 0 || draw > high_draw) { + high = i; + high_draw = draw; + } + } + return bucket->h.items[high]; +} + + static int crush_bucket_choose(struct crush_bucket *in, int x, int r) { dprintk(" crush_bucket_choose %d x=%d r=%d\n", in->id, x, r); @@ -255,12 +351,16 @@ static int crush_bucket_choose(struct crush_bucket *in, int x, int r) case CRUSH_BUCKET_STRAW: return bucket_straw_choose((struct crush_bucket_straw *)in, x, r); + case CRUSH_BUCKET_STRAW2: + return bucket_straw2_choose((struct crush_bucket_straw2 *)in, + x, r); default: dprintk("unknown bucket %d alg %d\n", in->id, in->alg); return in->items[0]; } } + /* * true if device is marked "out" (failed, fully offloaded) * of the cluster @@ -290,6 +390,7 @@ static int is_out(const struct crush_map *map, * @type: the type of item to choose * @out: pointer to output vector * @outpos: our position in that vector + * @out_size: size of the out vector * @tries: number of attempts to make * @recurse_tries: number of attempts to have recursive chooseleaf make * @local_retries: localized retries @@ -304,6 +405,7 @@ static int crush_choose_firstn(const struct crush_map *map, const __u32 *weight, int weight_max, int x, int numrep, int type, int *out, int outpos, + int out_size, unsigned int tries, unsigned int recurse_tries, unsigned int local_retries, @@ -322,6 +424,7 @@ static int crush_choose_firstn(const struct crush_map *map, int item = 0; int itemtype; int collide, reject; + int count = out_size; dprintk("CHOOSE%s bucket %d x %d outpos %d numrep %d tries %d recurse_tries %d local_retries %d local_fallback_retries %d parent_r %d\n", recurse_to_leaf ? "_LEAF" : "", @@ -329,7 +432,7 @@ static int crush_choose_firstn(const struct crush_map *map, tries, recurse_tries, local_retries, local_fallback_retries, parent_r); - for (rep = outpos; rep < numrep; rep++) { + for (rep = outpos; rep < numrep && count > 0 ; rep++) { /* keep trying until we get a non-out, non-colliding item */ ftotal = 0; skip_rep = 0; @@ -403,7 +506,7 @@ static int crush_choose_firstn(const struct crush_map *map, map->buckets[-1-item], weight, weight_max, x, outpos+1, 0, - out2, outpos, + out2, outpos, count, recurse_tries, 0, local_retries, local_fallback_retries, @@ -463,6 +566,7 @@ reject: dprintk("CHOOSE got %d\n", item); out[outpos] = item; outpos++; + count--; } dprintk("CHOOSE returns %d\n", outpos); @@ -654,6 +758,7 @@ int crush_do_rule(const struct crush_map *map, __u32 step; int i, j; int numrep; + int out_size; /* * the original choose_total_tries value was off by one (it * counted "retries" and not "tries"). add one. @@ -761,6 +866,7 @@ int crush_do_rule(const struct crush_map *map, x, numrep, curstep->arg2, o+osize, j, + result_max-osize, choose_tries, recurse_tries, choose_local_retries, @@ -770,11 +876,13 @@ int crush_do_rule(const struct crush_map *map, c+osize, 0); } else { + out_size = ((numrep < (result_max-osize)) ? + numrep : (result_max-osize)); crush_choose_indep( map, map->buckets[-1-w[i]], weight, weight_max, - x, numrep, numrep, + x, out_size, numrep, curstep->arg2, o+osize, j, choose_tries, @@ -783,7 +891,7 @@ int crush_do_rule(const struct crush_map *map, recurse_to_leaf, c+osize, 0); - osize += numrep; + osize += out_size; } } diff --git a/net/ceph/debugfs.c b/net/ceph/debugfs.c index 14d9995097cc..593dc2eabcc8 100644 --- a/net/ceph/debugfs.c +++ b/net/ceph/debugfs.c @@ -22,6 +22,7 @@ * .../monmap - current monmap * .../osdc - active osd requests * .../monc - mon client state + * .../client_options - libceph-only (i.e. not rbd or cephfs) options * .../dentry_lru - dump contents of dentry lru * .../caps - expose cap (reservation) stats * .../bdi - symlink to ../../bdi/something @@ -177,10 +178,24 @@ static int osdc_show(struct seq_file *s, void *pp) return 0; } +static int client_options_show(struct seq_file *s, void *p) +{ + struct ceph_client *client = s->private; + int ret; + + ret = ceph_print_client_options(s, client); + if (ret) + return ret; + + seq_putc(s, '\n'); + return 0; +} + CEPH_DEFINE_SHOW_FUNC(monmap_show) CEPH_DEFINE_SHOW_FUNC(osdmap_show) CEPH_DEFINE_SHOW_FUNC(monc_show) CEPH_DEFINE_SHOW_FUNC(osdc_show) +CEPH_DEFINE_SHOW_FUNC(client_options_show) int ceph_debugfs_init(void) { @@ -242,6 +257,14 @@ int ceph_debugfs_client_init(struct ceph_client *client) if (!client->debugfs_osdmap) goto out; + client->debugfs_options = debugfs_create_file("client_options", + 0600, + client->debugfs_dir, + client, + &client_options_show_fops); + if (!client->debugfs_options) + goto out; + return 0; out: @@ -252,6 +275,7 @@ out: void ceph_debugfs_client_cleanup(struct ceph_client *client) { dout("ceph_debugfs_client_cleanup %p\n", client); + debugfs_remove(client->debugfs_options); debugfs_remove(client->debugfs_osdmap); debugfs_remove(client->debugfs_monmap); debugfs_remove(client->osdc.debugfs_file); diff --git a/net/ceph/messenger.c b/net/ceph/messenger.c index a9f4ae45b7fb..967080a9f043 100644 --- a/net/ceph/messenger.c +++ b/net/ceph/messenger.c @@ -505,8 +505,6 @@ static int ceph_tcp_connect(struct ceph_connection *con) pr_err("connect %s error %d\n", ceph_pr_addr(&con->peer_addr.in_addr), ret); sock_release(sock); - con->error_msg = "connect error"; - return ret; } @@ -2145,12 +2143,10 @@ static int process_connect(struct ceph_connection *con) * to WAIT. This shouldn't happen if we are the * client. */ - pr_err("process_connect got WAIT as client\n"); con->error_msg = "protocol error, got WAIT as client"; return -1; default: - pr_err("connect protocol error, will retry\n"); con->error_msg = "protocol error, garbage tag during connect"; return -1; } @@ -2282,8 +2278,7 @@ static int read_partial_message(struct ceph_connection *con) crc = crc32c(0, &con->in_hdr, offsetof(struct ceph_msg_header, crc)); if (cpu_to_le32(crc) != con->in_hdr.crc) { - pr_err("read_partial_message bad hdr " - " crc %u != expected %u\n", + pr_err("read_partial_message bad hdr crc %u != expected %u\n", crc, con->in_hdr.crc); return -EBADMSG; } @@ -2313,7 +2308,7 @@ static int read_partial_message(struct ceph_connection *con) pr_err("read_partial_message bad seq %lld expected %lld\n", seq, con->in_seq + 1); con->error_msg = "bad message sequence # for incoming message"; - return -EBADMSG; + return -EBADE; } /* allocate message? */ @@ -2660,6 +2655,8 @@ more: switch (ret) { case -EBADMSG: con->error_msg = "bad crc"; + /* fall through */ + case -EBADE: ret = -EIO; break; case -EIO: @@ -2838,7 +2835,8 @@ static void con_work(struct work_struct *work) if (ret < 0) { if (ret == -EAGAIN) continue; - con->error_msg = "socket error on read"; + if (!con->error_msg) + con->error_msg = "socket error on read"; fault = true; break; } @@ -2847,7 +2845,8 @@ static void con_work(struct work_struct *work) if (ret < 0) { if (ret == -EAGAIN) continue; - con->error_msg = "socket error on write"; + if (!con->error_msg) + con->error_msg = "socket error on write"; fault = true; } @@ -2869,11 +2868,13 @@ static void con_work(struct work_struct *work) */ static void con_fault(struct ceph_connection *con) { - pr_warn("%s%lld %s %s\n", ENTITY_NAME(con->peer_name), - ceph_pr_addr(&con->peer_addr.in_addr), con->error_msg); dout("fault %p state %lu to peer %s\n", con, con->state, ceph_pr_addr(&con->peer_addr.in_addr)); + pr_warn("%s%lld %s %s\n", ENTITY_NAME(con->peer_name), + ceph_pr_addr(&con->peer_addr.in_addr), con->error_msg); + con->error_msg = NULL; + WARN_ON(con->state != CON_STATE_CONNECTING && con->state != CON_STATE_NEGOTIATING && con->state != CON_STATE_OPEN); @@ -3295,8 +3296,8 @@ static int ceph_con_in_msg_alloc(struct ceph_connection *con, int *skip) */ if (*skip) return 0; - con->error_msg = "error allocating memory for incoming message"; + con->error_msg = "error allocating memory for incoming message"; return -ENOMEM; } memcpy(&con->in_msg->hdr, &con->in_hdr, sizeof(con->in_hdr)); diff --git a/net/ceph/osdmap.c b/net/ceph/osdmap.c index b8c3fde5b04f..15796696d64e 100644 --- a/net/ceph/osdmap.c +++ b/net/ceph/osdmap.c @@ -122,6 +122,22 @@ bad: return -EINVAL; } +static int crush_decode_straw2_bucket(void **p, void *end, + struct crush_bucket_straw2 *b) +{ + int j; + dout("crush_decode_straw2_bucket %p to %p\n", *p, end); + b->item_weights = kcalloc(b->h.size, sizeof(u32), GFP_NOFS); + if (b->item_weights == NULL) + return -ENOMEM; + ceph_decode_need(p, end, b->h.size * sizeof(u32), bad); + for (j = 0; j < b->h.size; j++) + b->item_weights[j] = ceph_decode_32(p); + return 0; +bad: + return -EINVAL; +} + static int skip_name_map(void **p, void *end) { int len; @@ -204,6 +220,9 @@ static struct crush_map *crush_decode(void *pbyval, void *end) case CRUSH_BUCKET_STRAW: size = sizeof(struct crush_bucket_straw); break; + case CRUSH_BUCKET_STRAW2: + size = sizeof(struct crush_bucket_straw2); + break; default: err = -EINVAL; goto bad; @@ -261,6 +280,12 @@ static struct crush_map *crush_decode(void *pbyval, void *end) if (err < 0) goto bad; break; + case CRUSH_BUCKET_STRAW2: + err = crush_decode_straw2_bucket(p, end, + (struct crush_bucket_straw2 *)b); + if (err < 0) + goto bad; + break; } } diff --git a/net/core/dev.c b/net/core/dev.c index af4a1b0adc10..1796cef55ab5 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2713,7 +2713,7 @@ static struct sk_buff *validate_xmit_skb(struct sk_buff *skb, struct net_device if (unlikely(!skb)) goto out_null; - if (netif_needs_gso(dev, skb, features)) { + if (netif_needs_gso(skb, features)) { struct sk_buff *segs; segs = skb_gso_segment(skb, features); diff --git a/net/core/filter.c b/net/core/filter.c index b669e75d2b36..bf831a85c315 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -1175,12 +1175,27 @@ int sk_attach_bpf(u32 ufd, struct sock *sk) return 0; } +/** + * bpf_skb_clone_not_writable - is the header of a clone not writable + * @skb: buffer to check + * @len: length up to which to write, can be negative + * + * Returns true if modifying the header part of the cloned buffer + * does require the data to be copied. I.e. this version works with + * negative lengths needed for eBPF case! + */ +static bool bpf_skb_clone_unwritable(const struct sk_buff *skb, int len) +{ + return skb_header_cloned(skb) || + (int) skb_headroom(skb) + len > skb->hdr_len; +} + #define BPF_RECOMPUTE_CSUM(flags) ((flags) & 1) static u64 bpf_skb_store_bytes(u64 r1, u64 r2, u64 r3, u64 r4, u64 flags) { struct sk_buff *skb = (struct sk_buff *) (long) r1; - unsigned int offset = (unsigned int) r2; + int offset = (int) r2; void *from = (void *) (long) r3; unsigned int len = (unsigned int) r4; char buf[16]; @@ -1194,10 +1209,12 @@ static u64 bpf_skb_store_bytes(u64 r1, u64 r2, u64 r3, u64 r4, u64 flags) * * so check for invalid 'offset' and too large 'len' */ - if (unlikely(offset > 0xffff || len > sizeof(buf))) + if (unlikely((u32) offset > 0xffff || len > sizeof(buf))) return -EFAULT; - if (skb_cloned(skb) && !skb_clone_writable(skb, offset + len)) + offset -= skb->data - skb_mac_header(skb); + if (unlikely(skb_cloned(skb) && + bpf_skb_clone_unwritable(skb, offset + len))) return -EFAULT; ptr = skb_header_pointer(skb, offset, len, buf); @@ -1232,15 +1249,18 @@ const struct bpf_func_proto bpf_skb_store_bytes_proto = { #define BPF_HEADER_FIELD_SIZE(flags) ((flags) & 0x0f) #define BPF_IS_PSEUDO_HEADER(flags) ((flags) & 0x10) -static u64 bpf_l3_csum_replace(u64 r1, u64 offset, u64 from, u64 to, u64 flags) +static u64 bpf_l3_csum_replace(u64 r1, u64 r2, u64 from, u64 to, u64 flags) { struct sk_buff *skb = (struct sk_buff *) (long) r1; + int offset = (int) r2; __sum16 sum, *ptr; - if (unlikely(offset > 0xffff)) + if (unlikely((u32) offset > 0xffff)) return -EFAULT; - if (skb_cloned(skb) && !skb_clone_writable(skb, offset + sizeof(sum))) + offset -= skb->data - skb_mac_header(skb); + if (unlikely(skb_cloned(skb) && + bpf_skb_clone_unwritable(skb, offset + sizeof(sum)))) return -EFAULT; ptr = skb_header_pointer(skb, offset, sizeof(sum), &sum); @@ -1276,16 +1296,19 @@ const struct bpf_func_proto bpf_l3_csum_replace_proto = { .arg5_type = ARG_ANYTHING, }; -static u64 bpf_l4_csum_replace(u64 r1, u64 offset, u64 from, u64 to, u64 flags) +static u64 bpf_l4_csum_replace(u64 r1, u64 r2, u64 from, u64 to, u64 flags) { struct sk_buff *skb = (struct sk_buff *) (long) r1; u32 is_pseudo = BPF_IS_PSEUDO_HEADER(flags); + int offset = (int) r2; __sum16 sum, *ptr; - if (unlikely(offset > 0xffff)) + if (unlikely((u32) offset > 0xffff)) return -EFAULT; - if (skb_cloned(skb) && !skb_clone_writable(skb, offset + sizeof(sum))) + offset -= skb->data - skb_mac_header(skb); + if (unlikely(skb_cloned(skb) && + bpf_skb_clone_unwritable(skb, offset + sizeof(sum)))) return -EFAULT; ptr = skb_header_pointer(skb, offset, sizeof(sum), &sum); diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c index a3abb719221f..78fc04ad36fc 100644 --- a/net/core/net_namespace.c +++ b/net/core/net_namespace.c @@ -16,7 +16,6 @@ #include <linux/export.h> #include <linux/user_namespace.h> #include <linux/net_namespace.h> -#include <linux/rtnetlink.h> #include <net/sock.h> #include <net/netlink.h> #include <net/net_namespace.h> diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 3b6e5830256e..d1967dab9cc6 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -4124,19 +4124,21 @@ EXPORT_SYMBOL(skb_try_coalesce); */ void skb_scrub_packet(struct sk_buff *skb, bool xnet) { - if (xnet) - skb_orphan(skb); skb->tstamp.tv64 = 0; skb->pkt_type = PACKET_HOST; skb->skb_iif = 0; skb->ignore_df = 0; skb_dst_drop(skb); - skb->mark = 0; skb_sender_cpu_clear(skb); - skb_init_secmark(skb); secpath_reset(skb); nf_reset(skb); nf_reset_trace(skb); + + if (!xnet) + return; + + skb_orphan(skb); + skb->mark = 0; } EXPORT_SYMBOL_GPL(skb_scrub_packet); diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c index 5eaadabe23a1..079a224471e7 100644 --- a/net/dsa/dsa.c +++ b/net/dsa/dsa.c @@ -124,7 +124,7 @@ static ssize_t temp1_max_store(struct device *dev, return count; } -static DEVICE_ATTR(temp1_max, S_IRUGO, temp1_max_show, temp1_max_store); +static DEVICE_ATTR_RW(temp1_max); static ssize_t temp1_max_alarm_show(struct device *dev, struct device_attribute *attr, char *buf) @@ -159,8 +159,8 @@ static umode_t dsa_hwmon_attrs_visible(struct kobject *kobj, if (index == 1) { if (!drv->get_temp_limit) mode = 0; - else if (drv->set_temp_limit) - mode |= S_IWUSR; + else if (!drv->set_temp_limit) + mode &= ~S_IWUSR; } else if (index == 2 && !drv->get_temp_alarm) { mode = 0; } diff --git a/net/ipv4/fou.c b/net/ipv4/fou.c index af150b43b214..34968cd5c146 100644 --- a/net/ipv4/fou.c +++ b/net/ipv4/fou.c @@ -711,11 +711,10 @@ static int fou_nl_dump(struct sk_buff *skb, struct netlink_callback *cb) cb->nlh->nlmsg_seq, NLM_F_MULTI, skb, FOU_CMD_GET); if (ret) - goto done; + break; } mutex_unlock(&fn->fou_lock); -done: cb->args[0] = idx; return skb->len; } diff --git a/net/ipv4/inet_diag.c b/net/ipv4/inet_diag.c index 70e8b3c308ec..bb77ebdae3b3 100644 --- a/net/ipv4/inet_diag.c +++ b/net/ipv4/inet_diag.c @@ -111,6 +111,7 @@ int inet_sk_diag_fill(struct sock *sk, struct inet_connection_sock *icsk, const struct nlmsghdr *unlh) { const struct inet_sock *inet = inet_sk(sk); + const struct tcp_congestion_ops *ca_ops; const struct inet_diag_handler *handler; int ext = req->idiag_ext; struct inet_diag_msg *r; @@ -208,16 +209,31 @@ int inet_sk_diag_fill(struct sock *sk, struct inet_connection_sock *icsk, info = nla_data(attr); } - if ((ext & (1 << (INET_DIAG_CONG - 1))) && icsk->icsk_ca_ops) - if (nla_put_string(skb, INET_DIAG_CONG, - icsk->icsk_ca_ops->name) < 0) + if (ext & (1 << (INET_DIAG_CONG - 1))) { + int err = 0; + + rcu_read_lock(); + ca_ops = READ_ONCE(icsk->icsk_ca_ops); + if (ca_ops) + err = nla_put_string(skb, INET_DIAG_CONG, ca_ops->name); + rcu_read_unlock(); + if (err < 0) goto errout; + } handler->idiag_get_info(sk, r, info); - if (sk->sk_state < TCP_TIME_WAIT && - icsk->icsk_ca_ops && icsk->icsk_ca_ops->get_info) - icsk->icsk_ca_ops->get_info(sk, ext, skb); + if (sk->sk_state < TCP_TIME_WAIT) { + int err = 0; + + rcu_read_lock(); + ca_ops = READ_ONCE(icsk->icsk_ca_ops); + if (ca_ops && ca_ops->get_info) + err = ca_ops->get_info(sk, ext, skb); + rcu_read_unlock(); + if (err < 0) + goto errout; + } out: nlmsg_end(skb, nlh); diff --git a/net/ipv4/ip_forward.c b/net/ipv4/ip_forward.c index 939992c456f3..3674484946a5 100644 --- a/net/ipv4/ip_forward.c +++ b/net/ipv4/ip_forward.c @@ -82,6 +82,9 @@ int ip_forward(struct sk_buff *skb) if (skb->pkt_type != PACKET_HOST) goto drop; + if (unlikely(skb->sk)) + goto drop; + if (skb_warn_if_lro(skb)) goto drop; diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 18e3a12eb1b2..8c5cd9efebbc 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -520,8 +520,10 @@ unsigned int tcp_poll(struct file *file, struct socket *sock, poll_table *wait) /* Race breaker. If space is freed after * wspace test but before the flags are set, - * IO signal will be lost. + * IO signal will be lost. Memory barrier + * pairs with the input side. */ + smp_mb__after_atomic(); if (sk_stream_is_writeable(sk)) mask |= POLLOUT | POLLWRNORM; } @@ -2595,6 +2597,7 @@ void tcp_get_info(const struct sock *sk, struct tcp_info *info) const struct tcp_sock *tp = tcp_sk(sk); const struct inet_connection_sock *icsk = inet_csk(sk); u32 now = tcp_time_stamp; + u32 rate; memset(info, 0, sizeof(*info)); @@ -2655,10 +2658,11 @@ void tcp_get_info(const struct sock *sk, struct tcp_info *info) info->tcpi_total_retrans = tp->total_retrans; - info->tcpi_pacing_rate = sk->sk_pacing_rate != ~0U ? - sk->sk_pacing_rate : ~0ULL; - info->tcpi_max_pacing_rate = sk->sk_max_pacing_rate != ~0U ? - sk->sk_max_pacing_rate : ~0ULL; + rate = READ_ONCE(sk->sk_pacing_rate); + info->tcpi_pacing_rate = rate != ~0U ? rate : ~0ULL; + + rate = READ_ONCE(sk->sk_max_pacing_rate); + info->tcpi_max_pacing_rate = rate != ~0U ? rate : ~0ULL; } EXPORT_SYMBOL_GPL(tcp_get_info); diff --git a/net/ipv4/tcp_dctcp.c b/net/ipv4/tcp_dctcp.c index b504371af742..4376016f7fa5 100644 --- a/net/ipv4/tcp_dctcp.c +++ b/net/ipv4/tcp_dctcp.c @@ -277,7 +277,7 @@ static void dctcp_cwnd_event(struct sock *sk, enum tcp_ca_event ev) } } -static void dctcp_get_info(struct sock *sk, u32 ext, struct sk_buff *skb) +static int dctcp_get_info(struct sock *sk, u32 ext, struct sk_buff *skb) { const struct dctcp *ca = inet_csk_ca(sk); @@ -297,8 +297,9 @@ static void dctcp_get_info(struct sock *sk, u32 ext, struct sk_buff *skb) info.dctcp_ab_tot = ca->acked_bytes_total; } - nla_put(skb, INET_DIAG_DCTCPINFO, sizeof(info), &info); + return nla_put(skb, INET_DIAG_DCTCPINFO, sizeof(info), &info); } + return 0; } static struct tcp_congestion_ops dctcp __read_mostly = { diff --git a/net/ipv4/tcp_illinois.c b/net/ipv4/tcp_illinois.c index 1d5a30a90adf..67476f085e48 100644 --- a/net/ipv4/tcp_illinois.c +++ b/net/ipv4/tcp_illinois.c @@ -300,8 +300,7 @@ static u32 tcp_illinois_ssthresh(struct sock *sk) } /* Extract info for Tcp socket info provided via netlink. */ -static void tcp_illinois_info(struct sock *sk, u32 ext, - struct sk_buff *skb) +static int tcp_illinois_info(struct sock *sk, u32 ext, struct sk_buff *skb) { const struct illinois *ca = inet_csk_ca(sk); @@ -318,8 +317,9 @@ static void tcp_illinois_info(struct sock *sk, u32 ext, do_div(t, info.tcpv_rttcnt); info.tcpv_rtt = t; } - nla_put(skb, INET_DIAG_VEGASINFO, sizeof(info), &info); + return nla_put(skb, INET_DIAG_VEGASINFO, sizeof(info), &info); } + return 0; } static struct tcp_congestion_ops tcp_illinois __read_mostly = { diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index a7ef679dd3ea..3a4d9b34bed4 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -4845,6 +4845,8 @@ static void tcp_check_space(struct sock *sk) { if (sock_flag(sk, SOCK_QUEUE_SHRUNK)) { sock_reset_flag(sk, SOCK_QUEUE_SHRUNK); + /* pairs with tcp_poll() */ + smp_mb__after_atomic(); if (sk->sk_socket && test_bit(SOCK_NOSPACE, &sk->sk_socket->flags)) tcp_new_space(sk); diff --git a/net/ipv4/tcp_vegas.c b/net/ipv4/tcp_vegas.c index a6afde666ab1..c71a1b8f7bde 100644 --- a/net/ipv4/tcp_vegas.c +++ b/net/ipv4/tcp_vegas.c @@ -286,7 +286,7 @@ static void tcp_vegas_cong_avoid(struct sock *sk, u32 ack, u32 acked) } /* Extract info for Tcp socket info provided via netlink. */ -void tcp_vegas_get_info(struct sock *sk, u32 ext, struct sk_buff *skb) +int tcp_vegas_get_info(struct sock *sk, u32 ext, struct sk_buff *skb) { const struct vegas *ca = inet_csk_ca(sk); if (ext & (1 << (INET_DIAG_VEGASINFO - 1))) { @@ -297,8 +297,9 @@ void tcp_vegas_get_info(struct sock *sk, u32 ext, struct sk_buff *skb) .tcpv_minrtt = ca->minRTT, }; - nla_put(skb, INET_DIAG_VEGASINFO, sizeof(info), &info); + return nla_put(skb, INET_DIAG_VEGASINFO, sizeof(info), &info); } + return 0; } EXPORT_SYMBOL_GPL(tcp_vegas_get_info); diff --git a/net/ipv4/tcp_vegas.h b/net/ipv4/tcp_vegas.h index 0531b99d8637..e8a6b33cc61d 100644 --- a/net/ipv4/tcp_vegas.h +++ b/net/ipv4/tcp_vegas.h @@ -19,6 +19,6 @@ void tcp_vegas_init(struct sock *sk); void tcp_vegas_state(struct sock *sk, u8 ca_state); void tcp_vegas_pkts_acked(struct sock *sk, u32 cnt, s32 rtt_us); void tcp_vegas_cwnd_event(struct sock *sk, enum tcp_ca_event event); -void tcp_vegas_get_info(struct sock *sk, u32 ext, struct sk_buff *skb); +int tcp_vegas_get_info(struct sock *sk, u32 ext, struct sk_buff *skb); #endif /* __TCP_VEGAS_H */ diff --git a/net/ipv4/tcp_westwood.c b/net/ipv4/tcp_westwood.c index bb63fba47d47..b3c57cceb990 100644 --- a/net/ipv4/tcp_westwood.c +++ b/net/ipv4/tcp_westwood.c @@ -256,8 +256,7 @@ static void tcp_westwood_event(struct sock *sk, enum tcp_ca_event event) } /* Extract info for Tcp socket info provided via netlink. */ -static void tcp_westwood_info(struct sock *sk, u32 ext, - struct sk_buff *skb) +static int tcp_westwood_info(struct sock *sk, u32 ext, struct sk_buff *skb) { const struct westwood *ca = inet_csk_ca(sk); @@ -268,8 +267,9 @@ static void tcp_westwood_info(struct sock *sk, u32 ext, .tcpv_minrtt = jiffies_to_usecs(ca->rtt_min), }; - nla_put(skb, INET_DIAG_VEGASINFO, sizeof(info), &info); + return nla_put(skb, INET_DIAG_VEGASINFO, sizeof(info), &info); } + return 0; } static struct tcp_congestion_ops tcp_westwood __read_mostly = { diff --git a/net/sched/act_bpf.c b/net/sched/act_bpf.c index 4d2cede17468..dc6a2d324bd8 100644 --- a/net/sched/act_bpf.c +++ b/net/sched/act_bpf.c @@ -38,6 +38,9 @@ static int tcf_bpf(struct sk_buff *skb, const struct tc_action *act, struct tcf_bpf *prog = act->priv; int action, filter_res; + if (unlikely(!skb_mac_header_was_set(skb))) + return TC_ACT_UNSPEC; + spin_lock(&prog->tcf_lock); prog->tcf_tm.lastuse = jiffies; diff --git a/net/sched/act_mirred.c b/net/sched/act_mirred.c index 5953517ec059..3f63ceac8e01 100644 --- a/net/sched/act_mirred.c +++ b/net/sched/act_mirred.c @@ -157,7 +157,7 @@ static int tcf_mirred(struct sk_buff *skb, const struct tc_action *a, if (!(at & AT_EGRESS)) { if (m->tcfm_ok_push) - skb_push(skb2, skb2->dev->hard_header_len); + skb_push(skb2, skb->mac_len); } /* mirror is always swallowed */ diff --git a/net/sched/cls_bpf.c b/net/sched/cls_bpf.c index 5c4171c5d2bd..91bd9c19471d 100644 --- a/net/sched/cls_bpf.c +++ b/net/sched/cls_bpf.c @@ -66,6 +66,9 @@ static int cls_bpf_classify(struct sk_buff *skb, const struct tcf_proto *tp, struct cls_bpf_prog *prog; int ret = -1; + if (unlikely(!skb_mac_header_was_set(skb))) + return -1; + /* Needed here for accessing maps. */ rcu_read_lock(); list_for_each_entry_rcu(prog, &head->plist, link) { |