* [PATCH bpf-next 0/2] Split bpf_sk_lookup remote_port field @ 2022-02-07 13:14 Jakub Sitnicki 2022-02-07 13:14 ` [PATCH bpf-next 1/2] bpf: Make remote_port field in struct bpf_sk_lookup 16-bit wide Jakub Sitnicki 2022-02-07 13:14 ` [PATCH bpf-next 2/2] selftests/bpf: Cover 4-byte load from remote_port in bpf_sk_lookup Jakub Sitnicki 0 siblings, 2 replies; 9+ messages in thread From: Jakub Sitnicki @ 2022-02-07 13:14 UTC (permalink / raw) To: bpf Cc: netdev, Alexei Starovoitov, Daniel Borkmann, Andrii Nakryiko, kernel-team Following the recent split-up of the bpf_sock dst_port field, apply the same to technique to the bpf_sk_lookup remote_port field to make uAPI more user friendly. Jakub Sitnicki (2): bpf: Make remote_port field in struct bpf_sk_lookup 16-bit wide selftests/bpf: Cover 4-byte load from remote_port in bpf_sk_lookup include/uapi/linux/bpf.h | 3 ++- net/core/filter.c | 3 ++- tools/include/uapi/linux/bpf.h | 3 ++- tools/testing/selftests/bpf/progs/test_sk_lookup.c | 6 ++++++ 4 files changed, 12 insertions(+), 3 deletions(-) -- 2.31.1 ^ permalink raw reply [flat|nested] 9+ messages in thread
* [PATCH bpf-next 1/2] bpf: Make remote_port field in struct bpf_sk_lookup 16-bit wide 2022-02-07 13:14 [PATCH bpf-next 0/2] Split bpf_sk_lookup remote_port field Jakub Sitnicki @ 2022-02-07 13:14 ` Jakub Sitnicki 2022-02-07 19:14 ` Yonghong Song 2022-02-07 22:04 ` kernel test robot 2022-02-07 13:14 ` [PATCH bpf-next 2/2] selftests/bpf: Cover 4-byte load from remote_port in bpf_sk_lookup Jakub Sitnicki 1 sibling, 2 replies; 9+ messages in thread From: Jakub Sitnicki @ 2022-02-07 13:14 UTC (permalink / raw) To: bpf Cc: netdev, Alexei Starovoitov, Daniel Borkmann, Andrii Nakryiko, kernel-team remote_port is another case of a BPF context field documented as a 32-bit value in network byte order for which the BPF context access converter generates a load of a zero-padded 16-bit integer in network byte order. First such case was dst_port in bpf_sock which got addressed in commit 4421a582718a ("bpf: Make dst_port field in struct bpf_sock 16-bit wide"). Loading 4-bytes from the remote_port offset and converting the value with bpf_ntohl() leads to surprising results, as the expected value is shifted by 16 bits. Reduce the confusion by splitting the field in two - a 16-bit field holding a big-endian integer, and a 16-bit zero-padding anonymous field that follows it. Suggested-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: Jakub Sitnicki <jakub@cloudflare.com> --- include/uapi/linux/bpf.h | 3 ++- net/core/filter.c | 3 ++- 2 files changed, 4 insertions(+), 2 deletions(-) diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index a7f0ddedac1f..afe3d0d7f5f2 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -6453,7 +6453,8 @@ struct bpf_sk_lookup { __u32 protocol; /* IP protocol (IPPROTO_TCP, IPPROTO_UDP) */ __u32 remote_ip4; /* Network byte order */ __u32 remote_ip6[4]; /* Network byte order */ - __u32 remote_port; /* Network byte order */ + __be16 remote_port; /* Network byte order */ + __u16 :16; /* Zero padding */ __u32 local_ip4; /* Network byte order */ __u32 local_ip6[4]; /* Network byte order */ __u32 local_port; /* Host byte order */ diff --git a/net/core/filter.c b/net/core/filter.c index 99a05199a806..83f06d3b2c52 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -10843,7 +10843,8 @@ static bool sk_lookup_is_valid_access(int off, int size, case bpf_ctx_range(struct bpf_sk_lookup, local_ip4): case bpf_ctx_range_till(struct bpf_sk_lookup, remote_ip6[0], remote_ip6[3]): case bpf_ctx_range_till(struct bpf_sk_lookup, local_ip6[0], local_ip6[3]): - case bpf_ctx_range(struct bpf_sk_lookup, remote_port): + case offsetof(struct bpf_sk_lookup, remote_port) ... + offsetof(struct bpf_sk_lookup, local_ip4) - 1: case bpf_ctx_range(struct bpf_sk_lookup, local_port): case bpf_ctx_range(struct bpf_sk_lookup, ingress_ifindex): bpf_ctx_record_field_size(info, sizeof(__u32)); -- 2.31.1 ^ permalink raw reply related [flat|nested] 9+ messages in thread
* Re: [PATCH bpf-next 1/2] bpf: Make remote_port field in struct bpf_sk_lookup 16-bit wide 2022-02-07 13:14 ` [PATCH bpf-next 1/2] bpf: Make remote_port field in struct bpf_sk_lookup 16-bit wide Jakub Sitnicki @ 2022-02-07 19:14 ` Yonghong Song 2022-02-07 22:04 ` kernel test robot 1 sibling, 0 replies; 9+ messages in thread From: Yonghong Song @ 2022-02-07 19:14 UTC (permalink / raw) To: Jakub Sitnicki, bpf Cc: netdev, Alexei Starovoitov, Daniel Borkmann, Andrii Nakryiko, kernel-team On 2/7/22 5:14 AM, Jakub Sitnicki wrote: > remote_port is another case of a BPF context field documented as a 32-bit > value in network byte order for which the BPF context access converter > generates a load of a zero-padded 16-bit integer in network byte order. > > First such case was dst_port in bpf_sock which got addressed in commit > 4421a582718a ("bpf: Make dst_port field in struct bpf_sock 16-bit wide"). > > Loading 4-bytes from the remote_port offset and converting the value with > bpf_ntohl() leads to surprising results, as the expected value is shifted > by 16 bits. > > Reduce the confusion by splitting the field in two - a 16-bit field holding > a big-endian integer, and a 16-bit zero-padding anonymous field that > follows it. > > Suggested-by: Alexei Starovoitov <ast@kernel.org> > Signed-off-by: Jakub Sitnicki <jakub@cloudflare.com> Acked-by: Yonghong Song <yhs@fb.com> ^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [PATCH bpf-next 1/2] bpf: Make remote_port field in struct bpf_sk_lookup 16-bit wide 2022-02-07 13:14 ` [PATCH bpf-next 1/2] bpf: Make remote_port field in struct bpf_sk_lookup 16-bit wide Jakub Sitnicki 2022-02-07 19:14 ` Yonghong Song @ 2022-02-07 22:04 ` kernel test robot 2022-02-08 19:28 ` Alexei Starovoitov 1 sibling, 1 reply; 9+ messages in thread From: kernel test robot @ 2022-02-07 22:04 UTC (permalink / raw) To: Jakub Sitnicki, bpf Cc: kbuild-all, netdev, Alexei Starovoitov, Daniel Borkmann, Andrii Nakryiko, kernel-team Hi Jakub, I love your patch! Perhaps something to improve: [auto build test WARNING on bpf-next/master] url: https://github.com/0day-ci/linux/commits/Jakub-Sitnicki/Split-bpf_sk_lookup-remote_port-field/20220207-215137 base: https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next.git master config: sparc-randconfig-s031-20220207 (https://download.01.org/0day-ci/archive/20220208/202202080631.n8UjqRXy-lkp@intel.com/config) compiler: sparc64-linux-gcc (GCC) 11.2.0 reproduce: wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross chmod +x ~/bin/make.cross # apt-get install sparse # sparse version: v0.6.4-dirty # https://github.com/0day-ci/linux/commit/b859a90e4c32a55e71d2731dd8dae96d7ad1defe git remote add linux-review https://github.com/0day-ci/linux git fetch --no-tags linux-review Jakub-Sitnicki/Split-bpf_sk_lookup-remote_port-field/20220207-215137 git checkout b859a90e4c32a55e71d2731dd8dae96d7ad1defe # save the config file to linux build tree mkdir build_dir COMPILER_INSTALL_PATH=$HOME/0day COMPILER=gcc-11.2.0 make.cross C=1 CF='-fdiagnostic-prefix -D__CHECK_ENDIAN__' O=build_dir ARCH=sparc SHELL=/bin/bash net/bpf/ If you fix the issue, kindly add following tag as appropriate Reported-by: kernel test robot <lkp@intel.com> sparse warnings: (new ones prefixed by >>) >> net/bpf/test_run.c:1149:55: sparse: sparse: restricted __be16 degrades to integer vim +1149 net/bpf/test_run.c 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1111 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1112 int bpf_prog_test_run_sk_lookup(struct bpf_prog *prog, const union bpf_attr *kattr, 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1113 union bpf_attr __user *uattr) 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1114 { 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1115 struct bpf_test_timer t = { NO_PREEMPT }; 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1116 struct bpf_prog_array *progs = NULL; 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1117 struct bpf_sk_lookup_kern ctx = {}; 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1118 u32 repeat = kattr->test.repeat; 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1119 struct bpf_sk_lookup *user_ctx; 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1120 u32 retval, duration; 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1121 int ret = -EINVAL; 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1122 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1123 if (prog->type != BPF_PROG_TYPE_SK_LOOKUP) 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1124 return -EINVAL; 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1125 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1126 if (kattr->test.flags || kattr->test.cpu) 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1127 return -EINVAL; 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1128 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1129 if (kattr->test.data_in || kattr->test.data_size_in || kattr->test.data_out || 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1130 kattr->test.data_size_out) 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1131 return -EINVAL; 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1132 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1133 if (!repeat) 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1134 repeat = 1; 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1135 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1136 user_ctx = bpf_ctx_init(kattr, sizeof(*user_ctx)); 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1137 if (IS_ERR(user_ctx)) 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1138 return PTR_ERR(user_ctx); 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1139 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1140 if (!user_ctx) 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1141 return -EINVAL; 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1142 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1143 if (user_ctx->sk) 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1144 goto out; 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1145 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1146 if (!range_is_zero(user_ctx, offsetofend(typeof(*user_ctx), local_port), sizeof(*user_ctx))) 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1147 goto out; 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1148 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 @1149 if (user_ctx->local_port > U16_MAX || user_ctx->remote_port > U16_MAX) { 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1150 ret = -ERANGE; 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1151 goto out; 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1152 } 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1153 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1154 ctx.family = (u16)user_ctx->family; 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1155 ctx.protocol = (u16)user_ctx->protocol; 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1156 ctx.dport = (u16)user_ctx->local_port; 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1157 ctx.sport = (__force __be16)user_ctx->remote_port; 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1158 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1159 switch (ctx.family) { 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1160 case AF_INET: 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1161 ctx.v4.daddr = (__force __be32)user_ctx->local_ip4; 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1162 ctx.v4.saddr = (__force __be32)user_ctx->remote_ip4; 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1163 break; 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1164 --- 0-DAY CI Kernel Test Service, Intel Corporation https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org ^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [PATCH bpf-next 1/2] bpf: Make remote_port field in struct bpf_sk_lookup 16-bit wide 2022-02-07 22:04 ` kernel test robot @ 2022-02-08 19:28 ` Alexei Starovoitov 2022-02-08 20:42 ` Jakub Sitnicki 0 siblings, 1 reply; 9+ messages in thread From: Alexei Starovoitov @ 2022-02-08 19:28 UTC (permalink / raw) To: kernel test robot Cc: Jakub Sitnicki, bpf, kbuild-all, Network Development, Alexei Starovoitov, Daniel Borkmann, Andrii Nakryiko, kernel-team On Mon, Feb 7, 2022 at 2:05 PM kernel test robot <lkp@intel.com> wrote: > 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1148 > 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 @1149 if (user_ctx->local_port > U16_MAX || user_ctx->remote_port > U16_MAX) { Jakub, are you planning to respin and remove that check? ^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [PATCH bpf-next 1/2] bpf: Make remote_port field in struct bpf_sk_lookup 16-bit wide 2022-02-08 19:28 ` Alexei Starovoitov @ 2022-02-08 20:42 ` Jakub Sitnicki 2022-02-08 20:58 ` Alexei Starovoitov 0 siblings, 1 reply; 9+ messages in thread From: Jakub Sitnicki @ 2022-02-08 20:42 UTC (permalink / raw) To: Alexei Starovoitov Cc: kernel test robot, bpf, kbuild-all, Network Development, Alexei Starovoitov, Daniel Borkmann, Andrii Nakryiko, kernel-team On Tue, Feb 08, 2022 at 08:28 PM CET, Alexei Starovoitov wrote: > On Mon, Feb 7, 2022 at 2:05 PM kernel test robot <lkp@intel.com> wrote: >> 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1148 >> 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 @1149 if (user_ctx->local_port > U16_MAX || user_ctx->remote_port > U16_MAX) { > > Jakub, > are you planning to respin and remove that check? Yes, certainly. Just didn't get to it today. Can either respin the series or send a follow-up. Whatever works. ^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [PATCH bpf-next 1/2] bpf: Make remote_port field in struct bpf_sk_lookup 16-bit wide 2022-02-08 20:42 ` Jakub Sitnicki @ 2022-02-08 20:58 ` Alexei Starovoitov 0 siblings, 0 replies; 9+ messages in thread From: Alexei Starovoitov @ 2022-02-08 20:58 UTC (permalink / raw) To: Jakub Sitnicki Cc: kernel test robot, bpf, kbuild-all, Network Development, Alexei Starovoitov, Daniel Borkmann, Andrii Nakryiko, kernel-team On Tue, Feb 8, 2022 at 12:42 PM Jakub Sitnicki <jakub@cloudflare.com> wrote: > > On Tue, Feb 08, 2022 at 08:28 PM CET, Alexei Starovoitov wrote: > > On Mon, Feb 7, 2022 at 2:05 PM kernel test robot <lkp@intel.com> wrote: > >> 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 1148 > >> 7c32e8f8bc33a5f Lorenz Bauer 2021-03-03 @1149 if (user_ctx->local_port > U16_MAX || user_ctx->remote_port > U16_MAX) { > > > > Jakub, > > are you planning to respin and remove that check? > > Yes, certainly. Just didn't get to it today. > > Can either respin the series or send a follow-up. Whatever works. I think respin is better. ^ permalink raw reply [flat|nested] 9+ messages in thread
* [PATCH bpf-next 2/2] selftests/bpf: Cover 4-byte load from remote_port in bpf_sk_lookup 2022-02-07 13:14 [PATCH bpf-next 0/2] Split bpf_sk_lookup remote_port field Jakub Sitnicki 2022-02-07 13:14 ` [PATCH bpf-next 1/2] bpf: Make remote_port field in struct bpf_sk_lookup 16-bit wide Jakub Sitnicki @ 2022-02-07 13:14 ` Jakub Sitnicki 2022-02-07 19:20 ` Yonghong Song 1 sibling, 1 reply; 9+ messages in thread From: Jakub Sitnicki @ 2022-02-07 13:14 UTC (permalink / raw) To: bpf Cc: netdev, Alexei Starovoitov, Daniel Borkmann, Andrii Nakryiko, kernel-team Extend the context access tests for sk_lookup prog to cover the surprising case of a 4-byte load from the remote_port field, where the expected value is actually shifted by 16 bits. Signed-off-by: Jakub Sitnicki <jakub@cloudflare.com> --- tools/include/uapi/linux/bpf.h | 3 ++- tools/testing/selftests/bpf/progs/test_sk_lookup.c | 6 ++++++ 2 files changed, 8 insertions(+), 1 deletion(-) diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h index a7f0ddedac1f..afe3d0d7f5f2 100644 --- a/tools/include/uapi/linux/bpf.h +++ b/tools/include/uapi/linux/bpf.h @@ -6453,7 +6453,8 @@ struct bpf_sk_lookup { __u32 protocol; /* IP protocol (IPPROTO_TCP, IPPROTO_UDP) */ __u32 remote_ip4; /* Network byte order */ __u32 remote_ip6[4]; /* Network byte order */ - __u32 remote_port; /* Network byte order */ + __be16 remote_port; /* Network byte order */ + __u16 :16; /* Zero padding */ __u32 local_ip4; /* Network byte order */ __u32 local_ip6[4]; /* Network byte order */ __u32 local_port; /* Host byte order */ diff --git a/tools/testing/selftests/bpf/progs/test_sk_lookup.c b/tools/testing/selftests/bpf/progs/test_sk_lookup.c index 83b0aaa52ef7..bf5b7caefdd0 100644 --- a/tools/testing/selftests/bpf/progs/test_sk_lookup.c +++ b/tools/testing/selftests/bpf/progs/test_sk_lookup.c @@ -392,6 +392,7 @@ int ctx_narrow_access(struct bpf_sk_lookup *ctx) { struct bpf_sock *sk; int err, family; + __u32 val_u32; bool v4; v4 = (ctx->family == AF_INET); @@ -418,6 +419,11 @@ int ctx_narrow_access(struct bpf_sk_lookup *ctx) if (LSW(ctx->remote_port, 0) != SRC_PORT) return SK_DROP; + /* Load from remote_port field with zero padding (backward compatibility) */ + val_u32 = *(__u32 *)&ctx->remote_port; + if (val_u32 != bpf_htonl(bpf_ntohs(SRC_PORT) << 16)) + return SK_DROP; + /* Narrow loads from local_port field. Expect DST_PORT. */ if (LSB(ctx->local_port, 0) != ((DST_PORT >> 0) & 0xff) || LSB(ctx->local_port, 1) != ((DST_PORT >> 8) & 0xff) || -- 2.31.1 ^ permalink raw reply related [flat|nested] 9+ messages in thread
* Re: [PATCH bpf-next 2/2] selftests/bpf: Cover 4-byte load from remote_port in bpf_sk_lookup 2022-02-07 13:14 ` [PATCH bpf-next 2/2] selftests/bpf: Cover 4-byte load from remote_port in bpf_sk_lookup Jakub Sitnicki @ 2022-02-07 19:20 ` Yonghong Song 0 siblings, 0 replies; 9+ messages in thread From: Yonghong Song @ 2022-02-07 19:20 UTC (permalink / raw) To: Jakub Sitnicki, bpf Cc: netdev, Alexei Starovoitov, Daniel Borkmann, Andrii Nakryiko, kernel-team On 2/7/22 5:14 AM, Jakub Sitnicki wrote: > Extend the context access tests for sk_lookup prog to cover the surprising > case of a 4-byte load from the remote_port field, where the expected value > is actually shifted by 16 bits. > > Signed-off-by: Jakub Sitnicki <jakub@cloudflare.com> Acked-by: Yonghong Song <yhs@fb.com> ^ permalink raw reply [flat|nested] 9+ messages in thread
end of thread, other threads:[~2022-02-08 22:21 UTC | newest] Thread overview: 9+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2022-02-07 13:14 [PATCH bpf-next 0/2] Split bpf_sk_lookup remote_port field Jakub Sitnicki 2022-02-07 13:14 ` [PATCH bpf-next 1/2] bpf: Make remote_port field in struct bpf_sk_lookup 16-bit wide Jakub Sitnicki 2022-02-07 19:14 ` Yonghong Song 2022-02-07 22:04 ` kernel test robot 2022-02-08 19:28 ` Alexei Starovoitov 2022-02-08 20:42 ` Jakub Sitnicki 2022-02-08 20:58 ` Alexei Starovoitov 2022-02-07 13:14 ` [PATCH bpf-next 2/2] selftests/bpf: Cover 4-byte load from remote_port in bpf_sk_lookup Jakub Sitnicki 2022-02-07 19:20 ` Yonghong Song
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).