From mboxrd@z Thu Jan 1 00:00:00 1970 From: Calvin Owens Subject: [PATCH] net: Use SK_MEM_QUANTUM as minimum for tcp/udp rmem/wmem Date: Tue, 11 Aug 2015 21:54:20 -0700 Message-ID: <20150812045420.GA3908557@mail.thefacebook.com> References: <1438806414-751067-1-git-send-email-calvinowens@fb.com> <20150809.224114.818332231954008575.davem@davemloft.net> <20150811033406.GA1136819@mail.thefacebook.com> <20150810.204630.1903301700926701432.davem@davemloft.net> Mime-Version: 1.0 Content-Type: text/plain; charset="utf-8" Cc: , , , , , , , To: David Miller Return-path: Content-Disposition: inline In-Reply-To: <20150810.204630.1903301700926701432.davem@davemloft.net> Sender: linux-kernel-owner@vger.kernel.org List-Id: netdev.vger.kernel.org Commit 8133534c760d4083 ("net: limit tcp/udp rmem/wmem to SOCK_{RCV,SND}BUF_MIN") modified four sysctls to enforce that the values written to them are not less than SOCK_MIN_{RCV,SND}BUF. That change causes 4096 (or SK_MEM_QUANTUM) to no longer be accepted as a valid value for 'min' in tcp_wmem and udp_wmem_min. 4096 has been the default for both of those sysctls for a long time, and unfortunately seems to be an extremely popular setting. This change breaks a large number of sysctl configurations at FB. That commit referred to b1cb59cf2efe7971 ("net: sysctl_net_core: check SNDBUF and RCVBUF for min length"), which choose to use the SOCK_MIN constants as the lower limits to avoid nasty bugs. But AFAICS, a limit of SOCK_MIN_SNDBUF isn't necessary to do that: the BUG_ON cited in the commit message seems to have happened because unix_stream_sendmsg() expects a minimum of a full page (ie SK_MEM_QUANTUM) and the math broke, not because it had less than SOCK_MIN_SNDBUF allocated. Nothing seems to assume that it has at least SOCK_MIN_SNDBUF to play with, so I think enforcing a minimum of SK_MEM_QUANTUM avoids the sort of bugs 8133534c was trying to avoid, and it does so without breaking anybody's sysctl configurations. Fixes: 8133534c760d4083 ("net: limit tcp/udp rmem/wmem to SOCK_MIN...") Signed-off-by: Calvin Owens --- net/ipv4/sysctl_net_ipv4.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 433231c..a214b6a 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -41,8 +41,7 @@ static int tcp_syn_retries_min = 1; static int tcp_syn_retries_max = MAX_TCP_SYNCNT; static int ip_ping_group_range_min[] = { 0, 0 }; static int ip_ping_group_range_max[] = { GID_T_MAX, GID_T_MAX }; -static int min_sndbuf = SOCK_MIN_SNDBUF; -static int min_rcvbuf = SOCK_MIN_RCVBUF; +static int min_buf = SK_MEM_QUANTUM; /* Update system visible IP port range */ static void set_local_port_range(struct net *net, int range[2]) @@ -530,7 +529,7 @@ static struct ctl_table ipv4_table[] = { .maxlen = sizeof(sysctl_tcp_wmem), .mode = 0644, .proc_handler = proc_dointvec_minmax, - .extra1 = &min_sndbuf, + .extra1 = &min_buf, }, { .procname = "tcp_notsent_lowat", @@ -545,7 +544,7 @@ static struct ctl_table ipv4_table[] = { .maxlen = sizeof(sysctl_tcp_rmem), .mode = 0644, .proc_handler = proc_dointvec_minmax, - .extra1 = &min_rcvbuf, + .extra1 = &min_buf, }, { .procname = "tcp_app_win", @@ -758,7 +757,7 @@ static struct ctl_table ipv4_table[] = { .maxlen = sizeof(sysctl_udp_rmem_min), .mode = 0644, .proc_handler = proc_dointvec_minmax, - .extra1 = &min_rcvbuf, + .extra1 = &min_buf, }, { .procname = "udp_wmem_min", @@ -766,7 +765,7 @@ static struct ctl_table ipv4_table[] = { .maxlen = sizeof(sysctl_udp_wmem_min), .mode = 0644, .proc_handler = proc_dointvec_minmax, - .extra1 = &min_sndbuf, + .extra1 = &min_buf, }, { } }; -- 1.8.1