linux/net/smc/smc_sysctl.c

154 lines
3.9 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0
/*
* Shared Memory Communications over RDMA (SMC-R) and RoCE
*
* smc_sysctl.c: sysctl interface to SMC subsystem.
*
* Copyright (c) 2022, Alibaba Inc.
*
* Author: Tony Lu <tonylu@linux.alibaba.com>
*
*/
#include <linux/init.h>
#include <linux/sysctl.h>
#include <net/net_namespace.h>
#include "smc.h"
#include "smc_core.h"
#include "smc_llc.h"
#include "smc_sysctl.h"
static int min_sndbuf = SMC_BUF_MIN_SIZE;
static int min_rcvbuf = SMC_BUF_MIN_SIZE;
net/smc: Fix setsockopt and sysctl to specify same buffer size again Commit 0227f058aa29 ("net/smc: Unbind r/w buffer size from clcsock and make them tunable") introduced the net.smc.rmem and net.smc.wmem sysctls to specify the size of buffers to be used for SMC type connections. This created a regression for users that specified the buffer size via setsockopt() as the effective buffer size was now doubled. Re-introduce the division by 2 in the SMC buffer create code and level this out by duplicating the net.smc.[rw]mem values used for initializing sk_rcvbuf/sk_sndbuf at socket creation time. This gives users of both methods (setsockopt or sysctl) the effective buffer size that they expect. Initialize net.smc.[rw]mem from its own constant of 64kB, respectively. Internal performance tests show that this value is a good compromise between throughput/latency and memory consumption. Also, this decouples it from any tuning that was done to net.ipv4.tcp_[rw]mem[1] before the module for SMC protocol was loaded. Check that no more than INT_MAX / 2 is assigned to net.smc.[rw]mem, in order to avoid any overflow condition when that is doubled for use in sk_sndbuf or sk_rcvbuf. While at it, drop the confusing sk_buf_size variable from __smc_buf_create and name "compressed" buffer size variables more consistently. Background: Before the commit mentioned above, SMC's buffer allocator in __smc_buf_create() always used half of the sockets' sk_rcvbuf/sk_sndbuf value as initial value to search for appropriate buffers. If the search resorted to using a bigger buffer when all buffers of the specified size were busy, the duplicate of the used effective buffer size is stored back to sk_rcvbuf/sk_sndbuf. When available, buffers of exactly the size that a user had specified as input to setsockopt() were used, despite setsockopt()'s documentation in "man 7 socket" talking of a mandatory duplication: [...] SO_SNDBUF Sets or gets the maximum socket send buffer in bytes. The kernel doubles this value (to allow space for book‐ keeping overhead) when it is set using setsockopt(2), and this doubled value is returned by getsockopt(2). The default value is set by the /proc/sys/net/core/wmem_default file and the maximum allowed value is set by the /proc/sys/net/core/wmem_max file. The minimum (doubled) value for this option is 2048. [...] Fixes: 0227f058aa29 ("net/smc: Unbind r/w buffer size from clcsock and make them tunable") Co-developed-by: Jan Karcher <jaka@linux.ibm.com> Signed-off-by: Jan Karcher <jaka@linux.ibm.com> Reviewed-by: Wenjia Zhang <wenjia@linux.ibm.com> Reviewed-by: Tony Lu <tonylu@linux.alibaba.com> Signed-off-by: Gerd Bayer <gbayer@linux.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2023-08-04 17:06:23 +00:00
static int max_sndbuf = INT_MAX / 2;
static int max_rcvbuf = INT_MAX / 2;
static const int net_smc_wmem_init = (64 * 1024);
static const int net_smc_rmem_init = (64 * 1024);
static int links_per_lgr_min = SMC_LINKS_ADD_LNK_MIN;
static int links_per_lgr_max = SMC_LINKS_ADD_LNK_MAX;
static int conns_per_lgr_min = SMC_CONN_PER_LGR_MIN;
static int conns_per_lgr_max = SMC_CONN_PER_LGR_MAX;
static struct ctl_table smc_table[] = {
{
.procname = "autocorking_size",
.data = &init_net.smc.sysctl_autocorking_size,
.maxlen = sizeof(unsigned int),
.mode = 0644,
.proc_handler = proc_douintvec,
},
{
.procname = "smcr_buf_type",
.data = &init_net.smc.sysctl_smcr_buf_type,
.maxlen = sizeof(unsigned int),
.mode = 0644,
.proc_handler = proc_douintvec_minmax,
.extra1 = SYSCTL_ZERO,
.extra2 = SYSCTL_TWO,
},
{
.procname = "smcr_testlink_time",
.data = &init_net.smc.sysctl_smcr_testlink_time,
.maxlen = sizeof(int),
.mode = 0644,
.proc_handler = proc_dointvec_jiffies,
},
{
.procname = "wmem",
.data = &init_net.smc.sysctl_wmem,
.maxlen = sizeof(int),
.mode = 0644,
.proc_handler = proc_dointvec_minmax,
.extra1 = &min_sndbuf,
net/smc: Fix setsockopt and sysctl to specify same buffer size again Commit 0227f058aa29 ("net/smc: Unbind r/w buffer size from clcsock and make them tunable") introduced the net.smc.rmem and net.smc.wmem sysctls to specify the size of buffers to be used for SMC type connections. This created a regression for users that specified the buffer size via setsockopt() as the effective buffer size was now doubled. Re-introduce the division by 2 in the SMC buffer create code and level this out by duplicating the net.smc.[rw]mem values used for initializing sk_rcvbuf/sk_sndbuf at socket creation time. This gives users of both methods (setsockopt or sysctl) the effective buffer size that they expect. Initialize net.smc.[rw]mem from its own constant of 64kB, respectively. Internal performance tests show that this value is a good compromise between throughput/latency and memory consumption. Also, this decouples it from any tuning that was done to net.ipv4.tcp_[rw]mem[1] before the module for SMC protocol was loaded. Check that no more than INT_MAX / 2 is assigned to net.smc.[rw]mem, in order to avoid any overflow condition when that is doubled for use in sk_sndbuf or sk_rcvbuf. While at it, drop the confusing sk_buf_size variable from __smc_buf_create and name "compressed" buffer size variables more consistently. Background: Before the commit mentioned above, SMC's buffer allocator in __smc_buf_create() always used half of the sockets' sk_rcvbuf/sk_sndbuf value as initial value to search for appropriate buffers. If the search resorted to using a bigger buffer when all buffers of the specified size were busy, the duplicate of the used effective buffer size is stored back to sk_rcvbuf/sk_sndbuf. When available, buffers of exactly the size that a user had specified as input to setsockopt() were used, despite setsockopt()'s documentation in "man 7 socket" talking of a mandatory duplication: [...] SO_SNDBUF Sets or gets the maximum socket send buffer in bytes. The kernel doubles this value (to allow space for book‐ keeping overhead) when it is set using setsockopt(2), and this doubled value is returned by getsockopt(2). The default value is set by the /proc/sys/net/core/wmem_default file and the maximum allowed value is set by the /proc/sys/net/core/wmem_max file. The minimum (doubled) value for this option is 2048. [...] Fixes: 0227f058aa29 ("net/smc: Unbind r/w buffer size from clcsock and make them tunable") Co-developed-by: Jan Karcher <jaka@linux.ibm.com> Signed-off-by: Jan Karcher <jaka@linux.ibm.com> Reviewed-by: Wenjia Zhang <wenjia@linux.ibm.com> Reviewed-by: Tony Lu <tonylu@linux.alibaba.com> Signed-off-by: Gerd Bayer <gbayer@linux.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2023-08-04 17:06:23 +00:00
.extra2 = &max_sndbuf,
},
{
.procname = "rmem",
.data = &init_net.smc.sysctl_rmem,
.maxlen = sizeof(int),
.mode = 0644,
.proc_handler = proc_dointvec_minmax,
.extra1 = &min_rcvbuf,
net/smc: Fix setsockopt and sysctl to specify same buffer size again Commit 0227f058aa29 ("net/smc: Unbind r/w buffer size from clcsock and make them tunable") introduced the net.smc.rmem and net.smc.wmem sysctls to specify the size of buffers to be used for SMC type connections. This created a regression for users that specified the buffer size via setsockopt() as the effective buffer size was now doubled. Re-introduce the division by 2 in the SMC buffer create code and level this out by duplicating the net.smc.[rw]mem values used for initializing sk_rcvbuf/sk_sndbuf at socket creation time. This gives users of both methods (setsockopt or sysctl) the effective buffer size that they expect. Initialize net.smc.[rw]mem from its own constant of 64kB, respectively. Internal performance tests show that this value is a good compromise between throughput/latency and memory consumption. Also, this decouples it from any tuning that was done to net.ipv4.tcp_[rw]mem[1] before the module for SMC protocol was loaded. Check that no more than INT_MAX / 2 is assigned to net.smc.[rw]mem, in order to avoid any overflow condition when that is doubled for use in sk_sndbuf or sk_rcvbuf. While at it, drop the confusing sk_buf_size variable from __smc_buf_create and name "compressed" buffer size variables more consistently. Background: Before the commit mentioned above, SMC's buffer allocator in __smc_buf_create() always used half of the sockets' sk_rcvbuf/sk_sndbuf value as initial value to search for appropriate buffers. If the search resorted to using a bigger buffer when all buffers of the specified size were busy, the duplicate of the used effective buffer size is stored back to sk_rcvbuf/sk_sndbuf. When available, buffers of exactly the size that a user had specified as input to setsockopt() were used, despite setsockopt()'s documentation in "man 7 socket" talking of a mandatory duplication: [...] SO_SNDBUF Sets or gets the maximum socket send buffer in bytes. The kernel doubles this value (to allow space for book‐ keeping overhead) when it is set using setsockopt(2), and this doubled value is returned by getsockopt(2). The default value is set by the /proc/sys/net/core/wmem_default file and the maximum allowed value is set by the /proc/sys/net/core/wmem_max file. The minimum (doubled) value for this option is 2048. [...] Fixes: 0227f058aa29 ("net/smc: Unbind r/w buffer size from clcsock and make them tunable") Co-developed-by: Jan Karcher <jaka@linux.ibm.com> Signed-off-by: Jan Karcher <jaka@linux.ibm.com> Reviewed-by: Wenjia Zhang <wenjia@linux.ibm.com> Reviewed-by: Tony Lu <tonylu@linux.alibaba.com> Signed-off-by: Gerd Bayer <gbayer@linux.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2023-08-04 17:06:23 +00:00
.extra2 = &max_rcvbuf,
},
{
.procname = "smcr_max_links_per_lgr",
.data = &init_net.smc.sysctl_max_links_per_lgr,
.maxlen = sizeof(int),
.mode = 0644,
.proc_handler = proc_dointvec_minmax,
.extra1 = &links_per_lgr_min,
.extra2 = &links_per_lgr_max,
},
{
.procname = "smcr_max_conns_per_lgr",
.data = &init_net.smc.sysctl_max_conns_per_lgr,
.maxlen = sizeof(int),
.mode = 0644,
.proc_handler = proc_dointvec_minmax,
.extra1 = &conns_per_lgr_min,
.extra2 = &conns_per_lgr_max,
},
{
.procname = "limit_smc_hs",
.data = &init_net.smc.limit_smc_hs,
.maxlen = sizeof(int),
.mode = 0644,
.proc_handler = proc_dointvec_minmax,
.extra1 = SYSCTL_ZERO,
.extra2 = SYSCTL_ONE,
},
};
net/smc: fix compile warning for smc_sysctl kernel test robot reports multiple warning for smc_sysctl: In file included from net/smc/smc_sysctl.c:17: >> net/smc/smc_sysctl.h:23:5: warning: no previous prototype \ for function 'smc_sysctl_init' [-Wmissing-prototypes] int smc_sysctl_init(void) ^ and >> WARNING: modpost: vmlinux.o(.text+0x12ced2d): Section mismatch \ in reference from the function smc_sysctl_exit() to the variable .init.data:smc_sysctl_ops The function smc_sysctl_exit() references the variable __initdata smc_sysctl_ops. This is often because smc_sysctl_exit lacks a __initdata annotation or the annotation of smc_sysctl_ops is wrong. and net/smc/smc_sysctl.c: In function 'smc_sysctl_init_net': net/smc/smc_sysctl.c:47:17: error: 'struct netns_smc' has no member named 'smc_hdr' 47 | net->smc.smc_hdr = register_net_sysctl(net, "net/smc", table); Since we don't need global sysctl initialization. To make things clean and simple, remove the global pernet_operations and smc_sysctl_{init|exit}. Call smc_sysctl_net_{init|exit} directly from smc_net_{init|exit}. Also initialized sysctl_autocorking_size if CONFIG_SYSCTL it not set, this make sure SMC autocorking is enabled by default if CONFIG_SYSCTL is not set. Fixes: 462791bbfa35 ("net/smc: add sysctl interface for SMC") Reported-by: kernel test robot <lkp@intel.com> Signed-off-by: Dust Li <dust.li@linux.alibaba.com> Tested-by: Randy Dunlap <rdunlap@infradead.org> # build-tested Signed-off-by: David S. Miller <davem@davemloft.net>
2022-03-07 01:54:24 +00:00
int __net_init smc_sysctl_net_init(struct net *net)
{
size_t table_size = ARRAY_SIZE(smc_table);
struct ctl_table *table;
table = smc_table;
if (!net_eq(net, &init_net)) {
int i;
table = kmemdup(table, sizeof(smc_table), GFP_KERNEL);
if (!table)
goto err_alloc;
for (i = 0; i < table_size; i++)
table[i].data += (void *)net - (void *)&init_net;
}
net->smc.smc_hdr = register_net_sysctl_sz(net, "net/smc", table,
table_size);
if (!net->smc.smc_hdr)
goto err_reg;
net->smc.sysctl_autocorking_size = SMC_AUTOCORKING_DEFAULT_SIZE;
net->smc.sysctl_smcr_buf_type = SMCR_PHYS_CONT_BUFS;
net->smc.sysctl_smcr_testlink_time = SMC_LLC_TESTLINK_DEFAULT_TIME;
net/smc: Fix setsockopt and sysctl to specify same buffer size again Commit 0227f058aa29 ("net/smc: Unbind r/w buffer size from clcsock and make them tunable") introduced the net.smc.rmem and net.smc.wmem sysctls to specify the size of buffers to be used for SMC type connections. This created a regression for users that specified the buffer size via setsockopt() as the effective buffer size was now doubled. Re-introduce the division by 2 in the SMC buffer create code and level this out by duplicating the net.smc.[rw]mem values used for initializing sk_rcvbuf/sk_sndbuf at socket creation time. This gives users of both methods (setsockopt or sysctl) the effective buffer size that they expect. Initialize net.smc.[rw]mem from its own constant of 64kB, respectively. Internal performance tests show that this value is a good compromise between throughput/latency and memory consumption. Also, this decouples it from any tuning that was done to net.ipv4.tcp_[rw]mem[1] before the module for SMC protocol was loaded. Check that no more than INT_MAX / 2 is assigned to net.smc.[rw]mem, in order to avoid any overflow condition when that is doubled for use in sk_sndbuf or sk_rcvbuf. While at it, drop the confusing sk_buf_size variable from __smc_buf_create and name "compressed" buffer size variables more consistently. Background: Before the commit mentioned above, SMC's buffer allocator in __smc_buf_create() always used half of the sockets' sk_rcvbuf/sk_sndbuf value as initial value to search for appropriate buffers. If the search resorted to using a bigger buffer when all buffers of the specified size were busy, the duplicate of the used effective buffer size is stored back to sk_rcvbuf/sk_sndbuf. When available, buffers of exactly the size that a user had specified as input to setsockopt() were used, despite setsockopt()'s documentation in "man 7 socket" talking of a mandatory duplication: [...] SO_SNDBUF Sets or gets the maximum socket send buffer in bytes. The kernel doubles this value (to allow space for book‐ keeping overhead) when it is set using setsockopt(2), and this doubled value is returned by getsockopt(2). The default value is set by the /proc/sys/net/core/wmem_default file and the maximum allowed value is set by the /proc/sys/net/core/wmem_max file. The minimum (doubled) value for this option is 2048. [...] Fixes: 0227f058aa29 ("net/smc: Unbind r/w buffer size from clcsock and make them tunable") Co-developed-by: Jan Karcher <jaka@linux.ibm.com> Signed-off-by: Jan Karcher <jaka@linux.ibm.com> Reviewed-by: Wenjia Zhang <wenjia@linux.ibm.com> Reviewed-by: Tony Lu <tonylu@linux.alibaba.com> Signed-off-by: Gerd Bayer <gbayer@linux.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2023-08-04 17:06:23 +00:00
WRITE_ONCE(net->smc.sysctl_wmem, net_smc_wmem_init);
WRITE_ONCE(net->smc.sysctl_rmem, net_smc_rmem_init);
net->smc.sysctl_max_links_per_lgr = SMC_LINKS_PER_LGR_MAX_PREFER;
net->smc.sysctl_max_conns_per_lgr = SMC_CONN_PER_LGR_PREFER;
/* disable handshake limitation by default */
net->smc.limit_smc_hs = 0;
return 0;
err_reg:
if (!net_eq(net, &init_net))
kfree(table);
err_alloc:
return -ENOMEM;
}
net/smc: fix compile warning for smc_sysctl kernel test robot reports multiple warning for smc_sysctl: In file included from net/smc/smc_sysctl.c:17: >> net/smc/smc_sysctl.h:23:5: warning: no previous prototype \ for function 'smc_sysctl_init' [-Wmissing-prototypes] int smc_sysctl_init(void) ^ and >> WARNING: modpost: vmlinux.o(.text+0x12ced2d): Section mismatch \ in reference from the function smc_sysctl_exit() to the variable .init.data:smc_sysctl_ops The function smc_sysctl_exit() references the variable __initdata smc_sysctl_ops. This is often because smc_sysctl_exit lacks a __initdata annotation or the annotation of smc_sysctl_ops is wrong. and net/smc/smc_sysctl.c: In function 'smc_sysctl_init_net': net/smc/smc_sysctl.c:47:17: error: 'struct netns_smc' has no member named 'smc_hdr' 47 | net->smc.smc_hdr = register_net_sysctl(net, "net/smc", table); Since we don't need global sysctl initialization. To make things clean and simple, remove the global pernet_operations and smc_sysctl_{init|exit}. Call smc_sysctl_net_{init|exit} directly from smc_net_{init|exit}. Also initialized sysctl_autocorking_size if CONFIG_SYSCTL it not set, this make sure SMC autocorking is enabled by default if CONFIG_SYSCTL is not set. Fixes: 462791bbfa35 ("net/smc: add sysctl interface for SMC") Reported-by: kernel test robot <lkp@intel.com> Signed-off-by: Dust Li <dust.li@linux.alibaba.com> Tested-by: Randy Dunlap <rdunlap@infradead.org> # build-tested Signed-off-by: David S. Miller <davem@davemloft.net>
2022-03-07 01:54:24 +00:00
void __net_exit smc_sysctl_net_exit(struct net *net)
{
const struct ctl_table *table;
table = net->smc.smc_hdr->ctl_table_arg;
unregister_net_sysctl_table(net->smc.smc_hdr);
if (!net_eq(net, &init_net))
kfree(table);
}