| Message ID | 20200922133731.33478-5-jarod@redhat.com |
|---|---|
| State | Changes Requested |
| Delegated to: | David Miller |
| Headers | show |
| Series | bonding: rename bond components | expand |
On Tue, Sep 22, 2020 at 9:38 AM Jarod Wilson <jarod@redhat.com> wrote: > > By default, enable retaining all user-facing API that includes the use of > master and slave, but add a Kconfig knob that allows those that wish to > remove it entirely do so in one shot. > diff --git a/drivers/net/bonding/bond_procfs.c b/drivers/net/bonding/bond_procfs.c > index abd265d6e975..91ece68607b2 100644 > --- a/drivers/net/bonding/bond_procfs.c > +++ b/drivers/net/bonding/bond_procfs.c > @@ -7,6 +7,12 @@ > > #include "bonding_priv.h" > > +#ifdef CONFIG_BONDING_LEGACY_INTERFACES > +const char *linkdesc = "Slave"; > +#else > +const char *linkdesc = "Link"; > +#endif I've been asked if it would be okay to add extra lines to the /proc/net/bonding/<bond interface> output, so that for example, both "Slave Interface: <interface>" and "Link Interface: <interface>" are both in the default output, with the Slave bits then suppressed by the Kconfig option being unset, versus the Kconfig option currently swapping out Slave for Link when disabled. It would bloat the output by a fair number of lines, but all the same data would be there and parseable. Wasn't sure on this one, so I wanted to check on it. If it would be acceptable, I'll rework that bit of code.
On Tue, 22 Sep 2020 09:37:30 -0400 Jarod Wilson <jarod@redhat.com> wrote: > By default, enable retaining all user-facing API that includes the use of > master and slave, but add a Kconfig knob that allows those that wish to > remove it entirely do so in one shot. > > Cc: Jay Vosburgh <j.vosburgh@gmail.com> > Cc: Veaceslav Falico <vfalico@gmail.com> > Cc: Andy Gospodarek <andy@greyhouse.net> > Cc: "David S. Miller" <davem@davemloft.net> > Cc: Jakub Kicinski <kuba@kernel.org> > Cc: Thomas Davis <tadavis@lbl.gov> > Cc: netdev@vger.kernel.org > Signed-off-by: Jarod Wilson <jarod@redhat.com> Why not just have a config option to remove all the /proc and sysfs options in bonding (and bridging) and only use netlink? New tools should be only able to use netlink only. Then you might convince maintainers to update documentation as well. Last I checked there were still references to ifenslave.
Stephen Hemminger <stephen@networkplumber.org> wrote: >On Tue, 22 Sep 2020 09:37:30 -0400 >Jarod Wilson <jarod@redhat.com> wrote: > >> By default, enable retaining all user-facing API that includes the use of >> master and slave, but add a Kconfig knob that allows those that wish to >> remove it entirely do so in one shot. >> >> Cc: Jay Vosburgh <j.vosburgh@gmail.com> >> Cc: Veaceslav Falico <vfalico@gmail.com> >> Cc: Andy Gospodarek <andy@greyhouse.net> >> Cc: "David S. Miller" <davem@davemloft.net> >> Cc: Jakub Kicinski <kuba@kernel.org> >> Cc: Thomas Davis <tadavis@lbl.gov> >> Cc: netdev@vger.kernel.org >> Signed-off-by: Jarod Wilson <jarod@redhat.com> > >Why not just have a config option to remove all the /proc and sysfs options >in bonding (and bridging) and only use netlink? New tools should be only able >to use netlink only. I agree that new tooling should be netlink, but what value is provided by such an option that distros are unlikely to enable, and enabling will break the UAPI? >Then you might convince maintainers to update documentation as well. >Last I checked there were still references to ifenslave. Distros still include ifenslave, but it's now a shell script that uses sysfs. I see it used in scripts from time to time. -J --- -Jay Vosburgh, jay.vosburgh@canonical.com
On Tue, 22 Sep 2020 16:47:07 -0700 Jay Vosburgh <jay.vosburgh@canonical.com> wrote: > Stephen Hemminger <stephen@networkplumber.org> wrote: > > >On Tue, 22 Sep 2020 09:37:30 -0400 > >Jarod Wilson <jarod@redhat.com> wrote: > > > >> By default, enable retaining all user-facing API that includes the use of > >> master and slave, but add a Kconfig knob that allows those that wish to > >> remove it entirely do so in one shot. > >> > >> Cc: Jay Vosburgh <j.vosburgh@gmail.com> > >> Cc: Veaceslav Falico <vfalico@gmail.com> > >> Cc: Andy Gospodarek <andy@greyhouse.net> > >> Cc: "David S. Miller" <davem@davemloft.net> > >> Cc: Jakub Kicinski <kuba@kernel.org> > >> Cc: Thomas Davis <tadavis@lbl.gov> > >> Cc: netdev@vger.kernel.org > >> Signed-off-by: Jarod Wilson <jarod@redhat.com> > > > >Why not just have a config option to remove all the /proc and sysfs options > >in bonding (and bridging) and only use netlink? New tools should be only able > >to use netlink only. > > I agree that new tooling should be netlink, but what value is > provided by such an option that distros are unlikely to enable, and > enabling will break the UAPI? > > >Then you might convince maintainers to update documentation as well. > >Last I checked there were still references to ifenslave. > > Distros still include ifenslave, but it's now a shell script > that uses sysfs. I see it used in scripts from time to time. Some bleeding edge distros have already dropped ifenslave and even ifconfig. The Enterprise ones never will. The one motivation would be for the embedded folks which are always looking to trim out the fat. Although not sure if the minimal versions of commands in busybox are pure netlink yet.
Hi Jarod,
Thank you for the patch! Perhaps something to improve:
[auto build test WARNING on net-next/master]
url: https://github.com/0day-ci/linux/commits/Jarod-Wilson/bonding-rename-bond-components/20200922-214046
base: https://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next.git 92ec804f3dbf0d986f8e10850bfff14f316d7aaf
config: i386-randconfig-s031-20200921 (attached as .config)
compiler: gcc-9 (Debian 9.3.0-15) 9.3.0
reproduce:
# apt-get install sparse
# sparse version: v0.6.2-201-g24bdaac6-dirty
# save the attached .config to linux build tree
make W=1 C=1 CF='-fdiagnostic-prefix -D__CHECK_ENDIAN__' ARCH=i386
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 >>)
>> drivers/net/bonding/bond_procfs.c:11:12: sparse: sparse: symbol 'linkdesc' was not declared. Should it be static?
Please review and possibly fold the followup patch.
---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org
On Wed, Sep 23, 2020 at 12:15 AM kernel test robot <lkp@intel.com> wrote: > > Signed-off-by: kernel test robot <lkp@intel.com> > --- > bond_procfs.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/drivers/net/bonding/bond_procfs.c b/drivers/net/bonding/bond_procfs.c > index 91ece68607b23..9b1b37a682728 100644 > --- a/drivers/net/bonding/bond_procfs.c > +++ b/drivers/net/bonding/bond_procfs.c > @@ -8,7 +8,7 @@ > #include "bonding_priv.h" > > #ifdef CONFIG_BONDING_LEGACY_INTERFACES > -const char *linkdesc = "Slave"; > +static const char *linkdesc = "Slave"; > #else > const char *linkdesc = "Link"; > #endif Good attempt, robot, but you missed the #else. Will fold a full version into my set.
On Tue, Sep 22, 2020 at 8:01 PM Stephen Hemminger <stephen@networkplumber.org> wrote: > > On Tue, 22 Sep 2020 16:47:07 -0700 > Jay Vosburgh <jay.vosburgh@canonical.com> wrote: > > > Stephen Hemminger <stephen@networkplumber.org> wrote: > > > > >On Tue, 22 Sep 2020 09:37:30 -0400 > > >Jarod Wilson <jarod@redhat.com> wrote: > > > > > >> By default, enable retaining all user-facing API that includes the use of > > >> master and slave, but add a Kconfig knob that allows those that wish to > > >> remove it entirely do so in one shot. > > >> > > >> Cc: Jay Vosburgh <j.vosburgh@gmail.com> > > >> Cc: Veaceslav Falico <vfalico@gmail.com> > > >> Cc: Andy Gospodarek <andy@greyhouse.net> > > >> Cc: "David S. Miller" <davem@davemloft.net> > > >> Cc: Jakub Kicinski <kuba@kernel.org> > > >> Cc: Thomas Davis <tadavis@lbl.gov> > > >> Cc: netdev@vger.kernel.org > > >> Signed-off-by: Jarod Wilson <jarod@redhat.com> > > > > > >Why not just have a config option to remove all the /proc and sysfs options > > >in bonding (and bridging) and only use netlink? New tools should be only able > > >to use netlink only. > > > > I agree that new tooling should be netlink, but what value is > > provided by such an option that distros are unlikely to enable, and > > enabling will break the UAPI? Do you mean the initial proposed option, or what Stephen is suggesting? I think Red Hat actually will consider the former, the latter is less likely in the immediate future, since so many people still rely on the output of /proc/net/bonding/* for an overall view of their bonds' health and status. I don't know how close we are to having something comparable that could be spit out with a single invocation of something like 'ip' that would only be using netlink. It's entirely possible there's something akin to 'ip link bondX overview' already that outputs something similar, and I'm just not aware of it, but something like that would definitely need to exist and be well-documented for Red Hat to remove the procfs bits, I think. > > >Then you might convince maintainers to update documentation as well. > > >Last I checked there were still references to ifenslave. > > > > Distros still include ifenslave, but it's now a shell script > > that uses sysfs. I see it used in scripts from time to time. > > Some bleeding edge distros have already dropped ifenslave and even ifconfig. > The Enterprise ones never will. > > The one motivation would be for the embedded folks which are always looking > to trim out the fat. Although not sure if the minimal versions of commands > in busybox are pure netlink yet. Yeah, the bonding documentation is still filled with references to ifenslave. I believe Red Hat still includes it, though it's "deprecated" in documentation in favor of using ip. Similar with ifconfig. I could see them both getting dropped in a future major release of Red Hat Enterprise Linux, but they're definitely still here for at least the life of RHEL8.
Jarod Wilson <jarod@redhat.com> wrote: >On Tue, Sep 22, 2020 at 8:01 PM Stephen Hemminger ><stephen@networkplumber.org> wrote: >> >> On Tue, 22 Sep 2020 16:47:07 -0700 >> Jay Vosburgh <jay.vosburgh@canonical.com> wrote: >> >> > Stephen Hemminger <stephen@networkplumber.org> wrote: >> > >> > >On Tue, 22 Sep 2020 09:37:30 -0400 >> > >Jarod Wilson <jarod@redhat.com> wrote: >> > > >> > >> By default, enable retaining all user-facing API that includes the use of >> > >> master and slave, but add a Kconfig knob that allows those that wish to >> > >> remove it entirely do so in one shot. >> > >> >> > >> Cc: Jay Vosburgh <j.vosburgh@gmail.com> >> > >> Cc: Veaceslav Falico <vfalico@gmail.com> >> > >> Cc: Andy Gospodarek <andy@greyhouse.net> >> > >> Cc: "David S. Miller" <davem@davemloft.net> >> > >> Cc: Jakub Kicinski <kuba@kernel.org> >> > >> Cc: Thomas Davis <tadavis@lbl.gov> >> > >> Cc: netdev@vger.kernel.org >> > >> Signed-off-by: Jarod Wilson <jarod@redhat.com> >> > > >> > >Why not just have a config option to remove all the /proc and sysfs options >> > >in bonding (and bridging) and only use netlink? New tools should be only able >> > >to use netlink only. >> > >> > I agree that new tooling should be netlink, but what value is >> > provided by such an option that distros are unlikely to enable, and >> > enabling will break the UAPI? > >Do you mean the initial proposed option, or what Stephen is >suggesting? I think Red Hat actually will consider the former, the >latter is less likely in the immediate future, since so many people >still rely on the output of /proc/net/bonding/* for an overall view of >their bonds' health and status. I don't know how close we are to >having something comparable that could be spit out with a single >invocation of something like 'ip' that would only be using netlink. >It's entirely possible there's something akin to 'ip link bondX >overview' already that outputs something similar, and I'm just not >aware of it, but something like that would definitely need to exist >and be well-documented for Red Hat to remove the procfs bits, I think. At the present time, as much as the idea spurs the imagination, removing the bonding /proc and sysfs stuff wholesale is not feasible. As you explain, not everything in the proc file is available from other sources. I would rather freeze the /proc and sysfs bonding functionality and move to a netlink / iproute API for all of it, and then down the road remove the then-legacy APIs. Even though "down the road" may practically be "never" (because the removal breaks backwards compatibility for user space), unifying all of the configuration and reporting to one place would be worthwhile. For "initial proposed option," I'm not sure right off if that's referring to CONFIG_BONDING_LEGACY_INTERFACES or "duplicate lines in /proc/net/bonding." I'm not sure it matters, since both have the same problem, in that they create a Venn diagram of mutually incompatible bonding UAPIs. Portable user space code ends up having to handle all of the permutations. -J >> > >Then you might convince maintainers to update documentation as well. >> > >Last I checked there were still references to ifenslave. >> > >> > Distros still include ifenslave, but it's now a shell script >> > that uses sysfs. I see it used in scripts from time to time. >> >> Some bleeding edge distros have already dropped ifenslave and even ifconfig. >> The Enterprise ones never will. >> >> The one motivation would be for the embedded folks which are always looking >> to trim out the fat. Although not sure if the minimal versions of commands >> in busybox are pure netlink yet. > >Yeah, the bonding documentation is still filled with references to >ifenslave. I believe Red Hat still includes it, though it's >"deprecated" in documentation in favor of using ip. Similar with >ifconfig. I could see them both getting dropped in a future major >release of Red Hat Enterprise Linux, but they're definitely still here >for at least the life of RHEL8. As ifconfig is typically bundled in with the much-loved netstat in the net-tools package, it will be difficult to remove. Having an /sbin/ifenslave program isn't really the issue so much as its reliance on the bonding sysfs UAPI. It's a shell script, and could likely be reworked to use ip link. -J --- -Jay Vosburgh, jay.vosburgh@canonical.com
diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig index c3dbe64e628e..3640694be34d 100644 --- a/drivers/net/Kconfig +++ b/drivers/net/Kconfig @@ -56,6 +56,18 @@ config BONDING To compile this driver as a module, choose M here: the module will be called bonding. +config BONDING_LEGACY_INTERFACES + default y + bool "Maintain legacy interface names" + help + The bonding driver historically made use of the terms "master" and + "slave" to describe it's component members. This has since been + changed to "aggregator" and "link" as part of a broader effort to + remove the use of socially problematic language from the kernel. + However, removing all such cases requires breaking long-standing + user-facing interfaces in /proc and /sys, which will not be done, + unless you opt out of them here, by selecting 'N'. + config DUMMY tristate "Dummy net driver support" help diff --git a/drivers/net/bonding/bond_options.c b/drivers/net/bonding/bond_options.c index 437df9a207a6..7bf1a13a3c17 100644 --- a/drivers/net/bonding/bond_options.c +++ b/drivers/net/bonding/bond_options.c @@ -434,7 +434,7 @@ static const struct bond_option bond_opts[BOND_OPT_LAST] = { .values = bond_intmax_tbl, .set = bond_option_peer_notif_delay_set }, -/* legacy sysfs interface names */ +#ifdef CONFIG_BONDING_LEGACY_INTERFACES [BOND_OPT_PACKETS_PER_SLAVE] = { .id = BOND_OPT_PACKETS_PER_SLAVE, .name = "packets_per_slave", @@ -474,7 +474,7 @@ static const struct bond_option bond_opts[BOND_OPT_LAST] = { .flags = BOND_OPTFLAG_RAWVAL, .set = bond_option_links_set }, -/* end legacy sysfs interface names */ +#endif }; /* Searches for an option by name */ diff --git a/drivers/net/bonding/bond_procfs.c b/drivers/net/bonding/bond_procfs.c index abd265d6e975..91ece68607b2 100644 --- a/drivers/net/bonding/bond_procfs.c +++ b/drivers/net/bonding/bond_procfs.c @@ -7,6 +7,12 @@ #include "bonding_priv.h" +#ifdef CONFIG_BONDING_LEGACY_INTERFACES +const char *linkdesc = "Slave"; +#else +const char *linkdesc = "Link"; +#endif + static void *bond_info_seq_start(struct seq_file *seq, loff_t *pos) __acquires(RCU) { @@ -84,7 +90,7 @@ static void bond_info_show_aggregator(struct seq_file *seq) if (bond_uses_primary(bond)) { primary = rcu_dereference(bond->primary_link); - seq_printf(seq, "Primary Slave: %s", + seq_printf(seq, "Primary %s: %s", linkdesc, primary ? primary->dev->name : "None"); if (primary) { optval = bond_opt_get_val(BOND_OPT_PRIMARY_RESELECT, @@ -93,7 +99,7 @@ static void bond_info_show_aggregator(struct seq_file *seq) optval->string); } - seq_printf(seq, "\nCurrently Active Slave: %s\n", + seq_printf(seq, "\nCurrently Active %s: %s\n", linkdesc, (curr) ? curr->dev->name : "None"); } @@ -171,7 +177,7 @@ static void bond_info_show_link(struct seq_file *seq, { struct bonding *bond = PDE_DATA(file_inode(seq->file)); - seq_printf(seq, "\nSlave Interface: %s\n", link->dev->name); + seq_printf(seq, "\n%s Interface: %s\n", linkdesc, link->dev->name); seq_printf(seq, "MII Status: %s\n", bond_link_status(link->link_state)); if (link->speed == SPEED_UNKNOWN) @@ -189,7 +195,7 @@ static void bond_info_show_link(struct seq_file *seq, seq_printf(seq, "Permanent HW addr: %*phC\n", link->dev->addr_len, link->perm_hwaddr); - seq_printf(seq, "Slave queue ID: %d\n", link->queue_id); + seq_printf(seq, "%s queue ID: %d\n", linkdesc, link->queue_id); if (BOND_MODE(bond) == BOND_MODE_8023AD) { const struct port *port = &LINK_AD_INFO(link)->port; diff --git a/drivers/net/bonding/bond_sysfs.c b/drivers/net/bonding/bond_sysfs.c index 0a4d095b8c3d..9065f24e31c0 100644 --- a/drivers/net/bonding/bond_sysfs.c +++ b/drivers/net/bonding/bond_sysfs.c @@ -150,6 +150,7 @@ static const struct class_attribute class_attr_bonding_aggregators = { .store = bonding_store_bonds, }; +#ifdef CONFIG_BONDING_LEGACY_INTERFACES /* "show" function for the bond_masters attribute. * The class parameter is ignored. */ @@ -178,7 +179,6 @@ static ssize_t bonding_store_bonds_legacy(struct class *cls, return __bonding_store_bonds(bn, buffer, count); } -/* legacy sysfs interface name */ static const struct class_attribute class_attr_bonding_masters = { .attr = { .name = "bonding_masters", @@ -187,6 +187,7 @@ static const struct class_attribute class_attr_bonding_masters = { .show = bonding_show_bonds_legacy, .store = bonding_store_bonds_legacy, }; +#endif /* Generic "store" method for bonding sysfs option setting */ static ssize_t bonding_sysfs_store_option(struct device *d, @@ -771,7 +772,7 @@ static ssize_t bonding_show_ad_user_port_key(struct device *d, static DEVICE_ATTR(ad_user_port_key, 0644, bonding_show_ad_user_port_key, bonding_sysfs_store_option); -/* legacy sysfs interface names */ +#ifdef CONFIG_BONDING_LEGACY_INTERFACES static DEVICE_ATTR(slaves, 0644, bonding_show_links, bonding_sysfs_store_option); static DEVICE_ATTR(min_slaves, 0644, @@ -782,7 +783,7 @@ static DEVICE_ATTR(all_slaves_active, 0644, bonding_show_links_active, bonding_sysfs_store_option); static DEVICE_ATTR(packets_per_slave, 0644, bonding_show_packets_per_link, bonding_sysfs_store_option); -/* end legacy sysfs interface names */ +#endif static struct attribute *per_bond_attrs[] = { &dev_attr_links.attr, @@ -821,13 +822,13 @@ static struct attribute *per_bond_attrs[] = { &dev_attr_ad_actor_sys_prio.attr, &dev_attr_ad_actor_system.attr, &dev_attr_ad_user_port_key.attr, -/* legacy sysfs interface names */ +#ifdef CONFIG_BONDING_LEGACY_INTERFACES &dev_attr_slaves.attr, &dev_attr_active_slave.attr, &dev_attr_all_slaves_active.attr, &dev_attr_min_slaves.attr, &dev_attr_packets_per_slave.attr, -/* end legacy sysfs interface names */ +#endif NULL, }; @@ -874,6 +875,7 @@ int bond_create_sysfs(struct bond_net *bn) return ret; } +#ifdef CONFIG_BONDING_LEGACY_INTERFACES bn->class_attr_bonding_masters = class_attr_bonding_masters; sysfs_attr_init(&bn->class_attr_bonding_masters.attr); @@ -888,14 +890,17 @@ int bond_create_sysfs(struct bond_net *bn) ret = 0; } +#endif return ret; } /* Remove /sys/class/net/bonding_aggregators and _masters. */ void bond_destroy_sysfs(struct bond_net *bn) { +#ifdef CONFIG_BONDING_LEGACY_INTERFACES netdev_class_remove_file_ns(&bn->class_attr_bonding_masters, bn->net); +#endif netdev_class_remove_file_ns(&bn->class_attr_bonding_aggregators, bn->net); } diff --git a/drivers/net/bonding/bond_sysfs_link.c b/drivers/net/bonding/bond_sysfs_link.c index 595db312df10..f0cb43a9773b 100644 --- a/drivers/net/bonding/bond_sysfs_link.c +++ b/drivers/net/bonding/bond_sysfs_link.c @@ -48,8 +48,9 @@ static ssize_t link_failure_count_show(struct link *link, char *buf) return sprintf(buf, "%d\n", link->link_failure_count); } static LINK_ATTR_RO(link_failure_count); -/* legacy sysfs interface */ +#ifdef CONFIG_BONDING_LEGACY_INTERFACES static LINK_ATTR(slave_failure_count, 0444, link_failure_count_show); +#endif static ssize_t perm_hwaddr_show(struct link *link, char *buf) { @@ -119,8 +120,9 @@ static const struct link_attribute *link_attrs[] = { &link_attr_ad_aggregator_id, &link_attr_ad_actor_oper_port_state, &link_attr_ad_partner_oper_port_state, -/* legacy sysfs interface */ +#ifdef CONFIG_BONDING_LEGACY_INTERFACES &link_attr_slave_failure_count, +#endif NULL }; @@ -156,11 +158,12 @@ int bond_sysfs_link_add(struct link *link) if (err) goto err_kobject_put; -/* legacy sysfs interface */ +#ifdef CONFIG_BONDING_LEGACY_INTERFACES err = sysfs_create_link(&(link->dev->dev.kobj), &link->kobj, "bonding_slave"); if (err) goto err_kobject_put; +#endif for (a = link_attrs; *a; ++a) { err = sysfs_create_file(&link->kobj, &((*a)->attr)); @@ -182,8 +185,9 @@ void bond_sysfs_link_del(struct link *link) for (a = link_attrs; *a; ++a) sysfs_remove_file(&link->kobj, &((*a)->attr)); -/* legacy sysfs interface */ +#ifdef CONFIG_BONDING_LEGACY_INTERFACES sysfs_remove_link(&(link->dev->dev.kobj), "bonding_slave"); +#endif kobject_put(&link->kobj); } diff --git a/include/net/bond_options.h b/include/net/bond_options.h index 2c2e4a94bdc2..48c5bb20cf46 100644 --- a/include/net/bond_options.h +++ b/include/net/bond_options.h @@ -64,13 +64,13 @@ enum { BOND_OPT_AD_USER_PORT_KEY, BOND_OPT_NUM_PEER_NOTIF_ALIAS, BOND_OPT_PEER_NOTIF_DELAY, -/* legacy sysfs interface names */ +#ifdef CONFIG_BONDING_LEGACY_INTERFACES BOND_OPT_PACKETS_PER_SLAVE, BOND_OPT_MINSLAVES, BOND_OPT_ACTIVE_SLAVE, BOND_OPT_ALL_SLAVES_ACTIVE, BOND_OPT_SLAVES, -/* end legacy sysfs interface names */ +#endif BOND_OPT_LAST }; diff --git a/include/net/bonding.h b/include/net/bonding.h index af3fecc27a19..d43f1a7450b0 100644 --- a/include/net/bonding.h +++ b/include/net/bonding.h @@ -614,7 +614,9 @@ struct bond_net { struct proc_dir_entry *proc_dir; #endif struct class_attribute class_attr_bonding_aggregators; +#ifdef CONFIG_BONDING_LEGACY_INTERFACES struct class_attribute class_attr_bonding_masters; +#endif }; int bond_arp_rcv(const struct sk_buff *skb, struct bonding *bond, struct link *link);
By default, enable retaining all user-facing API that includes the use of master and slave, but add a Kconfig knob that allows those that wish to remove it entirely do so in one shot. Cc: Jay Vosburgh <j.vosburgh@gmail.com> Cc: Veaceslav Falico <vfalico@gmail.com> Cc: Andy Gospodarek <andy@greyhouse.net> Cc: "David S. Miller" <davem@davemloft.net> Cc: Jakub Kicinski <kuba@kernel.org> Cc: Thomas Davis <tadavis@lbl.gov> Cc: netdev@vger.kernel.org Signed-off-by: Jarod Wilson <jarod@redhat.com> --- drivers/net/Kconfig | 12 ++++++++++++ drivers/net/bonding/bond_options.c | 4 ++-- drivers/net/bonding/bond_procfs.c | 14 ++++++++++---- drivers/net/bonding/bond_sysfs.c | 15 ++++++++++----- drivers/net/bonding/bond_sysfs_link.c | 12 ++++++++---- include/net/bond_options.h | 4 ++-- include/net/bonding.h | 2 ++ 7 files changed, 46 insertions(+), 17 deletions(-)