In memcg_rstat_updated(), we iterate the memcg being updated and its
parents to update memcg->vmstats_percpu->stats_updates in the fast path
(i.e. no atomic updates). According to my math, this is 3 memory loads
(and potentially 3 cache misses) per memcg:
- Load the address of memcg->vmstats_percpu.
- Load vmstats_percpu->stats_updates (based on some percpu calculation).
- Load the address of the parent memcg.
Avoid most of the cache misses by caching a pointer from each struct
memcg_vmstats_percpu to its parent on the corresponding CPU. In this
case, for the first memcg we have 2 memory loads (same as above):
- Load the address of memcg->vmstats_percpu.
- Load vmstats_percpu->stats_updates (based on some percpu calculation).
Then for each additional memcg, we need a single load to get the
parent's stats_updates directly. This reduces the number of loads from
O(3N) to O(2+N) -- where N is the number of memcgs we need to iterate.
Additionally, stash a pointer to memcg->vmstats in each struct
memcg_vmstats_percpu such that we can access the atomic counter that all
CPUs fold into, memcg->vmstats->stats_updates.
memcg_should_flush_stats() is changed to memcg_vmstats_needs_flush() to
accept a struct memcg_vmstats pointer accordingly.
In struct memcg_vmstats_percpu, make sure both pointers together with
stats_updates live on the same cacheline. Finally, update
mem_cgroup_alloc() to take in a parent pointer and initialize the new
cache pointers on each CPU. The percpu loop in mem_cgroup_alloc() may
look concerning, but there are multiple similar loops in the cgroup
creation path (e.g. cgroup_rstat_init()), most of which are hidden
within alloc_percpu().
According to Oliver's testing [1], this fixes multiple 30-38%
regressions in vm-scalability, will-it-scale-tlb_flush2, and
will-it-scale-fallocate1. This comes at a cost of 2 more pointers per
CPU (<2KB on a machine with 128 CPUs).
[1] https://lore.kernel.org/lkml/ZbDJsfsZt2ITyo61@xsang-OptiPlex-9020/
[yosryahmed@google.com: fix struct memcg_vmstats_percpu size and alignment]
Link: https://lkml.kernel.org/r/20240203044612.1234216-1-yosryahmed@google.com
Link: https://lkml.kernel.org/r/20240124100023.660032-1-yosryahmed@google.com
Signed-off-by: Yosry Ahmed <yosryahmed@google.com>
Fixes: 8d59d2214c ("mm: memcg: make stats flushing threshold per-memcg")
Tested-by: kernel test robot <oliver.sang@intel.com>
Reported-by: kernel test robot <oliver.sang@intel.com>
Closes: https://lore.kernel.org/oe-lkp/202401221624.cb53a8ca-oliver.sang@intel.com
Acked-by: Shakeel Butt <shakeelb@google.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Muchun Song <muchun.song@linux.dev>
Cc: Roman Gushchin <roman.gushchin@linux.dev>
Cc: Greg Thelen <gthelen@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
The helper function nilfs_recovery_copy_block() of
nilfs_recovery_dsync_blocks(), which recovers data from logs created by
data sync writes during a mount after an unclean shutdown, incorrectly
calculates the on-page offset when copying repair data to the file's page
cache. In environments where the block size is smaller than the page
size, this flaw can cause data corruption and leak uninitialized memory
bytes during the recovery process.
Fix these issues by correcting this byte offset calculation on the page.
Link: https://lkml.kernel.org/r/20240124121936.10575-1-konishi.ryusuke@gmail.com
Signed-off-by: Ryusuke Konishi <konishi.ryusuke@gmail.com>
Tested-by: Ryusuke Konishi <konishi.ryusuke@gmail.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Commit c33c794828 ("mm: ptep_get() conversion") converted all (non-arch)
call sites to use ptep_get() instead of doing a direct dereference of the
pte. Full rationale can be found in that commit's log.
Since then, UFFDIO_MOVE has been implemented which does 7 direct pte
dereferences. Let's fix those up to use ptep_get().
I've asserted in the past that there is no reliable automated mechanism to
catch these; I'm relying on a combination of Coccinelle (which throws up a
lot of false positives) and some compiler magic to force a compiler error
on dereference. But given the frequency with which new issues are coming
up, I'll add it to my todo list to try to find an automated solution.
Link: https://lkml.kernel.org/r/20240123141755.3836179-1-ryan.roberts@arm.com
Fixes: adef440691 ("userfaultfd: UFFDIO_MOVE uABI")
Signed-off-by: Ryan Roberts <ryan.roberts@arm.com>
Reviewed-by: Suren Baghdasaryan <surenb@google.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: David Hildenbrand <david@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
After the recent changes nobody use siglock to read the values protected
by stats_lock, we can kill spin_lock_irq(¤t->sighand->siglock) and
update the comment.
With this patch only __exit_signal() and thread_group_start_cputime() take
stats_lock under siglock.
Link: https://lkml.kernel.org/r/20240123153359.GA21866@redhat.com
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Dylan Hatch <dylanbhatch@google.com>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
lock_task_sighand() can trigger a hard lockup. If NR_CPUS threads call
do_task_stat() at the same time and the process has NR_THREADS, it will
spin with irqs disabled O(NR_CPUS * NR_THREADS) time.
Change do_task_stat() to use sig->stats_lock to gather the statistics
outside of ->siglock protected section, in the likely case this code will
run lockless.
Link: https://lkml.kernel.org/r/20240123153357.GA21857@redhat.com
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Dylan Hatch <dylanbhatch@google.com>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Patch series "fs/proc: do_task_stat: use sig->stats_".
do_task_stat() has the same problem as getrusage() had before "getrusage:
use sig->stats_lock rather than lock_task_sighand()": a hard lockup. If
NR_CPUS threads call lock_task_sighand() at the same time and the process
has NR_THREADS, spin_lock_irq will spin with irqs disabled O(NR_CPUS *
NR_THREADS) time.
This patch (of 3):
thread_group_cputime() does its own locking, we can safely shift
thread_group_cputime_adjusted() which does another for_each_thread loop
outside of ->siglock protected section.
Not only this removes for_each_thread() from the critical section with
irqs disabled, this removes another case when stats_lock is taken with
siglock held. We want to remove this dependency, then we can change the
users of stats_lock to not disable irqs.
Link: https://lkml.kernel.org/r/20240123153313.GA21832@redhat.com
Link: https://lkml.kernel.org/r/20240123153355.GA21854@redhat.com
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Dylan Hatch <dylanbhatch@google.com>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
lock_task_sighand() can trigger a hard lockup. If NR_CPUS threads call
getrusage() at the same time and the process has NR_THREADS, spin_lock_irq
will spin with irqs disabled O(NR_CPUS * NR_THREADS) time.
Change getrusage() to use sig->stats_lock, it was specifically designed
for this type of use. This way it runs lockless in the likely case.
TODO:
- Change do_task_stat() to use sig->stats_lock too, then we can
remove spin_lock_irq(siglock) in wait_task_zombie().
- Turn sig->stats_lock into seqcount_rwlock_t, this way the
readers in the slow mode won't exclude each other. See
https://lore.kernel.org/all/20230913154907.GA26210@redhat.com/
- stats_lock has to disable irqs because ->siglock can be taken
in irq context, it would be very nice to change __exit_signal()
to avoid the siglock->stats_lock dependency.
Link: https://lkml.kernel.org/r/20240122155053.GA26214@redhat.com
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Reported-by: Dylan Hatch <dylanbhatch@google.com>
Tested-by: Dylan Hatch <dylanbhatch@google.com>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Patch series "getrusage: use sig->stats_lock", v2.
This patch (of 2):
thread_group_cputime() does its own locking, we can safely shift
thread_group_cputime_adjusted() which does another for_each_thread loop
outside of ->siglock protected section.
This is also preparation for the next patch which changes getrusage() to
use stats_lock instead of siglock, thread_group_cputime() takes the same
lock. With the current implementation recursive read_seqbegin_or_lock()
is fine, thread_group_cputime() can't enter the slow mode if the caller
holds stats_lock, yet this looks more safe and better performance-wise.
Link: https://lkml.kernel.org/r/20240122155023.GA26169@redhat.com
Link: https://lkml.kernel.org/r/20240122155050.GA26205@redhat.com
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Reported-by: Dylan Hatch <dylanbhatch@google.com>
Tested-by: Dylan Hatch <dylanbhatch@google.com>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
For shared memory of type SHM_HUGETLB, hugetlb pages are reserved in
shmget() call. If SHM_NORESERVE flags is specified then the hugetlb pages
are not reserved. However when the shared memory is attached with the
shmat() call the hugetlb pages are getting reserved incorrectly for
SHM_HUGETLB shared memory created with SHM_NORESERVE which is a bug.
-------------------------------
Following test shows the issue.
$cat shmhtb.c
int main()
{
int shmflags = 0660 | IPC_CREAT | SHM_HUGETLB | SHM_NORESERVE;
int shmid;
shmid = shmget(SKEY, SHMSZ, shmflags);
if (shmid < 0)
{
printf("shmat: shmget() failed, %d\n", errno);
return 1;
}
printf("After shmget()\n");
system("cat /proc/meminfo | grep -i hugepages_");
shmat(shmid, NULL, 0);
printf("\nAfter shmat()\n");
system("cat /proc/meminfo | grep -i hugepages_");
shmctl(shmid, IPC_RMID, NULL);
return 0;
}
#sysctl -w vm.nr_hugepages=20
#./shmhtb
After shmget()
HugePages_Total: 20
HugePages_Free: 20
HugePages_Rsvd: 0
HugePages_Surp: 0
After shmat()
HugePages_Total: 20
HugePages_Free: 20
HugePages_Rsvd: 5 <--
HugePages_Surp: 0
--------------------------------
Fix is to ensure that hugetlb pages are not reserved for SHM_HUGETLB shared
memory in the shmat() call.
Link: https://lkml.kernel.org/r/1706040282-12388-1-git-send-email-prakash.sangappa@oracle.com
Signed-off-by: Prakash Sangappa <prakash.sangappa@oracle.com>
Acked-by: Muchun Song <muchun.song@linux.dev>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Now that the driver core can properly handle constant struct bus_type,
move the pci_epf_bus_type variable to be a constant structure as well,
placing it into read-only memory which can not be modified at runtime.
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Suggested-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Ricardo B. Marliere <ricardo@marliere.net>
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Reviewed-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Link: https://lore.kernel.org/r/20240204-bus_cleanup-pci-v1-1-300267a1e99e@marliere.net
[mani: modified subject to reflect subsys prefix]
Signed-off-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
exit_batch_rtnl() is called while RTNL is held,
and devices to be unregistered can be queued in the dev_kill_list.
This saves one rtnl_lock()/rtnl_unlock() pair per netns
and one unregister_netdevice_many() call.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Antoine Tenart <atenart@kernel.org>
Link: https://lore.kernel.org/r/20240206144313.2050392-17-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
exit_batch_rtnl() is called while RTNL is held,
and devices to be unregistered can be queued in the dev_kill_list.
This saves one rtnl_lock()/rtnl_unlock() pair per netns
and one unregister_netdevice_many() call.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Antoine Tenart <atenart@kernel.org>
Link: https://lore.kernel.org/r/20240206144313.2050392-16-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
exit_batch_rtnl() is called while RTNL is held,
and devices to be unregistered can be queued in the dev_kill_list.
This saves one rtnl_lock()/rtnl_unlock() pair
and one unregister_netdevice_many() call.
This patch takes care of ipip, ip_vti, and ip_gre tunnels.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Antoine Tenart <atenart@kernel.org>
Link: https://lore.kernel.org/r/20240206144313.2050392-15-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
exit_batch_rtnl() is called while RTNL is held,
and devices to be unregistered can be queued in the dev_kill_list.
This saves one rtnl_lock()/rtnl_unlock() pair
and one unregister_netdevice_many() call.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Antoine Tenart <atenart@kernel.org>
Link: https://lore.kernel.org/r/20240206144313.2050392-14-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
exit_batch_rtnl() is called while RTNL is held,
and devices to be unregistered can be queued in the dev_kill_list.
This saves one rtnl_lock()/rtnl_unlock() pair
and one unregister_netdevice_many() call.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Antoine Tenart <atenart@kernel.org>
Link: https://lore.kernel.org/r/20240206144313.2050392-13-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
exit_batch_rtnl() is called while RTNL is held,
and devices to be unregistered can be queued in the dev_kill_list.
This saves one rtnl_lock()/rtnl_unlock() pair
and one unregister_netdevice_many() call.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Antoine Tenart <atenart@kernel.org>
Link: https://lore.kernel.org/r/20240206144313.2050392-12-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
exit_batch_rtnl() is called while RTNL is held,
and devices to be unregistered can be queued in the dev_kill_list.
This saves one rtnl_lock()/rtnl_unlock() pair
and one unregister_netdevice_many() call.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Antoine Tenart <atenart@kernel.org>
Link: https://lore.kernel.org/r/20240206144313.2050392-11-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
exit_batch_rtnl() is called while RTNL is held,
and devices to be unregistered can be queued in the dev_kill_list.
This saves one rtnl_lock()/rtnl_unlock() pair per netns
and one unregister_netdevice_many() call.
v4: (Paolo feedback : https://netdev-3.bots.linux.dev/vmksft-net/results/453141/17-udpgro-fwd-sh/stdout )
- Changed vxlan_destroy_tunnels() to use vxlan_dellink()
instead of unregister_netdevice_queue to propely remove
devices from vn->vxlan_list.
- vxlan_destroy_tunnels() can simply iterate one list (vn->vxlan_list)
to find all devices in the most efficient way.
- Moved sanity checks in a separate vxlan_exit_net() method.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Antoine Tenart <atenart@kernel.org>
Link: https://lore.kernel.org/r/20240206144313.2050392-10-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
unregister_nexthop_notifier() assumes the caller does not hold rtnl.
We need in the following patch to use it from a context
already holding rtnl.
Add __unregister_nexthop_notifier().
unregister_nexthop_notifier() becomes a wrapper.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Reviewed-by: Antoine Tenart <atenart@kernel.org>
Link: https://lore.kernel.org/r/20240206144313.2050392-9-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
exit_batch_rtnl() is called while RTNL is held,
and devices to be unregistered can be queued in the dev_kill_list.
This saves one rtnl_lock()/rtnl_unlock() pair per netns
and one unregister_netdevice_many() call per netns.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Antoine Tenart <atenart@kernel.org>
Link: https://lore.kernel.org/r/20240206144313.2050392-8-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
exit_batch_rtnl() is called while RTNL is held,
and devices to be unregistered can be queued in the dev_kill_list.
This saves one rtnl_lock()/rtnl_unlock() pair,
and one unregister_netdevice_many() call.
Note: it should be possible to remove the synchronize_net()
call from geneve_sock_release() in a future patch.
v4: move WARN_ON_ONCE(!list_empty(&gn->sock_list))
into geneve_exit_net(), after devices have been unregistered.
(Antoine Tenart feedback)
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Antoine Tenart <atenart@kernel.org>
Link: https://lore.kernel.org/r/20240206144313.2050392-7-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
exit_batch_rtnl() is called while RTNL is held,
and devices to be unregistered can be queued in the dev_kill_list.
This saves one rtnl_lock()/rtnl_unlock() pair,
and one unregister_netdevice_many() call.
v2: Added bond_net_pre_exit() method to make sure bond_destroy_sysfs()
is called before we unregister the devices in bond_net_exit_batch_rtnl
(Antoine Tenart : https://lore.kernel.org/netdev/170688415193.5216.10499830272732622816@kwain/)
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Jay Vosburgh <j.vosburgh@gmail.com>
Cc: Andy Gospodarek <andy@greyhouse.net>
Reviewed-by: Antoine Tenart <atenart@kernel.org>
Link: https://lore.kernel.org/r/20240206144313.2050392-6-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
exit_batch_rtnl() is called while RTNL is held,
and devices to be unregistered can be queued in the dev_kill_list.
This saves one rtnl_lock()/rtnl_unlock() pair,
and one unregister_netdevice_many() call.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Antoine Tenart <atenart@kernel.org>
Link: https://lore.kernel.org/r/20240206144313.2050392-5-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
exit_batch_rtnl() is called while RTNL is held.
This saves one rtnl_lock()/rtnl_unlock() pair.
We also need to create nexthop_net_exit()
to make sure net->nexthop.devhash is not freed too soon,
otherwise we will not be able to unregister netdev
from exit_batch_rtnl() methods.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Reviewed-by: Antoine Tenart <atenart@kernel.org>
Link: https://lore.kernel.org/r/20240206144313.2050392-4-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Many (struct pernet_operations)->exit_batch() methods have
to acquire rtnl.
In presence of rtnl mutex pressure, this makes cleanup_net()
very slow.
This patch adds a new exit_batch_rtnl() method to reduce
number of rtnl acquisitions from cleanup_net().
exit_batch_rtnl() handlers are called while rtnl is locked,
and devices to be killed can be queued in a list provided
as their second argument.
A single unregister_netdevice_many() is called right
before rtnl is released.
exit_batch_rtnl() handlers are called before ->exit() and
->exit_batch() handlers.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Antoine Tenart <atenart@kernel.org>
Link: https://lore.kernel.org/r/20240206144313.2050392-2-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Arınç ÜNAL says:
====================
MT7530 DSA Subdriver Improvements Act II
This is the second patch series with the goal of simplifying the MT7530 DSA
subdriver and improving support for MT7530, MT7531, and the switch on the
MT7988 SoC.
I have done a simple ping test to confirm basic communication on all switch
ports on MCM and standalone MT7530, and MT7531 switch with this patch
series applied.
MT7621 Unielec, MCM MT7530:
rgmii-only-gmac0-mt7621-unielec-u7621-06-16m.dtb
gmac0-and-gmac1-mt7621-unielec-u7621-06-16m.dtb
tftpboot 0x80008000 mips-uzImage.bin; tftpboot 0x83000000 mips-rootfs.cpio.uboot; tftpboot 0x83f00000 $dtb; bootm 0x80008000 0x83000000 0x83f00000
MT7622 Bananapi, MT7531:
gmac0-and-gmac1-mt7622-bananapi-bpi-r64.dtb
tftpboot 0x40000000 arm64-Image; tftpboot 0x45000000 arm64-rootfs.cpio.uboot; tftpboot 0x4a000000 $dtb; booti 0x40000000 0x45000000 0x4a000000
MT7623 Bananapi, standalone MT7530:
rgmii-only-gmac0-mt7623n-bananapi-bpi-r2.dtb
gmac0-and-gmac1-mt7623n-bananapi-bpi-r2.dtb
tftpboot 0x80008000 arm-zImage; tftpboot 0x83000000 arm-rootfs.cpio.uboot; tftpboot 0x83f00000 $dtb; bootz 0x80008000 0x83000000 0x83f00000
This patch series is the continuation of the patch series linked below.
https://lore.kernel.org/r/20230522121532.86610-1-arinc.unal@arinc9.com
Signed-off-by: Arınç ÜNAL <arinc.unal@arinc9.com>
====================
Link: https://lore.kernel.org/r/20240206-for-netnext-mt7530-improvements-2-v5-0-d7d92a185cb1@arinc9.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
There's no need to clear the config->supported_interfaces bitmap before
reporting the supported interfaces as all bits in the bitmap will already
be initialized to zero when the phylink_config structure is allocated. The
"config" pointer points to &dp->phylink_config, and "dp" is allocated by
dsa_port_touch() with kzalloc(), so all its fields are filled with zeroes.
There's no code that would change the bitmap beforehand. Remove it.
Acked-by: Daniel Golle <daniel@makrotopia.org>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Reviewed-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: Arınç ÜNAL <arinc.unal@arinc9.com>
Link: https://lore.kernel.org/r/20240206-for-netnext-mt7530-improvements-2-v5-7-d7d92a185cb1@arinc9.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
On the switch on the MT7988 SoC, as shown in Block Diagram 8.1.1.3 on page
125 of "MT7988A Wi-Fi 7 Generation Router Platform: Datasheet (Open
Version) v0.1", there are only 4 PHYs. That's port 0 to 3. Set the case for
ports which connect to switch PHYs to '0 ... 3'.
Port 4 and 5 are not used at all in this design.
Link: https://wiki.banana-pi.org/Banana_Pi_BPI-R4#Documents [1]
Acked-by: Daniel Golle <daniel@makrotopia.org>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Signed-off-by: Arınç ÜNAL <arinc.unal@arinc9.com>
Link: https://lore.kernel.org/r/20240206-for-netnext-mt7530-improvements-2-v5-6-d7d92a185cb1@arinc9.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The pad_setup function pointer was introduced with 88bdef8be9 ("net: dsa:
mt7530: Extend device data ready for adding a new hardware"). It was being
used to set up the core clock and port 6 of the MT7530 switch, and pll of
the MT7531 switch.
All of these were moved to more appropriate locations, and it was never
used for the switch on the MT7988 SoC. Therefore, this function pointer
hasn't got a use anymore. Remove it.
Acked-by: Daniel Golle <daniel@makrotopia.org>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Reviewed-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: Arınç ÜNAL <arinc.unal@arinc9.com>
Link: https://lore.kernel.org/r/20240206-for-netnext-mt7530-improvements-2-v5-5-d7d92a185cb1@arinc9.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
mt7530_pad_clk_setup() is called if port 6 is enabled. It used to do more
things than setting up port 6. That part was moved to more appropriate
locations, mt7530_setup() and mt7530_pll_setup().
Now that all it does is set up port 6, rename it to mt7530_setup_port6(),
and move it to a more appropriate location, under mt7530_mac_config().
Change mt7530_setup_port6() to void as there're no error cases.
Leave an empty mt7530_pad_clk_setup() to satisfy the pad_setup function
pointer.
This is the code path for setting up the ports before:
dsa_switch_ops :: phylink_mac_config() -> mt753x_phylink_mac_config()
-> mt753x_mac_config()
-> mt753x_info :: mac_port_config() -> mt7530_mac_config()
-> mt7530_setup_port5()
-> mt753x_pad_setup()
-> mt753x_info :: pad_setup() -> mt7530_pad_clk_setup()
This is after:
dsa_switch_ops :: phylink_mac_config() -> mt753x_phylink_mac_config()
-> mt753x_mac_config()
-> mt753x_info :: mac_port_config() -> mt7530_mac_config()
-> mt7530_setup_port5()
-> mt7530_setup_port6()
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Reviewed-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: Arınç ÜNAL <arinc.unal@arinc9.com>
Link: https://lore.kernel.org/r/20240206-for-netnext-mt7530-improvements-2-v5-4-d7d92a185cb1@arinc9.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
This code is from before this driver was converted to phylink API. Phylink
deals with the unsupported interface cases before mt7530_pad_clk_setup() is
run. Therefore, the default case would never run. However, it must be
defined nonetheless to handle all the remaining enumeration values, the
phy-modes.
Switch to if statement for RGMII and return which simplifies the code and
saves an indent.
Set P6_INTF_MODE, which is the three least significant bits of the
MT7530_P6ECR register, to 0 for RGMII even though it will already be 0
after reset. This is to keep supporting dynamic reconfiguration of the port
in the case the interface changes from TRGMII to RGMII.
Disable the TRGMII clocks for all cases. They will be enabled if TRGMII is
being used.
Read XTAL after checking for RGMII as it's only needed for the TRGMII
interface mode.
Reviewed-by: Daniel Golle <daniel@makrotopia.org>
Reviewed-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: Arınç ÜNAL <arinc.unal@arinc9.com>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Link: https://lore.kernel.org/r/20240206-for-netnext-mt7530-improvements-2-v5-3-d7d92a185cb1@arinc9.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The crystal frequency concerns the switch core. The frequency should be
checked when the switch is being set up so the driver can reject the
unsupported hardware earlier and without requiring port 6 to be used.
Move it to mt7530_setup(). Drop the unnecessary function printing.
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Reviewed-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: Arınç ÜNAL <arinc.unal@arinc9.com>
Link: https://lore.kernel.org/r/20240206-for-netnext-mt7530-improvements-2-v5-2-d7d92a185cb1@arinc9.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
There're two code paths for setting up port 5:
mt7530_setup()
-> mt7530_setup_port5()
mt753x_phylink_mac_config()
-> mt753x_mac_config()
-> mt7530_mac_config()
-> mt7530_setup_port5()
On the first code path, priv->p5_intf_sel is either set to
P5_INTF_SEL_PHY_P0 or P5_INTF_SEL_PHY_P4 when mt7530_setup_port5() is run.
On the second code path, priv->p5_intf_sel is set to P5_INTF_SEL_GMAC5 when
mt7530_setup_port5() is run.
Empty the default case which will never run but is needed nonetheless to
handle all the remaining enumeration values.
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Reviewed-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: Arınç ÜNAL <arinc.unal@arinc9.com>
Link: https://lore.kernel.org/r/20240206-for-netnext-mt7530-improvements-2-v5-1-d7d92a185cb1@arinc9.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
After 1c75c424bd43 ("leds: class: If no default trigger is given, make
hw_control trigger the default trigger") this line isn't needed any
longer.
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/r/3a9cd1a1-40ad-487d-8b1e-6bf255419232@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
1) IPSec global stats for xfrm and mlx5
2) XSK memory improvements for non-linear SKBs
3) Software steering debug dump to use seq_file ops
4) Various code clean-ups
-----BEGIN PGP SIGNATURE-----
iQEzBAABCAAdFiEEGhZs6bAKwk/OTgTpSD+KveBX+j4FAmXBgUEACgkQSD+KveBX
+j4qmQgAr5pNBUkxyylEKU0cB8kh21LaOzimVjpGrKCzPtb/Fz8h6E1h668icpFy
2G4iuyJ2/doA0Bn4FJC730i6Xk2sqkXiY85yclvMPObS7b0PckD72d6iH9UfzfYB
Vof1lg3HdVvTq1KhJ1T/sCEB2skBLzEGV/fZWmL39zPaxM9JfHA2o5xDxdW7/GhT
0LPg84MinikoKbfw9m45l2zp52yTRrLzTUbDubwXFC4Ltg/yM4uAsbTArQzA8VJ9
q95XGZ0ZmZDFsVbjp9suCJFaTVrKmh2yy16a3BpKpm72L+K8yciUTqcqDd5vH2wW
CQ0eKMqBtB8c1Jr16oKfQ42NFF0FIQ==
=0IjR
-----END PGP SIGNATURE-----
Merge tag 'mlx5-updates-2024-02-01' of git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux
Saeed Mahameed says:
====================
mlx5-updates-2024-02-01
1) IPSec global stats for xfrm and mlx5
2) XSK memory improvements for non-linear SKBs
3) Software steering debug dump to use seq_file ops
4) Various code clean-ups
* tag 'mlx5-updates-2024-02-01' of git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux:
net/mlx5e: XDP, Exclude headroom and tailroom from memory calculations
net/mlx5e: XSK, Exclude tailroom from non-linear SKBs memory calculations
net/mlx5: DR, Change SWS usage to debug fs seq_file interface
net/mlx5: Change missing SyncE capability print to debug
net/mlx5: Remove initial segmentation duplicate definitions
net/mlx5: Return specific error code for timeout on wait_fw_init
net/mlx5: SF, Stop waiting for FW as teardown was called
net/mlx5: remove fw reporter dump option for non PF
net/mlx5: remove fw_fatal reporter dump option for non PF
net/mlx5: Rename mlx5_sf_dev_remove
Documentation: Fix counter name of mlx5 vnic reporter
net/mlx5e: Delete obsolete IPsec code
net/mlx5e: Connect mlx5 IPsec statistics with XFRM core
xfrm: get global statistics from the offloaded device
xfrm: generalize xdo_dev_state_update_curlft to allow statistics update
====================
Link: https://lore.kernel.org/r/20240206005527.1353368-1-saeed@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Hangbin Liu says:
====================
selftests: bonding: use slowwait when waiting
There are a lot waitings in bonding tests use sleep. Let's replace them with
slowwait(added in the first patch). This could save much test time. e.g.
bond-break-lacpdu-tx.sh
before: 0m16.346s
after: 0m2.824s
bond_options.sh
before: 9m25.299s
after: 6m14.439s
bond-lladdr-target.sh
before: 0m7.090s
after: 0m6.148s
In total, we could save about 180 seconds.
====================
Link: https://lore.kernel.org/r/20240205130048.282087-1-liuhangbin@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Use slowwait instead of hard code sleep for bonding tests.
In function setup_prepare(), the client_create() will be called after
server_create(). So I think there is no need to sleep in server_create()
and remove it.
For lab_lib.sh, remove bonding module may affect other running bonding tests.
And some test env may buildin bond which can't be removed. The bonding
link should be removed by lag_reset_network() or netns delete.
Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Link: https://lore.kernel.org/r/20240205130048.282087-5-liuhangbin@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The purpose of grat_arp is testing commit 9949e2efb5 ("bonding: fix
send_peer_notif overflow"). As the send_peer_notif was defined to u8,
to overflow it, we need to
send_peer_notif = num_peer_notif * peer_notif_delay = num_grat_arp * peer_notify_delay / miimon > 255
(kernel) (kernel parameter) (user parameter)
e.g. 30 (num_grat_arp) * 1000 (peer_notify_delay) / 100 (miimon) > 255.
Which need 30s to complete sending garp messages. To save the testing time,
the only way is reduce the miimon number. Something like
30 (num_grat_arp) * 100 (peer_notify_delay) / 10 (miimon) > 255.
To save more time, the 50 num_grat_arp testing could be removed.
The arp_validate_test also need to check the mii_status, which sleep
too long. Use slowwait to save some time.
For other connection checkings, make sure active slave changed first.
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Link: https://lore.kernel.org/r/20240205130048.282087-4-liuhangbin@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Use tc filter to check if LACP was sent, which is accurate and save
more time.
No need to remove bonding module as some test env may buildin bonding.
And the bond link has been deleted.
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Link: https://lore.kernel.org/r/20240205130048.282087-3-liuhangbin@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Add slowwait functions to wait for some operations that may need a long time
to finish. The busywait executes the cmd too fast, which is kind of wasting
cpu in this scenario. At the same time, if shell debugging is enabled with
`set -x`. the busywait will output too much logs.
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Link: https://lore.kernel.org/r/20240205130048.282087-2-liuhangbin@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
According to latest release of SMCv2.1[1], the term 'virtual ISM' has
been changed to 'Emulated-ISM' to avoid the ambiguity of the word
'virtual' in different contexts. So the names or comments in the code
need be modified accordingly.
[1] https://www.ibm.com/support/pages/node/7112343
Signed-off-by: Wen Gu <guwen@linux.alibaba.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Reviewed-by: Wenjia Zhang <wenjia@linux.ibm.com>
Link: https://lore.kernel.org/r/20240205033317.127269-1-guwen@linux.alibaba.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
ksmbd_iov_pin_rsp_read() doesn't free the provided aux buffer if it
fails. Seems to be the caller's responsibility to clear the buffer in
error case.
Found by Linux Verification Center (linuxtesting.org).
Fixes: e2b76ab8b5 ("ksmbd: add support for read compound")
Cc: stable@vger.kernel.org
Signed-off-by: Fedor Pchelkin <pchelkin@ispras.ru>
Acked-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
The ksmbd_extract_sharename() function lacked a complete kernel-doc
comment. This patch adds parameter descriptions and detailed function
behavior to improve code readability and maintainability.
Signed-off-by: Yang Li <yang.lee@linux.alibaba.com>
Acked-by: Randy Dunlap <rdunlap@infradead.org>
Acked-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
Heiner Kallweit says:
====================
net: phy: realtek: complete 5Gbps support and replace private constants
Realtek maps standard C45 registers to vendor-specific registers which
can be accessed via C22 w/o MMD. For an unknown reason C22 MMD access
to C45 registers isn't supported for integrated PHY's.
However the vendor-specific registers preserve the format of the C45
registers, so we can use standard constants. First two patches are
cherry-picked from a series posted by Marek some time ago.
RTL8126 supports 5Gbps, therefore add the missing 5Gbps support to
rtl822x_config_aneg().
====================
Link: https://lore.kernel.org/r/31a83fd9-90ce-402a-84c7-d5c20540b730@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
RTL8126 as an evolution of RTL8125 supports 5Gbps. rtl822x_config_aneg()
is used by the PHY driver for the integrated PHY, therefore add 5Gbps
support to it.
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Link: https://lore.kernel.org/r/5644ab50-e3e9-477c-96db-05cd5bdc2563@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Drop the ad-hoc MDIO constants used in the driver and use generic
constants instead.
Signed-off-by: Marek Behún <kabel@kernel.org>
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/r/732a70d6-4191-4aae-8862-3716b062aa9e@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
