Skip to content
GitLab
  1. Jul 18, 2022
  3. Jul 16, 2022
  5. Jul 15, 2022
    • David S. Miller's avatar
      Merge branch 'stmmac-dwmac-mediatec-clock-fix' · 8f3184b9
      David S. Miller authored 
      

Biao Huang says:

====================
stmmac: dwmac-mediatek: fix clock issue

changes in v5:
1. add reivewd-by as Matthias's comments.
2. fix "warning: unused variable 'ret' [-Wunused-variable]" as Jakub's comments

changes in v4:
1. improve commit message and test ko insertion/remove as Matthias's comments.
2. add patch "net: stmmac: fix pm runtime issue in stmmac_dvr_remove()" to
   fix vlan filter deletion issue.
3. add patch "net: stmmac: fix unbalanced ptp clock issue in suspend/resume flow"
   to fix unbalanced ptp clock issue in suspend/resume flow.

changes in v3:
1. delete mediatek_dwmac_exit() since there is no operation in it,
as Matthias's comments.

changes in v2:
1. clock configuration is still needed in probe,
and invoke mediatek_dwmac_clks_config() instead.
2. update commit message.

v1:
remove duplicated clock configuration in init/exit.
====================

Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      8f3184b9
    • Biao Huang's avatar
      net: stmmac: fix unbalanced ptp clock issue in suspend/resume flow · f4c7d894
      Biao Huang authored 
      Current stmmac driver will prepare/enable ptp_ref clock in
stmmac_init_tstamp_counter().

The stmmac_pltfr_noirq_suspend will disable it once in suspend flow.

But in resume flow,
	stmmac_pltfr_noirq_resume --> stmmac_init_tstamp_counter
	stmmac_resume --> stmmac_hw_setup --> stmmac_init_ptp --> stmmac_init_tstamp_counter
ptp_ref clock reference counter increases twice, which leads to unbalance
ptp clock when resume back.

Move ptp_ref clock prepare/enable out of stmmac_init_tstamp_counter to fix it.

Fixes: 0735e639

 ("net: stmmac: skip only stmmac_ptp_register when resume from suspend")
Signed-off-by: default avatarBiao Huang <biao.huang@mediatek.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      f4c7d894
    • Biao Huang's avatar
      net: stmmac: fix pm runtime issue in stmmac_dvr_remove() · 0d9a1591
      Biao Huang authored 
      If netif is running when stmmac_dvr_remove is invoked,
the unregister_netdev will call ndo_stop(stmmac_release) and
vlan_kill_rx_filter(stmmac_vlan_rx_kill_vid).

Currently, stmmac_dvr_remove() will disable pm runtime before
unregister_netdev. When stmmac_vlan_rx_kill_vid is invoked,
pm_runtime_resume_and_get in it returns EACCESS error number,
and reports:

	dwmac-mediatek 11021000.ethernet eth0: stmmac_dvr_remove: removing driver
	dwmac-mediatek 11021000.ethernet eth0: FPE workqueue stop
	dwmac-mediatek 11021000.ethernet eth0: failed to kill vid 0081/0

Move the pm_runtime_disable to the end of stmmac_dvr_remove
to fix this issue.

Fixes: 64495203

 ("net: stmmac: properly handle with runtime pm in stmmac_dvr_remove()")
Signed-off-by: default avatarBiao Huang <biao.huang@mediatek.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      0d9a1591
    • Biao Huang's avatar
      stmmac: dwmac-mediatek: fix clock issue · fa4b3ca6
      Biao Huang authored 
      The pm_runtime takes care of the clock handling in current
stmmac drivers, and dwmac-mediatek implement the
mediatek_dwmac_clks_config() as the callback for pm_runtime.

Then, stripping duplicated clocks handling in old init()/exit()
to fix clock issue in suspend/resume test.

As to clocks in probe/remove, vendor need symmetric handling to
ensure clocks balance.

Test pass, including suspend/resume and ko insertion/remove.

Fixes: 3186bdad

 ("stmmac: dwmac-mediatek: add platform level clocks management")
Signed-off-by: default avatarBiao Huang <biao.huang@mediatek.com>
Reviewed-by: default avatarMatthias Brugger <matthias.bgg@gmail.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      fa4b3ca6
    • David S. Miller's avatar
      Merge branch 'net-sysctl-races-round2' · 782d86fe
      David S. Miller authored 
      

Kuniyuki Iwashima says:

====================
sysctl: Fix data-races around ipv4_net_table (Round 2).

This series fixes data-races around 15 knobs after ip_default_ttl in
ipv4_net_table.

These two knobs are skipped.
  - ip_local_port_range is safe with its own lock.
  - ip_local_reserved_ports uses proc_do_large_bitmap(), which will need
    an additional lock and can be fixed later.

So, the next round will start with igmp_link_local_mcast_reports.
====================

Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      782d86fe
    • Kuniyuki Iwashima's avatar
      tcp: Fix a data-race around sysctl_tcp_probe_interval. · 2a85388f
      Kuniyuki Iwashima authored 
      While reading sysctl_tcp_probe_interval, it can be changed concurrently.
Thus, we need to add READ_ONCE() to its reader.

Fixes: 05cbc0db

 ("ipv4: Create probe timer for tcp PMTU as per RFC4821")
Signed-off-by: default avatarKuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      2a85388f
    • Kuniyuki Iwashima's avatar
      tcp: Fix a data-race around sysctl_tcp_probe_threshold. · 92c0aa41
      Kuniyuki Iwashima authored 
      While reading sysctl_tcp_probe_threshold, it can be changed concurrently.
Thus, we need to add READ_ONCE() to its reader.

Fixes: 6b58e0a5

 ("ipv4: Use binary search to choose tcp PMTU probe_size")
Signed-off-by: default avatarKuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      92c0aa41
    • Kuniyuki Iwashima's avatar
      tcp: Fix a data-race around sysctl_tcp_mtu_probe_floor. · 8e92d442
      Kuniyuki Iwashima authored 
      While reading sysctl_tcp_mtu_probe_floor, it can be changed concurrently.
Thus, we need to add READ_ONCE() to its reader.

Fixes: c04b79b6

 ("tcp: add new tcp_mtu_probe_floor sysctl")
Signed-off-by: default avatarKuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      8e92d442