Skip to content

K
Kernel

openeuler / Kernel
1 年多 前同步成功

通知 8
Star 0
Fork 0
K
Kernel
切换分支/标签
  1. 09 12月, 2017 1 次提交
  3. 02 11月, 2017 1 次提交
  5. 01 11月, 2017 4 次提交
  7. 21 10月, 2017 2 次提交
  9. 20 10月, 2017 2 次提交
  11. 15 10月, 2017 2 次提交
  13. 04 10月, 2017 1 次提交
  15. 26 9月, 2017 1 次提交
  17. 23 9月, 2017 3 次提交
  19. 22 9月, 2017 1 次提交
  21. 21 8月, 2017 1 次提交
  23. 19 8月, 2017 1 次提交
  25. 15 8月, 2017 1 次提交
  27. 12 8月, 2017 1 次提交
  29. 08 8月, 2017 4 次提交
  31. 21 7月, 2017 1 次提交
  33. 12 7月, 2017 1 次提交
    • G
      cxgb4: fix BUG() on interrupt deallocating path of ULD · 6a146f3a
      由 Guilherme G. Piccoli 提交于 
      Since the introduction of ULD (Upper-Layer Drivers), the MSI-X
deallocating path changed in cxgb4: the driver frees the interrupts
of ULD when unregistering it or on shutdown PCI handler.

Problem is that if a MSI-X is not freed before deallocated in the PCI
layer, it will trigger a BUG() due to still "alive" interrupt being
tentatively quiesced.

The below trace was observed when doing a simple unbind of Chelsio's
adapter PCI function, like:
  "echo 001e:80:00.4 > /sys/bus/pci/drivers/cxgb4/unbind"

Trace:

  kernel BUG at drivers/pci/msi.c:352!
  Oops: Exception in kernel mode, sig: 5 [#1]
  ...
  NIP [c0000000005a5e60] free_msi_irqs+0xa0/0x250
  LR [c0000000005a5e50] free_msi_irqs+0x90/0x250
  Call Trace:
  [c0000000005a5e50] free_msi_irqs+0x90/0x250 (unreliable)
  [c0000000005a72c4] pci_disable_msix+0x124/0x180
  [d000000011e06708] disable_msi+0x88/0xb0 [cxgb4]
  [d000000011e06948] free_some_resources+0xa8/0x160 [cxgb4]
  [d000000011e06d60] remove_one+0x170/0x3c0 [cxgb4]
  [c00000000058a910] pci_device_remove+0x70/0x110
  [c00000000064ef04] device_release_driver_internal+0x1f4/0x2c0
  ...

This patch fixes the issue by refactoring the shutdown path of ULD on
cxgb4 driver, by properly freeing and disabling interrupts on PCI
remove handler too.

Fixes: 0fbc81b3 ("Allocate resources dynamically for all cxgb4 ULD's")
Reported-by: NHarsha Thyagaraja <hathyaga@in.ibm.com>
Signed-off-by: NGuilherme G. Piccoli <gpiccoli@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      6a146f3a
  35. 05 7月, 2017 2 次提交
  37. 24 6月, 2017 1 次提交
    • A
      cxgb4: Update T6 Buffer Group and Channel Mappings · 193c4c28
      由 Arjun Vynipadath 提交于 
      We were using t4_get_mps_bg_map() for both t4_get_port_stats()
to determine which MPS Buffer Groups to report statistics on for a given
Port, and also for t4_sge_alloc_rxq() to provide a TP Ingress Channel
Congestion Map.  For T4/T5 these are actually the same values (because they
are ~somewhat~ related), but for T6 they should return different values
(T6 has Port 0 associated with MPS Buffer Group 0 (with MPS Buffer Group 1
silently cascading off) and Port 1 is associated with MPS Buffer Group 2
(with 3 cascading off)).

Based on the original work by Casey Leedom <leedom@chelsio.com>
Signed-off-by: NArjun Vynipadath <arjun@chelsio.com>
Signed-off-by: NGanesh Goudar <ganeshgr@chelsio.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      193c4c28
  39. 19 6月, 2017 3 次提交
    • R
      cxgb4: notify uP to route ctrlq compl to rdma rspq · dec6b331
      由 Raju Rangoju 提交于 
      During the module initialisation there is a possible race
(basically race between uld and lld) where neither the uld
nor lld notifies the uP about where to route the ctrl queue
completions. LLD skips notifying uP as the rdma queues were
not created by then (will leave it to ULD to notify the uP).
As the ULD comes up, it also skips notifying the uP as the
flag FULL_INIT_DONE is not set yet (ULD assumes that the
interface is not up yet).

Consequently, this race between uld and lld leaves uP
unnotified about where to send the ctrl queue completions
to, leading to iwarp RI_RES WR failure.

Here is the race:

CPU 0                                   CPU1

- allocates nic rx queus
- t4_sge_alloc_ctrl_txq()
(if rdma rsp queues exists,
tell uP to route ctrl queue
compl to rdma rspq)
                                - acquires the mutex_lock
                                - allocates rdma response queues
                                - if FULL_INIT_DONE set,
                                  tell uP to route ctrl queue compl
                                  to rdma rspq
                                - relinquishes mutex_lock
- acquires the mutex_lock
- enable_rx()
- set FULL_INIT_DONE
- relinquishes mutex_lock

This patch fixes the above issue.

Fixes: e7519f99('cxgb4: avoid enabling napi twice to the same queue')
Signed-off-by: NRaju Rangoju <rajur@chelsio.com>
Acked-by: NSteve Wise <swise@opengridcomputing.com>
CC: Stable <stable@vger.kernel.org> # 4.9+
Signed-off-by: NGanesh Goudar <ganeshgr@chelsio.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      dec6b331
    • R
      cxgb4: notify uP to route ctrlq compl to rdma rspq · 91060381
      由 Raju Rangoju 提交于 
      During the module initialisation there is a possible race
(basically race between uld and lld) where neither the uld
nor lld notifies the uP about where to route the ctrl queue
completions. LLD skips notifying uP as the rdma queues were
not created by then (will leave it to ULD to notify the uP).
As the ULD comes up, it also skips notifying the uP as the
flag FULL_INIT_DONE is not set yet (ULD assumes that the
interface is not up yet).

Consequently, this race between uld and lld leaves uP
unnotified about where to send the ctrl queue completions
to, leading to iwarp RI_RES WR failure.

Here is the race:

CPU 0                                   CPU1

- allocates nic rx queus
- t4_sge_alloc_ctrl_txq()
(if rdma rsp queues exists,
tell uP to route ctrl queue
compl to rdma rspq)
                                - acquires the mutex_lock
                                - allocates rdma response queues
                                - if FULL_INIT_DONE set,
                                  tell uP to route ctrl queue compl
                                  to rdma rspq
                                - relinquishes mutex_lock
- acquires the mutex_lock
- enable_rx()
- set FULL_INIT_DONE
- relinquishes mutex_lock

This patch fixes the above issue.

Fixes: e7519f99('cxgb4: avoid enabling napi twice to the same queue')
Signed-off-by: NRaju Rangoju <rajur@chelsio.com>
Acked-by: NSteve Wise <swise@opengridcomputing.com>
CC: Stable <stable@vger.kernel.org> # 4.9+
Signed-off-by: NGanesh Goudar <ganeshgr@chelsio.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      91060381
    • G
      cxgb4: fix a NULL dereference · d427caee
      由 Ganesh Goudar 提交于 
      Avoid NULL dereference in setup_sge_queues() when the adapter is
in non offload mode.

Fixes: 0fbc81b3 ('chcr/cxgb4i/cxgbit/RDMA/cxgb4: Allocate resources dynamically for all cxgb4 ULD's')
Signed-off-by: NGanesh Goudar <ganeshgr@chelsio.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      d427caee
  41. 16 6月, 2017 1 次提交
    • J
      networking: make skb_put & friends return void pointers · 4df864c1
      由 Johannes Berg 提交于 
      It seems like a historic accident that these return unsigned char *,
and in many places that means casts are required, more often than not.

Make these functions (skb_put, __skb_put and pskb_put) return void *
and remove all the casts across the tree, adding a (u8 *) cast only
where the unsigned char pointer was used directly, all done with the
following spatch:

    @@
    expression SKB, LEN;
    typedef u8;
    identifier fn = { skb_put, __skb_put };
    @@
    - *(fn(SKB, LEN))
    + *(u8 *)fn(SKB, LEN)

    @@
    expression E, SKB, LEN;
    identifier fn = { skb_put, __skb_put };
    type T;
    @@
    - E = ((T *)(fn(SKB, LEN)))
    + E = fn(SKB, LEN)

which actually doesn't cover pskb_put since there are only three
users overall.

A handful of stragglers were converted manually, notably a macro in
drivers/isdn/i4l/isdn_bsdcomp.c and, oddly enough, one of the many
instances in net/bluetooth/hci_sock.c. In the former file, I also
had to fix one whitespace problem spatch introduced.
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      4df864c1
  43. 10 6月, 2017 1 次提交
  45. 08 6月, 2017 3 次提交
    • J
      net: propagate tc filter chain index down the ndo_setup_tc call · a5fcf8a6
      由 Jiri Pirko 提交于 
      We need to push the chain index down to the drivers, so they have the
information to which chain the rule belongs. For now, no driver supports
multichain offload, so only chain 0 is supported. This is needed to
prevent chain squashes during offload for now. Later this will be used
to implement multichain offload.
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      a5fcf8a6
    • D
      net: Fix inconsistent teardown and release of private netdev state. · cf124db5
      由 David S. Miller 提交于 
      Network devices can allocate reasources and private memory using
netdev_ops->ndo_init().  However, the release of these resources
can occur in one of two different places.

Either netdev_ops->ndo_uninit() or netdev->destructor().

The decision of which operation frees the resources depends upon
whether it is necessary for all netdev refs to be released before it
is safe to perform the freeing.

netdev_ops->ndo_uninit() presumably can occur right after the
NETDEV_UNREGISTER notifier completes and the unicast and multicast
address lists are flushed.

netdev->destructor(), on the other hand, does not run until the
netdev references all go away.

Further complicating the situation is that netdev->destructor()
almost universally does also a free_netdev().

This creates a problem for the logic in register_netdevice().
Because all callers of register_netdevice() manage the freeing
of the netdev, and invoke free_netdev(dev) if register_netdevice()
fails.

If netdev_ops->ndo_init() succeeds, but something else fails inside
of register_netdevice(), it does call ndo_ops->ndo_uninit().  But
it is not able to invoke netdev->destructor().

This is because netdev->destructor() will do a free_netdev() and
then the caller of register_netdevice() will do the same.

However, this means that the resources that would normally be released
by netdev->destructor() will not be.

Over the years drivers have added local hacks to deal with this, by
invoking their destructor parts by hand when register_netdevice()
fails.

Many drivers do not try to deal with this, and instead we have leaks.

Let's close this hole by formalizing the distinction between what
private things need to be freed up by netdev->destructor() and whether
the driver needs unregister_netdevice() to perform the free_netdev().

netdev->priv_destructor() performs all actions to free up the private
resources that used to be freed by netdev->destructor(), except for
free_netdev().

netdev->needs_free_netdev is a boolean that indicates whether
free_netdev() should be done at the end of unregister_netdevice().

Now, register_netdevice() can sanely release all resources after
ndo_ops->ndo_init() succeeds, by invoking both ndo_ops->ndo_uninit()
and netdev->priv_destructor().

And at the end of unregister_netdevice(), we invoke
netdev->priv_destructor() and optionally call free_netdev().
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      cf124db5
    • G
      cxgb4: Fix tids count for ipv6 offload connection · 1dec4cec
      由 Ganesh Goudar 提交于 
      the adapter consumes two tids for every ipv6 offload
connection be it active or passive, calculate tid usage
count accordingly.

Also change the signatures of relevant functions to get
the address family.
Signed-off-by: NRizwan Ansari <rizwana@chelsio.com>
Signed-off-by: NVarun Prakash <varun@chelsio.com>
Signed-off-by: NGanesh Goudar <ganeshgr@chelsio.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      1dec4cec
  47. 07 6月, 2017 1 次提交