]> www.infradead.org Git - users/jedix/linux-maple.git/log
users/jedix/linux-maple.git
9 years agoi40e/i40evf: Add a new offload for RSS PCTYPE V2 for X722
Anjali Singhai Jain [Thu, 19 Nov 2015 19:34:22 +0000 (11:34 -0800)]
i40e/i40evf: Add a new offload for RSS PCTYPE V2 for X722

Orabug: 22342532

X722 supports Expanded version of TCP, UDP PCTYPES for RSS.
Add a Virtchnl offload to support this.

Without this patch with X722 devices, driver will set wrong PCTYPES
for VF and UDP flows will not fan out.

Change-ID: I04fe4988253b7cd108c9179a643c969764efcb76
Signed-off-by: Anjali Singhai Jain <anjali.singhai@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit b9eacec3e66cd7e3db11882474024d1586d1f832)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: hush little warnings
Mitch Williams [Thu, 19 Nov 2015 19:34:21 +0000 (11:34 -0800)]
i40e: hush little warnings

Orabug: 22342532

These messages seem big and scary, but they're really not. The driver
can fully recover from any of these. The overflow error in particular
can happen when enabling a bunch of VFs and the VF driver is not
blacklisted.

Since these messages are really for debugging purposes, reclassify
them as such.

Change-ID: I628d0f5e135e7063450ba05393a50b7af23aa6d7
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 75eb73c127beaccf1b2c825f5d583ebca17ab2ba)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: Opcode and structures required by OEM Post Update AQ command and add new NVM...
Michal Kosiarz [Thu, 19 Nov 2015 19:34:20 +0000 (11:34 -0800)]
i40e: Opcode and structures required by OEM Post Update AQ command and add new NVM arq message

Orabug: 22342532

This is a part of implementation which contains data structures and
opcode for new AQ command. There's a new ARQ message that gets sent
near the end of the NVM update process that the driver should recognize
and ignore, rather than printing an Unknown Event error.

Change-ID: I04830a5bcae14823e16b9424cc4165e169336c1f
Signed-off-by: Michal Kosiarz <michal.kosiarz@intel.com>
Acked-by: Shannon Nelson <shannon.nelson@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 00ada50d8989ed14b160d831d25fe09bf9ea9115)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: use explicit cast from u16 to u8
Kamil Krawczyk [Thu, 19 Nov 2015 19:34:19 +0000 (11:34 -0800)]
i40e: use explicit cast from u16 to u8

Orabug: 22342532

Current implementation generates compilation warnings.

Change-ID: Icceefb50fe62aefaf90a64afb7192e08355a4ec5
Signed-off-by: Kamil Krawczyk <kamil.krawczyk@intel.com>
Acked-by: Shannon Nelson <shannon.nelson@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 6621e4b2518a2ea5e1f3aab041bfbb0e4fc91239)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40evf: check rings before freeing resources
Mitch Williams [Thu, 19 Nov 2015 19:34:18 +0000 (11:34 -0800)]
i40evf: check rings before freeing resources

Orabug: 22342532

If the driver gets unloaded during reset recovery, it's possible
that it will attempt to free resources when they're already free.

Add a check to make sure that the Tx and Rx rings actually exist
before dereferencing them to free resources.

Change-ID: I4d2b7e9ede49f634d421a4c5deaa5446bc755eee
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit fdb47ae87af537b24977a03bc69cfe1c5c55ca62)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: don't add zero MAC filter
Mitch Williams [Thu, 19 Nov 2015 19:34:17 +0000 (11:34 -0800)]
i40e: don't add zero MAC filter

Orabug: 22342532

When VFs are created, the MAC address defaults to all zeros, indicating
to the VF driver that it should use a random MAC address. However, the
PF driver was incorrectly adding this zero MAC to the filter table,
along with the VF's randomly generated MAC address.

Check for a good address before adding the default filter. While we're
at it, make the error message a bit more useful.

Change-ID: Ia100947d68140e0f73a19ba755cbffc3e79a8fcf
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit b7b713a8eaf325607d37229f024ad0b9f3e7f320)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: properly delete VF MAC filters
Mitch Williams [Thu, 19 Nov 2015 19:34:16 +0000 (11:34 -0800)]
i40e: properly delete VF MAC filters

Orabug: 22342532

The virtual channel interface was using incorrect semantics to remove
MAC addresses, which would leave incorrect filters active when using
VLANs. To correct this, add a new function that unconditionally removes
MAC addresses from all VLANs, and call this function when the VF
requests a MAC filter removal.

Change-ID: I69826908ae4f6c847f5bf9b32f11faa760189c74
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit b36e9ab59b7e3a5b14bf88dc0536e6579db7b54d)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: chomp the BIT(_ULL)
Jesse Brandeburg [Thu, 19 Nov 2015 19:34:14 +0000 (11:34 -0800)]
i40e: chomp the BIT(_ULL)

Orabug: 22342532

BIT_ULL was used on a u32 or less where it can simply be BIT. This
fixes some trivial static analyzer warnings. Chomp, chomp.

Tested with objdump of binary before and after, no changes to code.

Change-ID: I6245e9abd447192dbde1669c747aeb2878126c7d
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 75f5cea9ddef6a8f8cedb86c0ac8c696b258824d)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoRevert "i40e: remove CONFIG_I40E_VXLAN"
Singhai, Anjali [Fri, 4 Dec 2015 07:49:31 +0000 (23:49 -0800)]
Revert "i40e: remove CONFIG_I40E_VXLAN"

Orabug: 22342532

This reverts commit 8fe269991aece394a7ed274f525d96c73f94109a.
The case where VXLAN is a module and i40e driver is inbuilt
will not be handled properly with this change since i40e
will have an undefined symbol vxlan_get_rx_port in it.

v2: Add a signed-off-by.

Signed-off-by: Anjali Singhai Jain <anjali.singhai@intel.com>
Acked-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
(cherry picked from commit 6633d3815c099a0aaf28e4853f7a8994331b8c05)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: trivial fixes
Jesse Brandeburg [Thu, 19 Nov 2015 01:35:42 +0000 (17:35 -0800)]
i40e: trivial fixes

Orabug: 22342532

1) remove duplicate include of tcp.h
2) put an ampersand at the end of a line instead of the beginning
3) remove a useless dev_info
4) match declaration of function to the implementation
5) repair incorrect comment
6) correct whitespace
7) remove unused define

Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 4eeb1fff27f4578386955b0e88f4cffc3814df24)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e/i40evf: Bump version to 1.4.7 for i40e and 1.4.3 for i40evf
Catherine Sullivan [Fri, 6 Nov 2015 23:26:12 +0000 (15:26 -0800)]
i40e/i40evf: Bump version to 1.4.7 for i40e and 1.4.3 for i40evf

Orabug: 22342532

Bump.

Change-ID: Id8c83c64c973349a722bab40d285ad8ded8c28f7
Signed-off-by: Catherine Sullivan <catherine.sullivan@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 1e590660cb7ac675c735dd74d99dbca2103abf41)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e/i40evf: use logical operator
Mitch Williams [Fri, 6 Nov 2015 23:26:11 +0000 (15:26 -0800)]
i40e/i40evf: use logical operator

Orabug: 22342532

We shouldn't be using a bitwise operator here; it's not a bitwise
operation. Use a logical operator instead. Why doesn't c have a
logical-or-and-assign operator?

Change-ID: Id84f3ca884910bed7073c84b1e16a102e958d0de
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 44cdb791ae7bc40e0331b78a73992be44079a3ec)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: fix whitespace
Mitch Williams [Fri, 6 Nov 2015 23:26:10 +0000 (15:26 -0800)]
i40e: fix whitespace

Orabug: 22342532

Operators should have spaces around them.

Change-ID: I64735e9aa8618b9a5059a87ace1c999d6d3bfcfb
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit b82bc49ede1f012d37ba8777e8232cd7c01365cd)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: Remove separate functions gathering XOFF Rx stats
Neerav Parikh [Fri, 6 Nov 2015 23:26:09 +0000 (15:26 -0800)]
i40e: Remove separate functions gathering XOFF Rx stats

Orabug: 22342532

The separate functions to gather Flow control Rx XOFF stats was to
determine if the Tx for a queue was paused due to Link Flow Control(LFC)
or Priority Flow Control(PFC).

But, with recent change in the i40e driver the logic for checking th Tx
hang has been removed and these functions don't do anything meaningful.
Hence, there is no need to keep these separate functions to gather Rx
XOFF stats for LFC or PFC.

This patch removes these functions and moves the stat collection for
XOFF Rx to the i40e_update_pf_stats() that collects all the PF stats.

Change-ID: Iec1452dac3a6766f0d968e754cb407530d7c60cd
Signed-off-by: Neerav Parikh <neerav.parikh@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 95db239f4fc664d9ce4ba9a6a3e87d2a3d856a23)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: remove CONFIG_I40E_VXLAN
Jesse Brandeburg [Fri, 6 Nov 2015 23:26:08 +0000 (15:26 -0800)]
i40e: remove CONFIG_I40E_VXLAN

Orabug: 22342532

Instead of having our own custom symbol, we can just rely
on whether or not the kernel has the feature enabled.

In this case use IS_ENABLED(CONFIG_VXLAN) in order to handle
built-in or module in the current BKM way.

Change-ID: I5890fbb518ff8ed6bb07c3362fb0a8a829f9b241
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 8fe269991aece394a7ed274f525d96c73f94109a)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: use priv flags to control packet split
Jesse Brandeburg [Fri, 6 Nov 2015 23:26:07 +0000 (15:26 -0800)]
i40e: use priv flags to control packet split

Orabug: 22342532

Ethtool priv flags implementation to enable or disable packet split, which
is a hardware feature that inspects headers and will put headers in a
separate DMA buffer from the payload data.  The driver was automatically
choosing to enable packet split in some cases and this gives the user the
ability to turn it off/on explicitly.

to query state:
ethtool --show-priv-flags ethx

to enable:
ethtool --set-priv-flags ethx packet-split on
to disable:
ethtool --set-priv-flags ethx packet-split off

Why would anyone want this?
Because some environments benefit from header/data split in the receive
buffer, and the driver defaults to one or the other depending on
environment/kernel parameters.

Why didn't you implement a generic ethtool control for this feature?
Because Intel hardware is the only hardware that supports header/data
split.

Change-ID: I803121e1eecc9ccb2884031fd85dd1110b3af66d
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 827de39212298048e75309377c8262c4d7f2cb41)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40evf: use correct types
Mitch Williams [Fri, 6 Nov 2015 23:26:06 +0000 (15:26 -0800)]
i40evf: use correct types

Orabug: 22342532

Don't use uint32_t type the kernel. Use u32 instead. No functional
change.

Change-ID: I77bbf3b6464edaef747c7104b43534032a4dba63
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit d82acb353246d3989377761ca26f5489264a4793)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: propagate properly
Mitch Williams [Mon, 9 Nov 2015 23:35:50 +0000 (15:35 -0800)]
i40e: propagate properly

Orabug: 22342532

i40e_sync_vsi_filters() is the surly teenager of this driver. It says
it's going to report errors, but it doesn't actually do that most of the
time. And when it does, it leaves a mess.

Change this function to have a common exit point so it will properly
release the busy lock on the VSI. Propagate errors to the callers.
Finally, adjust a few callers to check for and deal with errors from
this function.

Change-ID: Ic6af4956491e72402ebb3c538a3c31a0ad7f8667
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit ea02e90b4b49782462d06a425e05c776909fbae4)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
Conflicts:
drivers/net/ethernet/intel/i40e/i40e_main.c

9 years agoi40evf: don't use atomic allocation
Mitch Williams [Fri, 6 Nov 2015 23:26:04 +0000 (15:26 -0800)]
i40evf: don't use atomic allocation

Orabug: 22342532

These allocations don't need to be at atomic level. GFP_KERNEL is fine
and they'll reduce stress on the allocator when the system is starved
for memory.

Change-ID: I3561d0399a681de0ad25291b6c848b224c1fde12
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit a85088d813a3ac44afa237113539585257127cfa)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: Fix memory leaks, sideband filter programming
Kiran Patil [Fri, 6 Nov 2015 23:26:03 +0000 (15:26 -0800)]
i40e: Fix memory leaks, sideband filter programming

Orabug: 22342532

This patch fixes the memory leak which would be seen otherwise when user
programs flow-director filter using ethtool (sideband filter programming).

When ethtool is used to program flow directory filter, 'raw_buf' gets
allocated and it is supposed to be freed as part of queue cleanup. But
check of 'tx_buffer->skb' was preventing it from being freed.

Change-ID: Ief4f0a1a32a653180498bf6e987c1b4342ab8923
Signed-off-by: Kiran Patil <kiran.patil@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit a42e7a369ea2b73a554a85dea7d6243af51cd4f0)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: Detection and recovery of TX queue hung logic moved to service_task from tx_timeout
Kiran Patil [Fri, 6 Nov 2015 23:26:02 +0000 (15:26 -0800)]
i40e: Detection and recovery of TX queue hung logic moved to service_task from tx_timeout

Orabug: 22342532

This patch contains following changes:
   - detection and recovery logic (issue SW interrupt) has been moved to
     service_task from timeout function.
   - added some more debug info from tx_timeout.

Logic to detect and recover TX queue hung is now two step process:
  - service_task detects TX queue hung and sets a bit(hung_detected) if
    it was not set.
  - if bit was set (means this is back-back hung condition detected),
    issue SW interrupt and clear the bit.
  - napi_poll clears the bit unconditionally since it cleans TX/RX queues.

Change-ID: Ieed03a48927c845a988b3ff375090bf37caeb903
Signed-off-by: Kiran Patil <kiran.patil@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 9c6c12595b73e580f103b0812837f015034d460e)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40evf: remove duplicate string
Mitch Williams [Fri, 6 Nov 2015 23:26:01 +0000 (15:26 -0800)]
i40evf: remove duplicate string

Orabug: 22342532

We already print the driver info string in probe, so don't print
it again in init. No need to repeat. No need to repeat.

Change-ID: Ief597997f580a8c54d5950e3a84c29f2075be66b
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 05281eb8e1754688ecf07a63844c56c9f76b38ad)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40evf: set real num queues
Mitch Williams [Fri, 6 Nov 2015 23:26:00 +0000 (15:26 -0800)]
i40evf: set real num queues

Orabug: 22342532

Use the helper function to set the real number of RX queues, and also
set the real number of TX queues.

Change-ID: I67982799de3f248fb4158ccdc9b1a74385f42ddd
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit e6c4cf6fb9323c0f65b8fef1c0c8c9e15b51007e)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40evf: increase max number of queues
Mitch Williams [Fri, 6 Nov 2015 23:25:59 +0000 (15:25 -0800)]
i40evf: increase max number of queues

Orabug: 22342532

Future devices will allow for more queue pairs, so allocate a netdev
that can handle them. While we're at it, get rid of the separate
MAX_TX/MAX_RX defines. Since we always get matched queue pairs, having
these makes no sense.

Change-ID: I0e3556cd9a962506e509eb7c0afa36b329e8cb51
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 1255b7a12ec4b6174a1c84f4aae4370b8ba1b1cf)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: Fix i40e_print_features() VEB mode output
Joe Perches [Thu, 3 Dec 2015 12:20:57 +0000 (04:20 -0800)]
i40e: Fix i40e_print_features() VEB mode output

Orabug: 22342532

Commit 7fd89545f337 ("i40e: remove BUG_ON from feature string building")
added defective output when I40E_FLAG_VEB_MODE_ENABLED was set in
function i40e_print_features.

Fix it.

Miscellanea:

- Remove unnecessary string variable
- Add space before not after fixed strings
- Use kmalloc not kzalloc
- Don't initialize i to 0, use result of first snprintf

Reported-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
(cherry picked from commit 3b195843f59b8aefdc15c689615745bf6758756c)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: remove unused argument
Jesse Brandeburg [Fri, 6 Nov 2015 01:01:02 +0000 (17:01 -0800)]
i40e: remove unused argument

Orabug: 22342532

With the final edition of the patches to remove sleeps from
the driver's entry points, the grab_rtnl argument is no
longer needed, so partially revert the commit that added it.

Change-ID: Ib9778476242586cc9e58b670f5f48d415cb59003
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 17652c6336fdca0f48906f0a1d9b5ac77b29e5a7)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: fix: do not sleep in netdev_ops
Jesse Brandeburg [Fri, 6 Nov 2015 01:01:01 +0000 (17:01 -0800)]
i40e: fix: do not sleep in netdev_ops

Orabug: 22342532

The driver was being called by VLAN, bonding, teaming operations
that expected to be able to hold locks like rcu_read_lock().

This causes the driver to be held to the requirement to not sleep,
and was found by the kernel debug options for checking sleep
inside critical section, and the locking validator.

Change-ID: Ibc68c835f5ffa8ffe0638ffe910a66fc5649a7f7
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 0e4425ed641f3eef67c892bc541949cd745a9ba9)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
Conflicts:
drivers/net/ethernet/intel/i40e/i40e_main.c

9 years agoi40e/i40evf: Bump i40e version to 1.4.4 and i40evf to 1.4.1
Catherine Sullivan [Mon, 26 Oct 2015 23:44:41 +0000 (19:44 -0400)]
i40e/i40evf: Bump i40e version to 1.4.4 and i40evf to 1.4.1

Orabug: 22342532

Bump.

Change-ID: I00ebbb2e5e5572f947502b8f6db4d94f666d6b14
Signed-off-by: Catherine Sullivan <catherine.sullivan@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit e1c227919548fb9d8ee771e715d2764f1d41e8e0)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40evf: allocate ring structs dynamically
Mitch Williams [Mon, 26 Oct 2015 23:44:40 +0000 (19:44 -0400)]
i40evf: allocate ring structs dynamically

Orabug: 22342532

Instead of awkwardly keeping a fixed array of pointers in the adapter
struct and then allocating ring structs individually, just keep a single
pointer and allocate a single blob for the arrays. This simplifies code,
shrinks the adapter structure, and future-proofs the driver by not
limiting the number of rings we can handle.

Change-ID: I31334ff911a6474954232cfe4bc98ccca3c769ff
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 0dd438d8ad5db85c1a805ce2244252e37d5586a0)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40evf: allocate queue vectors dynamically
Mitch Williams [Mon, 26 Oct 2015 23:44:39 +0000 (19:44 -0400)]
i40evf: allocate queue vectors dynamically

Orabug: 22342532

Change the queue_vector array from a statically-sized member of the
adapter structure to a dynamically-allocated and -sized array.

This reduces the size of the adapter structure, and allows us to support
any number of queue vectors in the future without changing the code.

Change-ID: I08dc622cb2f2ad01e832e51c1ad9b86524730693
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 7d96ba1a8b7c36d1f4a04dc40e031b6f8c677496)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40evf: quoth the VF driver, Nevermore
Mitch Williams [Mon, 26 Oct 2015 23:44:38 +0000 (19:44 -0400)]
i40evf: quoth the VF driver, Nevermore

Orabug: 22342532

If, upon a midnight dreary, the PF returns ERR_PARAM when the VF is
requesting resources, that's fatal. Either the firmware or NVM is badly,
badly misconfigured, or this VF has been disabled due to a previous VF
driver sending a bunch of bogus messages.

Either way, there is no recovery from this. Don't ponder weak and weary,
just quit.

Change-ID: I09d9f16cc4ee7fec3b57646a289d33838c1c5bf5
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit e743072fd16e6e772d5b5e4aef38873946f9a3e0)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: make error message more useful
Mitch Williams [Mon, 26 Oct 2015 23:44:37 +0000 (19:44 -0400)]
i40e: make error message more useful

Orabug: 22342532

If we get an invalid message from a VF, we should tell the user which VF
is being naughty, rather than making them guess.

Change-ID: I9252cef7baea3d8584043ed6ff12619a94e2f99c
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit e7ffb72d65c784ba445ac849bd46d2533365fae7)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: fix confusing message
Helin Zhang [Mon, 26 Oct 2015 23:44:36 +0000 (19:44 -0400)]
i40e: fix confusing message

Orabug: 22342532

This patch fixes the confusing kernel message of enabled RSS size,
by reporting it together with the hardware maximum RSS size.

Change-ID: I64864dbfbc13beccc180a7871680def1f3d5a339
Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit e36b0b111bcc651ce030e88e042f83b015dd04b8)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: Update error messaging
Carolyn Wyborny [Mon, 26 Oct 2015 23:44:35 +0000 (19:44 -0400)]
i40e: Update error messaging

Orabug: 22342532

This patch fixes an issue where adminq init failures always provided
a message that NVM was newer than expected.  This is not always the
case for init_adminq failures. Without this patch, if adminq init
fails for any reason, newer NVM message would be given.  This
problem is fixed by adding  a check for that specific error
condition and a different hopefully helpful message otherwise.

Change-ID: Iaeaebee4e398989eae40bb70f943ab66a3a521a5
Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 2b2426a760764017cb53ffae128c87b7808dcf11)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40evf: add new fields to store user configuration of RSS
Helin Zhang [Mon, 26 Oct 2015 23:44:34 +0000 (19:44 -0400)]
i40evf: add new fields to store user configuration of RSS

Orabug: 22342532

This patch adds new fields to i40e_vsi to store user configured
RSS config data and code to use it.

Change-ID: Ic5d3db8d9df52182b560248f8cdca9c5c7546879
Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 66f9af855a4df146713866f6076207edea44a4f5)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40evf: create a generic get RSS function
Helin Zhang [Mon, 26 Oct 2015 23:44:33 +0000 (19:44 -0400)]
i40evf: create a generic get RSS function

Orabug: 22342532

There are two ways to get RSS, this patch implements two functions
with the same input parameters, and creates a more generic function
for getting RSS configuration.

Change-ID: I12d3b712c21455d47dd0a5aae58fc9b7c680db59
Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 90b02b4392eb9778265ec6b4fef034010a937910)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40evf: create a generic config RSS function
Helin Zhang [Tue, 27 Oct 2015 20:15:06 +0000 (16:15 -0400)]
i40evf: create a generic config RSS function

Orabug: 22342532

There are two ways to configure RSS, this patch adjusts those two
functions with the same input parameters, and creates a more
generic function for configuring RSS.

Change-ID: Iace73bdeba4831909979bef221011060ab327f71
Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 2c86ac3c70794f0ecc3684b0ae0fd75b0cf0c1f6)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40evf: rename VF adapter specific RSS function
Helin Zhang [Mon, 26 Oct 2015 23:44:31 +0000 (19:44 -0400)]
i40evf: rename VF adapter specific RSS function

Orabug: 22342532

This patch renames old VF adapter specific RSS function to clarify
its scope.

Change-ID: Ie5253083a44c677ebb7709a8a3a18402ad2dc6a6
Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 96a8198652e5b80a2cfc87397a3512bf8e45cd63)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e/i40evf: prefetch skb data on transmit
Jesse Brandeburg [Mon, 26 Oct 2015 23:44:30 +0000 (19:44 -0400)]
i40e/i40evf: prefetch skb data on transmit

Orabug: 22342532

Issue a prefetch for data early in the transmit path.
This should not be generally needed for Tx traffic, but
it helps immensely for pktgen workloads and should help
for forwarding workloads as well.

Change-ID: Iefee870c20599e0c4240e1d8637e4f16b625f83a
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit b74118f08356fd8ab6cb5d1a15705a2760c9afdd)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e/i40evf: Fix RS bit update in Tx path and disable force WB workaround
Anjali Singhai Jain [Mon, 26 Oct 2015 23:44:29 +0000 (19:44 -0400)]
i40e/i40evf: Fix RS bit update in Tx path and disable force WB workaround

Orabug: 22342532

This patch fixes the issue of forcing WB too often causing us to not
benefit from NAPI.

Without this patch we were forcing WB/arming interrupt too often taking
away the benefits of NAPI and causing a performance impact.

With this patch we disable force WB in the clean routine for X710
and XL710 adapters. X722 adapters do not enable interrupt to force
a WB and benefit from WB_ON_ITR and hence force WB is left enabled
for those adapters.
For XL710 and X710 adapters if we have less than 4 packets pending
a software Interrupt triggered from service task will force a WB.

This patch also changes the conditions for setting RS bit as described
in code comments. This optimizes when the HW does a tail bump amd when
it does a WB. It also optimizes when we do a wmb.

Change-ID: Id831e1ae7d3e2ec3f52cd0917b41ce1d22d75d9d
Signed-off-by: Anjali Singhai Jain <anjali.singhai@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 6a7fded776a778f728b13d83a2c9fc893580c080)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: rename rss_size to alloc_rss_size in i40e_pf
Helin Zhang [Mon, 26 Oct 2015 23:44:28 +0000 (19:44 -0400)]
i40e: rename rss_size to alloc_rss_size in i40e_pf

Orabug: 22342532

This patch renames rss_size to alloc_rss_size in i40e_pf, which is
clearer and avoids confusion. It also adds comments to the other
related structure members to help clarify usage.

Change-ID: Ia90090609d006ab589cb639975bb8a0af795d16f
Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit acd65448f2e6b3407b8bb28b57d711ca417ea679)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: add new fields to store user configuration
Helin Zhang [Mon, 26 Oct 2015 23:44:27 +0000 (19:44 -0400)]
i40e: add new fields to store user configuration

Orabug: 22342532

This patch adds new fields to i40e_vsi to store user configured
RSS config data and code to use it.

Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Change-ID: I73886469dca9e9f6b16d842182a87f3f4009f95d
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 28c5869f2bc4e319027bc22a66a1aa9eefc69a16)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: Bump version to 1.4.2
Catherine Sullivan [Wed, 21 Oct 2015 23:56:24 +0000 (19:56 -0400)]
i40e: Bump version to 1.4.2

Orabug: 22342532

Bump.

Change-ID: I2d1ce93b2ce74e4eef2394c932aef52cba99713f
Signed-off-by: Catherine Sullivan <catherine.sullivan@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 1c2df9e5a763d6d88f9aea8ca76d0ca75753fbd5)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: create a generic configure rss function
Helin Zhang [Wed, 21 Oct 2015 23:56:23 +0000 (19:56 -0400)]
i40e: create a generic configure rss function

Orabug: 22342532

This patch renames the old pf-specific function in order to clarify
its scope. This patch also creates a more generic configure RSS
function with the old name.

This patch also creates a new more generic function to get RSS
configuration, using the appropriate method.

Change-ID: Ieddca2707b708ef19f1ebccdfd03a0a0cd63d3af
Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 043dd650efde3dfc65a6461e1cdc51cc87cb76f7)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: rework the functions to configure RSS with similar parameters
Helin Zhang [Wed, 21 Oct 2015 23:56:22 +0000 (19:56 -0400)]
i40e: rework the functions to configure RSS with similar parameters

Orabug: 22342532

Adjust the RSS configure functions so that there is a generic way to
hook to ethtool hooks.

Change-ID: If446e34fcfaf1bc3320d9d319829a095b5976e67
Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit e69ff813af354ae445518d44e299eeb85d5037e2)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: return the number of enabled queues for ETHTOOL_GRXRINGS
Helin Zhang [Wed, 21 Oct 2015 23:47:13 +0000 (19:47 -0400)]
i40e: return the number of enabled queues for ETHTOOL_GRXRINGS

Orabug: 22342532

This patch fixes a problem where using ethtool rxnfc command could
let RX flow hash be set on disabled queues. This patch fixes the
problem by returning the number of enabled queues before setting
rxnfc.

Change-ID: Idbac86b0b47ddacc8deee7cd257e41de01cbe5c0
Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 3e3aa21fe9265bdc45ff795dbcc90bcdd2f1017c)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40evf: handle many MAC filters correctly
Mitch Williams [Wed, 21 Oct 2015 23:47:12 +0000 (19:47 -0400)]
i40evf: handle many MAC filters correctly

Orabug: 22342532

When a lot (many hundreds) of MAC or VLAN filters are added at one time,
we can overflow the Admin Queue buffer size with all the requests.
Unfortunately, the driver would then calculate the message size
incorrectly, causing it to be rejected by the PF. Furthermore, there was
no mechanism to trigger another request to allow for configuring the
rest of the filters that didn't fit into the first request.

To fix this, recalculate the correct buffer size when we detect the
overflow condition instead of just assuming the max buffer size. Also,
don't clear the request bit in adapter->aq_required when we have an
overflow, so that the rest of the filters can be processed later.

Change-ID: Idd7cbbc5af31315e0dcb1b10e6a02ad9817ce65c
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 1418c3458118c6969d08e23aa377da7e2a7be36c)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e/i40evf: clean up error messages
Mitch Williams [Wed, 21 Oct 2015 23:47:11 +0000 (19:47 -0400)]
i40e/i40evf: clean up error messages

Orabug: 22342532

Clean up and enhance error messages related to VF MAC/VLAN filters.
Indicate which VF is having issues, and if possible indicate the MAC
address or VLAN involved.

Also, when an error is returned from the PF driver, print useful
information about what went wrong, for the most likely cases.

Change-ID: Ib3d15eef9e3369a78fd142948671e5fa26d921b8
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 8d8f2295d0752230e0a697af33e5af96561b64a0)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e/i40evf: Add comment to #endif
Helin Zhang [Wed, 21 Oct 2015 23:47:10 +0000 (19:47 -0400)]
i40e/i40evf: Add comment to #endif

Orabug: 22342532

Add a comment to the #endif to more easily match it with its #if.

Change-ID: I47eb0a60a17dc6d2f01a930e45006d2dc82e044f
Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 2f175f552ddb2498742c925332ce01b602fa029a)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: Move the saving of old link info from handle_link_event to link_event
Catherine Sullivan [Wed, 21 Oct 2015 23:47:09 +0000 (19:47 -0400)]
i40e: Move the saving of old link info from handle_link_event to link_event

Orabug: 22342532

The watchdog only calls link_event not handle_link_event which means
that we need to save the old information in link_event.

Previously when polling we were comparing current data to the old data
saved the last time we actually received a link event. This means that
the polling would only fix link status changes in one direction
depending on what the last old data saved off was.

Change-ID: Ie590f30fdbcb133d0ddad4e07e3eb1aad58255b3
Signed-off-by: Catherine Sullivan <catherine.sullivan@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 1f9610e4777c391e1e749ffc646b29dbff920834)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e/i40evf: Add a stat to track how many times we have to do a force WB
Anjali Singhai Jain [Wed, 21 Oct 2015 23:47:08 +0000 (19:47 -0400)]
i40e/i40evf: Add a stat to track how many times we have to do a force WB

Orabug: 22342532

When in NAPI with interrupts disabled, the HW needs to be forced to do a
write back on TX if the number of descriptors pending are less than a
cache line.

This stat helps keep track of how many times we get into this situation.

Change-ID: I76c1bcc7ebccd6bffcc5aa33bfe05f2fa1c9a984
Signed-off-by: Anjali Singhai Jain <anjali.singhai@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 164c9f54631beca4d174f306acdcaec2bdeef52e)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: remove BUG_ON from FCoE setup
Shannon Nelson [Wed, 21 Oct 2015 23:47:05 +0000 (19:47 -0400)]
i40e: remove BUG_ON from FCoE setup

Orabug: 22342532

There's no need to kill the kernel thread here. If this condition was
true, the probe() would have died long before we got here. In any case,
we'll get the same result when this code tries to use the VSI pointer
being checked.

Prompted by a recent Linus diatribe.

Change-ID: I62f531cac34d4fc28ff9657d5b2d9523ae5e33a4
Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit f9b26ebb6e46f2231adc22d69f4150dc53f19b74)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: remove BUG_ON from feature string building
Shannon Nelson [Wed, 21 Oct 2015 23:47:04 +0000 (19:47 -0400)]
i40e: remove BUG_ON from feature string building

Orabug: 22342532

There's really no reason to kill the kernel thread just because of a
little info string. This reworks the code to use snprintf's limiting to
assure that the string is never too long, and WARN_ON to still put out
a warning that we might want to look at the feature list length.

Prompted by a recent Linus diatribe.

Change-ID: If52ba5ca1c2344d8bf454a31bbb805eb5d2c5802
Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 7fd89545f3374a061669a279ae0e084b0ddbb53d)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: Change BUG_ON to WARN_ON in service event complete
Shannon Nelson [Wed, 21 Oct 2015 23:47:03 +0000 (19:47 -0400)]
i40e: Change BUG_ON to WARN_ON in service event complete

Orabug: 22342532

There's no need to kill the thread and eventually the kernel in this
case.  In fact, the remainder of the code won't hurt anything anyway,
so just complain that we're here and move along.

Prompted by a recent Linus diatribe.

Change-ID: Iec020d8bcfedffc1cd2553cc6905fd915bb3e670
Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit b875f99b4cb454e0aa1e68d2d8792112a1e07050)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e/i40evf: remove unused tunnel parameter
Shannon Nelson [Wed, 21 Oct 2015 23:47:02 +0000 (19:47 -0400)]
i40e/i40evf: remove unused tunnel parameter

Orabug: 22342532

Code was moved into a separate function some time ago.

Change-ID: Icabbe71ce05cf5d716d3e1152cdd9cd41d11bcb5
Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 9c883bd3ebd127785ef0538b04e2d4b4c64c4c2d)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agointel: i40e: fix confused code
Rasmus Villemoes [Sat, 17 Oct 2015 20:58:19 +0000 (22:58 +0200)]
intel: i40e: fix confused code

Orabug: 22342532

This code is pretty confused. The variable name 'bytes_not_copied'
clearly indicates that the programmer knew the semantics of
copy_{to,from}_user, but then the return value is checked for being
negative and used as a -Exxx return value.

I'm not sure this is the proper fix, but at least we get rid of the
dead code which pretended to check for access faults.

Signed-off-by: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Acked-by: Shannon Nelson <shannon.nelson@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 0286c67e710cc82260b8c826e5a7b75143fedbd6)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40evf: fix compiler warning of unused variable
Jesse Brandeburg [Sun, 4 Oct 2015 08:09:49 +0000 (01:09 -0700)]
i40evf: fix compiler warning of unused variable

Orabug: 22342532

Compiler complained of an unused variable, which the driver was just
using to store the result of a rd32 which is used to clear a register
unconditionally.  Just drop the unused variable and re-use one.

Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit cfbe4dba0d5891faa11bd4e7c1fcdc512aff2f5f)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: fix unconditional execution of cpu_to_le16()
Jean Sacren [Tue, 13 Oct 2015 07:06:32 +0000 (01:06 -0600)]
i40e: fix unconditional execution of cpu_to_le16()

Orabug: 22342532

The commit 3092e5e4cc79 ("i40e: add little endian conversion for
checksum") fixed the checksum bug on big-endian architecture.

But we should not execute cpu_to_le16() unconditionally. Thus, put
cpu_to_le16() under certain condition.

Cc: Jesse Brandeburg <jesse.brandeburg@intel.com>
Cc: Paul M Stillwell Jr <paul.m.stillwell.jr@intel.com>
Signed-off-by: Jean Sacren <sakiwit@gmail.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 2fc4cd52fa7ed86e9d1df94e459b6a2435bfa9c7)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: clean up local variable initialization
Jean Sacren [Tue, 13 Oct 2015 07:06:31 +0000 (01:06 -0600)]
i40e: clean up local variable initialization

Orabug: 22342532

In both i40e_calc_nvm_checksum() and i40e_update_nvm_checksum(), the
local variables designated by 'ret_code' are overwritten immediately. As
such, they should merely be declared.

Signed-off-by: Jean Sacren <sakiwit@gmail.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 0e5229c6c576969bf8f89f8ca74fa7ea355b434e)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40evf: clean up local variable initialization
Jean Sacren [Tue, 13 Oct 2015 07:06:30 +0000 (01:06 -0600)]
i40evf: clean up local variable initialization

Orabug: 22342532

In i40evf_msix_aq(), the first two lines of rd32() are mainly to clear
the registers. If we initialize 'val' at this point, it will be
overwritten immediately. We shall simply discard the return value here.

When we initialize 'val', we might as well include the mask in one step.

Signed-off-by: Jean Sacren <sakiwit@gmail.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit ed17f7e5d7918adb487debe49fadf57fd2b98d1c)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: add missing kernel-doc argument
Jean Sacren [Tue, 13 Oct 2015 07:06:28 +0000 (01:06 -0600)]
i40e: add missing kernel-doc argument

Orabug: 22342532

The following kernel-doc arguments for their respective functions are
missing:

1) @cd_type_cmd_tso_mss for i40e_tso();
2) @cd_type_cmd_tso_mss for i40e_tsyn();
3) @tx_ring for i40e_tx_enable_csum().

Add them all for the kernel-doc requirement.

Signed-off-by: Jean Sacren <sakiwit@gmail.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 554f4544afa541b25882cd0ca5fd6b01b4e2ce07)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40evf: add missing kernel-doc argument
Jean Sacren [Tue, 13 Oct 2015 07:06:27 +0000 (01:06 -0600)]
i40evf: add missing kernel-doc argument

Orabug: 22342532

@flush has been missing since the inception of i40evf_irq_enable(). Add
it for the kernel doc.

Signed-off-by: Jean Sacren <sakiwit@gmail.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 69c1d70ab6b5b53ec21919f4f8de84041d3f6cdd)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: re-use %*ph specifier to hexdump a data
Andy Shevchenko [Fri, 2 Oct 2015 09:18:16 +0000 (12:18 +0300)]
i40e: re-use %*ph specifier to hexdump a data

Orabug: 22342532

Instead of using a custom approach change the code to use %*ph format
specifier.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit a3524e95ac5339e9b65dba2228f3f7efd27648cf)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e/i40evf: Bump i40e to 1.3.46 and i40evf to 1.3.33
Catherine Sullivan [Thu, 1 Oct 2015 18:37:41 +0000 (14:37 -0400)]
i40e/i40evf: Bump i40e to 1.3.46 and i40evf to 1.3.33

Orabug: 22342532

Bump up the version...

Change-ID: Ib8d501021671ba20250115ed54330e2c182255b7
Signed-off-by: Catherine Sullivan <catherine.sullivan@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 0e320516b280a8cf7258d9d41e7140563dd7c027)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: Disable VEB bridge mode with SR-IOV failure
Akeem G Abodunrin [Thu, 1 Oct 2015 18:37:40 +0000 (14:37 -0400)]
i40e: Disable VEB bridge mode with SR-IOV failure

Orabug: 22342532

If a call to enable SR-IOV in the kernel failed, we need to disable
I40E_FLAG_VEB_MODE_ENABLED, so that bridge mode could fall back to
VEPA, which is a default.

Change-ID: I12b6f776769506db85b29bea94b9c88d0b5ee65e
Signed-off-by: Akeem G Abodunrin <akeem.g.abodunrin@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit de445b3dc29835ae840f202bc62fd3ae42da12ad)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: Fix an incorrect OEM version string
Carolyn Wyborny [Thu, 1 Oct 2015 18:37:39 +0000 (14:37 -0400)]
i40e: Fix an incorrect OEM version string

Orabug: 22342532

This patch fixes a problem where the driver output of the OEM
version string varied from the other tools.  The mask value
and the order of operations were incorrect, per the original
change request.  Without this patch, the version string will
appear incorrect from the driver.

Change-ID: Ie1ca6485284b4ce3b57e5a99b18b7641617c7ef7
Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 2efaad86b5cdd0f6d70436fa304a567df8a3cf6c)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: fix inconsistent statuses after a PF reset
Helin Zhang [Thu, 1 Oct 2015 18:37:38 +0000 (14:37 -0400)]
i40e: fix inconsistent statuses after a PF reset

Orabug: 22342532

This patch fixes a problem of possibly getting inconsistent flow control
statuses after a PF reset. Requested_mode was being set with a default
value during probing, but the initial HW state could be different from
this mode.

Change-ID: I772bf07b78616e87086418d4bd87954b66fa17cd
Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 58fc3267f151f6facfc94c6982bef09a53bb5219)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40evf: use correct struct for list manipulation
Mitch Williams [Thu, 1 Oct 2015 18:37:37 +0000 (14:37 -0400)]
i40evf: use correct struct for list manipulation

Orabug: 22342532

Not sure how this compiles at all. Use the correct struct for
manipulating the VLAN filter list. Without this, the VLAN filter
list doesn't get processed correctly, and VLAN filters will not
be re-enabled after any kind of reset.

Change-ID: Iceff2dc089f303058fb71ecb08419eed471e0e90
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 40d01366e6dd5ef1cb3505f11b2d208cfa3632f3)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: Fix VEB/VEPA bridge mode mismatch issue
Akeem G Abodunrin [Thu, 1 Oct 2015 18:37:36 +0000 (14:37 -0400)]
i40e: Fix VEB/VEPA bridge mode mismatch issue

Orabug: 22342532

Fix i40e_is_vsi_uplink_mode_veb to check if bridge is actually
in VEB mode before allowing LB in the add VSI routine, instead of
unconditionally returning VEB bridge mode.

Change-ID: I162397b1bdd02367735fe9baaeb51465be2a3ce9
Signed-off-by: Akeem G Abodunrin <akeem.g.abodunrin@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 09603eaa5ca91dd9a318d8722ff5de8feb1e36ab)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: fix a bug in debugfs with add/del macaddr
Anjali Singhai Jain [Thu, 1 Oct 2015 18:37:35 +0000 (14:37 -0400)]
i40e: fix a bug in debugfs with add/del macaddr

Orabug: 22342532

The new code flow requires us to grab the filter list lock before
adding/deleting the filter.

Change-ID: I4eaef508ab4da2d1b2e23f20f2a78d931d5b6aeb
Signed-off-by: Anjali Singhai Jain <anjali.singhai@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 10dc0358e802ea60f95432f89c8dd051767fc06f)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e/i40evf: Add a workaround to drop all flow control frames
Anjali Singhai Jain [Thu, 1 Oct 2015 18:37:34 +0000 (14:37 -0400)]
i40e/i40evf: Add a workaround to drop all flow control frames

Orabug: 22342532

This patch adds a workaround to drop any flow control frames from being
transmitted from any VSI. FW can still send flow control frames if flow
control is enabled.

With this patch in place a malicious VF cannot send flow control or PFC
packets out on the wire.

Change-ID: I4303b24e98b93066d2767fec24dfe78be591c277
Signed-off-by: Anjali Singhai Jain <anjali.singhai@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit e7358f54a3954df16d4f87e3cad35063f1c17de5)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e/i40evf: Bump i40e to 1.3.38 and i40evf to 1.3.25
Catherine Sullivan [Mon, 28 Sep 2015 18:17:00 +0000 (14:17 -0400)]
i40e/i40evf: Bump i40e to 1.3.38 and i40evf to 1.3.25

Orabug: 22342532

Bump.

Change-ID: Id0a7ecaa491f88ce94c9eba4901e592a56044ee0
Signed-off-by: Catherine Sullivan <catherine.sullivan@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit a1f192cf70b187ab01b8b03cdeea82b8820c927a)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: declare rather than initialize int object
Jean Sacren [Sat, 19 Sep 2015 11:08:45 +0000 (05:08 -0600)]
i40e: declare rather than initialize int object

Orabug: 22342532

'err' would be overwritten immediately, so we should declare it only
rather than initialize it to zero.

Signed-off-by: Jean Sacren <sakiwit@gmail.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 6f66a484f1ff644b4600368fe3091e31b65d3c50)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: fix kernel-doc argument name
Jean Sacren [Sat, 19 Sep 2015 11:08:43 +0000 (05:08 -0600)]
i40e: fix kernel-doc argument name

Orabug: 22342532

The second argument name in the kernel-doc argument list for
i40e_features_check() was slightly off. Fix it for the kernel doc.

Signed-off-by: Jean Sacren <sakiwit@gmail.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 2bc11c634eaebb936345b136a196f555045a47ea)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: Move error message to debug level
Catherine Sullivan [Mon, 28 Sep 2015 18:16:59 +0000 (14:16 -0400)]
i40e: Move error message to debug level

Orabug: 22342532

There is an error coming back from get_phy_capabilities that does not
seem to have any functional implications. We will continue looking into
why this error message is occurring, but in the meantime, we will move it
to debug to avoid confusion.

Change-ID: I9091754bf62c066ddedeb249923d85606e2d68ed
Signed-off-by: Catherine Sullivan <catherine.sullivan@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 52e9689e4ef070eff1e0f9e6e84a8451b86813b5)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: Fix order of checks when enabling/disabling autoneg in ethtool
Catherine Sullivan [Mon, 28 Sep 2015 18:16:58 +0000 (14:16 -0400)]
i40e: Fix order of checks when enabling/disabling autoneg in ethtool

Orabug: 22342532

We were previously checking if autoneg was allowed to change before
checking if autoneg was changing. We need to do this in the other order
or else we will erroneously return EINVAL when autoneg is not changing.

Change-ID: Iff9f7d1c9bddc1ad1e5d227d4f42754f90155410
Signed-off-by: Catherine Sullivan <catherine.sullivan@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 3ce12ee9d8f9ae245f38e6acbd32625d8e002c5b)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e/i40evf: Fix an accidental error with BIT_ULL replacement
Anjali Singhai Jain [Mon, 28 Sep 2015 18:16:57 +0000 (14:16 -0400)]
i40e/i40evf: Fix an accidental error with BIT_ULL replacement

Orabug: 22342532

A mask value of 0x1FF was accidentally replaced with a bit mask
causing flow director sideband to be broken.

Change-ID: Id3387f67dd1b567b41692b570b383c58671e1eae
Signed-off-by: Anjali Singhai Jain <anjali.singhai@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit a03dc368547a7a4e9fcedc6f64dd64a9ff5e0593)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: fix for PHY NVM interaction problem
Carolyn Wyborny [Mon, 28 Sep 2015 18:16:56 +0000 (14:16 -0400)]
i40e: fix for PHY NVM interaction problem

Orabug: 22342532

This patch fixes a problem where the NVMUpdate Tool, when using the PHY
NVM feature, gets bad data from the PHY because of contention on the
MDIO interface from get PHY capability calls from the driver during
regular operations.  The problem is fixed by adding a check if media
is available before calling get PHY capability function because that
bit is not set when device is in PHY interaction mode.

Change-ID: Ib89991b0f841808dd92410f5e8683d6ee3301cd0
Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 8589af70d0879a68c93ef77505a6234d22b1b1a7)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: Fix for Tools loopback test failing after driver load
Carolyn Wyborny [Mon, 28 Sep 2015 18:16:55 +0000 (14:16 -0400)]
i40e: Fix for Tools loopback test failing after driver load

Orabug: 22342532

This patch fixes a problem where our Tools MAC Loopback test, after
driver unbind would fail.  This was because the hw was configured
for multiqueue and unbind operation did not clear this configuration.
The problem is fixed by resetting this configuration in i40e_remove.

Change-ID: I130c05138319182ed1476d3a0b5222d6a6320af9
Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit bcab2db97d2423b97b248ba35eb2277168e71fcc)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e/i40evf: adjust interrupt throttle less frequently
Jesse Brandeburg [Mon, 28 Sep 2015 18:16:54 +0000 (14:16 -0400)]
i40e/i40evf: adjust interrupt throttle less frequently

Orabug: 22342532

The adaptive ITR (interrupt throttle rate) algorithm was adjusting
the hardware's interrupt rate too frequently.  This caused a lot
of variation in the interrupt rate for fairly constant workloads.

Change the code to have a counter and adjust only once every N
number of interrupts.

Change-ID: I0460f1f86571037484eca5aca36ac4d889cb8389
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit ee2319cf17ee64bbd0096f2f8f3f8390c93b1e39)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e/i40evf: change dynamic interrupt thresholds
Jesse Brandeburg [Mon, 28 Sep 2015 18:16:53 +0000 (14:16 -0400)]
i40e/i40evf: change dynamic interrupt thresholds

Orabug: 22342532

The dynamic algorithm, while now working, doesn't have good
performance in 40G mode.

One part of this patch addresses the high CPU utilization of some small
streaming workloads that the driver should reduce CPU in.

It also changes the minimum ITR that the dynamic algorithm
will settle on, causing our minimum latency to go from 12us
to about 14us, when using adaptive mode.

It also changes the BULK interrupt rate to allow maximum throughput
on a 40Gb connection with a single thread of transmit, clamping
interrupt rate to 8000 for TX makes single thread traffic go too
slow.

The new ULTRA bulk setting is introduced and is used
when the Rx packet rate on this queue exceeds 40000 packets per
second.  This value of 40000 was chosen because the automatic tuning
of minimum ITR=20us means that a single queue can't quite achieve
that many packets per second from a round-robin test.

Change-ID: Icce8faa128688ca5fd2c4229bdd9726877a92ea2
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit c56625d59726ee2dc4dfd91c8b6c22098abe1ac4)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e/i40evf: fix bug in throttle rate math
Jesse Brandeburg [Mon, 28 Sep 2015 18:16:52 +0000 (14:16 -0400)]
i40e/i40evf: fix bug in throttle rate math

Orabug: 22342532

The driver was using a value expressed in 2us increments
for the divisor to figure out our bytes/usec values.

Fix the usecs variable to contain a value in microseconds.

Change-ID: I5c20493103c295d6f201947bb908add7040b7c41
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 51cc6d9fccde27310b7dfba2be268ff5b8dcf52d)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e/i40evf: refactor IRQ enable function
Jesse Brandeburg [Mon, 28 Sep 2015 18:16:51 +0000 (14:16 -0400)]
i40e/i40evf: refactor IRQ enable function

Orabug: 22342532

This change moves a multi-line register setting into a function
which simplifies reading the flow of the enable function.

This also fixes a bug where the enable function was enabling
the interrupt twice while trying to update the two interrupt
throttle rate thresholds for Rx and Tx.

Change-ID: Ie308f9d0d48540204590cb9d7a5a7b1196f959bb
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 8f5e39ce9214888464d34ef7949e56bdc999b46c)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40evf: don't give up
Mitch Williams [Mon, 28 Sep 2015 18:16:50 +0000 (14:16 -0400)]
i40evf: don't give up

Orabug: 22342532

When the VF driver is unable to communicate with the PF, it just gives
up and never tries again. Aside from the obvious character flaw that
this shows, it's also a lousy user experience.

When PF communications fail, wait five seconds, and try again. And
again. Don't give up, little VF driver! Your prince will come!

Change-ID: Ia1378a39879883563b8faffce819f375821f9585
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit b9029e941dd66d2bc2ab529807a1eef4444f62f0)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e/i40evf: use napi_schedule_irqoff()
Alexander Duyck [Tue, 29 Sep 2015 22:19:50 +0000 (15:19 -0700)]
i40e/i40evf: use napi_schedule_irqoff()

Orabug: 22342532

The i40e_intr and i40e/i40evf_msix_clean_rings functions run from hard
interrupt context or with interrupts already disabled in netpoll.

They can use napi_schedule_irqoff() instead of napi_schedule()

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 5d3465a1e4c01be455ef00834ec365e578942d67)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: Fix basic support for X722 devices
Anjali Singhai [Thu, 24 Sep 2015 22:26:32 +0000 (15:26 -0700)]
i40e: Fix basic support for X722 devices

Orabug: 22342532

Acquire NVM, before issuing an AQ read nvm command for X722.
We need to acquire the NVM before issuing an AQ read to the NVM
otherwise we will get EBUSY from the FW. Also release when done.

This fixes the two X722 issues with respect to eeprom checksum verify
and reading NVM version info.

With this patch in place, i40e driver will provide basic support
for X722 devices.

Signed-off-by: Anjali Singhai Jain <anjali.singhai@intel.com>
Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 07f89be81fa51ec85456559d7db9bc71379a20c1)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40evf: fix overlong BIT defines
Jesse Brandeburg [Thu, 17 Sep 2015 02:01:08 +0000 (19:01 -0700)]
i40evf: fix overlong BIT defines

Orabug: 22342532

The defines from the RSS enabling call were mistakenly
missed in the patches to the i40e which should have been
to i40evf as well.

This is a follow up to (commit ed921559886dd40528) "fix
32 bit build warnings".

Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit d08f55585f22c38db5fcbed509a71d925e0509c1)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: Lock for VSI's MAC filter list
Kiran Patil [Wed, 30 Sep 2015 18:09:03 +0000 (14:09 -0400)]
i40e: Lock for VSI's MAC filter list

Orabug: 22342532

This patch introduces a spinlock which is to be used for synchronizing
access to VSI's MAC filter list.

This patch also synchronizes execution of other codepaths which are
accessing VSI's MAC filter list with execution of
service_task:sync_vsi_filters.

In function i40e_add_vsi, copied out LAA MAC address instead of cloning
MAC filter entry because only MAC address is needed to remove MAC VLAN
filter from FW/HW.

Change-ID: I0e10ac7c715d44aa994239642aa4d57c998573a2
Signed-off-by: Kiran Patil <kiran.patil@intel.com>
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 216590355cb0e79fa36c7220176ce850f0e3d276)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e/i40evf: Bump i40e to 1.3.34 and i40evf to 1.3.21
Catherine Sullivan [Mon, 28 Sep 2015 18:12:44 +0000 (14:12 -0400)]
i40e/i40evf: Bump i40e to 1.3.34 and i40evf to 1.3.21

Orabug: 22342532

Bump.

Change-ID: I7ec818a507554648675b9b245ced9e6b6bd9ed4e
Signed-off-by: Catherine Sullivan <catherine.sullivan@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit d1d39516e41d7ad107e807ebeee1519c412bb0e3)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: increase AQ work limit
Mitch Williams [Mon, 28 Sep 2015 18:12:43 +0000 (14:12 -0400)]
i40e: increase AQ work limit

Orabug: 22342532

With 64 VFs, we can easily overwhelm the AQ on the PF if we have too low
a limit on the number of AQ requests. This leads to ARQ overflow errors,
and occasionally VFs that fail to initialize.

Since we really only hit this condition on initial VF driver load, the
requests that we process are lightweight, so this extra work doesn't
cause problems for the PF driver.

Change-ID: I620221520d8af987df6ace9ba938ffaf22107681
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 628f096d8d244f54c3595a478b6e672cdb6c04ed)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40evf: relax and stagger init timing a bit
Mitch Williams [Mon, 28 Sep 2015 18:12:42 +0000 (14:12 -0400)]
i40evf: relax and stagger init timing a bit

Orabug: 22342532

On some devices, in some systems, in some configurations, the VFs would
fail to initialize the first time you loaded the driver.

To correct this, increase the delay time for the init task slightly, and
wait longer before giving up.

If we enable VFs and load the VF driver in the same kernel as the PF
driver, we can totally overwhelm the PF driver with AQ requests because
all of the instances try to initialize at the same time.

To help alleviate this, stagger the initial scheduling of the init task
using the PCIe function as a multiplier. We mask off the function to
only three bits so no instance has to wait too long.

With these two changes, initializing 128 VFs on a single device goes
from four minutes to just a few seconds.

Change-ID: If3d8720c1c4e838ab36d8781d9ec295a62380936
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 3f7e5c330e19da2ac822ea1d02b744a24497e61b)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: Recognize 1000Base_T_Optical phy type when link is up
Catherine Sullivan [Mon, 28 Sep 2015 18:12:41 +0000 (14:12 -0400)]
i40e: Recognize 1000Base_T_Optical phy type when link is up

Orabug: 22342532

1000Base_T_Optical got added to the function that figures out what
is supported when link is down but not when link is up. Add it in there
too so that we display the correct information.

Change-ID: I85ebcdfa7c02d898c44c673b1500552a53c8042e
Signed-off-by: Catherine Sullivan <catherine.sullivan@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 48becae60f7e461b4b204e6608dc6beebe6c8aff)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40evf: correctly populate vlan_features
Mitch Williams [Mon, 28 Sep 2015 18:12:40 +0000 (14:12 -0400)]
i40evf: correctly populate vlan_features

Orabug: 22342532

The vlan_features field was correctly being set to the same value as the
netdev features field. However, this was being done before the features
were actually being set up, leaving the vlan_features empty.

Also, after a reset, vlan_features will be incorrectly assigned the
previous netdev feature flags, which can contain VLAN feature bits. This
makes the VLAN code angry and will cause a stack dump.

To fix these issues, set up the netdev features first, then mask out the
VLAN feature bits when assigning vlan_features.

Change-ID: Ib0548869dc83cf6a841cb8697dd94c12359ba4d2
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit cc7e406cb967ccc84cd2141398bba24b07d68788)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: reset the invalid msg counter in vf when a valid msg is received
Jingjing Wu [Mon, 28 Sep 2015 18:12:39 +0000 (14:12 -0400)]
i40e: reset the invalid msg counter in vf when a valid msg is received

Orabug: 22342532

When the number of invalid messages from a VF is exceeded, the VF
will be disabled, due to the invalid messages.  This happens if
other VF drivers (like DPDK) send a message through the driver's
mailbox (aka virtchannel) interface, but the message is not
supported by the i40e pf driver, such as CONFIG_PROMISCUOUS_MODE.

This patch changes the num_invalid_msgs in struct i40e_vf to record
the continuous invalid msgs, and it will be reset when a valid msg
is received.

Change-ID: Iaec42fd3dcdd281476b3518be23261dd46fc3718
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 5d38c93e7111a68fc7b9f771bf420d9e026a21de)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e/i40evf: moderate interrupts differently
Jesse Brandeburg [Mon, 28 Sep 2015 18:12:37 +0000 (14:12 -0400)]
i40e/i40evf: moderate interrupts differently

Orabug: 22342532

The XL710 hardware has a different interrupt moderation design
that can support a limit of total interrupts per second per
vector, in addition to the "number of interrupts per second"
controls already established in the driver.  This combination
of hardware features allows us to set very low default latency
settings but minimize the total CPU utilization by not
making too many interrupts, should the user desire.

The current driver implementation is still enabling the dynamic
moderation in the driver, and only using the rx/tx-usecs
limit in ethtool to limit the interrupt rate per second, by default.

The new code implemented in this patch
2) adds init/use of the new "Interrupt Limit" register
3) adds ethtool knob to control/report the limits above

Usage is ethtool -C ethx rx-usecs-high <value> Where <value> is number
of microseconds to create a rate of 1/N interrupts per second,
regardless of rx-usecs or tx-usecs values. Since there is a credit based
scheme in the hardware, the rx-usecs and tx-usecs can be configured for
very low latency for short bursts, but once the credit runs out the
refill rate on the credits is limited by rx-usecs-high.

Change-ID: I3a1075d3296123b0f4f50623c779b027af5b188d
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit ac26fc136c24edee53f1719e490d896fb07cd79b)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: Add support for non-willing Apps
Greg Bowers [Mon, 28 Sep 2015 18:12:35 +0000 (14:12 -0400)]
i40e: Add support for non-willing Apps

Orabug: 22342532

Adds support for setting a new bit in the Set Local LLDP MIB AQ command
Type field.  When set to 1, the bit indicates to FW that Apps should be
treated as non-willing.  When 0, FW behaves as before.

Change-ID: I0d2101c1606c59c7188d3e6a0c7810e0f205233a
Signed-off-by: Greg Bowers <gregory.j.bowers@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 947570e800e2ae3cbf827918a999aa88ee1600b0)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: priv flag for controlling VEB stats
Shannon Nelson [Mon, 28 Sep 2015 18:12:34 +0000 (14:12 -0400)]
i40e: priv flag for controlling VEB stats

Orabug: 22342532

Add an ethtool priv flag to enable and disable printing
the VEB statistics.

Change-ID: I7654054a3a73b08aa8310d94ee8fce6219107dd8
Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 1cdfd88f2de89cbfffb5813dc92b7e711920c731)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: Removed unused defines
Greg Rose [Mon, 28 Sep 2015 18:12:33 +0000 (14:12 -0400)]
i40e: Removed unused defines

Orabug: 22342532

Two defines that are not used are causing customer confusion - remove
them.

Change-ID: Icef0325aca8e0f4fcdfc519e026bdd375e791200
Signed-off-by: Greg Rose <gregory.v.rose@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit d9d17cf74aed93add76c9cf648ce2f73b68ce410)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e: remove read/write failed messages from nvmupdate
Shannon Nelson [Mon, 28 Sep 2015 18:12:32 +0000 (14:12 -0400)]
i40e: remove read/write failed messages from nvmupdate

Orabug: 22342532

Allow the nvmupdate application to decide when a read or write error
should be exposed to the user.  Since the application needs to use
write probes to find the ReadOnly sections on a potentially unknown NVM
version in the HW and read probes to check the status of the last write,
some error messages are expected, but need not be shown to the users.
The driver doesn't know which are ignorable from real errors, so needs
to let the application make the decision.

Change-ID: I78fca8ab672bede11c10c820b83c26adfd536d03
Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 3c5c420535ab7e25a639db16f7b2b16a70e147ec)
Signed-off-by: Brian Maly <brian.maly@oracle.com>
9 years agoi40e/i40evf: Fix compile issue related to const string
Jingjing Wu [Mon, 28 Sep 2015 18:12:31 +0000 (14:12 -0400)]
i40e/i40evf: Fix compile issue related to const string

Orabug: 22342532

Add const to functions that return strings that aren't going to be
modified. This addresses some reported compile complaints.

Change-ID: Ic56b1e814ab4d23a50480e7fdec652445f776ee8
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
(cherry picked from commit 4e68adfeb91d684e5c7a0c6305af872120c71f23)
Signed-off-by: Brian Maly <brian.maly@oracle.com>