This patch replaces the existing Rx hot-path in the ixgbe driver with a new
implementation that is based on performing a double buffered receive.  The
ixgbe driver already had something similar in place for its' packet split
path, however in that case we were still receiving the header for the
packet into the sk_buff.  The big change here is the entire receive path
will receive into pages only, and then pull the header out of the page and
copy it into the sk_buff data.  There are several motivations behind this
approach.

First, this allows us to avoid several cache misses as we were taking a
set of cache misses for allocating the sk_buff and then another set for
receiving data into the sk_buff.  We are able to avoid these misses on
receive now as we allocate the sk_buff when data is available.

Second we are able to see a considerable performance gain when an IOMMU is
enabled because we are no longer unmapping every buffer on receive.
Instead we can delay the unmap until we are unable to use the page, and
instead we can simply call sync_single_range on the half of the page that
contains new data.

Finally we are able to drop a considerable amount of code from the driver
as we no longer have to support 2 different receive modes, packet split and
one buffer.  This allows us to optimize the Rx path further since less
branching is required.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NRoss Brattain <ross.b.brattain@intel.com>
Tested-by: NStephen Ko <stephen.s.ko@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This allows the NIC to receive all frames available, including
those with bad FCS, ethernet control frames, and more.

Tested by sending frames with bad FCS.
Signed-off-by: NBen Greear <greearb@candelatech.com>
Tested-by: NJeff Pieper <jeffrey.e.pieper@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Including bad FCS, used generate frames with bad FCS
to test other system's handling of RX of bad packets.
Signed-off-by: NBen Greear <greearb@candelatech.com>
Tested-by: NJeff Pieper <jeffrey.e.pieper@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Disabling and enabling DCB can cause FCoE hardware initialization to
occur on the incorrect traffic class when the up2tc mapping has not
yet been reconfigured.

Fix this by using the DCB configuration maps that are correct
and will be pushed at mqprio after DCB driver setup completes
successfully.
Signed-off-by: NJohn Fastabend <john.r.fastabend@intel.com>
Tested-by: NMarcus Dennis <marcusx.e.dennis@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

There was a race condition in the reset path where the RX buffer
could become corrupted during Fdir configuration.This is due to
a HW bug.The fix right now is to lock the buffer while we do the
fdir configuration.Since we were using similar workaround for another bug,
I moved the existing code to a function and reused it.HW team also recommended
that IXGBE_MAX_SECRX_POLL value be changed from 30 to 40.The erratum for this
bug will be published in the next release 82599 Spec Update
Signed-off-by: NAtita Shirwaikar <atita.shirwaikar@intel.com>
Acked-by: NPeter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Tested-by: NStephen Ko <stephen.s.ko@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

using the form min((int)var, ver)) is replaced by min_t(int, ...)
Signed-off-by: NJesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: NStephen Ko <stephen.s.ko@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This is clearly a typeo where we are not checking the return value from
get_link_capabilities but should.  This patch corrects that.
Signed-off-by: NDon Skidmore <donald.c.skidmore@intel.com>
Tested-by: NStephen Ko <stephen.s.ko@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

There isn't much point in using variables to store the values of eitr_low
and eitr_high since they are not user changeable.  As such I am replacing
them with the constants 10 and 20 in order to avoid any confusion on what
the values actually are.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NStephen Ko <stephen.s.ko@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

A previous fix had gone though and disabled relaxed ordering for Rx
descriptor read fetching.  This was not necessary as this functions
correctly and has no ill effects on the system.

In addition several of the defines used for the DCA control registers were
incorrect in that they indicated descriptor effects when they actually had
an impact on either data or header write back.  As such I have update these
to correctly reflect either DATA or HEAD.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NStephen Ko <stephen.s.ko@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This change makes it a bit easier to do the loopback frame creating and
testing.  Previously we were doing an and to drop the last bit, and then
dividing the frame_size by 2 in order to get locations for frame bytes and
testing.  Instead we can simplify it by just shifting the register one bit
to the right and using that for the frame offsets.

This change also replaces all instances of rx_buffer_info with just
rx_buffer since that is closer to the name of the actual structure being
used and can save a few extra characters.

In addition I have updated the logic for cleaning up a test frame so that
we pass an rx_buffer instead of the sk_buff.  The main motivation behind
this is changes that will replace the sk_buff with just a page in the
future.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NStephen Ko <stephen.s.ko@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Since there are multiple spots where we have to cycle through all of the
rings on a q_vector it makes sense to just add a function for iterating
through all of them.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NStephen Ko <stephen.s.ko@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This patch makes the rings a part of the q_vector directly instead of
indirectly.  Specifically on x86 systems this helps to avoid any cache
set conflicts between the q_vector, the tx_rings, and the rx_rings as the
critical stride is 4K and in order to cross that boundary you would need to
have over 15 rings on a single q_vector.

In addition this allows for smarter allocations when Flow Director is
enabled.  Previously Flow Director would set the irq_affinity hints based
on the CPU and was still using a node interleaving approach which on some
systems would end up with the two values mismatched.  With the new approach
we can set the affinity for the irq_vector and use the CPU for that
affinity to determine the node value for the node and the rings.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NStephen Ko <stephen.s.ko@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This patch is a minor cleanup to address the unnecessary use of
napi_schedule_prep in ixgbe_intr and to also remove a blank line that is
not needed since it is separating a comment from the line it is explaining.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NStephen Ko <stephen.s.ko@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

The old code had several errors in how it was determining the vector
budget.  In order to simplify things this patch updates the code so that it
will attempt to always allocated paired Rx/Tx vectors instead of attempting
to allocate individual vectors when the number of queues is less than the
number of CPUs.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NStephen Ko <stephen.s.ko@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This change corrects an issue in which Adaptive Interrupt Moderation was
not changing values due to the fact that we were performing an and
operation on the resultant value that was causing the value to never change
from the default 20K interrupts per second.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NStephen Ko <stephen.s.ko@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This change is meant to address the fact that the tx_itr_setting was
dropping to 0 when no separate Tx vectors were provided.  This had resulted
in the driver incorrectly configuring the Tx ring with a WTHRESH of 1 in
order to avoid Tx hangs even though that was not necessary. This change
makes it so that we instead take a look at the Tx ring's q_vector to
determine if the ring will have an ITR value less than 8us.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NStephen Ko <stephen.s.ko@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This change moves several frequently accessed items together into one cache
line in order to reduce cache misses in the hot-path.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NStephen Ko <stephen.s.ko@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

There isn't any need to clear the status bits in the descriptors due to the
fact that the eop_desc provides enough information for us to know
that we have cleaned to the last packet that the software has put on the
ring.  The status bits are cleared as a part of putting the frame on the
ring so as long as we do not read the descriptor bit prior to reading the
value eop_desc we should be able to guarantee that we will not clean beyond
the end of the current data stream.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NStephen Ko <stephen.s.ko@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This adds support for byte queue limits (BQL).

Based on patch from Eric Dumazet for igb.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NStephen Ko <stephen.s.ko@intel.com>

As noted by Ben Hutchings and David Miller, work limits for NAPI
should not be tied to interrupt moderation parameters.  This
should be handled by NAPI, possibly through sysfs.

Neil Horman & Stephen Hemminger are working on a solution for
NAPI currently.  In the meantime, remove this tie between
work limits and interrupt moderation.
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
CC: Ben Hutchings <bhutchings@solarflare.com>
Tested-by: NPhil Schmitt <phillip.j.schmitt@intel.com>

{g|s}etnumtcs() today returns a u8 that is only used by the DCB code
to verify no error occurred. Today the driver implementations return
negative error codes which end up being non-zero so the logic works
out but triggers some sparse warnings.

To fix the sparse warnings convert the return value to an int.

CC: Eilon Greenstein <eilong@broadcom.com>
Reported-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NJohn Fastabend <john.r.fastabend@intel.com>
Tested-by: NRoss Brattain <ross.b.brattain@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

dcb netlink code calls setup_tc to init hardware traffic classes
to use for DCB. At some call sites the return values are not
checked for errors and in one case may return -EINVAL back to
the net/dcbnl.c caller which is expecting a u8.

This fixes some smatch hits and although failures are never
seen in practive its best to check return codes.
Reported-by: NDan Carenter <dan.carpenter@oracle.com>
Signed-off-by: NJohn Fastabend <john.r.fastabend@intel.com>
Tested-by: NRoss Brattain <ross.b.brattain@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This patch corrects several comments that are either incorrect or formatted
incorrectly for multiline comments.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NPhil Schmitt <phillip.j.schmitt@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Correct spelling error caught with codespell.py.
Signed-off-by: NDon Skidmore <donald.c.skidmore@intel.com>
Tested-by: NPhil Schmitt <phillip.j.schmitt@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This patch is meant to address several minor issues in
ixgbe_xmit_frame_ring.  Specifically it adds a comment explaining the TXSW
flag, and correctly wraps a line over 80 characters.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NPhil Schmitt <phillip.j.schmitt@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Booleans should not be compared to true or false
but be directly tested or tested with !.

Done via cocci script:

@@
bool t;
@@
- t == true
+ t
@@
bool t;
@@
- t != true
+ !t
@@
bool t;
@@
- t == false
+ !t
@@
bool t;
@@
- t != false
+ t
Signed-off-by: NJoe Perches <joe@perches.com>
Reviewed-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This change combines a number of post-DMA Rx packet processing functions
into a single function.  The advantage of this is that it combines most of
the Rx descriptor processing into one spot so it should all be warm in the
cache.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NStephen Ko <stephen.s.ko@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

It doesn't make much sense to differentiate between advanced and legacy
descriptors when the only descriptors that ixgbe uses are advanced
descriptors.  As such we can drop the _ADV suffix since all ixgbe
descriptors are automatically advanced.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NStephen Ko <stephen.s.ko@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This change adds a small function for testing Rx status bits in the
descriptor.  The advantage to this is that we can avoid unnecessary
byte swaps on big endian systems.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NStephen Ko <stephen.s.ko@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This change makes it so that we always clear the status/error bits in the
Rx descriptor in the allocation path instead of the cleanup path.  The
advantage to this is that we spend less time modifying data.  As such we
can modify the data once and then let it go cold in the cache instead of
writing it, reading it, and then writing it again.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NStephen Ko <stephen.s.ko@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This patch is meant to address the fact that RSC has not been setting the
gso_size value on the skb.  As a result performance on lossy TCP
connections was negatively impacted.  This change resolves the issue by
setting gso_size to the average size for incoming packets.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NStephen Ko <stephen.s.ko@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This change addresses several issue.

First I had left the use of the next and prev skb pointers floating around
in the code and they were overdue to be pulled since I had rewritten the
RSC code in the out-of-tree driver some time ago to address issues brought
up by David Miller in regards to this.

I am also now defaulting to always leaving the first buffer unmapped on any
packet and then unmapping it after we read the EOP descriptor.  This allows
a simplification of the path with less branching.

Instead of counting packets received the code was changed some time ago to
track the number of buffers received.  This leads to inaccurate counting
when you compare numbers of packets received by the hardware versus what is
tracked by the software.  To correct this I am revising things so that the
append_cnt value for RSC accurately tracks the number of frames received.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NStephen Ko <stephen.s.ko@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

If the number of tx/rx queues changes the ethtool ioctl
ETHTOOL_GSTATS may overrun the userspace buffer. This
occurs because the general practice in user space to
query stats is to issue a ETHTOOL_GSSET cmd to learn the
buffer size needed, allocate the buffer, then call
ETHTOOL_GSTIRNGS and ETHTOOL_GSTATS. If the number of
real_num_queues is changed or flow control attributes
are changed after ETHTOOL_GSSET but before the
ETHTOOL_GSTRINGS/ETHTOOL_GSTATS a user space buffer
overrun occurs.

To fix the overrun always return the max buffer size
needed from get_sset_count() then return all strings
and stats from get_strings()/get_ethtool_stats().

This _will_ change the output from the ioctl() call
which could break applications and script parsing in
theory. I believe these changes should not break existing
tools because the only changes will be more {tx|rx}_queues
and the {tx|rx}_pb_* stats will always be returned.
Existing scripts already need to handle changing number
of queues because this occurs today depending on system
and current features. The {tx|rx}_pb_* stats are at the
end of the output and should be handled by scripts today
regardless.

Finally get_ethtool_stats and get_strings are free-form
outputs tools parsing these outputs should be defensive
anyways. In the end these updates are better then
having a tool segfault because of a buffer overrun.
Signed-off-by: NJohn Fastabend <john.r.fastabend@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Users expect the up2tc mapping to be maintained across a DCB
enable/disable/enable transition. And since we maintain all
the other DCB attributes we should do this for up2tc mappings
as well just to be consistent. Also without this we break
user space applications that expect this to occur that
previously worked.
Signed-off-by: NJohn Fastabend <john.r.fastabend@intel.com>
Tested-by: NStephen Ko <stephen.s.ko@intel.com>
Tested-by: NRoss Brattain <ross.b.brattain@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

If the netdev is already in NETREG_UNREGISTERING/_UNREGISTERED state, do not
update the real num tx queues. netdev_queue_update_kobjects() is already
called via remove_queue_kobjects() at NETREG_UNREGISTERING time. So, when
upper layer driver, e.g., FCoE protocol stack is monitoring the netdev
event of NETDEV_UNREGISTER and calls back to LLD ndo_fcoe_disable() to remove
extra queues allocated for FCoE, the associated txq sysfs kobjects are already
removed, and trying to update the real num queues would cause something like
below:

...
PID: 25138  TASK: ffff88021e64c440  CPU: 3   COMMAND: "kworker/3:3"
 #0 [ffff88021f007760] machine_kexec at ffffffff810226d9
 #1 [ffff88021f0077d0] crash_kexec at ffffffff81089d2d
 #2 [ffff88021f0078a0] oops_end at ffffffff813bca78
 #3 [ffff88021f0078d0] no_context at ffffffff81029e72
 #4 [ffff88021f007920] __bad_area_nosemaphore at ffffffff8102a155
 #5 [ffff88021f0079f0] bad_area_nosemaphore at ffffffff8102a23e
 #6 [ffff88021f007a00] do_page_fault at ffffffff813bf32e
 #7 [ffff88021f007b10] page_fault at ffffffff813bc045
    [exception RIP: sysfs_find_dirent+17]
    RIP: ffffffff81178611  RSP: ffff88021f007bc0  RFLAGS: 00010246
    RAX: ffff88021e64c440  RBX: ffffffff8156cc63  RCX: 0000000000000004
    RDX: ffffffff8156cc63  RSI: 0000000000000000  RDI: 0000000000000000
    RBP: ffff88021f007be0   R8: 0000000000000004   R9: 0000000000000008
    R10: ffffffff816fed00  R11: 0000000000000004  R12: 0000000000000000
    R13: ffffffff8156cc63  R14: 0000000000000000  R15: ffff8802222a0000
    ORIG_RAX: ffffffffffffffff  CS: 0010  SS: 0018
 #8 [ffff88021f007be8] sysfs_get_dirent at ffffffff81178c07
 #9 [ffff88021f007c18] sysfs_remove_group at ffffffff8117ac27
#10 [ffff88021f007c48] netdev_queue_update_kobjects at ffffffff813178f9
#11 [ffff88021f007c88] netif_set_real_num_tx_queues at ffffffff81303e38
#12 [ffff88021f007cc8] ixgbe_set_num_queues at ffffffffa0249763 [ixgbe]
#13 [ffff88021f007cf8] ixgbe_init_interrupt_scheme at ffffffffa024ea89 [ixgbe]
#14 [ffff88021f007d48] ixgbe_fcoe_disable at ffffffffa0267113 [ixgbe]
#15 [ffff88021f007d68] vlan_dev_fcoe_disable at ffffffffa014fef5 [8021q]
#16 [ffff88021f007d78] fcoe_interface_cleanup at ffffffffa02b7dfd [fcoe]
#17 [ffff88021f007df8] fcoe_destroy_work at ffffffffa02b7f08 [fcoe]
#18 [ffff88021f007e18] process_one_work at ffffffff8105d7ca
#19 [ffff88021f007e68] worker_thread at ffffffff81060513
#20 [ffff88021f007ee8] kthread at ffffffff810648b6
#21 [ffff88021f007f48] kernel_thread_helper at ffffffff813c40f4
Signed-off-by: NYi Zou <yi.zou@intel.com>
Tested-by: NRoss Brattain <ross.b.brattain@intel.com>
Tested-by: NStephen Ko <stephen.s.ko@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This patch fixes an issue in which RSC will generate corrupted frames when
PAGE_SIZE is larger than 8K.  Specifically it looks like that in 2.6.39 a
change was made so that GRO would always have at least 16 frags available
for coalescing, but the ixgbe RSC logic was not updated.  As such the RSC
feature would generate a frame larger than 64K and then overflow the value
in the IP length field.

To correct that I am now basing things on the PAGE_SIZE.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NStephen Ko <stephen.s.ko@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

A check for the number of VFs allocated should have used a greater than
equal operator instead of just greater than.  This caused allocation of
exactly 32 VFs to not enable the PF transmit and receive enables.
Signed-off-by: NGreg Rose <gregory.v.rose@intel.com>
Tested-by: NRobert E Garrett <robertX.e.garrett@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Recent addition of code to find already allocated VFs failed to take
account that systems with 2 or more multi-port SR-IOV capable controllers
might have already enabled VFs.  Make sure that the VFs the function is
finding are actually subordinate to the particular instance of the adapter
that is looking for them and not subordinate to some device that has
previously enabled SR-IOV.

This bug exists in 3.2 stable as well as 3.3 release candidates.

CC: stable@vger.kernel.org
Reported-by: NDavid Ahern <daahern@cisco.com>
Signed-off-by: NGreg Rose <gregory.v.rose@intel.com>
Tested-by: NRobert E Garrett <robertX.e.garrett@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Correct spelling "packtetbuf_num" to "packetbuf_num" in
drivers/net/ethernet/intel/ixgbe/ixgbe_common.h
Signed-off-by: NMasanari Iida <standby24x7@gmail.com>
Tested-by: NPhil Schmitt <phillip.j.schmitt@intel.com>

Signed-off-by: NStephen Hemminger <shemminger@vyatta.com>
Tested-by: NPhil Schmitt <phillip.j.schmitt@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>