| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/next-queue
Tony Nguyen says:
====================
1GbE Intel Wired LAN Driver Updates 2022-02-07
Corinna Vinschen says:
Fix the kernel warning "Missing unregister, handled but fix driver"
when running, e.g.,
$ ethtool -G eth0 rx 1024
on igc. Remove memset hack from igb and align igb code to igc.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/linux
Nguyen, Anthony L says:
====================
iwl-next Intel Wired LAN Driver Updates 2022-02-07
Dave adds support for ice driver to provide DSCP QoS mappings to irdma
driver.
[1] https://lore.kernel.org/netdev/20220202191921.1638-1-shiraz.saleem@intel.com/
* 'iwl-next' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/linux:
ice: add support for DSCP QoS for IDC
====================
Link: https://lore.kernel.org/r/20220207235921.1303522-1-anthony.l.nguyen@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
In some cases, pages cannot be reused by i40e because the page is busy. Add
a counter for this event.
Busy page count is accessible via ethtool.
Signed-off-by: Joe Damato <jdamato@fastly.com>
Tested-by: Dave Switzer <david.switzer@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
In some cases, pages can not be reused because they are not associated with
the correct NUMA zone. Knowing how often pages are waived helps users to
understand the interaction between the driver's memory usage and their
system.
Pass rx_stats through to i40e_can_reuse_rx_page to allow tracking when
pages are waived.
The page waive count is accessible via ethtool.
Signed-off-by: Joe Damato <jdamato@fastly.com>
Tested-by: Dave Switzer <david.switzer@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
Add a counter for new page allocations in the i40e RX path. This stat is
accessible with ethtool.
Signed-off-by: Joe Damato <jdamato@fastly.com>
Tested-by: Dave Switzer <david.switzer@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
rx page reuse was already being tracked by the i40e driver per RX ring.
Aggregate the counts and make them accessible via ethtool.
Signed-off-by: Joe Damato <jdamato@fastly.com>
Tested-by: Dave Switzer <david.switzer@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
Page reuse was being tracked from two locations:
- i40e_reuse_rx_page (via 40e_clean_rx_irq), and
- i40e_alloc_mapped_page
Remove the double count and only count reuse from i40e_alloc_mapped_page
when the page is about to be reused.
Signed-off-by: Joe Damato <jdamato@fastly.com>
Tested-by: Dave Switzer <david.switzer@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
On changing the RX ring parameters igb uses a hack to avoid a warning
when calling xdp_rxq_info_reg via igb_setup_rx_resources. It just
clears the struct xdp_rxq_info content.
Instead, change this to unregister if we're already registered. Align
code to the igc code.
Fixes: 9cbc948b5a20c ("igb: add XDP support")
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
Acked-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Tested-by: Sandeep Penigalapati <sandeep.penigalapati@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
Calling ethtool changing the RX ring parameters like this:
$ ethtool -G eth0 rx 1024
on igc triggers kernel warnings like this:
[ 225.198467] ------------[ cut here ]------------
[ 225.198473] Missing unregister, handled but fix driver
[ 225.198485] WARNING: CPU: 7 PID: 959 at net/core/xdp.c:168
xdp_rxq_info_reg+0x79/0xd0
[...]
[ 225.198601] Call Trace:
[ 225.198604] <TASK>
[ 225.198609] igc_setup_rx_resources+0x3f/0xe0 [igc]
[ 225.198617] igc_ethtool_set_ringparam+0x30e/0x450 [igc]
[ 225.198626] ethnl_set_rings+0x18a/0x250
[ 225.198631] genl_family_rcv_msg_doit+0xca/0x110
[ 225.198637] genl_rcv_msg+0xce/0x1c0
[ 225.198640] ? rings_prepare_data+0x60/0x60
[ 225.198644] ? genl_get_cmd+0xd0/0xd0
[ 225.198647] netlink_rcv_skb+0x4e/0xf0
[ 225.198652] genl_rcv+0x24/0x40
[ 225.198655] netlink_unicast+0x20e/0x330
[ 225.198659] netlink_sendmsg+0x23f/0x480
[ 225.198663] sock_sendmsg+0x5b/0x60
[ 225.198667] __sys_sendto+0xf0/0x160
[ 225.198671] ? handle_mm_fault+0xb2/0x280
[ 225.198676] ? do_user_addr_fault+0x1eb/0x690
[ 225.198680] __x64_sys_sendto+0x20/0x30
[ 225.198683] do_syscall_64+0x38/0x90
[ 225.198687] entry_SYSCALL_64_after_hwframe+0x44/0xae
[ 225.198693] RIP: 0033:0x7f7ae38ac3aa
igc_ethtool_set_ringparam() copies the igc_ring structure but neglects to
reset the xdp_rxq_info member before calling igc_setup_rx_resources().
This in turn calls xdp_rxq_info_reg() with an already registered xdp_rxq_info.
Make sure to unregister the xdp_rxq_info structure first in
igc_setup_rx_resources.
Fixes: 73f1071c1d29 ("igc: Add support for XDP_TX action")
Reported-by: Lennert Buytenhek <buytenh@arista.com>
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
Acked-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Tested-by: Dvora Fuxbrumer <dvorax.fuxbrumer@linux.intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
From 4.17 onwards the ixgbevf driver uses build_skb() to build an skb
around new data in the page buffer shared with the ixgbe PF.
This uses either a 2K or 3K buffer, and offsets the DMA mapping by
NET_SKB_PAD + NET_IP_ALIGN. When using a smaller buffer RXDCTL is set to
ensure the PF does not write a full 2K bytes into the buffer, which is
actually 2K minus the offset.
However on the 82599 virtual function, the RXDCTL mechanism is not
available. The driver attempts to work around this by using the SET_LPE
mailbox method to lower the maximm frame size, but the ixgbe PF driver
ignores this in order to keep the PF and all VFs in sync[0].
This means the PF will write up to the full 2K set in SRRCTL, causing it
to write NET_SKB_PAD + NET_IP_ALIGN bytes past the end of the buffer.
With 4K pages split into two buffers, this means it either writes
NET_SKB_PAD + NET_IP_ALIGN bytes past the first buffer (and into the
second), or NET_SKB_PAD + NET_IP_ALIGN bytes past the end of the DMA
mapping.
Avoid this by only enabling build_skb when using "large" buffers (3K).
These are placed in each half of an order-1 page, preventing the PF from
writing past the end of the mapping.
[0]: Technically it only ever raises the max frame size, see
ixgbe_set_vf_lpe() in ixgbe_sriov.c
Fixes: f15c5ba5b6cd ("ixgbevf: add support for using order 1 pages to receive large frames")
Signed-off-by: Samuel Mendoza-Jonas <samjonas@amazon.com>
Tested-by: Konrad Jankowski <konrad0.jankowski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/next-queue
Tony Nguyen says:
====================
40GbE Intel Wired LAN Driver Updates 2022-02-03
This series contains updates to the i40e client header file and driver.
Mateusz disables HW TC offload by default.
Joe Damato removes a no longer used statistic.
Jakub Kicinski removes an unused enum from the client header file.
Jedrzej changes some admin queue commands to occur under atomic context
and adds new functions for admin queue MAC VLAN filters to avoid a
potential race that could occur due storing results in a structure that
could be overwritten by the next admin queue call.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
No conflicts.
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
The ice driver provides QoS information to auxiliary drivers
through the exported function ice_get_qos_params. This function
doesn't currently support L3 DSCP QoS.
Add the necessary defines, structure elements and code to support
DSCP QoS through the IIDC functions.
Signed-off-by: Dave Ertman <david.m.ertman@intel.com>
Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
There was a race condition in access to hw->aq.asq_last_status
while adding and deleting MAC/VLAN filters causing
incorrect error status to be printed as ERROR OK instead of
the correct error.
Change calls to i40e_aq_add_macvlan in i40e_aqc_add_filters
and i40e_aq_remove_macvlan in i40e_aqc_del_filters
to _v2 versions that return Admin Queue status on the stack
to avoid race conditions in access to hw->aq.asq_last_status.
Signed-off-by: Sylwester Dziedziuch <sylwesterx.dziedziuch@intel.com>
Signed-off-by: Jedrzej Jagielski <jedrzej.jagielski@intel.com>
Tested-by: Gurucharan G <gurucharanx.g@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
ASQ send command functions are returning only i40e status codes
yet some calling functions also need Admin Queue status
that is stored in hw->aq.asq_last_status. Since hw object
is stored on a heap it introduces a possibility for
a race condition in access to hw if calling function is not
fast enough to read hw->aq.asq_last_status before next
send ASQ command is executed.
Add new _v2 version of i40e_aq_add_macvlan that is using
new _v2 versions of ASQ send command functions and returns
the Admin Queue status on the stack.
Signed-off-by: Sylwester Dziedziuch <sylwesterx.dziedziuch@intel.com>
Signed-off-by: Jedrzej Jagielski <jedrzej.jagielski@intel.com>
Tested-by: Gurucharan G <gurucharanx.g@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
ASQ send command functions are returning only i40e status codes
yet some calling functions also need Admin Queue status
that is stored in hw->aq.asq_last_status. Since hw object
is stored on a heap it introduces a possibility for
a race condition in access to hw if calling function is not
fast enough to read hw->aq.asq_last_status before next
send ASQ command is executed.
Add new versions of send ASQ command functions that return
Admin Queue status on the stack to avoid race conditions
in access to hw->aq.asq_last_status.
Add new _v2 version of i40e_aq_remove_macvlan that is using
new _v2 versions of ASQ send command functions and returns
the Admin Queue status on the stack.
Signed-off-by: Sylwester Dziedziuch <sylwesterx.dziedziuch@intel.com>
Signed-off-by: Jedrzej Jagielski <jedrzej.jagielski@intel.com>
Tested-by: Gurucharan G <gurucharanx.g@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
Change functions:
- i40e_aq_add_macvlan
- i40e_aq_remove_macvlan
- i40e_aq_delete_element
- i40e_aq_add_vsi
- i40e_aq_update_vsi_params
to explicitly use i40e_asq_send_command_atomic(..., true)
instead of i40e_asq_send_command, as they use mutexes and do some
work in an atomic context.
Without this change setting vlan via netdev will fail with
call trace cased by bug "BUG: scheduling while atomic".
Signed-off-by: Witold Fijalkowski <witoldx.fijalkowski@intel.com>
Signed-off-by: Jedrzej Jagielski <jedrzej.jagielski@intel.com>
Tested-by: Gurucharan G <gurucharanx.g@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
After commit 1a557afc4dd5 ("i40e: Refactor receive routine"),
rx_stats.realloc_count is no longer being incremented, so remove it.
The debugfs string was left, but hardcoded to 0. This is intended to
prevent breaking any existing code / scripts that are parsing debugfs
for i40e.
Signed-off-by: Joe Damato <jdamato@fastly.com>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Gurucharan G <gurucharanx.g@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
After loading driver hw-tc-offload is enabled by default.
Change the behaviour of driver to disable hw-tc-offload by default as
this is the expected state. Additionally since this impacts ntuple
feature state change the way of checking NETIF_F_HW_TC flag.
Signed-off-by: Norbert Zulinski <norbertx.zulinski@intel.com>
Signed-off-by: Przemyslaw Patynowski <przemyslawx.patynowski@intel.com>
Signed-off-by: Mateusz Palczewski <mateusz.palczewski@intel.com>
Tested-by: Dave Switzer <david.switzer@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/net-queue
Tony Nguyen says:
====================
Intel Wired LAN Driver Updates 2022-02-01
This series contains updates to e1000e driver only.
Sasha removes CSME handshake with TGL platform as this is not supported
and is causing hardware unit hangs to be reported.
* '1GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/net-queue:
e1000e: Handshake with CSME starts from ADL platforms
e1000e: Separate ADP board type from TGP
====================
Link: https://lore.kernel.org/r/20220201173754.580305-1-anthony.l.nguyen@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Handshake with CSME/AMT on none provisioned platforms during S0ix flow
is not supported on TGL platform and can cause to HW unit hang. Update
the handshake with CSME flow to start from the ADL platform.
Fixes: 3e55d231716e ("e1000e: Add handshake with the CSME to support S0ix")
Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Nechama Kraus <nechamax.kraus@linux.intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
We have the same LAN controller on different PCH's. Separate ADP board
type from a TGP which will allow for specific fixes to be applied for
ADP platforms.
Suggested-by: Kai-Heng Feng <kai.heng.feng@canonical.com>
Suggested-by: Dima Ruinskiy <dima.ruinskiy@intel.com>
Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Nechama Kraus <nechamax.kraus@linux.intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
Fix the crash in kernel while dereferencing the NULL pointer,
when the driver is unloaded and simultaneously the VSI rings
are being stopped.
The hardware requires 50msec in order to finish RX queues
disable. For this purpose the driver spins in mdelay function
for the operation to be completed.
For example changing number of queues which requires reset would
fail in the following call stack:
1) i40e_prep_for_reset
2) i40e_pf_quiesce_all_vsi
3) i40e_quiesce_vsi
4) i40e_vsi_close
5) i40e_down
6) i40e_vsi_stop_rings
7) i40e_vsi_control_rx -> disable requires the delay of 50msecs
8) continue back in i40e_down function where
i40e_clean_tx_ring(vsi->tx_rings[i]) is going to crash
When the driver was spinning vsi_release called
i40e_vsi_free_arrays where the vsi->tx_rings resources
were freed and the pointer was set to NULL.
Fixes: 5b6d4a7f20b0 ("i40e: Fix crash during removing i40e driver")
Signed-off-by: Slawomir Laba <slawomirx.laba@intel.com>
Signed-off-by: Sylwester Dziedziuch <sylwesterx.dziedziuch@intel.com>
Signed-off-by: Karen Sornek <karen.sornek@intel.com>
Tested-by: Gurucharan G <gurucharanx.g@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
There was an AQ error I40E_AQ_RC_EINVAL when trying
to reset bw limit as part of bw allocation setup.
This was caused by trying to reset bw limit with
DCB enabled. Bw limit should not be reset when
DCB is enabled. The code was relying on the pf->flags
to check if DCB is enabled but if only 1 TC is available
this flag will not be set even though DCB is enabled.
Add a check for number of TC and if it is 1
don't try to reset bw limit even if pf->flags shows
DCB as disabled.
Fixes: fa38e30ac73f ("i40e: Fix for Tx timeouts when interface is brought up if DCB is enabled")
Suggested-by: Alexander Lobakin <alexandr.lobakin@intel.com> # Flatten the condition
Signed-off-by: Sylwester Dziedziuch <sylwesterx.dziedziuch@intel.com>
Signed-off-by: Jedrzej Jagielski <jedrzej.jagielski@intel.com>
Reviewed-by: Alexander Lobakin <alexandr.lobakin@intel.com>
Tested-by: Imam Hassan Reza Biswas <imam.hassan.reza.biswas@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
For now, if the XDP prog returns XDP_PASS on XSK, the metadata
will be lost as it doesn't get copied to the skb.
Copy it along with the frame headers. Account its size on skb
allocation, and when copying just treat it as a part of the frame
and do a pull after to "move" it to the "reserved" zone.
net_prefetch() xdp->data_meta and align the copy size to speed-up
memcpy() a little and better match ixgbe_construct_skb().
Fixes: d0bcacd0a130 ("ixgbe: add AF_XDP zero-copy Rx support")
Suggested-by: Jesper Dangaard Brouer <brouer@redhat.com>
Suggested-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Alexander Lobakin <alexandr.lobakin@intel.com>
Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Tested-by: Sandeep Penigalapati <sandeep.penigalapati@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
{__,}napi_alloc_skb() allocates and reserves additional NET_SKB_PAD
+ NET_IP_ALIGN for any skb.
OTOH, ixgbe_construct_skb_zc() currently allocates and reserves
additional `xdp->data - xdp->data_hard_start`, which is
XDP_PACKET_HEADROOM for XSK frames.
There's no need for that at all as the frame is post-XDP and will
go only to the networking stack core.
Pass the size of the actual data only to __napi_alloc_skb() and
don't reserve anything. This will give enough headroom for stack
processing.
Fixes: d0bcacd0a130 ("ixgbe: add AF_XDP zero-copy Rx support")
Signed-off-by: Alexander Lobakin <alexandr.lobakin@intel.com>
Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Tested-by: Sandeep Penigalapati <sandeep.penigalapati@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
To not dereference bi->xdp each time in ixgbe_construct_skb_zc(),
pass bi->xdp as an argument instead of bi. We can also call
xsk_buff_free() outside of the function as well as assign bi->xdp
to NULL, which seems to make it closer to its name.
Suggested-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Alexander Lobakin <alexandr.lobakin@intel.com>
Tested-by: Sandeep Penigalapati <sandeep.penigalapati@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
{__,}napi_alloc_skb() allocates and reserves additional NET_SKB_PAD
+ NET_IP_ALIGN for any skb.
OTOH, igc_construct_skb_zc() currently allocates and reserves
additional `xdp->data_meta - xdp->data_hard_start`, which is about
XDP_PACKET_HEADROOM for XSK frames.
There's no need for that at all as the frame is post-XDP and will
go only to the networking stack core.
Pass the size of the actual data only (+ meta) to
__napi_alloc_skb() and don't reserve anything. This will give
enough headroom for stack processing.
Also, net_prefetch() xdp->data_meta and align the copy size to
speed-up memcpy() a little and better match igc_construct_skb().
Fixes: fc9df2a0b520 ("igc: Enable RX via AF_XDP zero-copy")
Signed-off-by: Alexander Lobakin <alexandr.lobakin@intel.com>
Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Tested-by: Nechama Kraus <nechamax.kraus@linux.intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
For now, if the XDP prog returns XDP_PASS on XSK, the metadata will
be lost as it doesn't get copied to the skb.
Copy it along with the frame headers. Account its size on skb
allocation, and when copying just treat it as a part of the frame
and do a pull after to "move" it to the "reserved" zone.
net_prefetch() xdp->data_meta and align the copy size to speed-up
memcpy() a little and better match ice_construct_skb().
Fixes: 2d4238f55697 ("ice: Add support for AF_XDP")
Suggested-by: Jesper Dangaard Brouer <brouer@redhat.com>
Suggested-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Alexander Lobakin <alexandr.lobakin@intel.com>
Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Tested-by: Kiran Bhandare <kiranx.bhandare@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
{__,}napi_alloc_skb() allocates and reserves additional NET_SKB_PAD
+ NET_IP_ALIGN for any skb.
OTOH, ice_construct_skb_zc() currently allocates and reserves
additional `xdp->data - xdp->data_hard_start`, which is
XDP_PACKET_HEADROOM for XSK frames.
There's no need for that at all as the frame is post-XDP and will
go only to the networking stack core.
Pass the size of the actual data only to __napi_alloc_skb() and
don't reserve anything. This will give enough headroom for stack
processing.
Fixes: 2d4238f55697 ("ice: Add support for AF_XDP")
Signed-off-by: Alexander Lobakin <alexandr.lobakin@intel.com>
Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Tested-by: Kiran Bhandare <kiranx.bhandare@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
In "legacy-rx" mode represented by ice_construct_skb(), we can
still use XDP (and XDP metadata), but after XDP_PASS the metadata
will be lost as it doesn't get copied to the skb.
Copy it along with the frame headers. Account its size on skb
allocation, and when copying just treat it as a part of the frame
and do a pull after to "move" it to the "reserved" zone.
Point net_prefetch() to xdp->data_meta instead of data. This won't
change anything when the meta is not here, but will save some cache
misses otherwise.
Suggested-by: Jesper Dangaard Brouer <brouer@redhat.com>
Suggested-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Alexander Lobakin <alexandr.lobakin@intel.com>
Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Tested-by: Kiran Bhandare <kiranx.bhandare@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
For now, if the XDP prog returns XDP_PASS on XSK, the metadata will
be lost as it doesn't get copied to the skb.
Copy it along with the frame headers. Account its size on skb
allocation, and when copying just treat it as a part of the frame
and do a pull after to "move" it to the "reserved" zone.
net_prefetch() xdp->data_meta and align the copy size to speed-up
memcpy() a little and better match i40e_construct_skb().
Fixes: 0a714186d3c0 ("i40e: add AF_XDP zero-copy Rx support")
Suggested-by: Jesper Dangaard Brouer <brouer@redhat.com>
Suggested-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Alexander Lobakin <alexandr.lobakin@intel.com>
Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Tested-by: Kiran Bhandare <kiranx.bhandare@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
{__,}napi_alloc_skb() allocates and reserves additional NET_SKB_PAD
+ NET_IP_ALIGN for any skb.
OTOH, i40e_construct_skb_zc() currently allocates and reserves
additional `xdp->data - xdp->data_hard_start`, which is
XDP_PACKET_HEADROOM for XSK frames.
There's no need for that at all as the frame is post-XDP and will
go only to the networking stack core.
Pass the size of the actual data only to __napi_alloc_skb() and
don't reserve anything. This will give enough headroom for stack
processing.
Fixes: 0a714186d3c0 ("i40e: add AF_XDP zero-copy Rx support")
Signed-off-by: Alexander Lobakin <alexandr.lobakin@intel.com>
Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
Tested-by: Kiran Bhandare <kiranx.bhandare@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
As stated in [1], dma_set_mask() with a 64-bit mask never fails if
dev->dma_mask is non-NULL.
So, if it fails, the 32 bits case will also fail for the same reason.
So, if dma_set_mask_and_coherent() succeeds, 'pci_using_dac' is known to be
1.
Simplify code and remove some dead code accordingly.
[1]: https://lkml.org/lkml/2021/6/7/398
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Tested-by: Konrad Jankowski <konrad0.jankowski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
As stated in [1], dma_set_mask() with a 64-bit mask never fails if
dev->dma_mask is non-NULL.
So, if it fails, the 32 bits case will also fail for the same reason.
So, if dma_set_mask_and_coherent() succeeds, 'pci_using_dac' is known to be
1.
Simplify code and remove some dead code accordingly.
[1]: https://lkml.org/lkml/2021/6/7/398
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Tested-by: Gurucharan G <gurucharanx.g@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
As stated in [1], dma_set_mask() with a 64-bit mask never fails if
dev->dma_mask is non-NULL.
So, if it fails, the 32 bits case will also fail for the same reason.
So, if dma_set_mask_and_coherent() succeeds, 'pci_using_dac' is known to be
1.
Simplify code and remove some dead code accordingly.
[1]: https://lkml.org/lkml/2021/6/7/398
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
As stated in [1], dma_set_mask() with a 64-bit mask never fails if
dev->dma_mask is non-NULL.
So, if it fails, the 32 bits case will also fail for the same reason.
Simplify code and remove some dead code accordingly.
[1]: https://lkml.org/lkml/2021/6/7/398
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Alexander Lobakin <alexandr.lobakin@intel.com>
Tested-by: Gurucharan G <gurucharanx.g@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
As stated in [1], dma_set_mask() with a 64-bit mask never fails if
dev->dma_mask is non-NULL.
So, if it fails, the 32 bits case will also fail for the same reason.
Simplify code and remove some dead code accordingly.
[1]: https://lkml.org/lkml/2021/6/7/398
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Alexander Lobakin <alexandr.lobakin@intel.com>
Tested-by: Konrad Jankowski <konrad0.jankowski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
As stated in [1], dma_set_mask() with a 64-bit mask never fails if
dev->dma_mask is non-NULL.
So, if it fails, the 32 bits case will also fail for the same reason.
So, if dma_set_mask_and_coherent() succeeds, 'pci_using_dac' is known to be
1.
Simplify code and remove some dead code accordingly.
[1]: https://lkml.org/lkml/2021/6/7/398
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Alexander Lobakin <alexandr.lobakin@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
As stated in [1], dma_set_mask() with a 64-bit mask never fails if
dev->dma_mask is non-NULL.
So, if it fails, the 32 bits case will also fail for the same reason.
Simplify code and remove some dead code accordingly.
[1]: https://lkml.org/lkml/2021/6/7/398
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Alexander Lobakin <alexandr.lobakin@intel.com>
Tested-by: Gurucharan G <gurucharanx.g@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
As stated in [1], dma_set_mask() with a 64-bit mask never fails if
dev->dma_mask is non-NULL.
So, if it fails, the 32 bits case will also fail for the same reason.
So, if dma_set_mask_and_coherent() succeeds, 'pci_using_dac' is known to be
1.
Simplify code and remove some dead code accordingly.
[1]: https://lkml.org/lkml/2021/6/7/398
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Alexander Lobakin <alexandr.lobakin@intel.com>
Tested-by: Gurucharan G <gurucharanx.g@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
As stated in [1], dma_set_mask() with a 64-bit mask never fails if
dev->dma_mask is non-NULL.
So, if it fails, the 32 bits case will also fail for the same reason.
So, if dma_set_mask_and_coherent() succeeds, 'pci_using_dac' is known to be
1.
Simplify code and remove some dead code accordingly.
[1]: https://lkml.org/lkml/2021/6/7/398
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Alexander Lobakin <alexandr.lobakin@intel.com>
Tested-by: Gurucharan G <gurucharanx.g@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
As stated in [1], dma_set_mask() with a 64-bit mask never fails if
dev->dma_mask is non-NULL.
So, if it fails, the 32 bits case will also fail for the same reason.
So, if dma_set_mask_and_coherent() succeeds, 'pci_using_dac' is known to be
1.
Simplify code and remove some dead code accordingly.
[1]: https://lkml.org/lkml/2021/6/7/398
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Alexander Lobakin <alexandr.lobakin@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
One of the things that commit 5574ff7b7b3d ("i40e: optimize AF_XDP Tx
completion path") introduced was the @xdp_tx_active field. Its usage
from i40e can be adjusted to ice driver and give us positive performance
results.
If the descriptor that @next_dd points to has been sent by HW (its DD
bit is set), then we are sure that at least quarter of the ring is ready
to be cleaned. If @xdp_tx_active is 0 which means that related xdp_ring
is not used for XDP_{TX, REDIRECT} workloads, then we know how many XSK
entries should placed to completion queue, IOW walking through the ring
can be skipped.
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Alexander Lobakin <alexandr.lobakin@intel.com>
Acked-by: Magnus Karlsson <magnus.karlsson@intel.com>
Link: https://lore.kernel.org/bpf/20220125160446.78976-9-maciej.fijalkowski@intel.com
|
|
Apply the logic that was done for regular XDP from commit 9610bd988df9
("ice: optimize XDP_TX workloads") to the ZC side of the driver. On top
of that, introduce batching to Tx that is inspired by i40e's
implementation with adjustments to the cleaning logic - take into the
account NAPI budget in ice_clean_xdp_irq_zc().
Separating the stats structs onto separate cache lines seemed to improve
the performance.
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Alexander Lobakin <alexandr.lobakin@intel.com>
Link: https://lore.kernel.org/bpf/20220125160446.78976-8-maciej.fijalkowski@intel.com
|
|
Commit 9610bd988df9 ("ice: optimize XDP_TX workloads") introduced
@next_dd and @next_rs to ice_tx_ring struct. Currently, their state is
not restored in ice_clean_tx_ring(), which was not causing any troubles
as the XDP rings are gone after we're done with XDP prog on interface.
For upcoming usage of mentioned fields in AF_XDP, this might expose us
to a potential dead Tx side. Scenario would look like following (based
on xdpsock):
- two xdpsock instances are spawned in Tx mode
- one of them is killed
- XDP prog is kept on interface due to the other xdpsock still running
* this means that XDP rings stayed in place
- xdpsock is launched again on same queue id that was terminated on
- @next_dd and @next_rs setting is bogus, therefore transmit side is
broken
To protect us from the above, restore the initial @next_rs and @next_dd
values when cleaning the Tx ring.
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Alexander Lobakin <alexandr.lobakin@intel.com>
Acked-by: Magnus Karlsson <magnus.karlsson@intel.com>
Link: https://lore.kernel.org/bpf/20220125160446.78976-7-maciej.fijalkowski@intel.com
|
|
Move desc_array from the driver to the pool. The reason behind this is
that we can then reuse this array as a temporary storage for descriptors
in all zero-copy drivers that use the batched interface. This will make
it easier to add batching to more drivers.
i40e is the only driver that has a batched Tx zero-copy
implementation, so no need to touch any other driver.
Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Alexander Lobakin <alexandr.lobakin@intel.com>
Link: https://lore.kernel.org/bpf/20220125160446.78976-6-maciej.fijalkowski@intel.com
|
|
XDP_TX workloads use a concept of Tx threshold that indicates the
interval of setting RS bit on descriptors which in turn tells the HW to
generate an interrupt to signal the completion of Tx on HW side. It is
currently based on a constant value of 32 which might not work out well
for various sizes of ring combined with for example batch size that can
be set via SO_BUSY_POLL_BUDGET.
Internal tests based on AF_XDP showed that most convenient setup of
mentioned threshold is when it is equal to quarter of a ring length.
Make use of recently introduced ICE_RING_QUARTER macro and use this
value as a substitute for ICE_TX_THRESH.
Align also ethtool -G callback so that next_dd/next_rs fields are up to
date in terms of the ring size.
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Alexander Lobakin <alexandr.lobakin@intel.com>
Acked-by: Magnus Karlsson <magnus.karlsson@intel.com>
Link: https://lore.kernel.org/bpf/20220125160446.78976-5-maciej.fijalkowski@intel.com
|
|
Currently, if ice_clean_rx_irq_zc() processed the whole ring and
next_to_use != 0, then ice_alloc_rx_buf_zc() would not refill the whole
ring even if the XSK buffer pool would have enough free entries (either
from fill ring or the internal recycle mechanism) - it is because ring
wrap is not handled.
Improve the logic in ice_alloc_rx_buf_zc() to address the problem above.
Do not clamp the count of buffers that is passed to
xsk_buff_alloc_batch() in case when next_to_use + buffer count >=
rx_ring->count, but rather split it and have two calls to the mentioned
function - one for the part up until the wrap and one for the part after
the wrap.
Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Alexander Lobakin <alexandr.lobakin@intel.com>
Link: https://lore.kernel.org/bpf/20220125160446.78976-4-maciej.fijalkowski@intel.com
|
|
With the upcoming introduction of batching to XSK data path,
performance wise it will be the best to have the ring descriptor count
to be aligned to power of 2.
Check if ring sizes that user is going to attach the XSK socket fulfill
the condition above. For Tx side, although check is being done against
the Tx queue and in the end the socket will be attached to the XDP
queue, it is fine since XDP queues get the ring->count setting from Tx
queues.
Suggested-by: Alexander Lobakin <alexandr.lobakin@intel.com>
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Alexander Lobakin <alexandr.lobakin@intel.com>
Acked-by: Magnus Karlsson <magnus.karlsson@intel.com>
Link: https://lore.kernel.org/bpf/20220125160446.78976-3-maciej.fijalkowski@intel.com
|
