| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
The ice driver has support for adaptive interrupt moderation, an
algorithm for tuning the interrupt rate dynamically. This algorithm
is based on various assumptions about ring size, socket buffer size,
link speed, SKB overhead, ethernet frame overhead and more.
The Linux kernel has support for a dynamic interrupt moderation
algorithm known as "dimlib". Replace the custom driver-specific
implementation of dynamic interrupt moderation with the kernel's
algorithm.
The Intel hardware has a different hardware implementation than the
originators of the dimlib code had to work with, which requires the
driver to use a slightly different set of inputs for the actual
moderation values, while getting all the advice from dimlib of
better/worse, shift left or right.
The change made for this implementation is to use a pair of values
for each of the 5 "slots" that the dimlib moderation expects, and
the driver will program those pairs when dimlib recommends a slot to
use. The currently implementation uses two tables, one for receive
and one for transmit, and the pairs of values in each slot set the
maximum delay of an interrupt and a maximum number of interrupts per
second (both expressed in microseconds).
There are two separate kinds of bugs fixed by using DIMLIB, one is
UDP single stream send was too slow, and the other is that 8K
ping-pong was going to the most aggressive moderation and has much
too high latency.
The overall result of using DIMLIB is that we meet or exceed our
performance expectations set based on the old algorithm.
Co-developed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Tested-by: Tony Brelinski <tonyx.brelinski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
Introduce several new helpers for writing ITR and GLINT_RATE
registers, and refactor the code calling them. This resulted
in removal of several duplicate functions and rolled a bunch
of simple code back into the calling routines.
In particular this removes some code that was doing both
a store and a set in a helper function, which seems better
done as separate tasks in the caller (and generally takes
less lines of code even with a tiny bit of repetition).
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Tony Brelinski <tonyx.brelinski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
Add two new VSI states, one to track if a netdev for the VSI has been
allocated and the other to track if the netdev has been registered.
Call unregister_netdev/free_netdev only when the corresponding state
bits are set.
Signed-off-by: Anirudh Venkataramanan <anirudh.venkataramanan@intel.com>
Tested-by: Tony Brelinski <tonyx.brelinski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
Remove the leading underscores in enum ice_pf_state. This is not really
communicating anything and is unnecessary. No functional change.
Signed-off-by: Anirudh Venkataramanan <anirudh.venkataramanan@intel.com>
Tested-by: Tony Brelinski <tonyx.brelinski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
The well-known IANA protocol port 3260 (iscsi-target 0x0cbc) and the
ether-types 0x8906 (ETH_P_FCOE) and 0x8914 (ETH_P_FIP) are already defined
in kernel header files. Use those definitions instead of open-coding the
same.
Signed-off-by: Bruce Allan <bruce.w.allan@intel.com>
Tested-by: Tony Brelinski <tonyx.brelinski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
In the nft_offload there is the mate flow_dissector with no
ingress_ifindex but with ingress_iftype that only be used
in the software. So if the mask of ingress_ifindex in meta is
0, this meta check should be bypass.
Fixes: 6d65bc64e232 ("net/mlx5e: Add mlx5e_flower_parse_meta support")
Signed-off-by: wenxu <wenxu@ucloud.cn>
Acked-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
Change register setting from bit number to bit mask.
Fixes: b5ede32d3329 ("net/mlx5e: Add support for FEC modes based on 50G per lane links")
Signed-off-by: Aya Levin <ayal@nvidia.com>
Reviewed-by: Eran Ben Elisha <eranbe@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
Prevent setting of devlink traps on the uplink while in switchdev mode.
In this mode, it is the SW switch responsibility to handle both packets
with a mismatch in destination MAC or VLAN ID. Therefore, there are no
flow steering tables to trap undesirable packets and driver crashes upon
setting a trap.
Fixes: 241dc159391f ("net/mlx5: Notify on trap action by blocking event")
Signed-off-by: Aya Levin <ayal@nvidia.com>
Reviewed-by: Moshe Shemesh <moshe@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux
Saeed Mahameed says:
====================
mlx5-updates-2021-04-13
mlx5 core and netdev driver updates
1) E-Switch updates from Parav,
1.1) Devlink parameter to control mlx5 metadata enablement for E-Switch
1.2) Trivial cleanups for E-Switch code
1.3) Dynamically allocate vport steering namespaces only when required
2) From Jianbo, Use variably sized data structures for Software steering
3) Several minor cleanups
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/net-queue
Tony Nguyen says:
====================
Intel Wired LAN Driver Updates 2021-04-14
This series contains updates to ixgbe and ice drivers.
Alex Duyck fixes a NULL pointer dereference for ixgbe.
Yongxin Liu fixes an unbalanced enable/disable which was causing a call
trace with suspend for ixgbe.
Colin King fixes a potential infinite loop for ice.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
This reverts commit 9c63faaa931e443e7abbbee9de0169f1d4710546, which
introduces a suspend/resume regression on Jetson TX2 boards that can be
reproduced every time. Given that the issue that this was supposed to
fix only occurs very sporadically the safest course of action is to
revert before v5.12 and then we can have another go at fixing the more
rare issue in the next release (and perhaps backport it if necessary).
The root cause of the observed problem seems to be that when the system
is suspended, some packets are still in transit. When the descriptors
for these buffers are cleared on resume, the descriptors become invalid
and cause a fatal bus error.
Link: https://lore.kernel.org/r/708edb92-a5df-ecc4-3126-5ab36707e275@nvidia.com/
Reported-by: Jonathan Hunter <jonathanh@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Normally, the bootloader will already initialize the MAC address
registers of the ENETC and the driver will just use them or generate a
random one, if it is not initialized.
Add a new way to provide the MAC address: via device tree. Besides the
usual 'mac-address' property, there is also the possibility to fetch it
via a NVMEM provider. The sl28 board stores the MAC address in the SPI
NOR flash OTP region. Having this will allow linux to fetch the MAC
address from there without being dependent on the bootloader.
No in-tree boards have the device tree properties set, thus for these,
this is a no-op.
Signed-off-by: Michael Walle <michael@walle.cc>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Fix the following coccicheck warning:
./drivers/net/ethernet/cavium/liquidio/cn66xx_regs.h:413:6-28:
duplicated argument to & or |
The CN6XXX_INTR_M1UPB0_ERR here is duplicate.
Here should be CN6XXX_INTR_M1UNB0_ERR.
Signed-off-by: Wan Jiabing <wanjiabing@vivo.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Commit a14d273ba159 ("net: macb: restore cmp registers on resume path")
introduces the restore of CMP registers on resume path. In case the IP
doesn't support type 2 screeners (zero on DCFG8 register) the
struct macb::rx_fs_list::list is not initialized and thus the
list_for_each_entry(item, &bp->rx_fs_list.list, list) loop introduced in
commit a14d273ba159 ("net: macb: restore cmp registers on resume path")
will access an uninitialized list leading to crash. Thus, initialize
the struct macb::rx_fs_list::list without taking into account if the
IP supports type 2 screeners or not.
Fixes: a14d273ba159 ("net: macb: restore cmp registers on resume path")
Signed-off-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Fix the following coccicheck warning:
./drivers/net/ethernet/sfc/enum.h:80:7-28: duplicated argument to |
Signed-off-by: Wan Jiabing <wanjiabing@vivo.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Remove the unnecessary napi_schedule() call in __ibmvnic_open() since
interrupt_rx() calls napi_schedule_prep/__napi_schedule during every
receive interrupt.
Fixes: ed651a10875f ("ibmvnic: Updated reset handling")
Signed-off-by: Lijun Pan <lijunp213@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
During adapter reset, do_reset/do_hard_reset calls ibmvnic_open(),
which will calls napi_schedule if previous state is VNIC_CLOSED
(i.e, the reset case, and "ifconfig down" case). So there is no need
for do_reset to call napi_schedule again at the end of the function
though napi_schedule will neglect the request if napi is already
scheduled.
Fixes: ed651a10875f ("ibmvnic: Updated reset handling")
Signed-off-by: Lijun Pan <lijunp213@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
__ibmvnic_open calls napi_disable without checking whether NAPI polling
has already been disabled or not. This could cause napi_disable
being called twice, which could generate deadlock. For example,
the first napi_disable will spin until NAPI_STATE_SCHED is cleared
by napi_complete_done, then set it again.
When napi_disable is called the second time, it will loop infinitely
because no dev->poll will be running to clear NAPI_STATE_SCHED.
To prevent above scenario from happening, call ibmvnic_napi_disable()
which checks if napi is disabled or not before calling napi_disable.
Fixes: bfc32f297337 ("ibmvnic: Move resource initialization to its own routine")
Suggested-by: Thomas Falcon <tlfalcon@linux.ibm.com>
Signed-off-by: Lijun Pan <lijunp213@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
It has been reported [0] that using pause frames in jumbo mode impacts
performance. There's no available chip documentation, but vendor
drivers r8168 and r8125 don't advertise pause in jumbo mode. So let's
do the same, according to Roman it fixes the issue.
[0] https://bugzilla.kernel.org/show_bug.cgi?id=212617
Fixes: 9cf9b84cc701 ("r8169: make use of phy_set_asym_pause")
Reported-by: Roman Mamedov <rm+bko@romanrm.net>
Tested-by: Roman Mamedov <rm+bko@romanrm.net>
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Cc: stable@vger.kernel.org
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
This adds support for the [g|s]et_pauseparam ethtool ops. It considers
that the chip doesn't support pause frame use in jumbo mode.
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.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:
====================
10GbE Intel Wired LAN Driver Updates 2021-04-13
This series contains updates to ixgbe and ixgbevf driver.
Jostar Yang adds support for BCM54616s PHY for ixgbe.
Chen Lin removes an unused function pointer for ixgbe and ixgbevf.
Bhaskar Chowdhury fixes a typo in ixgbe.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
The Synopsis MAC controller supports auxiliary snapshot feature that
allows user to store a snapshot of the system time based on an external
event.
This patch add supports to the above mentioned feature. Users will be
able to triggered capturing the time snapshot from user-space using
application such as testptp or any other applications that uses the
PTP_EXTTS_REQUEST ioctl request.
Cc: Richard Cochran <richardcochran@gmail.com>
Signed-off-by: Tan Tee Min <tee.min.tan@intel.com>
Co-developed-by: Wong Vee Khee <vee.khee.wong@linux.intel.com>
Signed-off-by: Wong Vee Khee <vee.khee.wong@linux.intel.com>
Acked-by: Richard Cochran <richardcochran@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Allow to create an RQ which is not registered as an XDP RQ. For example:
the trap-RQ doesn't register as an XDP RQ.
Fixes: 869c5f926247 ("net/mlx5e: Generalize open RQ")
Signed-off-by: Aya Levin <ayal@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
There should be no spaces at the start of the line.
Signed-off-by: Wenpeng Liang <liangwenpeng@huawei.com>
Signed-off-by: Weihang Li <liweihang@huawei.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
void function return statements are not generally useful.
Signed-off-by: Wenpeng Liang <liangwenpeng@huawei.com>
Signed-off-by: Weihang Li <liweihang@huawei.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
There should be a blank lines after declarations.
Signed-off-by: Wenpeng Liang <liangwenpeng@huawei.com>
Signed-off-by: Weihang Li <liweihang@huawei.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
The bit-wise and of the action field with MLX5_ACCEL_ESP_ACTION_DECRYPT
is incorrect as MLX5_ACCEL_ESP_ACTION_DECRYPT is zero and not intended
to be a bit-flag. Fix this by using the == operator as was originally
intended.
Addresses-Coverity: ("Logically dead code")
Fixes: 7dfee4b1d79e ("net/mlx5: IPsec, Refactor SA handle creation and destruction")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
The cmd size is 8K so use kvzalloc().
Signed-off-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
mlx5dr_action is a generally used data structure, and there is an
union for different types of actions in it. The size of mlx5dr_action
is about 72 bytes, but for those actions with fewer fields, most of
the allocated memory is wasted.
Remove this union, and mlx5dr_action becomes a generic action header.
Then actions are dynamically allocated with needed memory, the data
for each action is stored right after the header.
Signed-off-by: Jianbo Liu <jianbol@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
SF's hardware function id is already stored in mlx5_sf. Reuse it,
instead of querying the hw table.
Signed-off-by: Parav Pandit <parav@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
At many places in the code, device pointer is directly available. Make
use of it, instead of accessing it from the table.
Signed-off-by: Parav Pandit <parav@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
Currently eswitch flow steering (FS) namespace of vport's ingress and
egress ACL are enabled when FS layer is initialized. This is done even
when eswitch is diabled. This demands that total eswitch ports to be
known to FS layer without eswitch in use.
Given the FS core is not dependent on eswitch, make namespace init and
cleanup routines as helper routines to be invoked only when eswitch is
needed.
With this change, ingress and egress ACL namespaces are created only
when eswitch legacy/offloads mode is enabled.
Signed-off-by: Parav Pandit <parav@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
Currently eswitch offers two modes. Legacy and offloads.
Offloads code is already in its own file eswitch_offloads.c
However eswitch.c contains the eswitch legacy code and common
infrastructure code.
To enable future extensions and to better manage generic common eswitch
infrastructure code, move the legacy code to its own legacy.c file.
Signed-off-by: Parav Pandit <parav@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
Convert ESW_ALLOWED macro to a helper routine so that it can be used in
other eswitch files.
Signed-off-by: Parav Pandit <parav@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
Make cleanup sequence mirror of init sequence for cleaning up reps
and freeing vports.
Also when reps initialization fails, there is no need to perform reps
cleanup.
Signed-off-by: Parav Pandit <parav@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
Vport number is 16-bit field in hardware. Make it u16.
Move location of vport in the structure so that it reduces a hole
in the structure.
Signed-off-by: Parav Pandit <parav@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
With vhca events, SF state is queried through the VHCA events. Device no
longer expects SF bitmap in the query eswitch functions command.
Hence, remove it to simplify the code.
Signed-off-by: Parav Pandit <parav@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
Currently each packet inserted in eswitch is tagged with a internal
metadata to indicate source vport. Metadata tagging is not always
needed. Metadata insertion is needed for multi-port RoCE, failover
between representors and stacked devices. In many other cases,
metadata enablement is not needed.
Metadata insertion slows down the packet processing rate of the E-switch
when it is in switchdev mode.
Below table show performance gain with metadata disabled for VXLAN
offload rules in both SMFS and DMFS steering mode on ConnectX-5 device.
----------------------------------------------
| steering | metadata | pkt size | rx pps |
| mode | | | (million) |
----------------------------------------------
| smfs | disabled | 128Bytes | 42 |
----------------------------------------------
| smfs | enabled | 128Bytes | 36 |
----------------------------------------------
| dmfs | disabled | 128Bytes | 42 |
----------------------------------------------
| dmfs | enabled | 128Bytes | 36 |
----------------------------------------------
Hence, allow user to disable metadata using driver specific devlink
parameter. Metadata setting of the eswitch is applicable only for the
switchdev mode.
Example to show and disable metadata before changing eswitch mode:
$ devlink dev param show pci/0000:06:00.0 name esw_port_metadata
pci/0000:06:00.0:
name esw_port_metadata type driver-specific
values:
cmode runtime value true
$ devlink dev param set pci/0000:06:00.0 \
name esw_port_metadata value false cmode runtime
$ devlink dev eswitch set pci/0000:06:00.0 mode switchdev
Signed-off-by: Parav Pandit <parav@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Vu Pham <vuhuong@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
changelog:
v1->v2:
- added performance numbers in commit log
- updated commit log and documentation for switchdev mode
- added explicit note on when user can disable metadata in
documentation
|
|
s/Reprogam/Reprogram/
Signed-off-by: Bhaskar Chowdhury <unixbhaskar@gmail.com>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Acked-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
Remove the 'ixgbe_mc_addr_itr' typedef as it is not used.
Signed-off-by: Chen Lin <chen.lin5@zte.com.cn>
Tested-by: Dave Switzer <david.switzer@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
The Broadcom PHY is used in switches, so add the ID, and hook it up.
This upstreams the Linux kernel patch from the network operating system
SONiC from February 2020 [1].
[1]: https://github.com/Azure/sonic-linux-kernel/pull/122
Signed-off-by: Jostar Yang <jostar_yang@accton.com>
Signed-off-by: Guohan Lu <lguohan@gmail.com>
Signed-off-by: Paul Menzel <pmenzel@molgen.mpg.de>
Tested-by: Tony Brelinski <tonyx.brelinski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
Since we added the dpaa2_switch_acl_entry_add() function in the previous
patches to hide all the details of actually adding the ACL entry by
issuing a firmware command, let's use it also for adding a CPU trap for
the STP frames.
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Add support TC_SETUP_CLSMATCHALL by using the same ACL table entries
framework as for tc flower. Adding a matchall rule is done by installing
an entry which has a mask of all zeroes, thus matching on any packet.
This can be used as a catch-all type of rule if used correctly, ie the
priority of the matchall filter should be kept as the lowest one in the
entire filter block.
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
This patch adds support for tc flower hardware offload on the ingress
path. Shared filter blocks are supported by sharing a single ACL table
between multiple ports.
The following flow keys are supported:
- Ethernet: dst_mac/src_mac
- IPv4: dst_ip/src_ip/ip_proto/tos
- VLAN: vlan_id/vlan_prio/vlan_tpid/vlan_dei
- L4: dst_port/src_port
As per flow actions, the following are supported:
- drop
- mirred egress redirect
- trap
Each ACL entry (filter) can be setup with only one of the listed
actions.
A sorted single linked list is used to keep the ACL entries by their
order of priority. When adding a new filter, this enables us to quickly
ascertain if the new entry has the highest priority of the entire block
or if we should make some space in the ACL table by increasing the
priority of the filters already in the table.
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Change the default ACL trap rule for STP frames to have the highest
priority.
In the same ACL table will reside both default rules added by the driver
for its internal use as well as rules added with tc flower. In this
case, the default rules such as the STP one that we already have should
have the highest priority.
Also, remove the check for a full ACL table since we already know that
it's sized so that we don't hit this case. The last thing changes is
that default trap filters will not be counted in the acl_tbl's num_rules
variable since their number doesn't change.
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Introduce a new structure - dpaa2_switch_acl_tbl - to hold all data
related to an ACL table: number of rules added, ACL table id, etc.
This will be used more in the next patches when adding support for
sharing an ACL table between ports.
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
The copy_to_user() function returns the number of bytes that it wasn't
able to copy. We want to return -EFAULT to the user.
Fixes: fee6efce565d ("ionic: add hw timestamp support files")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Acked-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
We add the support of XDP ZC TX submission and cleaning into
stmmac_tx_clean(). The function is made to clean as many TX complete
frames as possible, i.e. limit by priv->dma_tx_size instead of NAPI
budget. For TX ring that is associated with XSK pool, the function
stmmac_xdp_xmit_zc() is introduced to TX frame buffers from XSK pool by
using xsk_tx_peek_desc(). To make stmmac_tx_clean() support the cleaning
of XSK TX frames, STMMAC_TXBUF_T_XSK_TX TX buffer type is introduced.
As stmmac_tx_clean() uses the return value to cue whether NAPI function
should continue to poll, we augment the caller of stmmac_tx_clean() to
pass NAPI budget instead of priv->dma_tx_size through 'budget' input and
made stmmac_tx_clean() to always clean up-to the TX ring size instead.
This allows us to use the return boolean status of stmmac_xdp_xmit_zc()
to decide if XSK TX work is done or not: If true, set 'xmits' to return
'budget - 1' so that NAPI poll may exit. Else, set 'xmits' to return
'budget' to make NAPI poll continue to poll since XSK TX work is not
done. Finally, at the end of stmmac_tx_clean(), the function now take
a maximum value between 'count' and 'xmits' so that status from both
TX cleaning and XSK TX (only for XDP ZC) is considered.
This patch adds a new NAPI poll called stmmac_napi_poll_rxtx() that is
meant to be enabled/disabled for RX and TX ring that are bound to XSK
pool. This NAPI poll function starts with cleaning TX ring, then submits
XSK TX frames to TX ring before proceed to perform RX operations, i.e.
, receiving RX frames and replenishing RX ring with RX free buffers
obtained from XSK pool. Therefore, during XSK RX and TX setup, the driver
enables stmmac_napi_poll_rxtx() for RX and TX operations, then during
XSK RX and TX pool tear-down, the driver reenables the exisiting
independent NAPI poll functions accordingly: stmmac_napi_poll_rx() and
stmmac_napi_poll_tx().
Signed-off-by: Ong Boon Leong <boon.leong.ong@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
This patch adds the support for receiving packet via AF_XDP zero-copy
mechanism.
XDP ZC uses 1:1 mapping of XDP buffer to receive packet, therefore the
use of split header is not used currently. The 'xdp_buff' is declared as
union together with a struct that contains 'page', 'addr' and
'page_offset' that are associated with primary buffer.
RX buffers are now allocated either via page_pool or xsk pool. For RX
buffers from xsk_pool they are allocated and deallocated using below
functions:
* stmmac_alloc_rx_buffers_zc(struct stmmac_priv *priv, u32 queue)
* dma_free_rx_xskbufs(struct stmmac_priv *priv, u32 queue)
With above functions now available, we then extend the following driver
functions to support XDP ZC:
* stmmac_reinit_rx_buffers()
* __init_dma_rx_desc_rings()
* init_dma_rx_desc_rings()
* __free_dma_rx_desc_resources()
Note: stmmac_alloc_rx_buffers_zc() may return -ENOMEM due to RX XDP
buffer pool is not allocated (e.g. samples/bpf/xdpsock TX-only). But,
it is still ok to let TX XDP ZC to continue, therefore, the -ENOMEM
is silently ignored to let the driver succcessfully transition to XDP
ZC mode for the said RX and TX queue.
As XDP ZC buffer size is different, the DMA buffer size is required
to be reprogrammed accordingly for RX DMA/Queue that is populated with
XDP buffer from XSK pool.
Next, to add or remove per-queue XSK pool, stmmac_xdp_setup_pool()
will call stmmac_xdp_enable_pool() or stmmac_xdp_disable_pool()
that in-turn coordinates the tearing down and setting up RX ring via
RX buffers and descriptors removal and reallocation through
stmmac_disable_rx_queue() and stmmac_enable_rx_queue(). In addition,
stmmac_xsk_wakeup() is added to initiate XDP RX buffer replenishing
by signalling user application to add available XDP frames back to
FILL queue.
For RX processing using XDP zero-copy buffer, stmmac_rx_zc() is
introduced which is implemented with the assumption that RX split
header is disabled. For XDP verdict is XDP_PASS, the XDP buffer is
copied into a sk_buff allocated through stmmac_construct_skb_zc()
and sent to Linux network GRO inside stmmac_dispatch_skb_zc(). Free RX
buffers are then replenished using stmmac_rx_refill_zc()
v2: introduce __stmmac_disable_all_queues() to contain the original code
that does napi_disable() and then make stmmac_setup_tc_block_cb()
to use it. Move synchronize_rcu() into stmmac_disable_all_queues()
that eventually calls __stmmac_disable_all_queues(). Then,
make both stmmac_release() and stmmac_suspend() to use
stmmac_disable_all_queues(). Thanks David Miller for spotting the
synchronize_rcu() issue in v1 patch.
Signed-off-by: Ong Boon Leong <boon.leong.ong@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Prepare stmmac_xdp_run_prog() for AF_XDP zero-copy support which will be
added by upcoming patches by splitting out the XDP verdict processing
into __stmmac_xdp_run_prog() and it callable for XDP ZC path which does
not need to verify bpf_prog is not NULL.
The stmmac_xdp_run_prog() is used for regular XDP Rx path which requires
bpf_prog to be verified.
Signed-off-by: Ong Boon Leong <boon.leong.ong@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
