e1000 writes to doorbells to post transmit descriptors and fill the
receive ring. After writing descriptors to memory but before
writing to doorbells, use dma_wmb() rather than wmb(). wmb() is more
heavyweight than necessary for a device to see descriptor writes.
On x86, this avoids SFENCEs before doorbell writes in both the
Tx and Rx paths. On ARM, this converts DSB ST -> DMB OSHST.
Tested: 82576EB / x86; QEMU (qemu emulates an 8257x)
Signed-off-by: Venkatesh Srinivas <venkateshs@google.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
The u32 variable rem is being shifted using u32 arithmetic however
it is being passed to div_u64 that expects the expression to be a u64.
The 32 bit shift may potentially overflow, so cast rem to a u64 before
shifting to avoid this. Also remove comment about overflow.
Addresses-Coverity: ("Unintentional integer overflow")
Fixes: cd45832069 ("ixgbe: implement support for SDP/PPS output on X550 hardware")
Fixes: 68d9676fc0 ("ixgbe: fix PTP SDP pin setup on X540 hardware")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Acked-by: Jacob Keller <jacob.e.keller@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
One of the more common cases of allocation size calculations is finding
the size of a structure that has a zero-sized array at the end, along
with memory for some number of elements for that array. For example:
struct foo {
int stuff;
struct boo entry[];
};
size = sizeof(struct foo) + count * sizeof(struct boo);
instance = alloc(size, GFP_KERNEL);
Instead of leaving these open-coded and prone to type mistakes, we can
now use the new struct_size() helper:
size = struct_size(instance, entry, count);
This code was detected with the help of Coccinelle.
Signed-off-by: "Gustavo A. R. Silva" <gustavo@embeddedor.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
If a packet which is utilizing the launchtime feature (via SO_TXTIME socket
option) also requests the hardware transmit timestamp, the hardware
timestamp is not delivered to the userspace. This is because the value in
skb->tstamp is mistaken as the software timestamp.
Applications, like ptp4l, request a hardware timestamp by setting the
SOF_TIMESTAMPING_TX_HARDWARE socket option. Whenever a new timestamp is
detected by the driver (this work is done in igb_ptp_tx_work() which calls
igb_ptp_tx_hwtstamps() in igb_ptp.c[1]), it will queue the timestamp in the
ERR_QUEUE for the userspace to read. When the userspace is ready, it will
issue a recvmsg() call to collect this timestamp. The problem is in this
recvmsg() call. If the skb->tstamp is not cleared out, it will be
interpreted as a software timestamp and the hardware tx timestamp will not
be successfully sent to the userspace. Look at skb_is_swtx_tstamp() and the
callee function __sock_recv_timestamp() in net/socket.c for more details.
Signed-off-by: Vedang Patel <vedang.patel@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The Header Parser allows identifying various fields in the packet
headers, used for various kind of filtering and classification
steps.
This is a re-entrant process, where the offset in the packet header
depends on the previous lookup results. This offset is represented in
the SRAM results of the TCAM, as a shift to be operated.
This shift can be negative in some cases, such as in IPv6 parsing.
This commit prevents overriding the sign bit when setting the shift
value, which could cause instabilities when parsing IPv6 flows.
Fixes: 3f518509de ("ethernet: Add new driver for Marvell Armada 375 network unit")
Suggested-by: Alan Winkowski <walan@marvell.com>
Signed-off-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Assign OF node to CPSW slave devices, otherwise it is not possible to
bind e.g. DSA switch to them. Without this patch, the DSA code tries
to find the ethernet device by OF match, but fails to do so because
the slave device has NULL OF node.
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: David S. Miller <davem@davemloft.net>
Cc: Ivan Khoronzhuk <ivan.khoronzhuk@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
If CMDQ ring is full, hclge_cmd_send may return directly, but IMP still
working and HW pointer changed, SW ring pointer do not match the HW
pointer. This patch update the SW pointer every time when the space is
full, so it can work normally next time if IMP and HW still working.
Signed-off-by: Peng Li <lipeng321@huawei.com>
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The HNS3_RXD_VLD_B bit has already been checked in hns3_add_frag
or hns3_handle_rx_bd before calling hns3_handle_bdinfo, so when
hns3_handle_bdinfo is called, the HNS3_RXD_VLD_B bit is always
set, which makes the checking in hns3_handle_bdinfo unnecessary.
This patch removes the RXD_VLD_B checking in hns3_handle_bdinfo.
Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com>
Signed-off-by: Peng Li <lipeng321@huawei.com>
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The frame column is based on rx_crc_errors and rx_frame_errors. So
l3l4 checksum error should not be counted by rx_crc_errors. Instead,
l3l4 checksum error should be counted in ifconfig error column.
Fixes: d3ec4ef669 ("net: hns3: refactor the statistics updating for netdev")
Signed-off-by: Yufeng Mo <moyufeng@huawei.com>
Signed-off-by: Peng Li <lipeng321@huawei.com>
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since some MSI-X interrupt's status may be cleared by hardware,
so when the driver receives the interrupt, reading
HCLGE_VECTOR0_PF_OTHER_INT_STS_REG register will get an empty
unknown interrupt. For this case, the irq handler should enable
vector0 interrupt. This patch also use dev_info() instead of
dev_dbg() in the hclge_check_event_cause(), since this information
will be useful for normal usage.
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: Peng Li <lipeng321@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The VF reset may fail for some probabilistic reasons,
such as wait for hardware reset timeout, wait for mailbox
response timeout, so this patch tries to re-schedule the
reset task when the number of reset failing is under
HCLGEVF_RESET_MAX_FAIL_CNT. This patch also add a function
hclgevf_reset_err_handle() to handle the reset failing.
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: Peng Li <lipeng321@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Local device and link partner config auto-negotiation on both,
local device config pause frame use as: rx on/tx off,
link partner config pause frame use as: rx off/tx on.
We except the result is:
Local device:
Autonegotiate: on
RX: on
TX: off
RX negotiated: on
TX negotiated: off
Link partner:
Autonegotiate: on
RX: off
TX: on
RX negotiated: off
TX negotiated: on
But actually, the result of Local device and link partner is both:
Autonegotiate: on
RX: off
TX: off
RX negotiated: off
TX negotiated: off
The root cause is that the supported flag is has only Pause,
reference to the function genphy_config_advert():
static int genphy_config_advert(struct phy_device *phydev)
{
...
linkmode_and(phydev->advertising, phydev->advertising,
phydev->supported);
...
}
The pause frame use of link partner is rx off/tx on, so its
advertising only set the bit Asym_Pause, and the supported is
only set the bit Pause, so the result of linkmode_and(), is
rx off/tx off.
This patch adds Asym_Pause to the supported flag to fix it.
Signed-off-by: Yonglong Liu <liuyonglong@huawei.com>
Signed-off-by: Peng Li <lipeng321@huawei.com>
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When setting -Wformat=2, there is a compiler warning like this:
hclge_main.c:xxx:x: warning: format not a string literal and no
format arguments [-Wformat-nonliteral]
strs[i].desc);
^~~~
This patch adds missing format parameter "%s" to snprintf() to
fix it.
Fixes: 46a3df9f97 ("Add HNS3 Acceleration Engine & Compatibility Layer Support")
Signed-off-by: Yonglong Liu <liuyonglong@huawei.com>
Signed-off-by: Peng Li <lipeng321@huawei.com>
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When hdev->tx_sch_mode is HCLGE_FLAG_VNET_BASE_SCH_MODE, the
hclge_tm_schd_mode_vnet_base_cfg calls hclge_tm_pri_schd_mode_cfg
with vport->vport_id as pri_id, which is used as index for
hdev->tm_info.tc_info, it will cause out of bound access issue
if vport_id is equal to or larger than HNAE3_MAX_TC.
Also hardware only support maximum speed of HCLGE_ETHER_MAX_RATE.
So this patch adds two checks for above cases.
Fixes: 848440544b ("net: hns3: Add support of TX Scheduler & Shaper to HNS3 driver")
Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com>
Signed-off-by: Peng Li <lipeng321@huawei.com>
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently when there is share buffer in the SSU(storage
switching unit), the low waterline for RX private buffer is
too low to keep the hardware running. Hardware may have
processed all the packet stored in the private buffer of the
low waterline before the new packet comes, because hardware
only tell the peer send packet again when the private buffer
is under the low waterline.
So this patch only allocate RX private buffer if there is
enough buffer according to hardware user manual.
This patch also reserve some buffer for reusing when TC num
is less than or equal to 2, and change PAUSE_TRANS_GAP &
HCLGE_NON_DCB_ADDITIONAL_BUF according to hardware user
manual.
Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com>
Signed-off-by: Peng Li <lipeng321@huawei.com>
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently when TC num is one, the DCB will be disabled no matter if
pfc_en is non-zero or not.
This patch enables the DCB if pfc_en is non-zero, even when TC num
is one.
Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com>
Signed-off-by: Peng Li <lipeng321@huawei.com>
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When change MTU or other operations, which just calling .reset_notify
to do HNAE3_DOWN_CLIENT and HNAE3_UP_CLIENT, then
the netdev_tx_reset_queue() in the hns3_clear_all_ring() will be
ignored. So the dev_watchdog() may misdiagnose a TX timeout.
This patch separates netdev_tx_reset_queue() from
hns3_clear_all_ring(), and unifies hns3_clear_all_ring() and
hns3_force_clear_all_ring into one, since they are doing
similar things.
Fixes: 3a30964a2e ("net: hns3: delay ring buffer clearing during reset")
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
On 32-bit architectures, putting an array of 256 u32 values on the
stack uses more space than the warning limit:
drivers/net/ethernet/huawei/hinic/hinic_main.c: In function 'hinic_rss_init':
drivers/net/ethernet/huawei/hinic/hinic_main.c:286:1: error: the frame size of 1068 bytes is larger than 1024 bytes [-Werror=frame-larger-than=]
I considered changing the code to use u8 values here, since that's
all the hardware supports, but dynamically allocating the array is
a more isolated fix here.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently, stmmac only supports 32 bits addressing for SKB. Enable the
support for upto 48 bits addressing in XGMAC core.
This avoids the use of bounce buffers and increases performance.
Changes from v1:
- Fallback to 32 bits in failure (Andrew)
Signed-off-by: Jose Abreu <joabreu@synopsys.com>
Cc: Joao Pinto <jpinto@synopsys.com>
Cc: David S. Miller <davem@davemloft.net>
Cc: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Cc: Alexandre Torgue <alexandre.torgue@st.com>
Cc: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
Before "sis900: fix TX completion" patch, TX completion was done on TxIDLE interrupt.
TX completion also was the only thing done on TxIDLE interrupt.
Since "sis900: fix TX completion", TX completion is done on TxDESC interrupt.
So it is not necessary any more to set and to check for TxIDLE.
Eliminate TxIDLE from sis900.
Correct some typos, too.
Signed-off-by: Sergej Benilov <sergej.benilov@googlemail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The new route handling in ip_mc_finish_output() from 'net' overlapped
with the new support for returning congestion notifications from BPF
programs.
In order to handle this I had to take the dev_loopback_xmit() calls
out of the switch statement.
The aquantia driver conflicts were simple overlapping changes.
Signed-off-by: David S. Miller <davem@davemloft.net>
Refactor the key layer calculation function, in particular the tunnel
key layer calculation by introducing helper functions. This is done
in preparation for supporting GRE tunnel offloads.
Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: John Hurley <john.hurley@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This commit adds support for AF_XDP zero-copy RX and TX.
We create a dedicated XSK RQ inside the channel, it means that two
RQs are running simultaneously: one for non-XSK traffic and the other
for XSK traffic. The regular and XSK RQs use a single ID namespace split
into two halves: the lower half is regular RQs, and the upper half is
XSK RQs. When any zero-copy AF_XDP socket is active, changing the number
of channels is not allowed, because it would break to mapping between
XSK RQ IDs and channels.
XSK requires different page allocation and release routines. Such
functions as mlx5e_{alloc,free}_rx_mpwqe and mlx5e_{get,put}_rx_frag are
generic enough to be used for both regular and XSK RQs, and they use the
mlx5e_page_{alloc,release} wrappers around the real allocation
functions. Function pointers are not used to avoid losing the
performance with retpolines. Wherever it's certain that the regular
(non-XSK) page release function should be used, it's called directly.
Only the stats that could be meaningful for XSK are exposed to the
userspace. Those that don't take part in the XSK flow are not
considered.
Note that we don't wait for WQEs on the XSK RQ (unlike the regular RQ),
because the newer xdpsock sample doesn't provide any Fill Ring entries
at the setup stage.
We create a dedicated XSK SQ in the channel. This separation has its
advantages:
1. When the UMEM is closed, the XSK SQ can also be closed and stop
receiving completions. If an existing SQ was used for XSK, it would
continue receiving completions for the packets of the closed socket. If
a new UMEM was opened at that point, it would start getting completions
that don't belong to it.
2. Calculating statistics separately.
When the userspace kicks the TX, the driver triggers a hardware
interrupt by posting a NOP to a dedicated XSK ICO (internal control
operations) SQ, in order to trigger NAPI on the right CPU core. This XSK
ICO SQ is protected by a spinlock, as the userspace application may kick
the TX from any core.
Store the pointers to the UMEMs in the net device private context,
independently from the kernel. This way the driver can distinguish
between the zero-copy and non-zero-copy UMEMs. The kernel function
xdp_get_umem_from_qid does not care about this difference, but the
driver is only interested in zero-copy UMEMs, particularly, on the
cleanup it determines whether to close the XSK RQ and SQ or not by
looking at the presence of the UMEM. Use state_lock to protect the
access to this area of UMEM pointers.
LRO isn't compatible with XDP, but there may be active UMEMs while
XDP is off. If this is the case, don't allow LRO to ensure XDP can
be reenabled at any time.
The validation of XSK parameters typically happens when XSK queues
open. However, when the interface is down or the XDP program isn't
set, it's still possible to have active AF_XDP sockets and even to
open new, but the XSK queues will be closed. To cover these cases,
perform the validation also in these flows:
1. A new UMEM is registered, but the XSK queues aren't going to be
created due to missing XDP program or interface being down.
2. MTU changes while there are UMEMs registered.
Having this early check prevents mlx5e_open_channels from failing
at a later stage, where recovery is impossible and the application
has no chance to handle the error, because it got the successful
return value for an MTU change or XSK open operation.
The performance testing was performed on a machine with the following
configuration:
- 24 cores of Intel Xeon E5-2620 v3 @ 2.40 GHz
- Mellanox ConnectX-5 Ex with 100 Gbit/s link
The results with retpoline disabled, single stream:
txonly: 33.3 Mpps (21.5 Mpps with queue and app pinned to the same CPU)
rxdrop: 12.2 Mpps
l2fwd: 9.4 Mpps
The results with retpoline enabled, single stream:
txonly: 21.3 Mpps (14.1 Mpps with queue and app pinned to the same CPU)
rxdrop: 9.9 Mpps
l2fwd: 6.8 Mpps
Signed-off-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Acked-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
structs mlx5e_{rq,sq,cq,channel}_param are going to be used in the
upcoming XSK RX and TX patches. Move them to a header file to make
them accessible from other C files.
Signed-off-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Acked-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Create new functions mlx5e_{open,close}_queues to encapsulate opening
and closing RQs and SQs, and call the new functions from
mlx5e_{open,close}_channel. It simplifies the existing functions a bit
and prepares them for the upcoming AF_XDP changes.
Signed-off-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Acked-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Use the existing mlx5e_get_linear_rq_headroom function to calculate the
headroom for mlx5e_xdp_max_mtu. This function takes the XSK headroom
into consideration, which will be used in the following patches.
Signed-off-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Acked-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
When an XDP program returns XDP_TX, and the RQ is XSK-enabled, it
requires careful handling, because convert_to_xdp_frame creates a new
page and copies the data there, while our driver expects the xdp_frame
to point to the same memory as the xdp_buff. Handle this case
separately: map the page, and in the end unmap it and call
xdp_return_frame.
Signed-off-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Acked-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Put the XDP SQ that is used for XDP_TX into the channel. It used to be a
part of the RQ, but with introduction of AF_XDP there will be one more
RQ that could share the same XDP SQ. This patch is a preparation for
that change.
Separate XDP_TX statistics per RQ were implemented in one of the previous
patches.
Signed-off-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Acked-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Currently, struct mlx5e_xdp_info has some issues that have to be cleaned
up before the upcoming AF_XDP support makes things too complicated and
messy. This structure is used both when sending the packet and on
completion. Moreover, the cleanup procedure on completion depends on the
origin of the packet (XDP_REDIRECT, XDP_TX). Adding AF_XDP support will
add new flows that use this structure even differently. To avoid
overcomplicating the code, this commit refactors the usage of this
structure in the following ways:
1. struct mlx5e_xdp_info is split into two different structures. One is
struct mlx5e_xdp_xmit_data, a transient structure that doesn't need to
be stored and is only used while sending the packet. The other is still
struct mlx5e_xdp_info that is stored in a FIFO and contains the fields
needed on completion.
2. The fields of struct mlx5e_xdp_info that are used in different flows
are put into a union. A special enum indicates the cleanup mode and
helps choose the right union member. This approach is clear and
explicit. Although it could be possible to "guess" the mode by looking
at the values of the fields and at the XDP SQ type, it wouldn't be that
clear and extendable and would require looking through the whole chain
to understand what's going on.
For the reference, there are the fields of struct mlx5e_xdp_info that
are used in different flows (including AF_XDP ones):
Packet origin | Fields used on completion | Cleanup steps
-----------------------+---------------------------+------------------
XDP_REDIRECT, | xdpf, dma_addr | DMA unmap and
XDP_TX from XSK RQ | | xdp_return_frame.
-----------------------+---------------------------+------------------
XDP_TX from regular RQ | di | Recycle page.
-----------------------+---------------------------+------------------
AF_XDP TX | (none) | Increment the
| | producer index in
| | Completion Ring.
On send, the same set of mlx5e_xdp_xmit_data fields is used in all
flows: DMA and virtual addresses and length.
Signed-off-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Acked-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Prepare to creation of the XSK RQ, which will require posting UMRs, too.
The same ICO SQ will be used for both RQs and also to trigger interrupts
by posting NOPs. UMR WQEs can't be reused any more. Optimization
introduced in commit ab966d7e4f ("net/mlx5e: RX, Recycle buffer of
UMR WQEs") is reverted.
Signed-off-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Acked-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Additional conditions introduced:
- XSK implies XDP.
- Headroom includes the XSK headroom if it exists.
- No space is reserved for struct shared_skb_info in XSK mode.
- Fragment size smaller than the XSK chunk size is not allowed.
A new auxiliary function mlx5e_get_linear_rq_headroom with the support
for XSK is introduced. Use this function in the implementation of
mlx5e_get_rq_headroom. Change headroom to u32 to match the headroom
field in struct xdp_umem.
Signed-off-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Acked-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
The PCI API for DMA is deprecated, and PCI_DMA_TODEVICE is just defined
to DMA_TO_DEVICE for backward compatibility. Just use DMA_TO_DEVICE.
Signed-off-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Acked-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Some drivers want to access the data transmitted in order to implement
acceleration features of the NICs. It is also useful in AF_XDP TX flow.
Change the xsk_umem_consume_tx API to return the whole xdp_desc, that
contains the data pointer, length and DMA address, instead of only the
latter two. Adapt the implementation of i40e and ixgbe to this change.
Signed-off-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Acked-by: Saeed Mahameed <saeedm@mellanox.com>
Cc: Björn Töpel <bjorn.topel@intel.com>
Cc: Magnus Karlsson <magnus.karlsson@intel.com>
Acked-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
